[Senate Hearing 108-1035]
[From the U.S. Government Publishing Office]
S. Hrg. 108-1035
BERING SEA/ALEUTIAN ISLAND
CRAB RATIONALIZATION PLAN
=======================================================================
HEARING
before the
COMMITTEE ON COMMERCE,
SCIENCE, AND TRANSPORTATION
UNITED STATES SENATE
ONE HUNDRED EIGHTH CONGRESS
FIRST SESSION
__________
MAY 20, 2003
__________
Printed for the use of the Committee on Commerce, Science, and Transportation
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
Available online: http://www.govinfo.gov
_________
U.S. GOVERNMENT PUBLISHING OFFICE
32-434 PDF WASHINGTON : 2018
SENATE COMMITTEE ON COMMERCE, SCIENCE, AND TRANSPORTATION
ONE HUNDRED EIGHTH CONGRESS
FIRST SESSION
JOHN McCAIN, Arizona, Chairman
TED STEVENS, Alaska ERNEST F. HOLLINGS, South
CONRAD BURNS, Montana Carolina, Ranking
TRENT LOTT, Mississippi DANIEL K. INOUYE, Hawaii
KAY BAILEY HUTCHISON, Texas JOHN D. ROCKEFELLER IV, West
OLYMPIA J. SNOWE, Maine Virginia
SAM BROWNBACK, Kansas JOHN F. KERRY, Massachusetts
GORDON H. SMITH, Oregon JOHN B. BREAUX, Louisiana
PETER G. FITZGERALD, Illinois BYRON L. DORGAN, North Dakota
JOHN ENSIGN, Nevada RON WYDEN, Oregon
GEORGE ALLEN, Virginia BARBARA BOXER, California
JOHN E. SUNUNU, New Hampshire BILL NELSON, Florida
MARIA CANTWELL, Washington
FRANK R. LAUTENBERG, New Jersey
Jeanne Bumpus, Republican Staff Director and General Counsel
Robert W. Chamberlin, Republican Chief Counsel
Kevin D. Kayes, Democratic Staff Director and Chief Counsel
Gregg Elias, Democratic General Counsel
C O N T E N T S
----------
Page
Hearing held on May 20, 2003..................................... 1
Statement of Senator Cantwell.................................... 5
Prepared statement........................................... 5
Statement of Senator Stevens..................................... 1
Witnesses
Duffy, Kevin C., Commissioner, Alaska Department of Fish and Game 6
Prepared statement........................................... 9
Fraser, Dave, Captain, Fishing Vessel Muir Milach................ 30
Prepared statement........................................... 33
Freed, Hon. Linda, City Manager, City of Kodiak.................. 16
Prepared statement........................................... 19
Kelty, Frank, Resource Analyst, City of Unalaska................. 56
Prepared statement........................................... 59
Murkowski, Hon. Lisa, U.S. Senator from Alaska................... 3
Prepared statement........................................... 4
Murray, Hon. Patty, U.S. Senator from Washington................. 2
Thomson, Arni, Executive Director, Alaska Crab Coalition......... 71
Prepared statement........................................... 73
Appendix
Anderson, Coleman A., prepared statement......................... 107
Burns, Michael F., President and Co-Owner, Blue North Fisheries,
prepared statement............................................. 107
Chevalier, Robert L., Sitka, AK, prepared statement.............. 108
Childers, Dorothy, Executive Director, Alaska Marine Conservation
Council, prepared statement.................................... 109
Garcia, Kale, Owner-Operator F/V Aquila, prepared statement...... 111
Harrington, Erin D., Kodiak, AK, prepared statement.............. 111
Halvorsen, Robert, Professor of Economics, Department of
Economics, University of Washington, prepared statement........ 112
Laukitis, Michael B., President, North Pacific Fisheries
Association, prepared statement................................ 147
Lestenkof, Phillip, President, Central Bering Sea Fishermen's
Association, prepared statement................................ 149
Letter dated May 19, 2003 to Hon. Ted Stevens from Larry Cotter,
CEO, Aleutian Pribilof Island Community Development Association 234
Letter dated May 20, 2003 from Michele Longo Eder to Hon. John
McCain......................................................... 235
Letter dated May 20, 2003 to Senator John McCain from Dennis
Petersen, Resident of Seattle, WA.............................. 236
Letter dated May 20, 2003 to Senator John McCain from Victor
Smith, Friday Harbor, WA....................................... 237
Letter dated May 25, 2003 to Senator John McCain from Victor
Smith, Friday Harbor, WA....................................... 238
Letter dated May 29, 2003 to Linda Freed, Manager, Kodiak City
Manager from Simeon Swetzof, Jr., Mayor, City of St. Paul...... 240
Letter dated May 30, 2003 to Senator John McCain from Jeffrey R.
Stephan, United Fishermen's Marketing Association, Inc......... 241
Letter dated June 6, 2003 to Senator John McCain from Victor
Smith, Friday Harbor, WA....................................... 245
Petersen, Rudy A., Fisherman, CEO, and Owner, Fishermen's Finest,
Inc., F/V American No. 1/F/V U.S. Intrepid, prepared statement. 151
Philemonoff, Ron, Chairman and CEO, Tanadgusix (TDX) Corporation
of St. Paul Island Alaska, prepared statement.................. 186
Pierce, Byron L., prepared statement............................. 188
Powell, Richard, Resident of Kodiak, Alaska, prepared statement.. 189
Pries, Robert E., Renton, WA, prepared statement................. 190
Soma, David, Deep Sea Fishermen's Union of the Pacific, prepared
statement...................................................... 191
Sorvik, Alf O., Owner F/V Rainier, F/V Paragon, F/V Isafjord,
prepared statement............................................. 196
Stewart, Capt. Gary, F/V Polar Lady, prepared statement.......... 196
Storrs, Bob, Vice President, Unalaska Native Fisherman
Association (UNFA), prepared statement......................... 197
Swetzof, Jr., Simeon, Mayor, City of Saint Paul, Alaska, prepared
statement...................................................... 215
Taufen, Stephen R., Founder, Groundswell Fisheries Movement,
prepared statement............................................. 222
Tolva, Mimi, Homer, AK, prepared statement....................... 223
Tyack, Peter, Senior Scientist and Walter A. and Hope Noyes Smith
Chair, Biology Department, Woods Hole Oceanographic Institution 224
Uri, Konrad S., Seattle, WA, prepared statement.................. 233
Hon. Ron Wyden, U.S. Senator from Oregon, prepared statement..... 246
Response to written questions submitted by Hon. Ron Wyden to:
Kevin Duffy.................................................. 246
Dave Fraser.................................................. 247
Linda Freed.................................................. 247
Frank Kelty.................................................. 248
Arni Thompson................................................ 248
BERING SEA/ALEUTIAN ISLAND
CRAB RATIONALIZATION PLAN
----------
TUESDAY, MAY 20, 2003
U.S. Senate,
Committee on Commerce, Science, and Transportation,
Washington, DC.
The Committee met, pursuant to notice, at 2:45 p.m. in room
SR-253, Russell Senate Office Building, Hon. Ted Stevens,
presiding.
OPENING STATEMENT OF HON. TED STEVENS,
U.S. SENATOR FROM ALASKA
Senator Stevens. Congress directed the North Pacific
Fisheries Management Council, in the Consolidated
Appropriations Act for 2001, to examine the fisheries under its
jurisdiction, particularly the Gulf of Alaska groundfish and
Bering Sea crab fisheries. The Council was directed to analyze
individual fishing quotas, processor quotas, cooperatives, and
quotas held by communities. In doing this, the Council was
expected to include economic analysis of the impact of all
options on the three components of the fishing industry
communities, processors, and fishing fees.
In 2002, the Council voted unanimously to create a three-
pie system of quota share for the harvesting sector, the
processing sector, and communities. 100 percent of the catch is
allocated to the harvesters and 90 percent to the processors,
with 10 percent open access. The rationalization concept of
processor quota share is a novel idea that has its roots in the
American Fisheries Act, AFA.
I worked closely with the Senators from Washington to
develop the AFA, a fisheries rationalization plan, for the
Bering Sea, Aleutian Islands pollock. AFA has proved to be a
good rationalization plan that has allowed for increased use of
target species, better execution of the fishery, instead of a
race for the fish, and a great success in conservation and
management of the fishery, and this has brought about
substantial growth in pollock stocks. The BSAI Crab
Rationalization Plan must benefit all Alaskan communities. This
fishery is prosecuted entirely off the coast of Alaska, and the
various communities that have a history under the Council's
plan must have a role in the future of this fishery.
I am encouraged to learn that the Council has completed
work on the trailing amendments and has provided additional
protections for the communities' impact by rationalization of
this fishery.
And I thank the witnesses for coming to Washington for this
hearing, and I look forward to your testimony. Again, my
apology for being slightly late.
Senator Murray, we are pleased to have your statement.
STATEMENT OF HON. PATTY MURRAY,
U.S. SENATOR FROM WASHINGTON
Senator Murray. Well, thank you very much, Mr. Chairman,
for holding this hearing today. I really appreciate your
leadership and for inviting me to testify today on the North
Pacific Crab Rationalization Plan.
The states of Alaska and Washington share many common
interests, and I appreciate the opportunity to work with you as
we address these important regional issues. The commercial
fishing industry is one example of the many interests that are
shared by the states we represent. On the issue of crab
rationalization which is before us now, I am confident that we
will be able to work together, as we have in the past, to
tackle this important subject.
Mr. Chairman, you and I are both here today for one simple
reason, the status quo is not working. The status quo means
more lives lost at sea, inefficient use of the resource, and
harm to fisheries. We know that overcapitalization plaguing the
industry was caused, in part, by the race for fish and short
fishing seasons, and we know that we can more efficiently
manage our resource.
Mr. Chairman, the race for fish puts our fisherman at an
unacceptably high risk. Too often, we hear of boats that risk
too rough a sea and never return home. We know from past
efforts that rationalization slows down the fishery and brings
tremendous environmental and safety benefits. Participants
avoid going to sea during bad weather, take time to treat the
resource carefully, and minimize waste. We do know, also, that
there is risk and uncertainty associated with employing new
public-policy tools, and, in many cases, we cannot know the
full impact of new approaches until we try them. I believe,
however, that the experience of the American Fisheries Act is a
testament to the potential benefits of crab rationalization.
The plan recommended by the Council may not be perfect, but
it does represent significant progress over the status quo. The
costs of continued delay, in lives and in resources, are high.
There are, of course, different perspectives on the crab
plan passed by the Council last June. Many of my constituents
are concerned about the plan because of its inclusion of
processor quotas and impacts for the industry and communities.
Most of the anxiety expressed by harvesters and communities is
focused on the proposed plan's guarantee of 90 percent
processor quotas.
Mr. Chairman, I appreciate these concerns, and I am
interested in exploring ways we might address them. We cannot
let the perfect become the enemy of the good. After years of
deliberation, the Council recommended its plan by a vote of
eleven to zero. I believe it was the intent of the Council to
send a strong message on this subject, and it did. Congress
cannot ignore this significant action.
Today, we will hear about the many benefits that can come
from rationalizing the crab fisheries. We will be urged to
enact legislation authorizing processor quotas so the Council
and the National Marine Fisheries Service can begin the arduous
work of implementing the plan.
I look forward to hearing from the witnesses today, in
hopes of hearing perspectives that will bring us new insights
about how we should proceed. After today's hearing, it is
important that Members of this Committee and Congress make crab
rationalization a priority this Congress so we can complete the
transition away from the status quo.
Mr. Chairman, I look forward to today's testimony and to
working with you on this issue in the weeks and months ahead.
And thank you, again, for allowing me to participate in this
important hearing.
Senator Stevens. Senator, I would be pleased if you would
join me here. I do expect another Senator, or more, could join
before we are through.
I should say to the audience--I know that most of you are
from home, or at least from the Pacific Northwest--and this is
one of those busy days in Washington, from the Senate. We have
the Armed Services bill on the floor, and we have several
competing conferences and hearings going on today. So I do not
expect we are going to have any other Senators join us, except
Senator Murkowski, if she gets cleared from her current
hearing.
We have, now, a panel which is made up of Mr. Kevin Duffy,
Commissioner of the Alaska Department of Fish and Game; Ms.
Linda Freed, City Manager of the City of Kodiak; Mr. Dave
Fraser, Captain of the fishing vessel Muir Milach; Mr. Frank
Kelty, Natural Resources Manager of the City of Unalaska; and
Mr. Arni Thomson, Executive Director of the Alaska Crab
Coalition, from Seattle. I failed to note that Mr. Fraser is
from Port Townsend, Washington.
I would appreciate it if you would come to the Council
table, and we will go through the hearing.
As you are coming up, many have questioned the time
involved in this hearing. I am not the Chairman of this
Subcommittee. The Chairman of this Subcommittee is otherwise
involved in other things and did not have time for this
hearing. He consented that we could have the hearing, but we
are limited, in terms of the time, because of the circumstances
I mentioned before, in terms of the bills that are before the
Senate. We may have votes during the time of this hearing, in
which we would have to stand in recess so that we could go cast
our votes and return. Senator, do you have an opening
statement?
STATEMENT OF HON. LISA MURKOWSKI,
U.S. SENATOR FROM ALASKA
Senator Murkowski. Mr. Chairman, thank you for the
opportunity to participate, to be here with so many Alaskans
this afternoon.
If I had to characterize my views on the proposed Bering
Sea Crab Rationalization Program at the moment, I would
probably say it is undecided. So I look forward to hearing from
today's witnesses, who represent a range of views, and who
will, I hope, help Congress make informed decisions on this.
The last decade has been marked by rapid change. Resource
levels, market conditions, and economic performance have all
fluctuated wildly. And it is clear that a race for fish is no
longer the optimum strategy for management, if it ever was. It
leads to adverse consequences for the resource, for the market,
and for the fishermen who risk their lives to pursue it. So
change is needed, but how much and what kind?
Mr. Chairman, I do have a rather lengthy opening statement,
but would like, more specifically, to hear from those who have
come to speak on this issue. I would be happy to include the
balance of my testimony for the record so that we can provide
the opportunity to hear from the panel this afternoon.
Senator Stevens. Well, thank you very much, Senator. All of
the statements that are filed by the witnesses and by yourself
and the statement made by Senator Murray will be printed in
full in the record as though completely read, and we appreciate
your courtesy of postponing the further testimony so that we
may turn to the witnesses now.
[The prepared statement of Senator Murkowski follows:]
Prepared Statement of Hon. Lisa Murkowski, U.S. Senator from Alaska
Mr. Chairman, thank you for the opportunity to participate in this
important hearing.
If I had to characterize my views on the proposed Bering Sea crab
rationalization program at this moment, the appropriate word would be
``undecided.''
I look forward to hearing from today's witnesses, who represent a
range of views, and who will, I hope, help Congress make informed
decisions.
The last decade has been marked by rapid change. Resource levels,
market conditions and economic performance all have fluctuated widely.
It is clear that a ``race for fish'' is no longer the optimum
strategy for management, if it ever was. It leads to adverse
consequences for the resource, for the market, and for the fishermen
who risk their lives to pursue it.
Change is needed. But how much? And what kind?
I see very substantial support for moving to Individual Fish Quotas
(IFQs) for crab harvesters. This alone is a radical change from the
open fisheries of the past, but IFQs have been shown to be beneficial
in other fisheries. They have slowed the race, improved product
quality, allowed more profitable operations, and improved safety.
But there have also been unintended consequences. When IFQs were
implemented in Alaska's halibut and sablefish industries, some
processors claimed they were harmed because their considerable capital
investments--necessary to compete in an open-access race for fish--were
no longer needed. I have no reason to doubt those losses. But neither
have I seen convincing analysis to demonstrate that the same impact
would necessarily occur in other fisheries.
The most controversial aspect of the proposed plan is that it
incorporates both IFQ's and Processor Quotas (PQs). Where IFQ's
allocate the right to compete for fish, PQs allocate the fish
themselves, after they have become private property of the fishermen.
One effectively provides a license to compete, while the other
effectively provides protection against competition. That is an
important distinction, and one which Congress must take very seriously.
Protection against competition is not a traditional role of the
government in a free marketplace, except where dealing with a limited
resource owned by the government, such as radio-frequency bandwidth.
Congress must now decide if it is a necessary step--or if it will
ultimately lead to more serious problems.
There is very little history to draw upon for guidance, at least in
the fishing world. Some point to the remarkably successful
restructuring of the Alaska Pollock fishery under the American
Fisheries Act (AFA), which also created a closed class of processors.
However, it is important to note that the AFA, crafted by Senator
Stevens and Senator Gorton with extensive input from involved fishermen
and processors, functions quite differently in the way it deals with
the processing issue. It requires that processors work with fishermens'
co-ops, and it offers no guarantees beyond the annual co-op agreements.
As an Alaskan representing my State in Congress, my primary concern
is the potential impact on my constituents and their communities. There
is strong support for the plan among many of the communities that have
traditionally processed the bulk of the Bering Sea crab catch. That is
understandable. Most of those communities would, like the processors
themselves, be protected by a program that mandates that harvested crab
be delivered to their existing processors.
Among the fishermen, however, and among other communities, there is
widespread opposition, and I believe we must fully evaluate those
arguments, as well.
The primary question is whether the proposed plan will benefit all
the fishery's participants--processors, fishermen and the processing
workers and support industries that form the backbone of our
communities. At least one community that has received Bering Sea crab
in the past will no longer be able to do so. Others, that may in the
future be positioned to take Bering Sea crab, will be prevented from
doing so.
I recognize that the North Pacific Council has attempted to adjust
the plan to provide for competition among processors for the 10 percent
of fishing shares that would not be allocated among processors. It has
attempted to preserve processing communities by geographic designations
and by providing communities with an opportunity to acquire processing
rights. And it has attempted to mitigate possible ex-vessel price
reductions by mandating a form of binding arbitration.
Are these steps enough?
Will the plan lead to renewed prosperity not just for today's
participants, but also to an expansion of activity to new companies and
new communities? Or will it lead to fewer economic opportunities in all
but a handful of lucky cases?
Will its impacts improve the ability of the fishing and processing
industry overall to adjust to changes in other fisheries? Or will it
simply cause stagnation?
Will it continue to allow healthy competition? Or lead to a radical
consolidation that leaves many communities marooned, with no way to
profit from future changes in the market or in the location, abundance,
or species mix of the Bering Sea crab resource?
Those are the questions that must be answered, and which leave me
wondering.
I hope that today's testimony will shed light on all of them, and
look forward to hearing from the experts on all sides.
Senator Stevens. I would call on, first--oh, wait a minute,
here is--Senator Cantwell, do you have an opening statement?
STATEMENT OF HON. MARIA CANTWELL,
U.S. SENATOR FROM WASHINGTON
Senator Cantwell. I do, Mr. Chairman, but I will have to
submit it for the record.
Senator Stevens. Thank you very much. We will submit it. We
have submitted the others. We are happy to have you join us.
[The prepared statement of Senator Cantwell follows:]
Prepared Statement of Hon. Maria Cantwell, U.S. Senator from Washington
Thank you Senator Stevens. I appreciate your continued leadership
on these issues and I look forward to working with you and Senator
Murray to ensure that the constituents of both of our great states
benefit from a fair crab rationalization plan.
As we heard today, managing America's Bering Sea and Aleutian
Island crab resources is critical to the continuation of this
multimillion-dollar industry.
However, I don't think anyone here today would dispute that the
current system is broken and that we must act.
Today's ``race to fish'' is dangerous, it leads to unsustainable
harvests, and it has forced an uneconomic overcapitalization of both
the fishing fleet and processor facilities.
The North Pacific Fishery Management Council's plan represents an
impressive investment of time and energy.
I would like to publicly thank the Council members for their work
and their consideration of the interests of all relevant stakeholders.
With its ``three-pie'' approach, the plan attempts to balance the
interests of harvesters, processors, and communities. However, Congress
must decide if this is the fairest and most effective way to
rationalize the North Pacific crab harvest.
I look forward to working with my Senate colleagues on this
critical issue in the weeks and months ahead.
Senator Stevens. It is probably not politically correct to
say, but obviously this is a regional issue, because there is
no one here from outside of our region. Mr. Duffy, let us hear
your statement first.
STATEMENT OF KEVIN C. DUFFY, COMMISSIONER,
ALASKA DEPARTMENT OF FISH AND GAME
Mr. Duffy. Thank you, Mr. Chairman. Good afternoon.
I am Kevin Duffy, Commissioner of the Alaska Department of
Fish and Game and also the State of Alaska representative on
the North Pacific Fishery Management Council. I appreciate the
opportunity to be here today to discuss the Council's preferred
alternative for the Bering Sea/Aleutian Island's Crab
Rationalization Plan.
Rationalization is the path to revitalize the economic
health of Bering Sea/Aleutian Island crab fisheries, provided
the policy recognize the partnership among harvesters,
processors, and communities. The Council sought to ensure this
rationalize program maintains the integrity of this partnership
by providing incentives for all parties to work toward mutually
beneficial goals.
There are two approaches that accomplish this objective.
The two approaches that maximize benefits to the Nation and
equitably distribute those benefits among the three partners
are the voluntary cooperatives and the two-pie allocation of
quota share, which, in combination with CDQ allocations and
other community elements, is in reality a three-pie scheme.
This voluntary three-pie approach approved by the Council in
June 2002 recognizes the prior economic interests among
harvesters, processors, and communities. This approach fits the
Bering Sea crab fishery.
The Council has a pioneering history of designing
rationalization programs unique to the fishery at hand. The
Council crafted an IFQ program for halibut and sable fish that
fit the small-vessel owner-onboard nature of that fishery. For
pollock, under direction from Congress, the mechanism was
processor-linked cooperatives.
We design rationalization programs to fit the dynamic needs
of any particular fishery, as a council. For the large-boat,
heavily industrialized corporate nature of the Bering Sea/
Aleutian Island crab fisheries, the Council found that a
voluntary three-pie coop structure fit best.
I want to emphasize conservation and safety are as much a
driving force to rationalize as the need to revitalize the
economic health of the industry and communities. Resource
protection is embedded within this plan. With extended slower
seasons, it is anticipated that dead-loss will be greatly
reduced and that already low bycatch levels will decline
further.
With more time to prosecute the fishery under IFQs, fewer
pots will be lost, and ghost fishing by lost or abandoned pots
will be reduced even beyond the current low levels.
To ensure that the resource, its protection, is maintained
and is a paramount factor under the Council's jurisdiction,
there will be increased monitoring under this program, and the
plan the Council developed includes a funding source for that
increased monitoring. Most importantly, this Council plan saves
lives. The current management scheme results in a derby-style
race for fish.
In terms of a brief background for the community, as the
Chairman indicated, Congress directed the Council to consider a
wide variety of approaches to a rationalized fishery. Early in
this process, the Council established a broad-based industry
working group to develop options to rationalize the Bering Sea
crab fisheries. This workgroup met over a 2-year period to
identify workable solutions and to develop a majority opinion.
Including the work of the Committee, the Council has studied
this issue for well over 3 years, with a detailed analysis of
all of the issues brought forward by the Committee, which did
include the concept of processor shares.
The first Council action was to unanimously adopt a problem
statement that any solution should consider the lack of
economic stability for harvesters, processors, and coastal
communities. After many revisions and exhaustive public review
of the analysis, the Council unanimously adopted a final motion
in June 2002, which responded to congressional direction. Since
then, the Council continued to work on several trailing
amendments and completed its work in April 2003.
Overall, the Crab Rationalization Program strikes a balance
among the three parties mentioned. Based on public input--
exhaustive public input, I might add--the program was
unanimously approved by this Council.
As a Council, we have an obligation to review program
performance, especially with respect to the impact on all three
partners. The Council is on record that we will make necessary
adjustments to achieve our stated goals and minimize unintended
consequences. Through program review, we can adjust elements of
the initial policy design. This flexibility is uncommon to most
market-based rationalization plans. Specifically, the Council
can adjust the percentage of these shares, adjust attributes
intended to protect community interests, adjust the other
mechanisms for community protection, and adjust specific design
features of the program to adapt to changes in stock status for
the Bering Sea/Aleutian Island crab.
All eligible fishermen under this program gain by having
value attached to their fishing history. The Council chose
years agreed to by the industry committee and recommended by
the advisory panel. Regardless of where a fisherman sits in the
qualifying period, the word of quota share generates immediate
value and economic gain over the race-for-fish scenario.
Furthermore, it is anticipated that consolidation and the
ability to stack quota, combined with a buy-back program, could
reduce the crab fleet significantly. This results in very
significant economic gains to the remaining participants.
Even though the economic gains are recognized by many crab
fisherman, there was still a need to address equity issues and
balancing of negotiating strength between harvesters and
processors. The Council has provided many protective measures
to ensure that vessel owners are not disadvantaged. They
include a discount of each processor's history, providing for B
shares, a system of binding arbitration to assure that no
harvester will be forced to accept a price by a processor
holding processor shares. Processors will be required, under
the program, to give regulators full, complete, and detailed
crab processing costs and crab sales revenues.
Caps on the amount of harvesting shares that can be owned
by or affiliated with processors are incorporated. This is
intended to ensure that the processor will continue to need
crab from independent harvesters. We have also prohibited
vessels owned processors from being awarded B shares. This was
done in order to further protect the interests of independent
harvesters.
While not a direct part of the Council's motion to protect
fishermen's interests, there is another key advantage that
accrues. Currently, harvesters cannot really withhold product
from processors for long periods of time, because the current
open-access model allows any catcher-processor or strikebreaker
to fish while others remain tied to the dock. Once IFQ is
issued to harvesters, every harvester can, either individually
or as a member of a coop under this plan, withhold product from
the processing sector without fear of someone else harvesting
it. As learned from the AFA model, the ability to withhold
product is a powerful tactic that fishermen can use to leverage
fair prices.
The next set of benefactors are the crab processors. Having
observed processor fallout from the harvester-only allocation
on IFQs for halibut and sable fish, and processor viability
under AFA coops, the Council made a deliberate choice to
include processors in this program. Processor shares will allow
the processing sector to more fully experience the efficiency
gains of rationalization; and, thus, remain viable. With the
ability to trade and purchase processing shares up to a set
limit, it is anticipated that the number of processors
available to buy crab may actually increase. At the very least,
award of processor shares should stabilize the crab-processing
sector so that the number of currently viable processors will
remain the same.
Last, the Council chose to provide specific community-
protection measures to recognize the investment of communities
in the relevant Bering Sea crab fisheries. Noting that the very
real likelihood that for those same communities to experience
irreparable harm without protection, National Standard 8 of the
Magnuson-Stevens Act takes into account the need for sustained
participation of fishing communities and minimizing adverse
effects on those communities.
There are a number of specific elements of the community
protection measures in this plan. In the interest of time, I
will pass those over, and please refer to my testimony.
The Council has provided Congress with two comprehensive
reports on the Bering Sea/Aleutian Island Crab Rationalization
Program. The first report, in August 2002, provided details on
the Council's Crab Rationalization Program that was unanimously
approved. The second report, dated May 6 of this year, provided
details on a set of trailing amendments that included a binding
arbitration program, a suite of community-protection measures,
details on captain and crew shares, sideboards, and details of
a comprehensive data-collection program.
In closing, let me emphasize that while imposing an
allocation scheme automatically has varying impacts among the
participants, the Council strove to make all sectors--
harvesters, processors, and communities--win from Bering Sea
crab rationalization. It was a challenging balancing act, to
say the least. However, through the work of the industry
committee, the Council staff, the advisory committee, and
community representatives, the Council's preferred alternative
does benefit all sectors. And, most importantly, the plan will
save lives and improve management.
Senator Stevens. I am going to have to cut you off there,
Kevin. Thank you very much.
Mr. Duffy. Yes, Mr. Chairman.
Senator Stevens. We have limited witnesses to 10 minutes
each so we can have some time for questions here. I would ask
you to finish. Yes, sir?
Mr. Duffy. Just one quick concluding statement, if I could,
Mr. Chairman.
Senator Stevens. Yes, sir.
Mr. Duffy. Thank you.
The State of Alaska supports implementation of the Crab
Rationalization Program as approved by the Council in 2002,
including the details addressed through a set of trailing
amendments completed in April 2003.
Thank you, Mr. Chairman.
Senator Stevens. Thank you.
[The prepared statement of Mr. Duffy follows:]
Prepared Statement of Kevin C. Duffy, Commissioner,
Alaska Department of Fish and Game
Good afternoon, Mr. Chairman and members of the Committee. I am
Kevin C. Duffy, Commissioner of the Alaska Department of Fish and Game.
I am also the State of Alaska's representative on the North Pacific
Fishery Management Council. I appreciate the opportunity to be here
today to discuss the North Pacific Fishery Management Council's
preferred alternative for the Bering Sea/Aleutian Islands (BSAI) Crab
Rationalization Plan.
Thirty years ago, foreign harvesters and processors prosecuted
Bering Sea crab fisheries. Then, the Magnuson-Stevens Fisheries Act
Americanized these fisheries. The U.S. industry started from scratch,
investing in both harvesting and processing capacity. At the same time,
Alaska communities invested in the labor and infrastructure to support
these fisheries.
The crab fisheries flourished, quickly becoming the most valuable
fisheries in the North Pacific. Investments in all sectors grew
together and soon there was too much capacity harvesting and processing
declining stocks of crab. None of the three partners--harvesters,
processors and communities--I currently consider financially healthy.
Of particular concern to the State of Alaska are the remote fishery-
dependent communities, which are now adversely impacted by the
diminished flow of economic activity.
Rationalization is the path to re-vitalize the economic health of
these fisheries, provided the policy recognizes the partnership among
harvesters, processors and communities. This partnership is like a
three-legged stool. Cut out any leg and the stool tips over.
The solution to keeping this stool strong and stable is to ensure
that any rationalization program maintains the integrity of this
partnership by providing incentives for all parties to work toward
mutually beneficial goals. There are two approaches that accomplish
this objective. The two approaches that maximize the benefits to the
Nation and equitably distribute these benefits among the three partners
are the Voluntary Cooperatives and the Two Pie allocation of quota
share, which in combination with the CDQ allocations and other
community elements, is in reality a Three Pie allocation scheme. This
Voluntary Three Pie Cooperative approach, approved by the North Pacific
Fisheries Management Council (hereafter referred to as the Council),
recognizes the prior economic interests among harvesters, processors
and communities. This three-pie approach fits the Bering Sea crab
fishery.
The Council has a pioneering history of designing rationalization
programs unique to the fishery at hand. The Council crafted an IFQ
program for halibut and sablefish that fit the small vessel, owner on-
board nature of that fishery. For pollock, the mechanism was processor-
linked cooperatives. We design rationalization programs to fit the
dynamics and needs of the particular fishery. For the large boat,
heavily industrialized, corporate nature of the BSAI crab fisheries,
the Council found that a voluntary, three-pie cooperative structure fit
best. Given the Council's history, I fully expect a different model
will emerge from the Council for the Gulf of Alaska rationalization
process that is currently underway.
I want to emphasize that conservation and safety are as much a
driving force to rationalize as the need to re-vitalize the economic
health of the industry and communities. Resource protection is imbedded
within the Council's plan. With extended, slower seasons it is
anticipated that deadloss will be greatly reduced and that the already
low bycatch will decline further. With more time to prosecute the
fishery, fewer pots will be lost and ghost fishing by lost or abandoned
pots will be reduced even beyond the low level that the current
regulatory safeguards have resulted in. To ensure that the resource
remains a paramount benefactor of rationalization, increased monitoring
(with fees to support it) will occur. With voluntary cooperatives, the
regulatory units will be smaller and hence administrators and resource
managers will benefit as well. Furthermore, ending the ``race for the
fish'' builds a conservation safety net that the Board of Fisheries and
the Department of Fish and Game will build and act upon.
And most importantly, the Council's plan saves lives. The current
management scheme of the BSAI crab fisheries results in a derby-style
``race for fish''. This system often puts pressure on participants to
fish in unsafe weather conditions, work continuously for long periods
without rest. There is also the potential to overload vessels with
fishing gear in an attempt to improve returns under these conditions.
Falling Guideline Harvest Levels (GHLs) and overcapacity in all crab
fisheries has reduced fishing periods and increased the ``race'' for
crab, thus exacerbating safety concerns for BSAI crab vessels.
According to the 1997 and 2001 Occupational Safety and Health
studies, the Alaskan commercial fishing industry had a fatality rate 28
times the national occupational average. Within this high-risk
industry, the BSAI crab fisheries were found to account for a
disproportionately high level of injuries and death. From 1990 to 2001,
there were a total of 25 vessels lost, resulting in 40 fatalities.
Twenty-one more lives were lost as a result of being swept overboard,
crushed by crab pots or entangled in lines or winches. These losses are
simply unacceptable. It is these losses that in part compelled the
Council to act now and end the dangerous ``race for fish.''
Examining safety statistics in the Alaskan halibut fishery suggests
that rationalization of that fishery has been a force in improving
safety. In the five years preceding implementation of the halibut IFQ
program (1990 to 1994), 17 fatalities occurred in the halibut fishery.
In the 7 years after implementation of the program (1995 to 2001) the
number of fatalities in the fishery declined to 4. While no one can
predict the number of lives saved, it is anticipated that rationalizing
BSAI crab, the most dangerous fishery in the U.S., will have a similar
affect on reducing fatalities.
It should also be noted that to the extent that a rationalized
management system results in a sustainable, economically viable
fishery, then improvements in safety should follow. In economically
viable fisheries, harvesters are able to make a profit, and vessel
owners are able to invest in better quality equipment, proper vessel
maintenance, and hire, train and keep professional skippers and crews.
All of these factors help improve safety.
Background on the Council Process
Congress directed the Council to consider a wide variety of
approaches to a rationalized fishery. As part of the Consolidated
Appropriations Act of 2001 (Pub. L. No. 106-554), Congress directed the
Council to examine fisheries under its jurisdiction to determine
whether rationalization was needed. The specific legislative language
is:
The North Pacific Fishery Management Council shall examine the
fisheries under its jurisdiction, particularly the Gulf of
Alaska groundfish and Bering Sea crab fisheries, to determine
whether rationalization is needed. In particular, the North
Pacific Council shall analyze individual fishing quotas,
processor quotas, cooperatives, and quotas held by communities.
The analysis should include an economic analysis of the impact
of all options on communities and processors as well as the
fishing fleets. The North Pacific Council shall present its
analysis to the appropriations and authorizing committees of
the Senate and House of Representatives in a timely manner.
Early in the rationalization process, the Council established a
broad-based industry working group to develop options to rationalize
the Bering Sea crab fisheries. This work group met over a two-year
period to identify workable solutions and to develop a majority
opinion. Including the work of the BSAI Crab Committee, the Council has
studied this issue for well over three years, with a detailed analysis
of all the issues brought forward by the Committee which included the
concept of processor shares. Additionally, the Council held hearings in
major ports and communities of Alaska, Washington, and Oregon.
The first Council action was to unanimously adopt a problem
statement that any solution should consider ``the lack of economic
stability for harvesters, processors and coastal communities''.
Adopting this problem statement allowed for the detailed analysis of
numerous options. To assist in our understanding of the impacts of the
various options on all participants, we employed several specialists to
work with Council staff. After several revisions and an exhaustive
public review of the analysis the Council then unanimously adopted a
final motion in June 2002, which responded to Congressional direction.
Since then, the Council continued to work on several trailing
amendments and completed its work on BSAI Crab Rationalization in April
2003.
Key Elements of the Council's Motion (includes trailing amendments)
Harvest shares will be allocated for 100 percent of the
total allowable catch (TAC). The qualifying years selected for
these shares were the years recommended by the BSAI Crab
Committee and the Advisory Panel.
Processing shares will be allocated for 90 percent of the
TAC. The qualifying years selected for these shares were the
years recommended by the BSAI Crab Committee.
Regional share designations will apply to all processor
shares and to the corresponding 90 percent harvest shares. The
regional designation recognizes historic delivery patterns and
distributes landings between two specific regions.
There is no linkage between harvest shares and processing
shares. Harvesters are free, within regions, to match up shares
with any processor that hold processing shares. Ten percent of
the harvester's allocation will remain available to deliver to
any processor (called B shares) whether or not that processor
was eligible to receive an initial allocation of processing
shares.
Voluntary harvester cooperatives are encouraged and once
formed should achieve efficiencies through the coordination of
harvest activities and deliveries to processors. It is
anticipated that these cooperatives will also serve as a forum
for profit-sharing arrangements similar to American Fisheries
Act (AFA) cooperatives.
Community Development Quota allocations will be increased
from 7.5 percent to 10 percent of the TAC. A community quota of
brown crab was also established for Adak, a similarly situated
community that lies outside the CDQ program.
A cool down period that would restrict the movement of
processing activity from crab dependent communities for a
period of two years following implementation of the program.
Allow communities to purchase harvest and processing shares
and grant crab dependent communities the right of first refusal
to purchase processing shares that may transfer outside of the
communities. Grant communities in the Northern Gulf of Alaska
the right of first refusal on the sale of processor shares from
communities that are not dependent on the crab fisheries.
A captain share allocation of 3 percent of the TAC was
established for exclusive use by captains and crew.
A vertical integration cap of 5 percent is used to control
processor accumulation of harvesting privileges.
Use and ownership caps on harvesting and processing
privileges are included to limit consolidation and excessive
control.
A mandatory binding arbitration program will be used to
settle any outstanding price disputes between harvesters and
processors.
A crew loan program will assist crewmember entry into the
fisheries.
Harvest restrictions, (called GOA sideboards) were included
to protect non-rationalized Gulf of Alaska fisheries.
Comprehensive data collection and program review will be
used to assess the success of the rationalization program and
to make modifications where needed.
As a Council, we have an obligation to review program performance,
especially with respect to impacts on all three partners. And we will
make necessary adjustments to achieve our stated goals and minimize
unintended consequences. Through program review we can adjust elements
of the initial policy design. This flexibility is uncommon to most
market based rationalization plans. Specifically, we can
1. adjust the percentage of B shares.
2. adjust attributes intended to protect community interests in a
region.
3. adjust the mechanism for community protection.
4. adjust specific design features to adapt to changes in stock
status.
Overall, the crab rationalization program strikes a balance among
the interests of harvesters, processors, and communities. The Council,
based on exhaustive public input over three years, unanimously approved
the crab rationalization program. In making this unanimous vote Council
members recognized the comprehensive, balanced approach that
incorporated both short and long-term safeguards to protect the
interests of all three partners.
Benefits for Harvesters, Processors and Communities
Now, I would like to address the how and why involved in creating a
stable three-legged stool, where harvesters, processors and communities
each benefit from crab rationalization. First in the line of
benefactors are the fishermen. All eligible fishermen gain by having
value attached to their fishing history. Some will have more value than
others. The years selected for qualifying quota share determine the
extent that each fishermen gains. The Council chose the years agreed to
by the industry committee and recommended by the Advisory Panel. More
recent years were also considered, and in a few cases adopted, based on
NOAA general counsel's advice to consider recency for qualification
determinations. Regardless of where a fisherman sits in the qualifying
period, the award of quota generates immediate value and economic gain
over the race-for-fish scenario. Furthermore, it is anticipated that
consolidation and the ability to stack quota, combined with a buy back
program, could reduce the crab fleet significantly. This results in
very significant economic gains to the remaining participants.
Even though these economic gains are recognized by many crab
fishermen, there was still a need to address ``equity issues'' and
balance of negotiating strength between fishermen and processors. With
respect to these issues, the Council provided many protective measures
to ensure that vessel owners are not disadvantaged. They include:
A discount of each processor's history, which leaves 10
percent of the harvest shares (B shares) open to delivery to
any processor. It is anticipated that these B shares will
provide adequate negotiating leverage for fishermen.
A system of binding arbitration that to ensure that no
harvester will be forced to accept a price by a processor
holding processor shares. The system will be conducted using an
independent third party arbitrator when or if there is a price
dispute between harvesters and a processor. It will include an
opportunity for groups of harvesters to bargain collectively,
as they have traditionally for Bering Sea crab prices, or to
enter arbitration on an individual basis. The binding
arbitration approach adopted by the Council also provides for
independent market analysis and a non-binding pre-season price
formula.
Processors will be required to give the regulators full,
complete and detailed crab processing costs and crab sales
revenues. This information will be useful for the Council to
monitor the way revenues are shared with vessel owners.
Caps on the amount of harvesting shares that can be owned by
or affiliated with processors. This is intended to ensure that
the processor will continue to need crab from independent
harvesters.
Prohibiting vessels owned by processors from being awarded
``B'' harvest shares. This was done in order to further protect
the interests of independent harvesters.
While not a direct part of the Council's motion to protect
fishermen's interest, there is another key advantage that accrues to
fishermen as a result of rationalization. Currently, harvesters cannot
really withhold product from processors for long periods of time
because the current open-access model allows any catcher processor or
other strike breaker to fish while others remain tied to the dock. Once
IFQ is issued, every harvester can--either individually or as a member
of a cooperative--withhold product from the processing sector without
fear of someone else harvesting it. As learned from the AFA model, the
ability to withhold product is a powerful tactic that fishermen can use
to leverage fair prices.
It is also important to point out that the Council instituted caps
on the amount of consolidation that may occur among the processors. The
consolidation rules are far more restrictive than what would be
required under normal antitrust laws, and they are intended to ensure
multiple market opportunities. Furthermore, when crab stocks are low, a
quota system that includes processors should result in less
concentration of processing facilities and more buying entities as
firms could choose to lease quota or have their shares custom
processed. These measures to keep the existing processing base healthy
while allowing for new entrants should expand, not limit, market
opportunities for fishermen.
The next set of benefactors are the crab processors. Having
observed processor fallout from the harvester only allocation on IFQs
for halibut and sablefish, and processor viability under the AFA
cooperatives, the Council made a deliberate choice to include
processors in the rationalization program. This decision to protect
processor's investments also responds to the Congressional direction
given to the Council.
Under the sablefish IFQ program (note: sablefish is a more
appropriate comparison than halibut because sablefish did not change
product form from frozen to fresh) only 25 of the 67 pre-IFQ firms
survived the harvester only allocation of quota according to thorough
analysis done by Washington State University. Additionally, most of the
surviving processing firms lost market share. Processor shares will
allow the processing sector to more fully experience the efficiency
gains of rationalization and thus remain viable. With the ability to
trade and purchase processing shares, up to a set limit, it is
anticipated that the number of processors available to buy crab may
actually increase. In the very least, award of processor shares should
stabilize the crab processing sector so that the number of currently
viable processors will remain the same.
Similar to fishermen, processors have concern that they will be
``price takers''. Currently, some processors estimate that fishermen
capture upwards of 87 percent of the wholesale price of crab. Crab
harvesters are organized under the 1934 Act, into a monopolistic
bargaining association and as such, negotiate price as one entity
represented by the Alaska Marketing Association. The processors are not
similarly organized and argue that the bargaining strength is all in
the fishermen's hands. This perspective suggests that like beauty,
bargaining strength is in the eyes of the beholder, or the potential
loser in this case. Nonetheless, the Council's approach to binding
arbitration, data collection and program review are aimed to protect
the processing sector as well.
Lastly, the Council chose to provide specific community protection
measures to recognize the investment of communities in the development
of the Bering Sea crab fisheries, noting the very real likelihood for
those same communities to experience irreparable harm without
protection. As you know, National Standard 8 of the Magnuson-Stevens
Act directs Fishery Management Councils to take into account (a) the
need for sustained participation of fishing communities and (b)
minimizing adverse economic impacts to such communities. The Council
chose to address communities and National Standard 8 through a ``three
pie'' approach to rationalizing the crab fisheries in the Bering Sea.
There are two key elements to the community third pie--(a) market-based
allocations to community interests and (b) measures to protect
communities from potentially adverse effects. Following are the
specific elements of the ``community third pie''.
Market-based Allocations to Community Interests include:
Expanding the CDQ program (includes 65 Western Alaska
communities) to include all crab fisheries approved under the
rationalization program with the exception of Western AI brown
king crab.
Increasing the CDQ allocation of crab from 7.5 percent to 10
percent.
Allocating the percentage of Western AI brown king crab not
used during the base period to the community of Adak and
require that up to 50 percent of the ``A'' share brown king
crab be processed in Adak.
Allowing for purchase of processing and harvesting shares by
crab dependent communities not included in the CDQ program.
Give these same communities the right of first refusal for
processing shares that may leave their communities after a two-
year cool down period. Grant communities in the Northern Gulf
of Alaska the right of first refusal on the sale of processor
shares from communities that are not dependent on the crab
fisheries.
Community Protection Measures include:
Establishing two regions to protect historical, primary
delivery patterns of various crab species and attach regional
designations to crab harvested with Class A shares and to
processing quota shares.
Maximizing the benefits of regionalization on the
communities of St. Paul and St. George by limiting B shares to
10 percent of the total amount of crab.
Imposing a two-year (post implementation) cool down period
where processing shares must remain in the originating
community.
Imposing processing use caps to assure that at least two
plants will always be operating in the Northern Region and four
plants in the Southern Region.
Limiting the future expansion of the offshore catcher
processor fleet so that more crab might be processed in
community based plants.
Requiring that a portion (equivalent to the increase
received) of CDQ crab be delivered to shore-based plants.
Allowing for the transfer of catcher processor shares to
shore-based processors.
Providing a 3 percent allocation to help captains, some of
who reside in coastal communities, to leverage future
employment opportunities in a rationalized crab fishery.
Excluding small, underutilized crab stocks from the
rationalization program so that small vessel, state water
fisheries could be developed near communities.
The direct allocation of harvesting and processing shares, and
purchase of processing shares by communities, translates into direct
benefits to Alaska's coastal communities. These actions in conjunction
with the regionalization elements comprise the ``meat of the third
pie''.
Although processor shares are typically viewed as part of the
second pie, the issuance of such shares has direct implications for the
protection of a community's processing base. As noted above, experience
to date with the issuance of harvester only allocation in Alaska
reveals a dramatic reduction and shift in shore based processing
operations. While the causes for this shift and reduction are complex
and debatable, it is good public policy to try and minimize these
impacts, as the more processors that survive the transition to a
`rationalized fishery' the better the community's economic base will
be. Such a system inherently benefits communities as well as
processors. Furthermore, the establishment of processor shares allowed
for the creation of additional community protection measures. In
essence, the processor pie helps makes the community pie whole.
Catcher Processors
There remains one more sector that benefits from the Council's
action--catcher processors. Catcher processors have long had an
advantage over all other sectors participating in the harvesting and
processing of BSAI crab. This advantage comes about from being 100
percent vertically integrated operations, which do not have to purchase
all their crab. The Council's action solidified this sector of the
industry through the creation of catcher processor shares. However, the
Council chose not to allow this sector to grow at the expense of the
onshore sector and disenfranchise Alaska's coastal communities. Hence,
the Council's program limits catcher processors to their historical
aggregate level. In essence, the Council made the catcher processor
sector whole without being punitive and for some fisheries such as the
Aleutian Islands brown king crab, the Council chose qualifying years
that advantaged catcher processors.
Criticism of Council Action
Despite the balanced approach of the Council motion and extensive
efforts to be inclusive of all affected parties, the Council action
still faces criticism. I will take some time to address the following
concerns:
Not including the year 2000 for processing shares
Not creating a higher portion of B shares
Antitrust implications of processing shares.
Year 2000
It is reported that including 2000 would have benefited Kodiak
based processors. While is true that the year 2000 was one of Kodiak's
better years in terms of BSAI crab landings, the Council analysis shows
that Bristol Bay Red King Crab comprised 1.8 percent of the total value
of fish landed at the port of Kodiak and Bering Sea Opilio Crab
represented 1.35 percent of the total value landed in the year 2000.
It is important to note that the year 2000 was not included in the
recommendations made by the industry committee which, incidentally,
included Kodiak representatives who did not bring this concern forward.
The committee noted that the year 2000 was not included for a number of
reasons: (a) it was the only year that had a delayed start date of
April 1 due to inclement weather and some vessels did not participate
with the season change date; (b) it was the only potentially qualifying
year affected by the AFA processor sideboards; and (c) a large storm
greatly disadvantaged some small Alaskan vessels. As such, the year
2000 was seen as an anomaly and not included in the recommendations or
in the Council actions.
Nonetheless, processors in Kodiak will still benefit from the
rationalization program, as they will no longer just be the port of
last load delivery. With a slower, extended fishery fishermen will have
the time to deliver mid-season to Kodiak processors. There are
processors in Kodiak who will receive processor shares. Other
processors in Kodiak have the option of buying processor shares or
receiving ``open delivery'' B-shares of crab. In fact, Kodiak, home to
Alaska's largest crab fleet, is the port that is expected to benefit
most from the `open delivery' arrangement. To help ensure this, the
Council made B-shares off limits to catcher processors.
Kodiak is not the only community to gain under the Council's
rationalization plan. Dutch Harbor, St. Paul, St. George, Adak, Sand
Point and all the communities under the CDQ program gain in the
Council's three-pie approach to rationalization.
Why 10% B Shares
In determining the appropriate amount of B shares, the Council
first looked to the experience under AFA. In the AFA there is a similar
10 percent discount on processing history, i.e.; vessels may deliver up
to 10 percent of their catch to any processor. Immediately after AFA
passage, many harvesters tried to get the Council to amend the AFA
agreement. They did so under the Dooley-Hall proposal. There were
claims that the processors would capture all the profits. Today, if one
were to ask key Dooley-Hall advocates if they are disenfranchised and
would prefer no AFA or AFA as originally passed, their answer is
unequivocal: AFA as enacted by Congress. This suggests that a 10
percent discount could serve as adequate negotiation leverage in a
large-boat industrialized fishery.
Based on the cost and wholesale data provided by crab processors,
it does not make sense to discount processor history any more than the
10 percent level selected by the Council. Data was provided to the
Council in public testimony that shows that processor margins are very
thin, and probably non-existent in the small quota, slow-paced crab
fisheries. With these low margins, processors will be hungry for all
deliveries of crab, especially the crab delivered last. Once the start-
up and initial operation costs have been covered, the remaining loads
of delivered crab are where the margins increase and profits are
realized.
Additionally, crab fishermen (through their bargaining association)
now allocate crab as a reward for a company setting the price, a price
that in turn is adopted by the rest of the companies. The amount
allocated has been less than 5 percent to achieve this advantage. This
demonstrates the power of having 10 percent (or less) of the crab to
allocate to processors as a reward for higher price. As such, the
notion that the fishermen delivering last will become a price hostage
is a notion unsupported by existing economic reality. No one will fish
A-shares without a contract. There will be no ``last harvester leaving
the grounds.'' Furthermore, with the adoption of voluntary cooperatives
between fishermen and in discussions with a processor of their choice,
it is anticipated that price and profit-sharing arrangements will: (a)
cover all the crab the harvester intends to deliver, including the 10
percent B shares portion; (b) address profit sharing across vessels
participating in the cooperative; and (c) be agreed to well in advance
of the season.
Public testimony from the community of St. Paul also pointed out
that a higher amount of B shares would negatively impact their economic
base. The balancing act of community interests required the Council to
consider these concerns as well when determining the appropriate amount
of B shares.
Antitrust Issues
The Department of Justice has not found any anti-competitive
effects in the division of whiting allocations among catcher-processors
or the division of pollock allocations among AFA cooperatives. It is
expected that the Department will view the crab rationalization plan in
a similar light. The Congressional Research Service has already made a
similar review of the Council's action to create processor shares and
concluded ``that the Federal antitrust laws are likely to be deemed
irrelevant in the context of the subject proposal''.
Even with this green light from the Congressional Research Service
it is important to note that the same anti-competitive issues that are
being raised under crab rationalization were brought up under the AFA.
Since the processor provisions of the AFA had no anti-competitive
effects, then the less restrictive processor elements included in the
Council approach on BSAI Crab Rationalization should be viewed
similarly. The less restrictive provisions of the Bering Sea/Aleutian
Islands (BSAI) crab rationalization program include: (1) no closed
class of processors (new processors can purchase `open delivery' B
shares or allocated shares from a qualifying processor); and (2) no
direct linkage between fishermen and processors.
The 1934 Fisherman's Marketing Act contains an explicit statutory
exemption from the antitrust laws allowing fishermen to bargain
collectively in price negotiations with fish processors. The fishermen
are allowed to decide, as a group, on their bargaining position, an
activity that would violate the Federal antitrust laws without the
exemption.
The Whiting Cooperative (vertically integrated catcher processors)
and the AFA inshore cooperatives have each submitted to the Department
of Justice a request for a ``business review letter'' concerning
planned cooperative activities. Section 210 of the AFA requires each
cooperative to request a business review letter prior to submitting the
cooperative contract to the Council and the Secretary. Each cooperative
described its proposed activities of allocating pollock quotas among
member vessels. In response, the Department of Justice has issued the
requested business review letters stating that the Department of
Justice finds that the described activities are not anti-competitive
and the Department intends to take no enforcement action. However, the
AFA does not require the cooperative to have received a business review
letter and the AFA does not grant an exemption to the antitrust laws.
While we anticipate a similar ruling for crab cooperatives, if Congress
approves the Council's crab rationalization plan, this is a non-issue.
Concluding Remarks
In closing let me emphasize that while imposing an allocation
scheme automatically has varying impacts among the participants, the
Council strove to make all sectors--harvesters, processors, and
communities--win from BSAI crab rationalization. It was a challenging
balancing act to say the least. However, through the hard work of the
industry committee, the Council staff, the Advisory Panel and the
community representatives, the Council's preferred alternative does
benefit all sectors. And most importantly, BSAI Crab Rationalization
will save lives and improve management while reducing bycatch and
deadloss.
The program is a Voluntary Three Pie Cooperative with unique
protections and opportunities for communities and captains. The novelty
of the program compelled the Council to include several safeguards into
the program, including a binding arbitration program for the resolution
of price disputes and extensive data collection and review programs to
assess the success of the rationalization program. These safeguards
demonstrate the Council's commitment to a fair and equitable
rationalization program that will protect the interests of all sectors
which depend on these crab fisheries.
In unanimously adopting the crab rationalization plan, the Council
made it very clear that the program was one crafted specifically for
the crab fisheries of the Bering Sea/Aleutian Islands, and that in so
doing no one should assume a similar system for other fisheries under
its jurisdiction. We have met the charge given to us by Congress. We
have acted responsibly and creatively to meet all the standards of the
Magnuson-Stevens Act. It is time for Congress to move the plan to
rationalize these crab fisheries forward.
Senator Stevens. Ms. Freed?
STATEMENT OF HON. LINDA FREED, CITY MANAGER,
CITY OF KODIAK
Ms. Freed. Thank you.
Senator Stevens. Could you pull that mike up toward you?
Ms. Freed. Is that close enough for you?
Senator Stevens. Not quite. Thank you.
Ms. Freed. Thank you.
For the record, my name is Linda Freed. I have been a
resident of Kodiak, Alaska, for 23 years, and I now have the
privilege of serving as the City Manager of the City of Kodiak.
I would like to note, for your information, that in the
audience today is the Honorable Carolyn L. Floyd, Mayor of the
City of Kodiak, and Dave Woodruff, a council member from the
City of Kodiak. In addition, Dick Powell, a long-time Kodiak
City resident and a Bering Sea crab fisherman, is also here.
Kodiak is a major fishing port in the United States. It has
been consistently in the top communities for both product
landed, the volume of product landed, and the value of that
product. Kodiak has the largest resident fishing fleet in the
State of Alaska and the largest resident population of
processing workers.
Crab rationalization and the issue of processor quota
shares are critical to the City of Kodiak and all its
residents. Kodiak has already lost 70 children from our school
system within this last year due to the closure of processing
plants and reductions in our fishing fleet. Kodiak is a fishing
town, and we cannot afford to see our primary source of
natural-resource employment taken from our residents by either
fisheries managers or Congress.
The City of Kodiak is opposed to the North Pacific
Fisheries Management Council's request that Congress pass
special legislation to create individual processor quotas, or
IPQs to establish individual monopoly markets for processing
crab harvested from the Bering Sea. If enacted, the Council's
proposal would grant each member of a select group of companies
the exclusive right to process specific percentages of the
Bering Sea crab after it has been harvested by independent
fishermen. Under the Council plan, each fisherman would be
forced to deliver 90 percent of their catch to one of a select
group of processing companies that are issued an IPQ.
The City of Kodiak is not alone in its opposition to the
Council's unprecedented proposal. All of the regional
Governments with communities where crab is processed--the
Aleutians East Borough, the Lake and Peninsula Borough, the
Kodiak Island Borough, and the Kenai Peninsula Borough--are all
seriously concerned about this proposal. The Aleutians East
Borough and the Kenai Peninsula Borough have gone further and
passed formal resolutions opposing the Council plan, as have
the cities of Sand Point, Chignik, Homer, and Seldovia, just to
name a few. In fact, the only communities on record supporting
the Council proposal are Dutch Harbor and Akutan. Even St.
Paul, a community in the middle of the Bering Sea crab fishing
grounds, is divided about whether to support or oppose IPQs. In
fact, TDX, the village corporation for St. Paul, and the City
of Kodiak find ourselves in similar circumstances. We both own
processing plants that have processed crab. We are not eligible
to have IPQs because we did not run those plants ourselves.
The adverse impacts of IPQs on Alaskans are tremendous.
Fishing communities that depend on independent fishermen and
local processing plants for the mainstay of our economies will
be devastated if Congress enacts this IPQ proposal. IPQs will
mean the loss of innovation and competition within the
processing industry. Companies with IPQs will not have to
remain efficient, establish new product lines, or offer to
process less valuable species in order to attract more
deliveries. With a guaranteed share of the harvest, companies
with IPQs will maximize their profits by minimizing what they
pay for crab and reducing their investment in their facilities.
The City of Kodiak alone stands to lose between $3 million
and $10 million a year in revenues to our community and lost
fish taxes if the Council's plan is implemented. I must say
that we had to have that analysis done on our own, because the
State of Alaska did not provide that kind of economic analysis
for our community.
The community protection provisions that have been proposed
by the Council do not protect Kodiak and many other coastal
communities. One provision is a right of first refusal on a
proposed sale of IPQs between two companies. However, a
community must meet all of the terms and conditions of the
agreement between the two private parties and provide full
payment within 120 days. This means that the two private
parties, one of whom is a buyer, has a vested interested in
making sure a community cannot afford to exercise its option,
and they have every incentive to include provisions in the
agreement that are poison pills to the community. Most
communities would need 120 days just to start financing
negotiations, much less make full payment of potentially
millions of dollars. And there is no appeal process under the
Council plan if a community does, in fact, try to exercise its
right and is unable to do so. The Council's IPQ plan
effectively returns Alaska fishing communities to the days
before statehood, when our State fishery resources were under
the control of a few large processing companies. The Council's
own documents show the extent to which a few processors will
control Alaska's fisheries if this plan is enacted.
In 1998, Congress enacted the American Fisheries Act, which
created a closed class of processors for the inshore delivery
of Bering Sea pollock. The AFA granted six companies exclusive
right to process the entire 50 percent of the Bering Sea
pollock that was AFA decreed to be delivered inshore. All six
of those AFA companies also process crab. According to the
Council's own preliminary analysis, those six AFA processing
companies will be granted IPQs approximately equal to 75
percent of the Bering Sea opilio and red king crab harvest.
Perhaps coincidentally, five of those AFA companies are also
defendants in the ongoing Bristol Bay salmon antitrust lawsuit
in Anchorage.
As many Members of the Committee are aware, Kodiak and
other Alaskan coastal communities are in the midst of a long-
term crisis in the salmon industry. IPQs will make this even
worse. The combined effect of the closed class of pollock
processors and IPQs will virtually ensure that there is no
solution to the salmon problem. One of the AFA companies, Wards
Cove, has used its guaranteed pollock revenue to enable it to
withdraw from salmon processing marketing entirely. They no
longer need the marginal income from their salmon operations.
Other AFA companies have reduced their salmon buying in the
years since the AFA passed despite the considerable increase in
revenues and financial stability provided by the AFA. Add the
guaranteed revenue from crab IPQs, and these companies have
even less incentive to expend money to gain marginal revenue
from salmon.
We need to attract new and innovative processors to our
salmon industry. The Council plan just locks in the same old
companies that have created the mess we currently find
ourselves in.
Despite the overlap in companies that benefit under the AFA
and the Council's proposal, it needs to be stressed that the
benefits granted to processors under the AFA and the IPQ plan
are not the same. The fact that Congress passed the AFA
provides no basis for Congress to pass IPQs. IPQs confer more
benefits and far greater market power to crab processors than
pollock processors received under the AFA.
And I would like to refer you to an attachment to my
testimony, by Dr. Halvorsen, that points out the significant
differences between the AFA and the plan proposed by the
Council.
The simple truth is that IPQs are nothing more than an
economic allocation of markets. They are a ransom that
processors are demanding before they will allow the crab
fishery to be rationalized to improve conservation and fishing
safety. There is no fishery management or community-protection
justification for IPQs.
The push for IPQs will not stop with crab. The Council is
already considering plans to rationalize the Gulf of Alaska
groundfish fisheries, and some Gulf of Alaska fish processors
are demanding that IPQs be included. Further, a recent report
on the crisis in salmon fisheries suggested that IPQs may be
needed for salmon fisheries. Congress has already passed
special legislation for pollock and is being asked again to
pass special legislation for crab. If Congress authorizes crab
IPQs, the floodgates will be opened, and Congress will be
besieged with requests for special legislation for each fishery
off Alaska.
Having Congress legislate fishery by fishery is precisely
what the Magnuson-Stevens Act was designed to prevent. Congress
needs to send a clear message that IPQs will not be allowed.
Once that message is delivered, it will be easy for the Council
to quickly implement an individual fishing-quota program under
its existing authority that will address the legitimate
concerns of fishermen, communities, and processors.
Thank you for the opportunity to present this testimony.
And I would ask if the Committee could keep the record open for
another 10 days, because we do have some additional information
we would like to present you with.
Thank you.
[The prepared statement of Ms. Freed follows:]
Prepared Statement of Linda Freed, City Manager, City of Kodiak, Alaska
My name is Linda Freed. I have been a resident of Kodiak, Alaska
for 23 years and now have the pleasure of serving as the City Manager
for the City of Kodiak. Kodiak is a major fishing port in the United
States, and has consistently been among the top three fishing ports for
both volume and value. Kodiak has the largest resident fishing fleet in
Alaska and also the largest resident population of processing workers.
Crab rationalization and the issue of processor quota shares are
critical to the City of Kodiak and all of its residents. Kodiak has
already seen 70 children leave our schools in the last year due to the
closure of processing plants and reductions in the fishing fleet.
Kodiak is a fishing town, and we can not afford to see our primary
source of natural resource employment taken from our residents by
fishery managers or the Congress.
The City of Kodiak is opposed to the North Pacific Fishery
Management Council's request that Congress pass special legislation to
create Individual Processor Quotas--IPQs--to establish individual
monopoly markets for processing crab harvested in the Bering Sea. If
enacted, the Council's proposal would grant each member of a select
group of companies the exclusive right to process a specific percentage
of the Bering Sea crab after it has been harvested by independent
fishermen. Under the Council plan each fisherman would be forced to
deliver 90 percent of their catch to one of the select group of
processing companies that are issued an IPQ.
Many Alaskans are frankly stunned that Congress would seriously
consider the Council's proposal for IPQs. An analogy to another
independent American industry might help the Committee understand our
dilemma. Congress established permits for ranchers to graze their
cattle on Federal land when the west was fenced a century ago. Those
permits are akin to individual fishing quotas. What the Council has
proposed to address the ``harm'' to processors caused by individual
fishing quotas is the equivalent of granting a select few companies the
exclusive right for each of them to process a set percentage of all of
the beef products made from cattle grazed on Federal land, and then
requiring each rancher to deliver 90 percent of the cattle grazed on
Federal land to one of those companies--all in the name of protecting
the meat packing companies from the harm caused by granting grazing
rights on Federal land to independent cattle ranchers.
The City Okuda is not alone in its opposition to the Council's
unprecedented proposal. All of the regional governments with
communities where crab is processed the Aleutians East Borough, Lake
and Peninsula Borough, Kenai Peninsula Borough, and Kodiak Island
Borough-are seriously concerned about the Council proposal. The
Aleutians East Borough and the Kenai Peninsula Borough have passed
formal resolutions opposing the Council plan, as have the cities of
Sand Point, Chignik, Homer, and Seldovia, to name a few. In fact, the
only communities on record supporting the Council proposal are Dutch
Harbor and Akutan. Even St. Paul, a community located in the middle of
the crab fisheries, is divided about whether to support or oppose IPQs.
The adverse impacts of IPQs on Alaska are tremendous. Fishing
communities that depend on independent fishermen and local processing
plants for the mainstay of their economies will be devastated if
Congress enacts this IPQ proposal. IPQs will mean the loss of
innovation and competition within the processing industry. Companies
with IPQs will not have to remain efficient, establish new product
lines, or offer to process less valuable species in order to attract
more deliveries. With a guaranteed share of the harvest, companies with
IPQs will maximize their profits by minimizing what they pay for crab
and reducing their investment in facilities. To enable the IPQ
companies to do just that the Council included in their plan a binding
arbitration proposal supported by the processors, community protection
provisions that are mere window dressing, and explicit language to
ensure that processors could consolidate facilities by transferring and
leasing IPQs.
The effects are not limited simply to fishermen and processing
plant workers. The City of Kodiak alone stands to lose between
$3,000,000 and $10,000,000 a year in revenues to the community and lost
fish taxes if the Council's plan is implemented. Because neither the
Council nor the State has undertaken an economic impact analysis of the
Council proposal, the City had to hire its own economist to provide
these figures. A copy of that analysis by Richard Tremaine is attached
to my testimony.
Other communities face an even more daunting prospect--under the
Council plan a processor could transfer its IPQ out of their community
and leave them with a crab processing plant which no one can use
because there are no IPQs available to allow that plant to operate. The
community would be legally barred by Congress from getting a new
processor to come in and operate in their community.
Kodiak fishermen and processors are also directly harmed by this
proposal. Alaska Fresh Seafood, a long time Kodiak processor that
helped pioneer the Bering Sea crab fisheries, would be unable to
process crab caught by the fishermen that are part owners of the plant
because, under the restrictive criteria selected by the Council, Alaska
Fresh would receive IPQs equal to far less than what they historically
processed. In fact, the Council's own analysis of their proposal shows
that the 12 largest processors receiving IPQ under the plan would get
nearly 20 percent more than they have historically processed-to the
detriment of Kodiak processors who have already been locked out of
entering the Bering Sea pollock processing market.
Another Kodiak example of the absurd nature of the Council's IPQ
plan is illustrated by Dick Powell, a long time Kodiak fisherman who
operates both a crab catcher processor and a crab fishing boat. Under
the Council plan Dick would be unable to use his own catcher processor
to process crab from his own fishing boat. Once again the Council's
plan operates to thwart competition and free enterprise.
Alaska coastal communities fare no better under the Council plan.
The so-called ``community protection provisions'' are nothing of the
kind. What the Council has offered to communities is a first right of
refusal on a proposed sale of IPQs between two private companies.
However, a community must meet ALL of the terms and conditions of the
agreement between the two private parties, and provide full payment
within 120 days. This means that the two private parties, one of whom
(the buyer) has a vested interest in making sure a community can't
afford to exercise its option, have every incentive to include
provisions in the agreement that are poison pills to the community.
Most communities would need 120 days just to start financing
negotiations, much less make a full payment of millions of dollars. And
there is no appeal process under the Council plan if a community does
in fact try to exercise its right and is unable to do so.
I was a member of the Community Protection committee appointed by
the Council chairman, and have to say that the process was nothing more
than a sham. The committee was under instruction not to consider
anything that was outside the scope of the IPQ alternative that the
Council had already adopted unless \3/4\ percent of the Committee
members agreed to discuss it. As a result the Committee was unable to
have any discussion of, let alone take any action on, substantive
proposals that would have addressed the impacts of IPQs on Kodiak and
other coastal communities.
The City of Kodiak also takes exception to the May 6, 2003 letter
the Chairman of the North Pacific Fishery Management Council sent to
Congress. The letter once again attempts to mislead Congress about the
support for this plan by referencing an 11 to 0 vote that occurred in
June of 2002, before all of the details of the Council plan were known
to the public or even to Council members. Since that time the Council
has had 6 to 5 votes approving various elements of the ``safeguards''
promised to communities and fishermen at that June meeting. Now that
the details of the entire plan are known, only one fishing group
controlled by processors and two Alaska communities remain in support
of the IPQ plan.
Contrary to the claim of the Council chairman, the Council has NOT
taken final action on the entire suite of crab rationalization options.
The Council was scheduled at their upcoming meeting in Kodiak this June
to vote to put out for public review a draft analysis of the entire
package that includes the work done over the past year. But the Council
chairman chose to pull crab rationalization from the June agenda rather
than have to report to Congress a less than unanimous vote in support
of the entire IPQ plan. In addition, by delaying action until the fall,
the Council chairman has denied the public and the Congress access to
the Council's detailed analysis of the IPQ alternative and an
individual fishing quota only alternative that is also being considered
as part of the National Environmental Policy Act process.
The Council's IPQ plan effectively returns Alaska fishing
communities to the days before Statehood, when our State fishery
resources were under the control of a few large processing companies.
IPQs grant an exclusive right to process a specific percentage of the
crab harvest to a particular company, and the company may chose to
process all of their percentage at any facility that they own within
the region. In fact, the Council has gone out of its way to ensure that
processors have an unrestricted right to consolidate their IPQs at any
facility they own within the two regions--north and south--that are
created under the Council plan. Consolidation for processors is nothing
more than a euphemism for closing plants in remote Alaska coastal
communities.
The Council's own documents show the extent to which a few
processors will control Alaska's fisheries if this plan is enacted. In
1998 Congress enacted the American Fisheries Act (AFA), which created a
closed class of processors for the inshore delivery of Bering Sea
pollock. The AFA granted six companies the exclusive right to process
the entire 50 percent of the Bering Sea pollock that the AFA decreed
must be delivered inshore. No other processing companies are allowed to
process that pollock, regardless of their history in Alaska's other
fisheries. All six of those AFA companies also process crab. According
to the Council's own preliminary analysis, those six AFA processing
companies will be granted IPQs equal to approximately 75 percent of the
Bering Sea opilio and red King crab harvest. Perhaps coincidentally,
five of those AFA companies are also defendants in the ongoing Bristol
Bay salmon anti-trust lawsuit in Anchorage.
As many members of the Committee are aware, Kodiak and other
Alaskan coastal communities are in the midst of a long term crisis in
the salmon industry. IPQs will make this crisis even worse. The
combined effect of the closed class of pollock processors and IPQs will
virtually ensure that there is no solution to the salmon problem. One
of the AFA companies--Wards Cove--has used its guaranteed pollock
revenue to enable it to withdraw from the salmon processing market
entirely--they no longer need the marginal income from their salmon
operations. Other AFA companies have reduced their salmon buying in the
years since the AFA passed, despite the considerable increase in
revenues and financial stability provided by the AFA. Add the
guaranteed revenue from crab IPQs, and these companies have even less
incentive to expend money to gain marginal revenue from salmon.
Even more to the point, Nippon Suisan, an AFA company that will
receive IPQs equal to roughly 15 percent of the Bering Sea crab under
the Council plan, is also among the largest owners of Chilean salmon
farms. They clearly have no interest in seeing Alaska salmon compete in
the market place with their more expensive farmed product. How does
Congress expect to ever solve the salmon problem if they allocate
exclusive access and specific market share rights to the very same
companies that dominate the dying salmon industry. We need to attract
new and innovative processors in salmon. The Council plan just locks in
the same old companies that have created the mess we are in.
Perhaps the greatest irony is that if IPQs had been granted 30
years ago, two of the six AFA companies--Trident and Icicle--would
never have come into existence. Both of these U.S. processing companies
were formed by fishermen who were unhappy with the price being paid by
the existing processors. In fact, many of the fishing companies
operating in Kodiak and throughout Alaska were formed by fishermen who
wanted to get a better price for their fish. The Council proposal would
end that competitive market process for fish processing, because no new
entrant can hope to compete against companies like Trident, Nippon
Suisan, and Maruha when those companies have a closed class for pollock
and an exclusive right to process crab.
Despite the overlap in companies that benefit under the AFA and the
Council's proposal, it needs to be stressed that the benefits granted
to processors under the AFA and the IPQ plan are not the same. The fact
that Congress passed the AFA provides no basis for Congress to pass
IPQs. IPQs confer more benefits, and far greater market power, to crab
processors than pollock processors received under the AFA. Under the
AFA Congress created a closed class of inshore processing facilities
and linked the set aside of a coop's fish to an agreement to deliver 90
percent of a coop's harvest to a particular processing facility. But
Congress did not guarantee the owner of each facility in the closed
class a particular share of the inshore pollock harvest, nor did they
force the fishermen to join the coops if they chose not to. In
addition, each coop could deliver 10 percent of their collective
harvest to another inshore processing facility, which means that each
coop has the ability to let one or more boats transfer and take their
history to another processor if they so choose.
Under the Council's IPQ plan each processing company is guaranteed
a particular share of the crab harvest, which means that company can
consolidate processing facilities and has no risk of losing market
share if they don't compete on price and terms. Further, each fisherman
must deliver 90 percent of their catch to a company with IPQs, which
makes delivery of the remaining 10 percent to a different processing
company economically and practically infeasible. A more detailed
summary of four key differences between the AFA and the Council's IPQ
proposal prepared by Dr. Robert Halverson of the University of
Washington is also attached to my testimony.
What is particularly troubling about this proposal is that neither
the Council nor the State has ever provided any justification for the
proposal other than that certain processors demanded it. IPQs are not
used anywhere in the world. Dr. Scott Matulich of Washington State
University is the only academic who has supported the IPQ proposal. He
has promoted IPQs as the only way to ensure that all parties--
processors, fishermen, and communities--benefit from rationalization of
a fishery. The reality is that IPQs benefit only one party--the
processors--at the expense of fishermen and communities.
Independent reviews of Dr. Matulich's work and the Council proposal
have failed to find merit in the IPQ concept and have demonstrated the
harm they are likely to cause. The National Academy of Sciences
considered IPQs as part of their comprehensive review of individual
fishing quota plans and recommended the use of other methods--for
example allocating individual fishing quotas to processors, which the
Council plan does already--instead of IPQs. The Council's own
independent analysis by Drs. Milon and Hamilton of the University of
Central Florida found that crab IPQs were anti-competitive and would
have adverse impacts on fishermen. Independent reviews of Dr.
Matulich's work by the General Accounting Office and Dr. Robert
Halvorsen of the University of Washington have concluded that the
analysis done by Dr. Matulich for the State of Alaska used a flawed
methodology and cannot be validated. Yet the Council persists in
pushing IPQs as necessary for crab rationalization to occur.
Congress does not lightly create monopoly markets. When Congress
does, they create regulatory commissions to oversee those monopolies,
complete with investigative authority and detailed enforcement powers,
in order to protect the public from the adverse economic impacts of
those monopolies. None of those safeguards are included in the Council
proposal. The binding arbitration provision is enforced, if at all,
through private contract. The right of first refusal offered to protect
communities is ephemeral at best--the reality is that no community
could ever expect to meet the conditions the Council has set. There is
no appeal to a government agency if problems arise. The Council plan
makes no provision for oversight and enforcement by the Secretary of
Commerce or any other government agency. The Council has neither the
expertise nor the legal ability to oversee the monopoly market
structures they are asking Congress to create in the name of fishery
management. Antitrust remedies would simply not exist, because Congress
will have legislatively created these monopolies.
The simple truth is that IPQs are nothing more than economic
allocation of markets. They are a ransom that processors are demanding
before they will allow the crab fishery to be rationalized to improve
conservation and fishing safety. There is no fishery management or
community protection justification for IPQs. In fact, IPQs are
precisely what National Standard 5 of the Magnuson-Stevens Act
expressly prohibits they are a measure that has economic allocation as
its SOLE purpose.
Congress needs to realize that the push for IPQs will not stop with
crab. The Council is already considering plans to rationalize the Gulf
of Alaska groundfish fisheries, and some Gulf of Alaska fish processors
are demanding that IPQs be included. Further, a recent report on the
crisis in the salmon fisheries suggested that IPQs may be needed for
the salmon fisheries. Congress has already passed special legislation
for pollock, and is being asked again to pass special legislation for
crab. If Congress authorizes crab IPQs, the floodgates will be open and
Congress will be besieged with requests for special legislation for
each fishery off Alaska. Having Congress legislate fishery by fishery
is precisely what the Magnuson-Stevens Act was designed to prevent.
IPQs have dominated the Council's discussion of much needed
individual fishing quota proposals for far too long. Congress needs to
send a clear message that IPQs will not be allowed. Once that message
is delivered it will be easy for the Council to quickly implement an
individual fishing quota program under its existing authority that will
address the legitimate concerns of fishermen, communities, and
processors.
Thank you for the opportunity to present this testimony. I would be
happy to answer any questions at the hearing, and ask that the hearing
record be kept open for ten business days so that the City of Kodiak
may submit additional information in support of our testimony.
Attachments: Estimation of Losses to the Kodiak Community
Major Differences Between the AFA and the Council's IPQ Plan
______
Attachment
Estimation of Losses to the Kodiak Community from Bristol Bay Red King
Crab Rationalization
During the past several years a number of Kodiak crab fishermen
have delivered their Bristol Bay red king crab to Kodiak rather than a
port closer to the grounds. This is due to a number of factors
including investments some have made in a local processor, responses to
higher prices offered in Kodiak and relationships with local
processors.
Under the NPFMC proposed crab rationalization program, crab
processors will be granted a processing quota based on their activity
during 1998-9. Harvesters will be granted corresponding quota of which
90 percent (``A'' shares) must be delivered to processors holding
processing quota and the remaining 10 percent (``B'' shares) may be
delivered to any processor. Analysis of the rationalization program
suggests that the ``B'' quota deliverable to any processor will
initially equal about 12.9 percent of each independent (i.e., non-
processor controlled) harvester's allocation.
The impact of the reduction in deliveries to Kodiak can be
estimated based on information provided by NPFMC. The following table
lists the landings and value to Kodiak, overall landings, and the
estimated impacts to the Kodiak community based on first sale of
processed products.
Based on Based on
Year 2000 2001 1999 1990
Kodiak BBRKC 824,780 817,916
deliveries (lb)
Overall BBRKC 7,546,145 7,786,420 11,000,000 20,000,000
harvest (lb)
Kodiak percentage 10.93% 10.50% 10.72% 10.72%
3.8% of total 258,078 266,296 376,200 684,000
harvest ``A'' share
harvest
3.8% plus 12.9 291,370 300,648 424,730 772,236
percent of ``A''
shares (IQ)
Difference between 533,410 517,268 754,152 1,371,185
actual (projected)
and IQ
Ex-vessel prices $4.80 $4.92 $4.86 $4.86
Converted to first $5.48 $5.60 $5.54 $5.54
wholesale product
value
Difference in value $2,921,379 $2,895,047 $4,175,587 $7,591,977
processed products
Community weighted 1.232 1.232 1.232 1.232
average multiplier
Foregone community $3,599,138 $3,566,698 $5,144,324 $9,353,316
multiplied impact
Total Kodiak 94.5 80.5 87.86 88.82
deliveries ($
million) all
species
Difference % of 2.709% 3.161% 4.172% 7.503%
total deliveries
Total fisheries $1,287,344 $796,393 $1,041,869 $1,041,869
taxes
Fisheries taxes on $34,879 $25,177 $43,464 $78,167
difference
Total Effect $3,634,017 $3,591,876 $5,187,788 $9,431,483
(Wholesale +
Multiplier + Taxes)
In 2000, 824,780 lbs of Bristol Bay red king crab were delivered to
Kodiak from a total harvest of 7,546,145 lbs. This represents 10.93
percent of the total harvest. Kodiak processors will receive only 3.8
percent of the processor IQs (which are 90 percent of the total
harvest) which would have translated into 258,078 lbs of deliveries in
2000. In addition, harvesters making those deliveries would have
received 12.9 percent of that amount in non-processor specific IFQs. If
this amount is also assumed to be delivered to Kodiak a total of
291,370 lbs would have been delivered in 2000--a difference of 533,410
lbs less than actual deliveries. The fishery average ex-vessel price
was $4.80 per pound in 2000. The round weight equivalent average
difference between ex-vessel and wholesale product value is $0.6867 per
lb for a wholesale price of $5.48/lb. This results in potential product
loss of $2,921,379 to processors. The best available economic
multiplier for Kodiak is 1.232 or a local increase of 23.2 percent for
every new dollar into the community. Total deliveries to Kodiak in 2000
from all fisheries were valued at $94.5 million. The proposed loss in
BB red king crab deliveries would amount to 2.709 percent of the total.
Kodiak collected $1,287,344 in fisheries landing and business taxes for
2000 landings. The straight percentage of that simply allocated to the
crab difference is a loss of $34,879. This calculated total loss to the
community of Kodiak from regulatorily prohibited Bristol Bay red king
crab is $3,634,017. For 2001 the same calculation estimates a loss due
to processor restrictions of $3,591,876.
The same calculation can be used to estimate loss to the Kodiak
community when crab harvest levels increase from current low levels. In
1999, 11,090,930 lbs of Bristol Bay red king crab were harvested
overall and in 1990 20,362,342 pounds were harvested. Rounding these
numbers to the nearest million pounds and projecting the 2000/2001
averages for other variables (Kodiak percentage of landings, ex-vessel
price, conversion to wholesale price, total Kodiak deliveries, and
total taxes modified only by the increase in crab landings) allows
derivation of losses to Kodiak. As the amount of crab harvest increases
its relative importance to the Kodiak economy likewise increases. For
instance, recent delivery percentages applied to 1990 crab harvest
levels suggest that Bristol Bay red king crab would contribute over 7.5
percent of the total value of deliveries of all species to Kodiak.
Restrictive processor quotas applied to the 1990 harvest would cause an
annual loss of $9.4 million to the Kodiak community.
Background
When product is prohibited from delivery to a community the result
is an economic loss not only to the persons buying the product for
further processing but also to the community as a whole. Since the
local industry is not able to manufacture processed product it is not
able to employ workers or to purchase other local goods and services
required in the manufacturing process. In turn, there are lower wages
and profits throughout the community due to economic multiplier
effects. This analysis derives these cumulative impacts based on
readily available information. Since the restrictions on trade due to
crab processor quotas can only be estimated, and since market trends
have shown increased economic crab processing activity in Kodiak in
recent years, these estimates are viewed as conservative and at the
lower end of probable impacts.
Estimating the economic impact to the Kodiak economy from reduced
king crab deliveries must be based on processed product value.
Processed product value is derivable from deliveries and comparison
between ex-vessel and processed product prices.
Fishery purchases by Kodiak processors have decreased dramatically,
both in terms of volume and value, during the past few years. This is
related to changing ocean conditions, changing world markets, and
regulatory changes designed to protect Seller sea lions.
Recent deliveries to Kodiak
Accurate data for 2002 deliveries are not yet available from A.
However, the change between 1997 and 2001 is illustrative. Both overall
landings at Kodiak and ex-vessel value of all species combined are
similar between 1997 and 2001. However, the relative importance of
species, such as Bristol Bay red king crab, have changed. Red king crab
changed from 0.1 percent of overall landings and 1.5 percent of overall
value in 1998 to 0.3 percent of overall landings and 4.8 percent of
overall value in 2001.
Ex-Vessel Landings of Seafood at Kodiak, 1997-2001
[Millions of Pounds]
------------------------------------------------------------------------
Species 1997 1998 1999 2000 2001
------------------------------------------------------------------------
Bristol Bay King Crab 0.4 0.3 0.5 0.9 0.8
(0.1%) (0.1%) (0.2%) (0.3%) (0.3%)
------------------------------------------------------------------------
Total Crab 1.1 1.2 1.4 2.7 1.4
------------------------------------------------------------------------
Pollock 83.3 165.8 130.5 102.2 90.8
------------------------------------------------------------------------
Pacific Cod 73.1 72.0 85.0 64.9 54.7
------------------------------------------------------------------------
Total Groundfish 184.2 263.4 237.6 200.4 176.3
------------------------------------------------------------------------
Salmon 57.8 105.6 70.5 61.8 78.8
------------------------------------------------------------------------
Halibut 11.0 9.1 9.9 9.3 8.5
------------------------------------------------------------------------
Sablefish 3.9 3.6 3.2 3.4 2.2
------------------------------------------------------------------------
Other Species 8.7 5.7 4.1 3.3 3.3
------------------------------------------------------------------------
Total All Species 267.0 388.6 326.7 281.0 270.5
------------------------------------------------------------------------
Source: McDowell Group. 2002. Analysis of Economic Impacts From Fishing
Restrictions on the Kodiak Island Borough Economy. Prepared for Kodiak
Island Borough. November 2002. Table 11.
Ex-Vessel Landings of Seafood at Kodiak, 1997-2001
[Millions of Dollars]
------------------------------------------------------------------------
Species 1997 1998 1999 2000 2001
------------------------------------------------------------------------
Bristol Bay King Crab $1.3 $1.2 $1.7 $1.7 $3.9
(1.6%) (1.5%) (1.6%) (1.8%) (4.8%)
------------------------------------------------------------------------
Total Crab 3.1 2.0 2.8 3.4 4.9
------------------------------------------------------------------------
Pollock 8.1 11.6 13.1 8.7 12.7
------------------------------------------------------------------------
Pacific Cod 15.5 13.7 25.5 24.0 15.9
------------------------------------------------------------------------
Total Groundfish 27.8 28.6 41.2 36.8 32.5
------------------------------------------------------------------------
Salmon 18.8 29.8 31.1 21.5 18.8
------------------------------------------------------------------------
Halibut 21.0 10.0 20.6 23.1 16.2
------------------------------------------------------------------------
Sablefish 8.0 5.2 5.7 7.0 6.9
------------------------------------------------------------------------
Other Species 4.2 3.7 2.7 2.7 1.2
------------------------------------------------------------------------
Total All Species $82.9 $79.3 $103.9 $94.5 $80.5
------------------------------------------------------------------------
Source: McDowell Group. 2002. Analysis of Economic Impacts From Fishing
Restrictions on the Kodiak Island Borough Economy. Prepared for Kodiak
Island Borough. November 2002. Table 12.
Total Bristol Bay red king crab harvests and Kodiak deliveries for
2000 and 2001 are available from ADF&G.\1\ In 2000, 10.93 percent of
the Bristol Bay red king crab were delivered to Kodiak and in 2001,
10.5 percent were delivered there. Kodiak processors are expected to be
allocated slightly less than 3.8 percent of the Bristol Bay red king
crab IPQ.\2\
---------------------------------------------------------------------------
\1\ Fishery Information Packet for the Bristol Bay Red King Crab
Fishery, 2002. ADF&G RIR 4K02-45. Kodiak data--NMFS data as presented
at http://www.kodiak.org/seafood.html.
\2\ Communication from NPFMC staff Dr. Mark Fina. Includes one
additional processor for confidentiality reasons.
---------------------------------------------------------------------------
Deliveries to Kodiak under crab rationalization
Thirty-one red king crab harvest vessels are reported to be
affiliated with processors.\3\ This represents 12.1 percent of the
total 256 harvest vessels which qualify.\4\ The cumulative quota due
these 31 vessels amounts to 12.6 percent of the red king crab IFQ.
Using the data from the analysis, a maximum of 225 vessels will be
eligible to receive the ``B'' shares. This number is likely to be lower
since there continues to be consolidation in the harvesting fleet. This
consolidation will only become greater under the proposed buyout
program. Likewise, ``control'', when defined, will be more restrictive
than the 10 percent ownership ``affiliation'' criteria used in the
NPFMC analysis.
---------------------------------------------------------------------------
\3\ North Pacific Fishery Management Council (NPFMC). 2002. Bering
Sea Crab Rationalization Program Alternatives: Public Review Draft.
May, 2002. Table 3.4-27.
\4\ NPFMC, 2002. Table A3-32.
---------------------------------------------------------------------------
Supposing for this analysis that the percentages listed above are
correct, the independent harvesters will control86.4 percent of the
``A'' shares (77.76 percent of all IFQs) and all 10 percent of the
``B'' shares. Therefore, each independent harvester will receive ``B''
shares in an amount equal to 12.9 percent of their ``A'' shares (rather
than 11.1 percent if all harvesters received ``B'' shares). Conversely,
these ``B'' shares will equal 11.6 percent of their total annual
harvest allocation (rather than 10 percent).
In order to simplify the analysis of probable loss to Kodiak, it is
assumed that the full3.8 percent of the overall ``A'' share harvest
would be delivered to Kodiak along with the ``B'' share deliveries
associated with that IQ (an additional 12.9 percent of the ``A''
shares). This equates to 3.86 percent of the overall harvest. The
difference between deliveries and the percentage that would be
allocated to Kodiak processors in 2000 and 2001 is calculated at
533,410 lbs and 517,268 lbs, respectively.
Wholesale processed product value
In order to calculate the economic impacts of reductions in
processed crab to the Kodiak economy, it is first necessary to
determine the value of crab and how that value relates to ex vessel
value. The following table derives the difference of crab value between
first sale of wholesale product (sections) and ex-vessel value.
Shellfish sections represent 98 percent or more of the processed
product type for red king crab from 1996-2000.\5\ Therefore, the first
wholesale crab prices for shellfish sections is used as a proxy for all
first wholesale prices. In order to convert wholesale prices into
equivalent delivered prices it is necessary to multiply by the product
recovery rate. In this case, a product recovery rate of 60 percent is
used. \6\ Comparing these back cast, delivered wholesale prices to
actual ex-vessel prices allows a difference or processor markup to be
determined. This difference varies year to year but an average of the
five most recent years available is $0.6768 per pound delivered weight.
---------------------------------------------------------------------------
\5\ NPFMC, 2002. Appendix 2-3, Table 2.
\6\ Chuck Crapo, Brian Paust, and Jerry Babbitt. 1988. Recoveries
and Yields from Pacific Fish and Shellfish. Alaska Sea Grant, Marine
Advisory Bulletin No. 37. King crab, raw-whole, converted to cooked
sections.
Wholesale 1st Backcast
Sale (Sections) Delivered (Whole) Actual Weighted Difference $/lb
$/lb $/lb Ex-Vessel $/lb
1996 $ 8.53 $5.12 $4.01 $1.11
1997 $ 6.15 $3.69 $3.26 $0.43
1998 $ 5.52 $3.31 $2.61 $0.70
1999 $11.25 $6.75 $6.26 $0.49
2000 $ 9.11 $5.47 $4.81 $0.66
Average $0.6768
Sources: First wholesale crab price--NPFMC, 2002, Appendix 2-3, Table 1. Conversion ratio of 60 percent for
whole fresh crab to sections--Crapo, 1988. Ex-vessel average price--NPFMC, 2002, Table 2.5-1.
The reported average ex-vessel prices for BB king crab in 2000 and
2001 are $4.80 and $4.92, respectively.\7\ Converting these to
processed sections and back calculating a wholesale price results in
estimates of $5.48 per delivered pound at the wholesale level for 2000
and $5.60 per pound for 2001. Based on these estimated processed
product values, the foregone value at the first wholesale level would
have been $2,921,379 in 2000 and $2,895,047 in 2001.
---------------------------------------------------------------------------
\7\ ADF&G data by year http://www.cf.adfg.state.ak.us/geninfo/
shellfsh/shelhome.htm#catch
---------------------------------------------------------------------------
Economic multiplier impacts
Gross revenue received from the sale of processed crab at the first
wholesale level, is termed a direct effect (output value). In addition
to these direct effects of revenue, a community also incurs indirect
and induced effects from the sale of crab. ``Indirect effects are
changes that occur because producers change the amounts of goods and
services they purchase, such as raw fish, fuel, utilities and packaging
supplies. Induced effects are generated as income from direct and
indirect expenditures work their way through the economy. An example of
an induced effect would be a reduction in restaurant sales as
processing plant workers choose to cut back on personal expenditures. .
. . [I]t is assumed that there is an overall U.S. multiplier of 2.0 for
all direct effects--with indirect and induced effects that accrue
either inside the local economy or leak out to non-local economies. In
other words, it is assumed that for every dollar of direct output
effect, there is an additional dollar of indirect and induced effects
that is generated somewhere in the U.S. economy.'' \8\
---------------------------------------------------------------------------
\8\ Northern Economics. 2001. Assessment of Economic Impacts of
Federal Actions to Protect Steller Sea Lions on Alaska Groundfish
Fishery Participants. Prepared for Southwest Alaska Municipal
Conference. August 2001. Section 8.4.
---------------------------------------------------------------------------
The most recent estimate of community wide economic impacts for
Kodiak Island are a weighted average aggregate multiplier of 1.232.
This compares to similarly derived multipliers of 1.185 for the
Aleutians West Census area and 1.085 for the Aleutians East Borough.\9\
Applying this economic multiplier to the potential value of foregone
wholesale processed product results in a cumulative impact of
$3,599,379 for 2000 and $3,566,698 for 2001.
---------------------------------------------------------------------------
\9\ Estimated using IMPLAN impact analysis software by Northern
Economics, 2001.
---------------------------------------------------------------------------
Local fisheries taxes
Taxes are derived directly from the fishing industry both via a
shared landing tax and a fisheries business tax. A simplified method is
used to determine the amount related to a difference in crab landings.
Total landing and fisheries business taxes for Kodiak are reported as
$1,287,344 in 2000 and $796,393 in 2001.\10\ Total landings for all
seafood products in Kodiak during the same two years were valued at
$94.5 million and $80.5 million, respectively.\11\ The potential loss
in ex-vessel value is calculated at 2.709 percent of the overall
landings for 2001 or $34,879. The same calculation for 2001 results in
a potential loss of $25,177.
---------------------------------------------------------------------------
\10\ McDowell Group. 2002. Analysis of Economic Impacts From
Fishing Restrictions on the Kodiak Island Borough Economy. Prepared for
Kodiak Island Borough. November 2002. Table 17.
\11\ McDowell Group, 2002. Table 17.
---------------------------------------------------------------------------
There are other taxes and municipal income associated with crab
processing and with indirect and induced economic activity that are not
accounted for in this analysis. For instance, reduced processing and
the concomitant reductions in purchases and spending result in lower
overall sales tax revenues.\12\
---------------------------------------------------------------------------
\12\ The City of Kodiak currently has a 6 percent sales tax.
---------------------------------------------------------------------------
Cumulative potential impacts to Kodiak, 2000 and 2001
The estimated total effects of prohibiting crab deliveries to
Kodiak processors includes, at a minimum, the loss of wholesale
product, multiplied community impacts, and foregone taxes. The
estimated potential minimum impacts for 2000 and 2001 are thereby
calculated at $3,634,017 for 2000 and $3,591,876 for 2001.
Cumulative impact to Kodiak based on historic crab harvest levels
The Bristol Bay red king crab harvests in 2000 and 2001 were at the
low end of recent harvests. There was no fishery at all during 1983,
1994 and 1995. The only years during the past three decades when the
harvest was lower than 2000/2001 was in 1982, 1984, 1985, and 1988.
During the past three decades, harvests have ranged from 4.2 million
pounds to 130 million pounds.\13\ The harvests from 1990 through 2002
have averaged 11.6 million pounds.
---------------------------------------------------------------------------
\13\ Harvests of 3 million pounds in 1982, 4.2 million pounds in
1984 and 1985, and 7.4 million pounds in 1988. Fishery Information
Packet for the Bristol Bay Red King Crab Fishery, 2002. Table 1.
---------------------------------------------------------------------------
In order to arrive at a more realistic expectation of economic
losses to the Kodiak community under crab rationalization it is
illustrative to analyze recent delivery rates compared to crab
rationalization restrictions using examples from the last ten years of
harvest. In order to make the analysis within current trends, the
projection uses several simplifications. Overall Bristol Bay red king
crab harvests are rounded off to 11 million and 20 million pounds as
proxies for actual harvests in 1999 (11,090,930 pounds) and 1990
(20,362,342 pounds). The average harvest delivery percentage to Kodiak
from 2000 and 2001, 10.72 percent, is used to compare to restricted
deliveries under crab rationalization. The average ex-vessel value from
2000/2001 of $4.86 is used to derive a first wholesale back calculated
price of $5.54 per delivered pound. Likewise, total deliveries to
Kodiak are assumed to be the same as average in 2000/2001 except for
the increase in deliveries that would be made to Kodiak due to the
increased crab harvest. Total taxes are under-estimated at the average
of 2000 and 2001.
The analysis projects that an 11 million pound crab harvest with
processor quotas would result in a loss to the community of Kodiak of
$5,187,788 million. A harvest of 20 million pounds, based on that
experienced in 1990, would result in a loss to Kodiak of $9,431,483
million. The losses would represent 4.172 percent and 7.503 percent of
overall value of landings to the community, respectively.
These calculations are, at best, conservative estimates. They
derive expected losses based on actual deliveries to Kodiak during 2000
and 2001 compared to projected deliveries. However, the expected
deliveries are likely to be even less than used in this analysis and
therefore the impacts greater. For example, processors are free to
transfer processing quota between processing plants. It is entirely
possible that processors with larger plants in Unalaska/Dutch Harbor
would direct deliveries to those plants at the expense of deliveries to
Kodiak. Likewise, vessels that are ``controlled'' by processors, such
as several fishermen who have investments in both crab harvesters and a
processor at Kodiak, would not receive ``B'' harvest shares and
therefore crab deliveries to Kodiak would decrease accordingly.
______
Major Differences between AFA Inshore Cooperatives and the Council's
Proposed Rationalization Program for the BSAI Crab Fisheries
Robert Halvorsen, Professor of Economics, Department of Economics,
University of Washington Seattle, Washington
I. Protection of Processors' Market Shares
Under the proposed rationalization program for the crab fisheries,
a processor would receive processor quota share equal to 90 percent of
its historic processing share. This amount would be guaranteed to the
processor because harvesters could deliver their Class A allocation
only to processors holding processing share. Therefore the only way
that a processor could lose more than 10 percent of its historic market
share would be if it set the ex vessel price so low that vessels would
prefer to forego fishing rather than deliver fish to it, because that
would be their only alternative.
Under the AFA, each processor has the right to process 90 percent
of its cooperative's total harvest, but this does not guarantee that it
will receive 90 percent of its historic processing share, because
vessels have alternatives to remaining in the cooperative. One
alternative for a vessel is to fish in the open access portion of the
fishery and deliver its fish to another processor. It could then either
remain in open access or join the cooperative of that processor.
Another option is to qualify for another cooperative without going
through open access by delivering its fish to the alternative processor
as part of the 10 percent of the cooperative's total harvest that can
be delivered to any processor. And if enough vessels defected, the
cooperative itself might be dissolved, eliminating any processing
rights under the AFA.
Therefore protection of processors' market shares would be much
greater under the proposed program for the crab fisheries than under
the AFA, thereby giving the processors much greater bargaining power.
In particular, under the AFA a processor will retain market share only
if it offers an ex vessel price that is competitive with the price
being offered by other processors, whereas under processor quota shares
a processor could retain 90 percent of its market share even if it
offered a price just slightly higher than the cost of catching fish.
II. Regionalization
Under the proposed rationalization program for the crab fisheries,
Class A harvest shares and processor shares for each crab fishery would
be regionally designated, whereas under the AFA the entire inshore
sector is treated as a single region. This difference has important
implications both for the net economic benefits that can be realized
from rationalization and for the distributional consequences of
rationalization.
The Council itself recognizes that regionalization reduces net
economic benefits by restricting consolidation of activities that are
desirable for reducing capacity and gaining efficiency in both the
harvesting and processing sectors under rationalization (Report to
Congress August 2002, page 18). The lack of such constraints under the
AFA increased the total net economic benefits that were available to be
shared by harvesters and processors.
Regionalization also has implications for the distributional
consequences of rationalization because it subdivides the markets for
crab and thereby increases the already high degree of concentration
among processors. It also creates an incentive for processors to
consolidate their market shares on a regional basis, which would
increase the degree of concentration still more. The greater bargaining
power attained by processors can be expected to adversely affect the
price received by harvesters for Class B allocations as well as for
Class A allocations, both because it might be difficult logistically to
deliver to different markets and because processors might be able to
require bundling deliveries of the two classes of fish.
III. Complexity
The ``three-pie voluntary cooperative program'' being recommended
for the crab fisheries is much more complex than the rationalization
program implemented under the AFA. The greater complexity can be
expected to have serious negative consequences both with respect to the
cost of management and with respect to the functioning of the market
for fish and for quota shares.
Implementation of the proposed rationalization program for the crab
fisheries would require the determination of share allocations in each
region of each fishery for each individual vessel and processor.
Ongoing management measures would include annual monitoring and
enforcement measures at the same level of detail. Eventual formation of
voluntary cooperatives might reduce some of the management costs with
respect to harvesting, but the extra costs of managing processing
activities would continue.
More importantly, the increased complexity of the system might make
the determination of prices through a decentralized market structure
impracticable. For each regional market in each fishery the prices that
would have to be determined include the ex vessel price of Class A
fish, the ex vessel price of Class B fish, the price of Class A
harvesting quota, the price of Class B harvesting quota, and the price
of processing quota.
Attaining equilibrium prices in such a complex system would be
difficult in even in large, well-functioning, markets, and the markets
in the crab fisheries would be both thin and imperfectly competitive.
In addition, the large fluctuations in total allowable catch would
complicate the determination of equilibrium prices and hinder the
ability of the system to converge to stable values. In recognition of
the possibility of the price system breaking down, the rationalization
plan includes a binding arbitration program. However, the necessity of
such a procedure increases the cost of managing the fisheries under the
proposed rationalization plan, and even a well-designed arbitration
procedure is not an adequate substitute for a well-functioning market.
IV. Net Benefits from Rationalization
Rationalization of the pollock fishery under the AFA created large
net economic benefits for the inshore sector, which made it feasible
for both the harvesting and processing sectors to benefit from the
program. As already noted, the regionalization requirement under the
proposed plan for the crab fisheries would decrease the potential net
economic benefits to be obtained by rationalization. But even if this
provision did not exist, the total net economic benefits of
rationalization in the crab fisheries could not be expected to be as
large as they were under the AFA, because participants in the inshore
pollock fishery benefited both from a large increase in the sector's
total allocation and from large rationalization benefits from the
formation of cooperatives.
The sector's total allocation was increased first by an increase in
its share of the total directed pollock fishery from 35 percent to 50
percent, and subsequently by an increase in the total allowable catch
for the pollock fishery. The combined result was that the inshore
sector's total allocation has increased by 80 percent from the pre-AFA
level in 1998 to the present.
Large efficiency benefits were realized from the formation of the
AFA cooperatives. Rationalization under the AFA permitted improved
targeting of pollock during the peak roe season, resulting in greatly
increased ex vessel prices during this season. The value of output was
also increased because slowing the race for fish permitted an increase
in the recovery rate and a shift to higher valued products. Harvesting
costs have been reduced by the transfer of quota shares from less
efficient vessels to more efficient vessels, and easy transferability
of allocation within a cooperative has facilitated the full harvesting
of the available allocations.
In contrast, the proposed rationalization program for the crab
fisheries does not include an increase in the total allocations
available to these fisheries. It does incorporate a buyback program,
but the efficacy of the buyback program has yet to be determined, and
in any case could not result in benefits equivalent to the 80 percent
increase in total allocation experienced by the inshore pollock
fishery. Similarly, increases in the value of output due to
rationalization are not anticipated to be as large for the crab
fisheries, and increases in harvesting efficiency are likely to be
hindered by the restrictions imposed by the proposed program.
Senator Stevens. Yes, we can keep the record open for 10
days. We are going on recess on Friday, so that would be--we
should make that so it would be a week from the coming--not the
coming Monday. I forget the date that will be, but that is the
date we will come back into session. That would be the date we
will close the record.
Ms. Freed. Thank you.
Senator Stevens. The second of June.
Ms. Freed. Thank you.
Senator Stevens. Thank you very much.
Our next witness is Dave Fraser, Captain of the fishing
vessel Muir Milach.
STATEMENT OF DAVE FRASER, CAPTAIN,
FISHING VESSEL MUIR MILACH
Mr. Fraser. Thank you, Mr. Chairman, Members of the
Committee. I appreciate the opportunity to testify on behalf of
the Crab Rationalization and Buyback Group.
The Crab Group represents over 100 crab vessels, and the
Crab Group strongly supports rationalization of the Bering Sea
and Aleutian Island crab fisheries. We agree that the status
quo system has left harvesters on the brink of bankruptcy, it
has led to the demise of most of the non-AFA crab processors
already, and it is also, and worst of all, been a murderous
management system. We lose, on average, five fishermen a year.
And that is clearly unacceptable, and we need to move forward
into a rationalized fishery.
That said, Mr. Chairman, the Council has recommended to you
two distinct programs, one of which is a crab-management
program that manages the harvest of crab. It includes skipper
and crew provisions. It includes CDQ provisions,
regionalization, a number of other protections for processors
inherent in the program. But separate from that, they have
recommended to you a different program. It is one to manage
markets, manage the marketplace for crab and the sale of crab.
And that is what requires congressional authorization. That is
the program to which the Crab Group objects.
Mr. Chairman, after the June meeting, I went home and was
explaining to my neighbor, who is a plumber, what the Council
had recommended, and just started out at the very basics and
explained that they were recommending a program that divided up
the right to buy crab from fishermen and assign each processor,
or actually a small handful of processors, specific amounts of
crab they could buy. He interrupted me after about 30 seconds
and said, ``Well, you are going to get paid less for your crab
that way, are not you?'' Now, neither Harold, my neighbor, nor
I are economists, but I think he is right.
In 1996, Mr. Chairman, you, in reauthorizing the Magnuson-
Stevens Act, asked for advice, and you did not ask for it from
a plumber or me; you asked for it from the National Academy of
Sciences. The National Academy of Sciences spent a long time
preparing its advice, and they looked specifically, on the
basis of your direction, at crab processing quotas, or at
processing quotas. And they said, quote, ``We find no
compelling reason for establishing a separate processor quota
system.''
Mr. Chairman, the Council, at your direction, was also, as
you pointed out at the beginning of this session, asked to
analyze the impacts of processor quotas in looking at crab
rationalization. Now, the Council staff new a hot potato when
they were handed one, and they contracted out that portion of
the crab rationalization analysis to two independent economists
from Florida. Those economists said, quote, under a processor
quota system, ``processors will do better and harvesters will
do worse as the ratio of A to B shares increases.''
Now, that did not go over real well, and the Council took
that out of the analysis and put it in an appendix, but that
was the only analysis done on the impact of processor quotas;
and the rest of the document was about four pages out of
several hundred pages devoted to the issue of processor-quota
impacts.
Mr. Chairman, when the Council got to June and made its
recommendation, it did not have any analysis of the impacts of
processor quota before it, and it did not make any motions to
consider any other alternatives, such as 50-50 or 70-30 or
anything other than 90-10 split on processing quotas. The
argument--there was no debate over this, but the argument
essentially boils down to, ``It is our way or no way.''
Processor quotas must be included or processors will block
rationalization of the crab fishery and these lives will be
held hostage.
What the economists that were contracted by the Council
described as ``monopsony ex-vessel pricing,'' I described as a
game of musical chairs. If you, as a fisherman, are unhappy in
the situation you are in with a processor and want to move to a
different processor, under an IPQ system you have got to move
somebody else out of their chair, and they have to agree to go
to someplace else. The only other open chair is the one you
left that you found intolerable.
Now, Commissioner Duffy has suggested that 10 percent B
shares, open-market shares, are enough to correct the problem,
and he has also equated that, and others have equated that, to
the 10 percent rule in the AFA. I think it is absolutely
critical that you realize the essential difference between the
AFA, which is an excellent program, and the IPQ program for
crab, which is not. And that is the under the AFA, you do have
stability. Processors have a planning horizon. They contract
with members of their coop on an annual basis. But it is not
musical chairs. If a harvester chooses to move to another
processor by going through a year of open access, he can do so
and take his chair with him. No processor under the AFA is
guaranteed any fixed amount of quota from year to year. And
that is what keeps the marketplace honest. It is the element of
competition.
Mr. Chairman, the Council did recognize the impacts of
processor quota and what it would do to a competitive
marketplace. And as a result, in recognizing that, they
initiated a couple of trailing amendments. Those were to deal
with community protections and the binding arbitration program.
As Ms. Freed has suggested, the community-protections element
ended up being--ephemeral, I think that was the word she used,
or a sham. Likewise, the binding-arbitration provisions were
the only basis on which some harvesters were willing to swallow
the idea of processing quota.
The Council set up a committee. It worked for several
months, came back with two models, one of which was supported
by processors, the other by harvesters. And the bottom line,
Mr. Chairman, is that the Council, on a 6-5 vote, chose the one
that the processors favored and rejected the one that the
harvesters favored. And after that meeting, every single group
of crab harvesters has expressed the fact that the binding
arbitration model chosen by the Council will not provide a
meaningful safety net or the assurance of an outside option
that equates to a competitive marketplace.
Now, Mr. Chairman, Arni will testify later. They believe
that this can be corrected, and will be corrected, but they are
the only group that is taking that position at this point in
time.
So binding arbitration has not fixed the problem of
destroying a competitive marketplace, and it becomes a tar
baby. Arni will be asking for continual oversight by Council
and Congress of the binding-arbitration process. Unless you
write them a blank check, that means the Council has to come
back to Congress for more authority if they want a different
kind of binding arbitration than the one currently proposed.
If Congress authorizes IPQs and gives broad authorization
for changes in binding arbitration, it will transform the
Council's task from managing fisheries to managing markets, and
they will need to invest the Council with the powers and
experiences of the National Labor Relations Board, the Federal
Trade Commission, Department of Justice Antitrust Division, and
Public Utility Regulatory Commissions.
It is a tar baby, Mr. Chairman, and I think it is a path
Congress should think very seriously about going down, because
the changes that will occur while attempting to fix the impacts
from a flawed system are going to be irreversible changes. You
cannot put Humpty-Dumpty back together again.
Mr. Chairman, I suggested at the beginning that the
Council's program, absent processing quota shares, already does
include elements to protect processors. We do not disagree with
the need for protection of processors under a rationalized
fishery, but some of the elements that are already embedded in
the program for processors are separate classes of quota for
catcher vessels and catcher processors so that catcher vessels
cannot start processing their own catch; they have to use
existing processors. There is regional restrictions on
deliveries. Processors are allowed to acquire and own harvest
quota; and, in fact, a processing company is allowed to hold
five times as much harvest quota as any individual harvesting
company is allowed to hold. That limitation for harvesters of
only 1 percent, versus the processors' 5 percent, means that
there will be a diverse supply on the harvest side for
processors.
Mr. Chairman, I think that the AFA provides a workable
alternative to PQs, but the AFA is not equivalent to processor
quotas. Congress should reject the Council's request for a
radical expansion of its duties beyond fisheries management and
into market regulation represented by the request for
authorization to adopt the processor-quota element. If Congress
does, in fact, authorize processor quotas, it is critical to do
so in a manner that guarantees improved community protections
and a better binding-arbitration system, and that those
elements are done first.
Thank you, Mr. Chairman.
[The prepared statement of Mr. Fraser follows:]
Prepared Statement of David Fraser, Captain, FV Muir Milach,
Crab Rationalization and Buyback Group
Executive Summary
1.0 What Happened in Dutch Harbor and Why the 11-0 Vote?
2.0 Are Processor Quotas Necessary or Prudent?
2.1 National Academy of Science Recommendations
2.1.1 ``Sharing the Fish'' on Processor Quota
2.1.2 Distribution of Benefits of Quota Shares--Initial
Allocation
2.1.3 Distribution of Benefits and Consolidation
2.1.4 Foreign Ownership
2.2 Economists' Views on Processor Quotas
2.2.1 The GAO On Matulich
2.2.2 Economists on Processor Quota--Milon and Hamilton
2.2.3 Economists on Processor Quota--Halverson
2.2.4 Economists on Processor Quota--Christy and Anderson
3.0 Do We Believe in the Value of a Competitive Marketplace?
3.1 Price Formation Under Status Quo versus Under Processor Quota
3.1.1 Price Formation Under Status Quo
3.1.2 Price Formation Under Processor Quota
3.2 The Nature of the Right or Privilege represented by the
Processing Quota
4.0 Making a Square Wheel Round--Fixing the Impacts of PQs
4.1 Will 10% B Shares Serve Their Intended Purpose?
4.1.1 A Simplified Answer
4.1.2 The Missing Analysis
4.2 Binding Arbitration--A Substitute for Competition?
4.2.1 Need for Arbitration
4.2.2 Contrasting the Alternatives
4.2.2.1 What's In a Name?--Fleetwide Arbitration
4.2.2.2 What's In a Name?--Last Best Offer Arbitration
4.2.2.4 Dr. Plott's Analysis
4.2.2.5 Information available to the arbitrator(s)
4.2.2.6 Arbitration timelines
4.2.2.7 Arbitration administration
4.2.3 Quality issues
4.2.4 Data Verification
4.2.5 The Council's Revised LBO Model
4.3 Community Protections
4.4 Government in the Marketplace
5.0 Alternatives to Processor Quotas to Protect Processors
5.1 Existing Programs
5.1.1 AFA Pollock in the Shoreside Sector
5.1.2 Halibut & Sablefish IFQs
5.1.3 British Columbia's IVQ Groundfish
5.1.4 Newfoundland Crab
5.2 Alternatives discussed in the NPFMC's Advisory Panel
6.0 Public Comment--Who Supports Processor Quotas
7.0 Where Angels Fear to Tread
______
Executive Summary
The CRAB Group supports rationalizing the Bering Sea/Aleutians
Islands crab fisheries. The status quo system has left harvesters on or
over the brink of bankruptcy, and has led to the demise of most of the
non-AFA (American Fisheries Act) crab processors. Worst of all, the
murderous ``race for fish'' in the middle of winter kills an average of
five fishermen a year. We are also aware that asking Congress to
overrule a unanimous Council request puts us in a difficult position.
However, we believe a fair rationalization program can be developed
without resorting to Processor Quotas (PQs). Processor quotas will
result in a segmented monopsony and require endless government
regulatory involvement to simulate a competitive market.
It is the position of the CRAB Group and hundreds of others that
Congress should reject the Council's request for a radical expansion of
its authority.
The harvester component of the crab rationalization program is
legal now. It can be adopted together with the CDQ provisions, the
regionalization provisions, and the skipper and crew provisions
designed by the Council, but without PQs.
IFQs allocate access to a share of a public resource, which becomes
private property only after it is captured. PQs grant a right or
privilege to process a fixed portion of the harvest. Thus PQs direct
the disposition of private property, rather than a public resource.
IFQs insure that public resources are harvested in a safe and
efficient manner. PQs eliminate or restrain competition among
processors and create a regulated marketplace which requires creating a
substitute mechanism for price formation.
The crab rationalization process included nearly two years of
committee meetings in which the various elements were crafted. Both
harvesters and processors participated in these meetings. Throughout
this process, the processing sector was adamant that rationalization
was not going forward without processor quota. In the end all the
arguments for PQ come down to this: ``Our way, or no way.''
The Council spent three days debating nearly all the proposed
elements. In the end it created a balanced harvester IFQ system with
regional delivery requirements, skipper and crew provisions, and a
balance of recent and historic participation credit.
However during this time, not a single motion was made regarding
the allocation of 90 percent or the procession rights. There was no
debate regarding the impacts of creating this system in spite of
considerable testimony opposing processor quota shares.
It is our firm belief that an effective lobbying effort cannot
substitute for building a low cost and effective rationalization
program within the current framework of the Magnuson-Stevens Fisheries
Management and Conservation Act.
National Academy of Science Recommendations
Congress should heed the advice it requested from the National
Academy of Science. In the NAS report, ``Sharing the Fish,'' they said
there was no:
``. . . compelling reason to establish a separate,
complementary processor quota system (the ``two-pie''
system).''
It also noted with regard to foreign ownership:
``If there is a consistent congressional policy, it can be
characterized as resistance to foreign ownership of fishing
vessels and foreign exploitation of fish resources with the
U.S.''
The heart of the controversy over PQs is the impact on price formation.
PQs destroy the harvesters' ability to benefit from collective
bargaining under the 1934 Fishermen's Marketing Act.
Ninety percent of a harvester's crab must be delivered to a
processor holding unused PQ. The result is a game of musical chairs
which encourages harvesters to accept a sub -optimal price to avoid
being the last one standing. In order to change processors, a harvester
must find a new chair . Such harvesters must sell at a lower price than
the person they are displacing in order to buy this chair. Clearly this
creates a downward price spiral.
Issuance of PQs divides the market and takes competition for
product out of the equation. My neighbor owns a plumbing business and
has no experience with fishing, but it took him less than thirty
seconds to figure out that PQs will serve one purpose well . . . and
that fishermen will be paid less for their crab as a result.
Lower ex-vessel prices mean lower crew wages and reduced landing
taxes for the State of Alaska.
The Magical 10 percent Solution?--B Shares
The State of Alaska's Issue Paper has states that giving harvesters
``B'' shares to sell 10 percent of their crab on an open market will
restore harvesters bargaining power and guarantee a fair price for all
crab deliveries.
The Issue Paper also claim the 10 percent B shares allow new
processors to enter the fishery and increase the share of communities
like Kodiak with limited qualified ``A'' share processors.
Simply asserting that a 10 percent of the crab can do all this
doesn't make it so. Indeed, these assertions lack supporting analysis.
Unless B shares are like the biblical loaves and fishes, it is
impossible to believe these shares will provide enough crab to leverage
price by offering them to PQ endowed processors and provide a pool of
product for disenfranchised processors and communities.
In fact, the Council recognized 10 percent was inadequate and
initiated a series of trailing amendments to deal with community
protections and with price formation through binding arbitration.
Trailing Amendments--Binding Arbitration
PQs would segment and allocate 90 percent of the Bering Sea crab
market. Without specific legislative permission, that allocation would
constitute a per se violation of antitrust law, equivalent to price
fixing. It is a ``hard-core cartel agreement'' that would otherwise be
prosecuted criminally by the Department of Justice. This inherently
anti-competitive effect of PQs is the problem that requires Binding
Arbitration.
The Council Chairman appointed a committee of fishermen and
processor representatives to address the issues of binding arbitration.
Over several months of meetings committee developed two distinct
approaches.
The Council chose the version supported by processors on a 6-5
vote. That version fails to provide a meaningful safety net. Nor does
it provide assurance of an outside option approximating a competitive
marketplace.
Real Time Oversight
The Council has repeatedly stated its intent to modify the program
to respond to unintended or unanticipated impacts. To do so will
require far broader authority than simple legislation to implement the
preferred alternative. It is also naive to imagine that some of the
effects will be reversible.
Without additional authority, the Council lacks the tools required
to address many of the problems that will arise. It is difficult to
imagine that Congress will relish taking on the role of real-time
program manager. It is also difficult to believe that additional
authority should be granted outside the context of the Magnuson-Stevens
Fisheries Conservation Management Act.
If Congress authorizes PQs, it will transform Councils' task from
managing fisheries to managing markets. They will need the powers and
experience of the National Labor Relations Board, Federal Trade
Commission, Department of Justice Anti-trust Division, and Public
Utility Regulatory Commissions.
Alternatives to Processor Quotas
There are many alternative approaches that have been utilized to
deal with the concerns of processors in a variety of rationalized
fisheries. Even without PQs, the crab rationalization plan gives
processors substantial protection by program elements. These include:
Separate Catcher Vessel and Catcher Processor classes of
quota, so fishermen can't process their own crab.
Regional restrictions on deliveries.
Processors are allowed to acquire and own harvest quota.
Limits on consolidation of harvest quota, preserving a
diverse supply for processors.
The Council's Advisory Panel also offered a number of alternatives
to PQs which didn't receive adequate consideration. If analysis did
show there was further necessity to protect processors, there are less-
damaging alternatives.
The American Fisheries Act provides one such workable alternative
to PQs. The critical difference is that while the AFA coops provide a
large measure of stability through the requirement for annual coop
contracts with an eligible processor, no processor is guaranteed a
fixed share of the harvest for more than one year, and ultimately it is
competition that governs whether a vessel will remain with a processor
or move its quota to another processor.
Legitimate processor concerns can be addressed without authorizing
PQs and segmented markets.
Recommendations:
Congress should reject the Council's request for a radical
expansion of its duties beyond fisheries management and into market
regulation represented by the request for authorization to adopt the
Processor Quota element of their preferred alternative.
If Congress does authorize PQs it is critical to do so in a manner
that guarantees that improved community protections and a better
binding arbitration process are accomplished first.
Introduction
The CRAB Group supports Crab Rationalization including measures to
protect communities and processors. The status quo system has left
harvesters on or over the brink of bankruptcy, and has already led to
the demise of most of the non-AFA crab processors. Worst of all, the
murderous ``race for fish'' in the middle of winter kills an average of
five fishermen a year.
We believe a fair rationalization program can be developed without
resorting to Processor Quotas (PQs) which will result in a segmented
monopsony and require endless government regulatory involvement to
simulate a competitive market.
Our concerns are detailed in the following discussion.
1.0 What Happened in Dutch Harbor and Why the 11-0 Vote?
The North Pacific Fisheries Management Council staff prepared a 436
page analysis for the crab plan of which just four pages were devoted
to the impacts of Processor Quotas. Recognizing the controversial
nature of PQs, the Council contracted outside economists Milon and
Hamilton from Florida for further analysis of the impacts. These
economists produced a 35 page document which highlighted the negative
impacts on harvesters. Unfortunately, the Council chose to remove it
from the analysis.
At the June meeting where the Council picked its preferred
alternative, it passed 20 amendments (and considered many more) to the
proposed plan over three days of debate on the element and options of
the program dealing with the harvest sector. Not a single motion was
made on the regarding the level of PQ, and so there wasn't a word of
debate regarding the impacts of creating a PQ system governing the
marketing of 90 percent of the crab.
Aside from the PQ element, the Council did an excellent job of
putting together a balanced IFQ system with regional community
protection, skipper and crew provisions, a reasonable balance of recent
and historic participation credit, and excessive share provisions for
the harvest sector. All of which could be adopted without the PQs now
the IFQ moratorium has lapsed. However, the processing lobby made it
quite clear throughout the process that nothing was going to happen
unless and until they got processing quota. In the end all the
arguments for PQ come down to this: ``Our way, or no way.''
This powerful group has made it clear to everyone seeking to
rationalize fisheries, that regardless of the economic cost to other
non-diversified processors, fishing communities, vessel owners, or,
indeed, the cost in human life for those who work in the nations most
dangerous occupation, they will block any action that doesn't give them
control of the harvester's market choices.
It's my opinion that the Council, tired of having blood on its
hands, chose what they hoped would be the lesser of two evils. Since
the Council vote last June Congress allowed the moratorium on IFQs to
expire. It is now up to Congress to decide whether Processor Quotas are
good policy.
2.0 Are Processor Quotas Necessary or Prudent?
2.1 National Academy of Science Recommendations
In the 1996 Magnuson-Stevens Reauthorization Congress directed the
National Academy of Science to provide advice and recommendations on
IFQ programs and specifically directed the evaluation of processor
allocations. Section 303(d)(5) of the M-S Act directs Councils to
consider the recommendations for the NAS report (Sharing the Fish).
2.1.1 ``Sharing the Fish'' on Processor Quota
Page 205 of ``Sharing the Fish'' contains a two part recommendation
relative to processors and quota. The first part speaks to allocating a
portion of the IFQs to processors; the second speaks to creating a
``two pie'' or PQ system:
``On a national basis, the committee found no compelling reason
to recommend the inclusion or exclusion of processors from
eligibility to receive initial (fishing) quota shares''
``Nor did the committee find a compelling reason to establish a
separate, complementary processor quota system (the ``two-pie''
system).''
Page 153-155 of ``Sharing the Fish '' provides a more extensive and
very useful discussion of the issues surrounding processor quota
allocations. The NAS concluded:
``The committee was not convinced, however, that the solution
to the perceived problems lies in the allocation of either
harvesting or processing quota to processors.''
2.1.2 Distribution of Benefits of Quota Shares--Initial Allocation
``Sharing the Fish''--the report to Congress by the National
Academy of Science recommended a broad distribution of the benefits of
Quota share programs. The benefits are broadly distributed in the
initial allocation under the harvester IFQ portion of the Council's
plan. However, the benefits of the Processor Quota are highly
concentrated.
About 250 harvester vessels would get allocations as IFQs.
Only 21 of the 80 processors who operated in opilio in the
last 10 years would receive PQs.
According to the Council's analysis ``the top 12 would
receive more quota allocation than they historically processed
(96.4 percent compared to 75.66 percent).''
The top four opilio processors will be guaranteed 57.6
percent of the PQ.
2.1.3 Distribution of Benefits and Consolidation
In contrast to the harvest sector where caps were set at levels
that would maintain a minimum fleet size of about 100 boats, processing
caps were set so as to allow consolidation to only 2 processing
facilities.
Though the processor ownership cap is set at 30 percent, the action
would also grandfather in initial recipients of IPQ, not as of the date
of Council action in June of 2002, but as of a date in the future when
IPQ is actually issued--thus inviting consolidation in the interim.
Use caps were set for only one fishery in one region, (opilio in
the northern region) at 60 percent of the IPQ. There was no definition
of the duration of ``use'' (i.e., leasing) as time limited. Thus even
if the 30 percent ownership caps did provide for a minimum of 4
processors per fishery, the lack of ``use'' caps allows the ownership
caps to be neatly circumvented by 99 year leases.
In addition to the lack of meaningful consolidation limits for PQs
holders, individual processing companies are allowed to own up to five
percent of the harvest quota, while harvesting companies are limited to
one percent.
The Council's lack of meaningful action on ownership and use caps
opens the door to unconstrained consolidation of the processor sector
before and after implementation. There is a fundamental inconsistency
between the concerns imbedded in the MS-FCMA over excessive shares and
promoted in the NAS report, and the creation of PQs.
2.1.4 Foreign Ownership
On page 155 of ``Sharing the Fish'' the NAS notes:
``If there is a consistent congressional policy, it can be
characterized as resistance to foreign ownership of fishing vessels and
foreign exploitation of fish resources with the U.S. EEZ (e.g., 16
U.S.C. 1812[a], 1824[b][6]). The concerns giving rise to the exclusion
of foreign interests fall within several categories:
Fear of foreign domination of the maritime industry and
fisheries;
Difficulties in regulating foreign-owned businesses;
Threats to the social values of U.S. fishing communities;
and
Loss of potential economic benefits.
The Council's analysis (page 393) showed that between 37 percent
and 49 percent of Processor Quota would be allocated to foreign
processors. This does not take into account the amounts to be allocated
to domestic `'shell'' corporations, formed to qualify vessel ownership
under MARAD rules. [e.g., Peter Pan Seafoods, a Japanese-owned company,
would receive an estimated 14 percent of initial PQS for opilio. A
subsidiary which owns a processing ship, Steller Sea, would receive as
much as 5 percent additional allocation, which has not been accounted
as allocation to foreign processors.
It is unlikely that foreign owned processors could be precluded
from being issued Processor Quota, because they would rely on treaty
protections to demand equal treatment (as occurred under the AFA). It
is ironic that we restrict the allocation of IFQs to U.S. citizens
only, when Processor Quota would require some of those U.S. citizens to
sell their property/catch to foreign owned processing companies.
Processor Quotas are inconsistent with the recommendations of the
National Academy of Sciences found in ``Sharing the Fish.''
2.2 Economists' Views on Processor Quotas
The entire theoretical underpinning of Processor Quotas rests on
the work of one economist--Scott Matulich. It is his belie f that in a
free market, fishers with IFQs will ``expropriate the quasi-rents
rightfully belonging to processors'' because harvesters would no longer
fear that company owned boats would pre-empt their catch if they were
to go on strike.
Matulich has been able to parlay this diagnosis into a prescription
for a particular cure of his own design called the ``2-pie'' or PQ
system. The Council bought off on this prescription at a particular
dosage level PQs for 90 percent of each catcher boats' harvest.
Unfortunately there is no FDA to require testing on this new medicine
to determine a safe dosage level, before it is administered to the crab
fleet.
To judge whether the side effects of Matulich's cure are likely to
be worse than the disease, it is necessary to turn to other economists.
As noted in the preceding section the National Academy of Science
considered and rejected Matulich's prescription. They were not alone.
2.2.1 The GAO on Matulich
In December of 2002 the GAO provided this committee with a report
on IFQs which contained a very critical review of a paper by Matulich
purporting to provide an empirical basis for his theory in the context
of the existing Halibut and Sablefish IFQ program. They questioned the
methodology and the potential for bias in the survey design for
gathering data.
2.2.2 Economists on Processor Quota--Milon and Hamilton
In a paper prepared under contract for the Council by Florida
economists J. Walter Milon and Stephan F. Hamilton (A Comparative
Analysis of Alternative Rationalization Models for the Bering Sea/
Aleutian Islands Crab Fisheries--March 2002) the authors describe the
impacts of a ``segmented monopsony.''
In discussing the IPQ model Milon and Hamilton noted:
``The (PQ) quota allocation defines a property right of each
processor to serve a perfectly segmented market, and, with a
fixed quantity of harvest, each processor maximizes his profits
by paying the lowest ex-vessel price that supports harvester
delivery of this quantity. The outcome is regional monopsony
ex-vessel pricing . . . Accordingly, the delineation of
processor quota rights subsumes all economic rent from the ITQ
program in the harvest sector . . . With a two pie permit
distribution that allocates the full processing quota, the
value of harvester permits are driven to zero . . . With
completely defined property rights in the processing sector,
the allocation of property rights in a harvest sector ITQ
program becomes redundant.''
The Council stopped just short of completely defining property
rights in the processing sector, leaving 10 percent of the catcher boat
harvest in an open market.
Milon and Hamilton went on to observe that in a system where some
percent of the harvest share remains ``free market'' (such as the 10
percent recommended in the Council action) the outcome is a blend that:
``. . . results in a continuum of market segmentation levels.
Consequently, all possible two-pie permit distributions have
identical implications for economic efficiency, but differ in
the degree to which the policy rent is shared between market
participants. Processors are likely to fare better, and
harvesters fare worse, as the ratio of A to B permits increases
in the proposed fishery management system.''
Cartels are precluded by existing anti-trust laws. It is ironic
that the same outcome (monopsony pricing) would be legally achievable
under Processor Quotas. The only functional difference is that when a
legal Processor Quota system segments the market, it will be more
effective than if a group of processors had conspired to set prices. In
the latter instance there is always hope that a new processor could
enter destabilize the cartel by offering competitive prices.
2.2.3 Economists on Processor Quota--Halvorsen
Economist Dr. Halverson, who was contracted by the Council for an
earlier analysis of the distribution of bargaining power under
different ``game'' rules for American Fisheries Act coops, was also
critical of the Matulich 2 Pie theory. Dr. Halvorsen presented a paper
to a hearing of the U.S. House Resources Committee explaining the
theoretical deficiencies of the Matulich theory. Additional analysis by
Dr. Halvorsen have been submitted to this committee by Mayor Freed.
2.2.4 Economists on Processor Quota--Christy and Anderson
Two other very prominent fisheries economists served on the NMFS
Advisory Panel to the NAS when ``Sharing the Fish'' was prepared, Lee
Anderson (chairman of the NMFS East Coast AP) and Francis Christy.
Christy, who worked in fisheries for many years for the UN-FAO, is
considered to be the economist who came up with the idea for IFQs. Lee
Anderson, who wrote a seminal text book on IFQs and economic theory,
was a member of the Mid-Atlantic Fisheries Management Council when the
1st IFQ program was adopted. Both economists have been very critical of
the Matulich theory and of the idea of PQs. While Anderson recognizes
the potential for negative impacts on processors from IFQs to the
extent that their capital is non-malleable, he doesn't advocate PQs as
the appropriate fix for that potential problem.
3.0 Do We Believe in the Value of a Competitive Marketplace?
3.1 Price Formation Under Status Quo versus Under Processor Quota
The heart of the controversy over Processor Quota goes to its
impact on price formation.
3.1.1 Price Formation Under Status Quo
Given the depressed state of crab stock, the last Bristol Bay Red
King Crab fishery lasted just over three days. The last opilio season
was a matter of weeks. These are the two major crab fisheries. Price
has been negotiated pre-season by a marketing association. The derby
nature of the fishery makes it difficult to ``shop around''--crabbers
have generally been price takers in a world market for crab.
The opilio catcher boat fleet has gone on strike the last couple
years. However, because there are a number of catcher processors who
fish a common quota with the catcher boats, striking means foregoing a
portion of the harvest. One of the major processing companies owns 4 of
the catcher processors, so their reaction to a strike is ``throw me in
the briar patch.''
With three day seasons in the Red King crab fishery, strikes would
be economic suicide for catcher boats. Matulich is right about one
thing, shortened seasons are better for processors than they are for
harvesters.
3.1.2 Price Formation Under Processor Quota
The program the Council passed and has requested Congress to make
legal is very different. It allows harvesters to sell 10 percent of
their catch to the processor of their choice. The other 90 percent must
be delivered only to a processor holding unused IPQ. This results in a
game of ``musical chairs'' where the ``last man standing'' has no
choice about where to sell--and as a consequence there is an urgency to
``sit down'' early at a sub-optimal price to avoid being the ``last man
standing.''
If a harvester wishes to move to a different processor because they
are unhappy with the way they are being treated, there is only one way
to do it. They must displace someone who is working for a different
processor. The only way to do that is to offer to fish at a lower price
than the person you are displacing. This fundamental alteration of the
dynamics into a game of musical chairs destroys the ability of fishers
to benefit from collective bargaining as provided under the 1934
Fishermen's Marketing Act.
3.2 The Nature of the Right or Privilege represented by the Processing
Quota
There is a fundamental difference in purpose between IFQs and PQs.
The purpose of PQ is to direct the transfer of private property. The
purpose of IFQs is to allocate access to a share of free swimming
critters, which up to the point of capture, are a public trust
resource.
IFQs are generally understood to be a privilege to harvest a fixed
portion of the common property public trust resource. The result of
being allowed to harvest that resource is that it is converted to
private property at the point of harvest.
A PQ is a right or privilege to process a fixed portion of the
harvest. Congress has been clear that they regard Harvest Quota shares
as a privilege, but there is a spectrum between ``privilege'' and
``right'' that has yet to be debated with regard to PQs. The wrinkle
here is that crab, once harvested, have been converted to private
property. Thus, it appears that the PQ directs the disposition of
private property, rather than the disposition of a public resource.
The introduction of PQs for the purpose of eliminating or
restraining competition among processors creates a regulated
marketplace and the need to provide a substitute mechanism for price
formation.
4.0 Making a Square Wheel Round--Fixing the Impacts of PQs
The Council failed to analyze and debate the impacts of a 90/10 PQ
system up front and made the assumption that allowing 10 percent free
market or ``B shares'' would serve to simultaneously allow opportunity
for disenfranchised processors and communities as well as provide
harvesters leverage for obtaining a fair price. The Council also set in
motion a process of ``trailing amendments'' to address price formation
through ``binding arbitration'' and ``community protections.''
4.1 Will 10% B Shares Serve Their Intended Purpose?
The State of Alaska's Issue Paper has said that ``B'' shares will
Protect harvester's bargaining power and guarantee a fair
price for all crab deliveries.
Provide a pool of product for new processors to enter the
fishery.
Increase the share of communities which have limited
qualified ``A'' share processors.
The National Environmental Policy Act requires the evaluation of
regulatory actions by examining contrasting alternatives, so that
decisions rely on analysis rather than unsupported assertions.
In order for the above assertions to be possible, it must be
plausible that being a non-PQ endowed processor is a viable business.
This raises the threshold question, ``How does a non-PQ endowed
processor attract B share deliveries?''
4.1.1 A Simplified Answer
The following is a very simple modeling exercise that shows the
answer is ``no.'' Assume a base ex-vessel price of $1/lb in the PQ
sector.
Assume a Processor 1 is PQ endowed with 1,000,000 lbs and takes A
share deliveries from 10 vessels with 100,000 lbs each.
Assume Processor 2, not PQ endowed needs 100,000 lbs to justify
operating a crab line.
In order to attract deliveries from 10 vessels with 10,000 lb each
of B share crab Processor 2 will have to pay some sort of incentive
bonus. If Processor 2 determines it can pay $1.10 (a 10 cent
``competitive bonus'') and still show a profit, will doing so attract
deliveries of B shares?
In order to retain the deliveries of the 100,000 lbs of B share
crab from its 10 vessels Processor 1 will have to pay some sort of
``loyalty bonus.'' If Processor 1 determines it is willing to pay $1.01
(a 1 cent ``loyalty bonus'') pro-rated over both A and B deliveries,
why would the vessels deliver B shares to Processor 2?
Both Processor 1 and 2 are paying an ``extra'' $10,000 to get the B
share deliveries. The difference is that Processor 1 is amortizing that
$10,000 over 1,000,000 lbs and Processor 2 is amortizing over just
100,000 lbs. This gives the PQ endowed processor a 10:1 advantage over
the non-PQ processor. (If B shares had been set at 20 percent the PQ
endowed processor would still have a 5:1 advantage, or about a 3:1
advantage if B shares had been set at 30 percent)
Entry by a non-PQ endowed processor is unlikely to occur unless PQ
endowed processors are indifferent to retaining B share deliveries. If
any new processors did enter, they would quickly be driven out by the
endowed processors price leverage. When the game is this clearly
rigged, very few will make the mistake of playing.
If there are no non-PQ endowed B share processors, they can't
fulfill the variety of functions asserted in the ``Issue Papers.''
4.1.2 The Missing Analysis
One of the stated purposes of PQs is to address the transitional
costs associated with non-malleable capital in the processing sector.
The analysis currently lacks any quantitative analysis of the crab
specific fixed capital (malleable or otherwise) in the processing
sector.
Part of the reason for the lack of analysis is that Section
303(b)(7) of the M-S Act exempts processors from the requirement to
submit economic data. As a result they are free to claim harm, but the
analysts don't have the ability to verify their claims.
If analysis shows that there is only 10 cents on the dollar of
bargaining power at stake in the choice between PQs at levels between 0
percent to 100 percent, that difference represents a difference of $10-
50 million per year in ex-vessel revenue. That difference in revenue in
turn affects the raw fish tax collected by the state of Alaska as well
as wages for crew flowing into communities. The analysis should have
included an evaluation of the level and duration of the PQ necessary to
compensate the transitional costs of the processing sector; but again,
processors have hidden behind the lack of data.
4.2 Binding Arbitration--A Substitute for Competition?
The Council recognized that PQs would have a profound impact on
price formation and so they initiated a trailing amendment to deal with
Binding Arbitration. The NPFMC Chairman appointed a committee of
fishermen and processor representatives to address the issues of
binding arbitration. This committee developed two distinct approaches.
4.2.1 Need for Arbitration
Binding arbitration is necessary to address the inherently anti-
competitive effect of the PQ component of crab rationalization. The
magnitude of the problem is proportionate to the A/B share split, and
at 90/10 the Binding Arbitration process plays a crucial role in
substituting for the removal of a competitive market for harvesters.
PQs would effectively segment and allocate 90 percent of the market
into which crab harvests will be delivered. That action would radically
shift negotiating leverage between harvesters and processors relative
to status quo. Without a specific legislative exemption, that action
would constitute a ``per se'' violation of antitrust law equivalent to
price fixing. It is a ``hard-core cartel agreement'' that is prosecuted
criminally by the Department of Justice.\1\
---------------------------------------------------------------------------
\1\ By contrast, fishermen have had an exemption from antitrust law
that allows them to collectively harvest, process, market and/or sell
their catch since the adoption of the Fishermen's Collective Marketing
Act in 1934. Through a qualifying fishermen's cooperative, it is legal
for fishermen to allocate among themselves harvest shares of a fishery,
and to collectively negotiate the prices at which they are willing to
sell their catch. It is also legal for Councils to adopt individual
fishing quota programs which allocate harvest shares by regulation.
---------------------------------------------------------------------------
Binding arbitration is intended to address failed price
negotiations, and to reintroduce parity lost through processor market
segmentation. Binding arbitration is not intended to be a substitute
for consensual price negotiation, and we expect and intend that most if
not all crab delivery contracts will be settled through negotiation.
However, all negotiations are conducted against the parties' outside
option if negotiations fail, which in this case is the price that would
be set under arbitration. Therefore, the results of the model chosen
will directly condition the results of such negotiations.
While it is not panacea that will undo the anti-competitive impact
of PQs, the ``Fleetwide'' conventional arbitration model more
appropriately addresses the negotiation leverage shift associated with
the Council's preferred rationalization alternative. T he segmented
``Last Best Offer'' model, does not.
4.2.2 Contrasting the Alternatives
The Council tasked the Binding Arbitration Committee with
developing an arbitration program, and the committee came back to the
Council with two models referred to as the ``Fleetwide Minimum Price''
(FW) and the ``Last Best Offer'' (LBO) models.
The names of the two Binding Arbitration models don't capture the
range of differences.
4.2.2.1 What's In a Name?--Fleetwide Arbitration
We believe the FW model works better than the segmented LBO model
in part because it is more closely models the current price formation
process, by setting price before a harvester is required to make an
irrevocable commitment to deliver.
As its name implies the ``Fleetwide'' arbitration model is designed
to create a minimum price that is available to the whole fleet as a
``safety net.'' It allows the arbitrator to ``cut the baby'' and
establish a minimum price that is between the positions taken by either
processors or harvesters in the preliminary negations.
4.2.2.2 What's In a Name?--Last Best Offer Arbitration
Under the segmented LBO model, harvesters must irrevocably commit
to deliver their crab to a processor to trigger a price arbitration.
This is a highly unusual departure from standard commercial practice.
It is akin to entering a contract to buy a house before the price has
been revealed. It will have adverse economic and psychological effects
on harvesters entering the negotiation process.
Given that the purpose of Binding Arbitration is to compensate for
the creation of a segmented market, providing a reasonable outside
option makes more sense than stripping the harvester of the right to
strike.
The ``Last Best Offer'' model involves separate isolated
arbitrations for each processor and for individual harvesters or groups
of harvesters delivering to a particular processor. While LBO
arbitration forces parties to narrow the range of proposals submitted
to the arbitrator(s), also disadvantages the more risk averse party,
and invites strategic gaming by processors. Crab harvesters, who
typically depend heavily on their crab revenues for survival, are
likely to be much more risk averse in crab price negotiations than crab
processors who have other sources of income (such as AFA pollock
processors).
Harvesters in these circumstances have proportionately more to lose
than their processor counterpart. As a consequence, they may well have
a strong incentive to buy their way out of arbitration at a discount,
rather than enter a process under which an arbitrator is constrained to
accepting one or the other of the parties' price proposals, rather than
having the latitude to frame an equitable result.
4.2.2.4 Dr. Plott's Analysis
The FW model was deliberately designed to produce an environment in
which the parties are encouraged to collaborate to produce additional
value. In Dr. Plott's experimental analysis for the Council, it appears
to have done so On the other hand, the segmented LBO model
fractionalizes parties, and in Dr. Plott's experiments produced a more
contentious negotiating environment, with fewer well timed deliveries,
and at least one instance of a harvester choosing not to ``deliver.''
This is an important consideration, if we are hoping to obtain
additional value from our crab resources through rationalization.
4.2.2.5 Information available to the arbitrator(s)
It is critical under either model that the arbitrators have access
to data concerning historical and current crab transactions. Under the
FW model, all arbitrations are conducted by the same arbitrator or
arbitration panel. However, under the segmented LBO model there is no
assurance of such information exchange.
This is important, as there are unresolved legal and policy issues
which may prevent the arbitrators from accessing the database being
established in connection with the program to verify the accuracy of
information submitted in the discrete arbitrations. Section 303(b)(7)
and 402(b) of the M-S Act prevent the collection of necessary
verification data from processors, and would prohibit its release to an
arbitrator even if it were collected.
4.2.2.6 Arbitration timelines
The FW model was designed to prevent the arbitration process from
being time constrained. The segmented LBO system has set what we
perceive to be an extremely tight time frame within which all
arbitrations would take place. We are concerned that the resulting time
constraint may shift negotiating leverage inappropriately, and prevent
an effective exchange of data between arbitrations.
The compressed 15 day timeline for LBO arbitration is likely to
require separate arbitrators as well. This means that any individual
arbitrator will have a limited information base to evaluate the market.
A single Arbitrator with access to a broad view of the full market
under the FW model is better situated to determine a fair minimum
price. If each PQ owner's arbitrations are happening in a vacuum, the
arbitrator will have a narrow frame of reference for ground-truthing
the PQ owner's data.
4.2.2.7 Arbitration administration
Given the significance of bin ding arbitration in the context of a
segmented and allocated processing market, we think it is far more
critical that the system function well than be cheap. In any case, we
do not believe that the relative costs of the two systems would be
substantially different. The FW model would generally use one
arbitrator or a single panel over a longer time, while the segmented
LBO model would use more arbitrators or arbitrator panels over a
shorter time frame.
4.2.3 Quality issues
Quality affects price, and in the absence of well defined quality
standards and a quick, efficient and equitable enforcement system,
quality issues could skew negotiating leverage notwithstanding
arbitration system design. This remains as an issue that needs to be
referred to the Binding Arbitration Committee for further work.
4.2.4 Data Verification
The Analysis states on page 3.7-8 that the arbitrator will need to
invest substantial time and effort into development of the historic
division of revenues standard, and to determine both historic ex vessel
prices and first wholesale prices. It notes:
``The magnitude of this problem is not likely to be fully
understood until the arbitrator begins the process of
calculating the division of revenues.''
The Analysis points to a long list of complicating factors in the
arbitrators' task. It states that determining the historic first
wholesale prices and revenues division will also be:
``complicated by the lack of uniformity of processors and the
different products those processors sell into different
markets.''
``complicated by vertical integration of the processing
sector.''
``complicated by several other factors . . .
``. . . sensitive to the production levels of specific
products . . .''
``. . . sensitive to changes in total harvest . . .''
``. . . (and) location of landings''
``complicate(d) . . . (because) Commercial Operator Annual
Reports (COAR) . . . distinguish species, but not fishery.''
Without the ability to access the Data Collection system, it is
imperative that the Arbitrator will not have the ability to ``ground
truth'' information provide by participant. There is a need for
retroactive gathering of data to establish a baseline for measuring
impacts of PQs and the efficacy of Binding Arbitration. Unfortunately
this may be an impossible task.
NOAA-GC has said that the Data Collection program will be unable to
either provide the necessary data, or even verify the accuracy of data
provided to the Arbitrator by participants, without changes to section
303(b)(7) and 402(b) of the M-S Act.
4.2.5 The Council's Revised ``Last Best Offer'' Model
The Council ultimately adopted a revised LBO model which
incorporated a watered down version of what is called the ``Steele
amendment.'' The Steele amendment would have utilized the highest
arbitrated price (reflecting a minimum of 7 percent of the market)
resulting from the LBO process to establish a fleetwide price. In one
significant way this reflects current practice where the Alaska
Marketing Association negotiates with a number of processors and signs
a contract with the processor with the best offer, and then that offer
is ultimately matched by other processors.
The success of the Steele amendment version of LBO would have
depended on the continued survival of a number of processors in the 7
percent market share range. Under the consolidation rules, these
smaller IPQ holders may disappear over time, and with them the
potential of the revised LBO to achieve its goal of a quasi-competitive
fleetwide price.
However, the Council's revised LBO alternative merely uses the
information on the highest price representing 7 percent of the market
as information available for the arbitrator's consideration in the
subsequent year.
The Binding Arbitration process recommended by the Council fails to
provide a meaningful safety net. It does not provide assurance of an
outside option approximating a competitive marketplace.
4.4 Community Protections
The second set of trailing amendments dealt with community
protections. The CRAB Group concurs with the testimony of Mayor Freed
of Kodiak.
The supposed ``right'' of first refusal is an ephemeral and
unworkable ``protection'' to a problem that would not exist but for the
inclusion of PQs in the program.
PQs facilitate consolidation and the lack of meaningful processor
consolidation limits in the program ultimately means plant closings in
remote Alaskan coastal communities.
4.4 Government in the Marketplace
The Council has repeatedly stated its intent to modify the program
to respond to unintended impacts. If it is to do so it will need far
broader authority than legislation which simply implements the
preferred alternative. This implies the Council is asking for a blank
check from Congress. If Congress doesn't provide a blank check, then
Congress must exercise real time oversight of the program as only
Congress will have the power to amend it. It is naive to imagine that
Congress can micro-manage the impacts of PQs in real time, or that the
impacts will be reversible.
When government steps it to try to simulate the function of a
competitive market it is sticking its foot in a tar baby. If Congress
authorizes PQs, it will transform Councils from fish management bodies
into a hybrid of, or delegating the authority of, the National Labor
Relations Board, Federal Trade Commission, Department of Justice Anti-
trust Division, and a Public Utility Regulatory Commission. Council
members are not selected for their competence in these areas, which is
why they do not currently have the authority to adopt PQs.
5.0 Alternatives to Processor Quotas to Protect Processors
There are many alternative approaches that have been utilized
elsewhere to deal with the concerns of processors in a variety of
rationalized fisheries. These include elements in a number of existing
programs, as well as proposed alternatives that didn't receive adequate
consideration by the Council.
In the crab rationalization plan, processors were given substantial
protection by various program elements including the following:
Processors are allowed to own and acquire IFQs.
Catcher Vessel IFQ holders must deliver their crab to
processors rather than processing themselves as Catcher
Processors.
Regional restriction on deliveries, which favor existing
processors.
Limitations on consolidation of IFQ ownership at 1 percent
each for harvesters, which preserve a diverse supply for
processors.
Processors are allowed up to 5 percent each of the harvest
IFQ, in contrast to 1 percent limit for harvesters.
Without analysis of the adequacy of these provisions, nor
discussion or debate, the NPFMC added the provision of Processor Quota.
If analysis shows there is further necessity to protect processors,
there are less-damaging alternatives in existing programs such as the
AFA.
5.1 Existing Programs
5.1.1 AFA Pollock in the Shoreside Sector
AFA shoreside processors were collectively guaranteed an
increased share of the pie.
AFA shoreside processors were provided a closed class.
AFA shoreside processors were provided with a degree of
stability in the design of the coop rule.
AFA catcher vessels are only guaranteed their history as a member
of a coop with a processor partner. 90 percent of the catch history of
the coop had to be delivered to the processor partner in a given year.
Though vessels are able to move between processors annually,
disincentives were built in that discouraged movement between coops,
where the alternative to being in a coop was an open access derby for
one year.
The critical difference between the AFA processor protections and
PQs is that while the AFA coops provide a large measure of stability
through the requirement for annual coop contracts with an eligible
processor, no processor is guaranteed a fixed share of the harvest for
more than one year, and ultimately it is competition that governs
whether a vessel will remain with a processor or move its quota to
another processor.
5.1.2 Halibut & Sablefish IFQs
Halibut and sablefish shoreside processors were protected from
competing with freezer boats.
5.1.3 British Columbia's IVQ Groundfish
In the BC Canada groundfish ITQ, the allocation of 10 percent of a
vessel's catch history is conditional on community and processor
concerns.
5.1.4 Eastern Canada Opilio Crab
In the ``harvester only'' IFQ program for snow crab, binding
arbitration was instituted to set a base price. It is worth noting that
crab processing there is still profitable enough that it has attracted
a number of new entrants.
5.2 Alternatives discussed by the NPFMC's Advisory Panel
The Council's Advisory Panel offered a number of alternatives to
Processor Quotas which didn't receive adequate consideration.
Processors could be allocated a portion of the harvest ITQ
commensurate with their relative proportion of fishery specific
non-malleable capital.
A quasi closed class of processor, guaranteeing a percentage
of the harvest to be delivered to the class of eligible
processors based on their aggregate processing history.
An AFA style coop with disincentives for not joining or for
leaving a coop.
There are many options for addressing processor concerns without
adopting PQs and a segmented market.
6.0 Public Comment--Who Supports Processor Quotas
Letters submitted to the Council concerning Crab Rationalization at
the June 2002 meeting were overwhelmingly against Processor Quota.
The single harvesters association endorsing Processor Quota was
Alaska Crab Coalition, who supported it at the 80 percent level, with
the remaining 20 percent to be ``free market.'' Their written rationale
for limiting PQs to 80 percent provided to the Advisory Panel at the
June Council meeting makes an excellent case as to the problems with
Processor Quota at any level. Their recent (albeit temporary)
retraction of support for the PQ program following the April Council
action on Binding Arbitration stated:
``The ACC cannot accept the arbitration approach adopted by the
Council, and accordingly is forced to oppose statutory
authorization of the BSAI rationalization plan, until and
unless the Council adopts a system that protects harvesters
against market distortions that would otherwise result from
processors shares. Whether that can be achieved within the
context of the 90/10 formula is an open question.''
Though ACC is now ``confident that crab harvesters will be provided
the comfort level they expect and deserve'' in the ongoing process. The
open question of the moment is what changes ACC expects w ill be
adopted, but it is clear that they recognize the reality that Processor
Quotas will result in ``market distortions'' and that even they
question whether 90/10 isn't going too far.
The CRAB Group, representing over 100 vessels, supports
rationalization of the crab fishery but is opposed to PQ for all the
reasons outlined in this document.
There are several associations representing segments of the crab
fleet in addition to ACC and the CRAB Group.
United Fishermen's Marketing Association, Inc. (UFMA) based
in Kodiak strongly opposes PQs in the crab fishery or any other
fishery.
Tom Casey, representing the 30 crab vessels in the Alaska
Fisheries Conservation Group (AFCG), has stated that their
members find the 90 percent PQ program so untenable that they
would prefer status quo.
Jake Jacobs representing the Alaska Marketing Association
testified at the April Council meeting that ``Last Best Offer
Arbitration is an open sore.''
A petition signed by over 1,000 Alaskans was published this week in
the Anchorage Daily News. Other petitions circulated in the Seattle
area with several hundred signatures were submitted to members of this
committee by Fishermen's Finest company.
Changing the M-S Act to allow for Processor Quotas in the Bering
Sea crab fishery is only the allowing the camels nose into the tent. If
it is allowed in this fishery, the pressure to allow it in all
federally managed fisheries is inevitable. Does Congress really want to
open up this can of worms and transform fisheries management Councils
in to bodies with the authority to regulate trade? If so, then expect
the amount of public comment on how to regulate markets to escalate
exponentially.
7.0 Where Angels Fear to Tread--Recommendations
It is the position of the CRAB Group and others that Congress
should reject the Council's request for a radical expansion of its
duties beyond fisheries management and into market regulation
represented by the request for authorization to adopt the Processor
Quota element of their preferred alternative.
The harvester component of the crab rationalization program is
legal now that the moratorium on IFQs has expired. It can be adopted
together with the CDQ provisions, the regionalization provisions, and
the skipper and crew provisions designed by the Council, but without
PQs.
The need for the complex elements on binding arbitration and
additional community protections are largely a response to the impacts
of market segmentation resulting from PQs.
If Congress does authorize PQs it is critical to do so in a manner
that guarantees that improved community protections and stronger
binding arbitration process are all done at the same time and done
right.
Thank you for the privilege of submitting this testimony to your
committee.
______
August 2, 2002
The Bering Sea Crab Rationalization Preferred Alternative Or Anti-
Trust: Why Opposition to the Processing Quota Plan is Growing
Summary: During the North Pacific Fisheries Management Council's
(NPFMC) June meeting, the State of Alaska presented a comprehensive
crab rationalization motion, which with surprisingly few amendments,
was unanimously approved. This plan contains several elements that
require congressional approval, so a report to Congress is being
drafted detailing the council's preferred crab rationalization plan.
One of the elements that will require congressional action is Processor
Quota (PQ). This element is also being debated in the Magnuson-Stevens
Act reauthorization.
Recommendation:
Congress Should Not Authorize Processing Quota for the Crab Plan or
in the MSFCMA.
Because of Processor Quota, the crab plan is opposed by more than
half the vessels in the affected fisheries, and three of the four
associations of vessel owners in these fisheries. Because of processor
quota, one community publicly supports the crab plan while two
Boroughs, two native associations, three cities and more processors
than not--oppose it. We expect that as the salmon seasons wind down,
more communities, native associations, and fishing groups will join the
opposition.
Discussion:
The State of Alaska has circulated a document containing a number
of issue papers describing the Bering Sea Crab Rationalization Plan in
glowing terms.
In fact, the plan actually provides a new mechanism for complete
control of the market by a single processor.
Consolidation of so-called excess processing capacity is
guaranteed. Many of the processors involved have already benefited from
the American Fisheries Act.
Proponents urge that Congress, should not attempt to second guess
the NPFMC. They compare the NPFMC plan to a three-legged stool, which
is designed to balance. However, a relatively small group of interests
dominated the stool's construction. Those who would not concede that
processor quota was an essential element were completely subsumed to
those interests represented at the NPFMC table.
Background:
The Council could not properly design the ``processor leg of the
stool,'' because the processors represented at the Council table acted
in concert to frustrate Council implementation of a rationalization
plan for the BSAI crab fisheries.
Rationalization proceeded in incremental steps beginning with the
1992 moratorium on new vessels entering the Federal fisheries in the
North Pacific. In 1995 the council adopted a limited entry program,
followed by a recency requirement in 1998, and in 2001 by a not yet
implemented buyback program.
Individual Fishing Quotas, a controversial element of this long-
range plan, have become generally accepted by the crab fleets. In
developing a consensus on appropriate crab rationalization measures,
IFQ resistance has been mitigated by including provisions for elements
imposition of strict ``caps'' on ownership and use of IFQs.
Processor representatives made it clear that they would not allow
rationalization of the crab fisheries without Processing Quota. At a
NMFS workshop on IFQ programs in Galveston this summer, these interests
declared that they would block any IFQ programs, if they were not given
Processing Quota in the BSAI crab fisheries.
Processing Quota became the price that fishermen had to pay in
order to slow down the world's most dangerous of fisheries. It's not
surprising, therefore, that a small group of fishermen became willing
to pay this price.
Part I--The Case of Processor Quotas--Is it Reasonable?
Why Processor Shares Are Not Necessary:
The market division and consolidation sought by proponents of
processor quota are not allowed under the Sherman Act, and consequently
would require special enabling legislation. What is the justification
for this?
Processors have offered several justifications for Processor Quota
in the BSAI crab fisheries. These include issues of:
stranded capital,
lower profitability,
potential impacts on isolated rural communities,
compensation for those who will leave the industry because
of excess processing capacity,
concerns about competition with new processors attracted to
the rationalized fisheries.
None of these arguments ultimately hold up to scrutiny.
Stranded Capital
Advocates claim that processing quota is justified on the grounds
that a transition from open access to an IFQ program in the halibut/
sablefish fisheries caused problems for processors.
The National Academy of Sciences report to Congress, Sharing the
Fish stated that:
``Adversely affected processors assert that harvester-only IFQs
may result in stranded capital, lower profitability and
significant impacts on isolated rural communities.'' In
addition, ``The arguments for allocating harvesting or
processing quota to processors derive from a desire to
compensate those who will leave the industry because of excess
processing capacity.''
However, rather than accept the assertion that processing quota was
a reasonable solution to the problems expressed, the NAS panel
recommended:
``If the regional councils determine that processors may be
unacceptably disadvantaged by an IFQ program because of changes
in the policy or management structure, there are means, such as
buyouts, for mitigating these impacts without resorting to the
allocation of some different type of quota . . . For example,
coupling an IFQ program with an inshore-offshore allocation
would preserve the access of shore-based processors to fishery
resources.''
Stranded Capital is also a Function of the Derby
The Council analysis notes:
``The problem of stranded capital in the processing sector is
difficult to assess. . . . In recent years, some crab
processing facilities have been removed from service and are
currently idle. This suggests that crab rationalization (and
the current low stocks) have fishery declines and crab
rationalization, however, is somewhat problematic. Since
support facilities are often developed for use in multiple
fisheries, it [sic] changes in crab fisheries might not be the
sole cause of stranded capital. Declines in crab fisheries,
however, are certainly a contributor to stranded capital in the
processing sector.''
Successful rebuilding of the fisheries, a key element of the
rationalization plan, will produce a need for far more processing
capacity than is presently utilized.
Under an IFQ, smaller, currently closed, fisheries can be safely
opened.
Impacts on Isolated Rural Communities
The Council accomplished this by partitioning the allocation of IFQ
into Northern and Southern shares. This regionalization in the crab
plan partially addresses the problem of potentially significant impacts
from an IFQ on isolated rural communities. Regionalization doesn't
require IPQs to function.
Compensated Exit
How well does the NPFMC crab plan compensate those who will leave
the industry because of excess processing capacity?
The plan qualifies 21 processors in the opilio fishery, even
though 80 different processors worked in the fishery between
1991 and 2000.
The plan reallocates history from the smaller processors to
the largest.
Compensation for forced exit in the processing industry
resulting from excess capacity is not provided by the crab
plan.
Forced exit is institutionalized by the crab plan.
Fear of New Competition
Processors claim that facilities will become surplus because there
will be new processors attracted to the rationalized fisheries. These
processors, it is said, will operate low cost operations, because there
will not be a necessity to handle the volumes that were formerly
required.
This fear of competition by new entrants is itself an indication
that there is potential profit for processors in a rationalized fishery
without Processor Quotas.
Ending the Race for Fish
Processors have claimed representative the race for fish was driven
by processors as well as fishermen, and therefore there would not be an
end to the race for fish, unless processors also stopped competing. He
advocated that Processing Quota was the way to end processor
competition.
Experience of the halibut sablefish IFQ programs and other IFQ
fisheries clearly shows PQs aren't necessary to end the race for fish.
The desire of processors to end competition by segmenting the
market has one objective--to provide the mechanism for paying a lower
price to harvesters.
Power to Control the Process
Justifications of Processor Quota in the BSAI crab fisheries based
upon the reasonable arguments of stranded capital, lower profitability,
potential impacts on isolated rural communities, and compensation to
for exit due to excess processing capacity, do not stand up to
scrutiny.
Processor representatives have repeatedly told the fleets that they
``will not allow'' rationalization of the crab fisheries without
processing quota. Power, rather than reason, carried the day with the
Council.
The motion to allocate processor quota passed with out any
discussion of alternative methods to mitigate processor concerns.
Part II--The State of Alaska Issue Papers--Facts or Fallacies?
It is important to distinguish between the expressions of intent
set forth in the State of Alaska's compilations of Issue Papers and the
probable impacts that legislation of this crab plan would produce. In a
number of instances the impacts were examined, and are at variance with
the conclusions drawn by the State of Alaska, and consequently, with
the expressed intent of the crab plan.
Discounted Processing History--Fallacy 1:
The Issue Papers refer to the effect of the 10 percent open market
or B shares as ``discounted processing history'' since PQ is allocated
for only 90 percent of the harvest.
``No processor will receive 100 percent of their processing
history. They will receive 90 percent, leaving each processor
to have to compete with other processors to secure the
remaining 10 percent of processor shares.''
providing opportunities for new processors and increased harvester
leverage in the price formation process,'' and ``The intent of this
discount is to counteract any potential gain in processing power.''
[Emphasis in the original.]
Discounted Processing History--Fact:
The Council's own analysis states:
``the top 12 [processors] (with one exception) would receive
more quota allocation than they historically processed'' (99.4
percent compared to 75.66 percent)
A narrow qualification window for processors, excludes over
50 processors who have bought crab in the last 10 years. This
reallocates ``processing history'' to the ``top 12''.
Nearly one quarter of the processing history is first
redistributed from smaller and disqualified processors to the
largest processor quota share holders. Only then is it
``discounted'' by 10 percent.
The allocation of processing history' is a reduction of
broader market opportunity, and a bonus above the historic
share for the largest of the processors.
By contrast, the harvester program reallocates shares from
those with greater history, to those with less, leaving the
average harvester with about 88 percent of their own historic
catch.
The Council had no discussion, and built no record, for
their unanimous selection of this skewed allocation of
processing shares, or of the 90/10 split between A and B
harvest shares.
Discounted Processing History--More Fallacies:
The Issue Papers refer to the effect of the 10 percent open market
or B shares as ``discounted processing history'' since PQ is allocated
for only 90 percent of the harvest.
``the remaining 10 percent of processor shares . . .
[provide] opportunities for new processors and increased
harvester leverage in the price formation process,''
``The intent of this discount is to counteract any potential
gain in processing power''
``Based on the cost and wholesale data provided by crab
processors, it does not make sense to discount processor
history any more than the 10 percent level selected by the
Council. Data . . . shows processor margins are very thin, . .
. With these low margins, processors will be hungry for all
deliveries of crab, especially the crab delivered last.''
Discounted Processing History--More Facts:
The writer makes several critically flawed statements:
That a ``discount'' of Processor Quota shares actually
exists.--Which the Council Analysis shows is not the case.
That any such market competition would be permanent, not
ephemeral.--But, if processors ``compete'' to buy the `B' quota
shares for themselves, rather than B share harvest from
fishers, competitive market behavior ceases.
That the data received from processors was definitive.--
However, this ``data'' presented confidentially to the NPFMC,
without public review.
That processors have ``thin margins.''--By contrast,
harvesters had testified that recent crab fisheries had
resulted in losses for those with mean (average) catch, or
less. Thin margins' are certainly better than none
That ``processors will be hungry for all deliveries of crab,
especially the crab delivered last.''--This statement ignores
the impact of a fixed market share. To the first crab delivered
is no different than the last--the processor receives exactly
the amount of ``A class'' harvest for which he owns PQ. This
for increased margins by relieving the processor of the need to
compete based on price.
IFQs Result in Fewer Processors--Fallacy
The Issue Paper, Council Action on Processor Shares,'' claims that,
``only 25 of the 67 pre-IFQ firms survived the harvester-only
allocation of [sablefish] quota according to thorough analysis
done by Washington State University. . . . Experience to date
shows that harvester only allocation of shares results in fewer
processors.''
IFQs Result in Fewer Processors--Fact
This is conclusion not supported by the ``WSU study'', which showed
new processors entering following the halibut IFQ or other real world
experience.
The study, itself, cautions that the ITQ allocations ``may
not be causal.''
Attrition of processors in ``open access'' crab fisheries
occurred at a similar rate during the same time frame.
The NPFMC analysis shows 80 different processors operated in
the opilio fishery during the same period as that of the WSU
study.
The Council crab plan qualified only 21 processors in the
fishery.
Processors fared better under the halibut/sablefish IFQ
plan, than under either the ``open access'' fishery regime for
crab, or in the NPFMC crab IPQ plan.
The allocation of processor quota invites consolidation, it
does not require that recipients actually process in the
future.
Processing Quota = Legal Monopoly of Domestic Supply:
Processing Quota, as defined in the NPFMC crab plan, is presently
prohibited under U.S. law. The prohibition occurs under terms of
antitrust statute, and Congressional authorization of the crab plan
would provide, intentionally or otherwise, exemption from current
antitrust provisions. The specific nature of the exemption and the
impacts that such exemption could be expected to produce has not been
examined by the NPFMC in the rush to produce a report to Congress.
``AFA Coop Analogy''--Fallacy:
The State of Alaska document includes a one page Issue Paper titled
``Antitrust Issues'', which states:
``The same anti-competitive issues that were brought up under
consideration of the American Fisheries Act (AFA) are being
raised under crab rationalization''
``AFA Coop Analogy''--Fact:
The AFA did not create a fixed market division between processors,
as the crab plan does. The vessels were only allowed to enjoy the
benefit of this catch history within the operation of a cooperative.
Vessels are required to form a cooperative with a specific
processor, depending upon the deliveries of the vessel in the year
prior to coop formation. A vessel may leave the coop and move to
another, following a year in the ``open access'' fishery, reducing the
amount of market share of one processor, and increasing the market
share of the new processor.
In contrast to AFA, under the crab plan vessels are to receive a
direct allocation of fishing history. As a consequence, processor-owned
vessels will also receive a direct allocation, whereas under the AFA
processor-owned vessels could only operate in a rationalized manner if
they were allowed to participate in a coop.
``Dooley Hall'' and ``Wild Claims''--Fallacy:
The Issue Paper on ``Processor Shares'' states,
``Immediately after AFA passage, many harvesters tried to get
the Council to amend the AFA agreement. They did so under the
Dooley-Hall proposal, using precisely the same rhetoric, like
company towns and surfs [sic] working for feudal lords. There
were all sorts of wild claims the processors would capture all
the profits.''
``Dooley Hall'' and ``Wild Claims''--Facts:
The Issue Paper writer misses the very important balance that was
achieved, and which constitutes the dynamic point about which the AFA
shoreside agreements operate. The fears of the harvesters were genuine,
and were justified.
Without a mechanism for harvester movement between AFA coops,
harvesters would have been critically disadvantaged. The ability of a
coop to collectively allocate some or all of that ``10 percent'' to
effectively accommodate the movement of a vessel, allowing the vessel
to transition without enduring the `open access' year, had not yet been
established. The precise role of `processor association' with
cooperatives had not been determined, and one of the issues was whether
a processor could act, within the coop, to block such internal
allocations.
The interplay of this collective 10 percent ``open market''
allocation and ability of vessels to move between cooperatives was a
very important determinant in the balance that has been achieved under
the AFA.
The Dooley-Hall proposal, which would facilitate freedom of
movement of harvesters, was not rejected by the Council, it was tabled.
This action created a potential sword, which remains in place, hung
above the table. The Issue Papers do not recognize the finesse with
which the Council resolved this particular difficulty.
There were also wild claims at the time by Scott Matulich, the
processors' favorite economist and author of the WSU study. He claimed
that AFA as passed by Congress would cause Unisea to got out of
business, because the AFA gave harvesters too much power. However,
Unisea, an AFA processor, supported the removal of the AFA sunset,
which suggests they are happy with the benefits the AFA gives them.
Antitrust Relates to AFA Coops--Fallacy
An antitrust issue that did arise under the AFA is related to coop
formation. The Issue Paper ``Antitrust Issues,'' breezes past this:
``The 1934 Fisherman's Marketing Act contains an explicit
statutory exemption from the antitrust laws allowing fishermen
to bargain collectively in price negotiations with fish
processors. The fishermen are allowed to decide, as a group, on
their bargaining position, an activity that would violate
Federal antitrust laws without the exemption.''
``Since the processor provisions of the AFA had no anti-
competitive effects, then the less restrictive processor
elements included in the Council motion on BSAI Crab
Rationalization should be viewed similarly.''
Antitrust Relates to AFA Coops--Fact
A Congressional Research Service memorandum [CRS memo] prepared at
the request of Sen. Murray's staff, references a court opinion which
underscores the importance of the protections of the Fishermen's
Collective Marketing Act to harvesters.
In United States v. Hinote, 823 F. Supp. 1350 (S.D. Miss. 1993),
the court, concluded that catfish processors could not take advantage
of the antitrust exemption under the FCMA solely by purchasing or
leasing some interest in a catfish farming operation. Id. at 1359. The
court reasoned that if it were to come to the opposite conclusion,
``large integrated agribusinesses organized to market and sell
agricultural products could exempt themselves from the
antitrust laws by the simple expedient of purchasing and/or
leasing some interest in a farming operation, no matter how de
minimis the interest. Such a result, however, would undermine
Congress' express purpose in enacting both the Sherman and
Capper-Volstead Acts''
While the ``intent language'' of Senators Stevens and Gorton, the
makers of the AFA, indicated that processor owned vessels were intended
to participate in the coops, the final Department of Justice letter
ruling on the issue provided that the implicit exemption was by not a
blanket one:
``footnote 6. Our conclusion that processor-owned vessels may
participate in FCMA cooperatives under the AFA is therefore
unlikely to lead to anticompetitive results. Nevertheless, to
minimize the possibility of negative effects on the fishing
industry, Congress included within the AFA several provisions
designed to eliminate potentially adverse economic
consequences. See, e.g., 213(c)(1) (granting the North
Pacific Council the authority to recommend conservation and
management measures ``that supersede the provisions of this
title . . . to mitigate adverse effects in fisheries or on
owners of fewer than three vessels in the directed pollock
fishery caused by . . . fishery cooperatives in the directed
pollock fishery''); see also 144 Cong. Rec. S12,708 (daily ed.
Oct. 20, 1998) (statement of Sen. Murray) (``In the inter est
of ensuring that small, independent fishermen are the true
beneficiaries of fishery cooperatives, the bill includes a
number of requirements for fishery cooperatives in all three
sectors which are designed to provide these small, independent
fishermen with sufficient leverage in the negotiations to
protect their interests.''). Thus, should shoreside processors
in the BSAI fishery affiliate with catcher vessels for no
purpose other than to engage in anticompetitive conduct under
the umbrella of antitrust exemption, the AFA would appear to
give the Council the authority to check such abuses.''
Crab Coops vs AFA Coops--Fallacy
An important feature of the potential legislation should not be
hidden under the Issue Papers gloss:
``The Department of Justice has not found any anti-competitive
effects in the division of . . . pollock allocations among AFA
Coops. It is expected that the Department will view the crab
plan in a similar light.''
The crab plan includes a provision for the formation of
cooperatives of harvesters ``in association with a processor.'' This
should not become a means of allowing processor owned crab vessels to
join 1934 FCMA coops.
Crab Coops vs AFA Coops--Fact
It is clearly evident that processor-owned vessels already enjoy
the benefits of vertical integration. Since these vessels are also
directly allocated their catch history under the crab plan, there is no
necessity for the processor owned vessels to participate in the crab
plan coops.
There should be no exemption, direct or implied, granting processor
owned vessels participation in crab coops.
Antitrust Issues Under Crab--Fallacy
There are other anti-trust issues that are unique to the crab plan.
The Issue Papers radically understate the anti-competitive effects of
the action that the NPFMC is recommending to Congress.
Antitrust Issues Under Crab--Fact
Among the issues ignored, is the manner in which the NPFMC crab
plan creates a horizontal division of the entire market for the
domestic production of harvest vessels of snow crab, and more than 97
percent of the domestic production of harvest vessels of king crab.
This type of horizontal division of the market would represent a per se
violation of the Sherman Act, under existing law, if it were not
exempted by congressional action. This was never an issue under the
AFA.
Authority to Give Processors Antitrust Exemptions--Fallacy
The Issue Paper on ``Antitrust Issues'' concludes:
``If Congress says yes to the Council's crab rationalization
plan, this is a non-issue.''
Authority to Give Processors Antitrust Exemptions--Fact
Congress could enact legislation that would exempt the NPFMC from
prohibitions on almost any otherwise illegal activity. The question is,
not whether Congress can, but why should Congress do such a thing?
This is certainly true that:
It is within the power of the NPFMC, as a Federal advisory
body, to recommend a program that would otherwise be
prohibited.
It is within the power of Congress to provide this
exemption.
If it ``says yes,'' then Congress will have enacted legislation
which provides an implied antitrust exemption allowing BSAI crab
processors:
to divide up the market,
to consolidate control over that market to a single entity,
despite provisions of the Sherman Act and other antitrust
statutes.
It is that this power to reverse the present antitrust protections
that is of concern.
If Congress says ``yes'' without a more rigorous examination of the
issues, the results could be a irreversible disaster for harvesters,
communities, and the public.
The ``Antitrust Issues'' Issue Paper states that there is no
``closed class'' of processors [as is recognized to exist, under the
AFA] because new processors can ``purchase `open delivery' B shares or
allocated shares from a qualifying processor.''
No Closed Class of Processors?--Fact
This is a semantic device word play. The crab plan does result in a
closed class relative to A shares.
Under the NPFMC crab plan, as in the AFA, new entrants to
the processing sector must purchase a ``right'' from a ``closed
class'' of processors in order to buy ``A share'' deliveries.
In the AFA, membership in the `closed class' does not guarantee
market share.
In the NPFMC crab plan, closed class (``qualified'')
processors would be gifted with permanent market share, due to
participation in processing activity during a very narrowly
defined qualifying period.
The crab plan is more restrictive upon new entry of
processors, because the market share, as well as the right to
process, must be acquired from `qualified' processors.
The ``open delivery'' B shares referred to as source of product for
``new'' processors. It is true any processor (except catcher
processors) can legally purchase B share harvests. However, B share
fail to represent a viable basis for a new processor to enter the
fishery for several reasons :
This is a 10 percent market share segmented into more than
250 pieces, in each of the major fisheries.
Many of these pieces are already in control of the ``closed
class'' of processors that will receive Processing Quota under
the plan.
Processors are allowed, under the plan, to purchase and
control B shares.
B shares are intended to be used as leverage to obtain a
``fair price'' from the processor for A share deliveries.
B shares are intended to enhance deliveries to communities
such as Kodiak.
No Direct Linkage of Harvester to Processor?--Fallacy
The ``Antitrust Issues'' Issue Paper argues that the crab plan is
less restrictive than AFA for harvesters, stating:
``there is no direct linkage between fishermen and processors
(fishermen are free to deliver to any processor holding
processor shares).''--and that the plan,
``assures multiple processors in each crab fishery,''
No Direct Linkage of Harvester to Processor?--Facts
NPFMC has provided a system that is ultimately even more
restrictive than the AFA.
The AFA provides specific provisions to accommodate
harvester movement from one processor to another.
The crab plan requires harvesters to match quota shares with
a processor for 100 percent of the harvester's ``A'' designated
quota shares.
The bar to movement is that a harvester wishing to move must
identify a processor willing to discharge another harvester
already delivering to that processor, in order to ``free up''
Processing Quota shares, to accept delivery.
The crab plan allows processors to transfer PQ freely,
without effective caps upon the consolidation of quota shares
to a processing facility. Thus, a harvester may be directed by
the terms of a processing quota share trade, to deliver to a
processor, without the harvester having choice.
The Council action, in fact, allows consolidation of control
over all processing to a single entity.
Consolidation Rules Ensure Multiple Markets?--Fallacy
The Issue Paper, ``Council Action on Processor Shares,'' states:
``The Council instituted caps on the amount of consolidation
that may occur among the processors. The consolidation rules
are far more restrictive than what would be required under the
normal antitrust laws, and they are intended to ensure multiple
market opportunities.''
Consolidation Rules Ensure Multiple Markets?--Facts
The disparity between this unsupported assertion, and the reality
of the mechanism of the crab plan, was enough to cause the Alaska
Fishermen's Journal to write an editorial titled: ``Earth to Council:
Abort! Abort!''
Dealing with the fallacies of this claim in the Issue Paper
requires a full review of the Analysis and the actual options selected
by the Council.
The NPFMC Analysis of Antitrust Issues--and the NPFMC Motion:
The Analytical Document prepared for NPFMC consideration of the
program developed in the Council motion contains the following brief
reference titled: ``Antitrust Issues Related to the Issuance of
Processing Shares and Regionalization of the Fishery.''
``Generally speaking, congressional action will be needed to
resolve any potential anti-trust issues associated with the
regionalization and processor share provisions. Antitrust
issues, however, could also arise in a harvester IFQ program,
if constraints are not placed on consolidation of QS. These
issues are discussed with more specificity in the appropriate
sections below.''
What are ``the appropriate sections below?''
the section on QS ownership caps and excessive share issues
(3.4.1.2),
the section on processor caps (3.4.2.6),
the discussion of regionalization (3.6), and
the discussion of competition in the fisheries (3.16).
Section 3.4.1.2 Ownership and Use Caps
This section discusses the role of ownership and use caps in
regulating the degree of consolidation of harvester shares (QS, in the
analysis):
``As noted in the NRC study `Sharing the Fish,' ownership and
use caps are generally favored as a means to prevent excessive
shares. . . .'' Further on, ``The options for capping the
ownership of QS [in the] Bristol Bay red king crab, the BS
opilio, the BS bairdi [crab fisheries] . . . are 1, 5 and 8
percent of the QS pool. These caps would limit consolidation of
the fishery to 100, 20, and 13 QS owners, respectively, if all
QS holders own quota amounts equal to the cap.''
The Council's discussion of harvest caps cited the necessity to
start with ``very conservative guidelines.''
NPFMC settled the ``ownership caps'' for these fisheries at
1 percent of total allocated.
``Use caps'' allowing additional quota to be leased, provide
that a total ``use'' (owned and leased) of 2 percent of QS.
However, processors are also allowed to own up to 5 percent
of the fishing quota in a given fishery, despite that all other
owners are restricted to the 1 percent caps earlier referenced.
This ``control on vertical integration'' establishes that
processors are entitled to a larger portion of fishing quota
than any other QS holders.
Section 3.4.2.6 Consolidation
This section discusses the role of ownership and use caps in
regulating the degree of consolidation of processing shares
``Caps on ownership and use of processing shares could be used
to prevent consolidation of market power in a few firms. These
caps might be favored as a means to ensure competition in the
processing labor market. In addition, share concentration could
influence the market power of processors with respect to
harvesters. Harvesters are concerned that if processing shares
become consolidated in the hands of a few firms, those firms
could have the ability to control the ex vessel price of crab.
In addition, caps on ownership could be used to facilitate a
market for processing shares, contributing to entry of
processors to these fisheries.''
A footnote to the analysis states:
``Ownership and use caps, together with an allocation of
processing shares, are the only options that would guarantee a
minimum number of participants in the processing sector.''
Another footnote to the analysis states:
``Some of the companies listed in Appendix 3-3 have common
owners . . . Depending on the rules chosen for determining
ownership for purposes of applying caps, these companies with
common owners might be considered a single entity.''
Application of the Rule
The NPFMC action provides ownership and use caps that do not
guarantee a minimum number of participants in the processing sector.
The Council didn't choose the rule that would implement caps that
``apply individually and collectively.'' The NPFMC chose that ``No
ownership [is] to exceed 30 percent of the total PQS . . . at the
company level.'' In general, this allows an individual entity to own 60
percent of the total PQS pool, provided the ownership is held in at
least two separate companies.
Use Caps
With respect to processing ``use'' caps, the NPFMC motion applies
three rules:
``In the Northern Region, annual use caps will be at 60
percent for the opilio crab fishery.''
``50 percent of the WAI IPQ brown king crab QS shall be
processed in the WAI region.''
``Custom processing would continue to be allowed within this
rationalization proposal.''
Taken together with the ownership cap rules, this means:
While ``ownership'' may not exceed 30 percent of the total
PQS, this calculation is made at the ``company level.'' Thus,
provided companies are not ``merged,'' an individual may
acquire ownership of ALL PQS.
With only two exceptions and there are no ``use caps'' on
IPQS. Thus, all PQS may be processed in one facility, whether
or not ``companies'' are owned in common.
For the Northern Region opilio crab fishery, there must be
at least two processing companies to which harvesters may sell
crab.
However, these companies, may be owned by the same
individual.
Also, these companies may also be allowed to operate
from the same processing facility, if amounts delivered
above the ``60 percent use cap'' are processed according to
``custom processing'' agreements.
For the Western Aleutian Islands IPQ brown king crab
fishery, 50 percent of the catch must be processed in the
region. This IPQ could be processed by the same company.
Other Sections of the Council Analysis
Section 3.16, particularly 3.16.2 which contains a brief survey of
the literature of conflicting accounts of competitive impacts projected
by economists. It includes a summary of a discussion paper prepared for
the NPFMC by economists Milon and Hamilton. They conclude that:
Segmentation of the market limits harvester negotiating
power, and
PQs ``could allow processors to capture efficiency gains
realized by the harvesting sector since harvesters would be
required to deliver harvests to processors holding processing
shares.''
Are Harvesters Willing to Accept Processor Quota as the Price of
IFQs?--Fallacy:
In the Issue Paper, ``Council Action on Processor Shares,'' the
writer asserts that the Council ``asked crabbers which they would
prefer? Answer: the Council motion is preferred to open access.''
Are Harvesters Willing to Accept Processor Quota as the Price of
IFQs?--Fact:
This recollection is inconsistent with the tapes of record.
The Council motion had not yet been made at the time public
comment was taken.
Only one of four harvester associations endorsed Processing
Quota.
The C.R.A.B. Group's testimony to the Council opposed processing
quota and supported rationalizing the fisheries, that we represented
nearly half of the crab harvesters, and that the Council decision would
determine whether opposition to Processing Quota would grow. Opposition
to Processing Quota is growing.
An Unrecognized Conflict of Interest by the State of Alaska
Nowhere in the praises heaped upon the crab plan in the Issue
Papers is there a hint of a direct financial interest in the outcome by
the Alaska Department of Fish and Game [ADF&G]. Perhaps it is through a
lack of self-consciousness by the author of the Issue Papers, but an
undeclared conflict of interest is glossed over. Consider the NPFMC
motion:
``Section 5, Program Elements, Option 5. A proportional share
of fees charged to the harvesting sectors and processing
sectors for management and enforcement of the IFQ/IPQ program
shall be forwarded to the State of Alaska for use in management
and observer programs for BSAI crab fisheries.''
This is likely to be a sum considerably exceeding the amount
presently authorized by the Alaska Legislature. It is certain that
ADF&G needs the money. The fishing industry has supported the budget
requests of ADF&G to the Alaska Legislature for years, only to
experience debilitating cuts, year after year.
Part IV--Our Request to Congress
The Council has recommended two distinct programs to Congress in a
single package, a harvester IFQ and a Processor Quota program. However,
the two programs are not equally well supported by the record.
The Harvester Program
After a day-and-a-half of discussion, the NPFMC chose a suite of
options that determined the harvester elements of the crab plan. These
elements were based upon years of experience with the halibut/sablefish
plan, were selected from a broad array of possibilities, and included
provisions that would mitigate each of the impacts noted by the Select
Panel in ``Sharing the Fish'' including options for skippers and
communities, and the prevention of accumulation of excessive shares of
fishing quota.
The Processor Program
After discussions ranging from little to none, the NPFMC chose a
suite of options that determined a Processing Quota provision. There
were no real alternatives presented, there was no effective cap on
consolidation, no practical experience except for tenuous analogy to
provisions of the American Fisheries Act, theoretic guidance based on
highly controversial and unproven assertions, and no demonstrated
understanding of the anti-competitive aspects of the provision. The
NPFMC crab plan has been billed as bold step into new territory. It is
instead, a reactionary slide into a system that has not been seen in
Alaska since the Seward Purchase--a State sanctioned monopoly of
fisheries resources.
Is the Ultimate Question ``Processors Shares or Nothing?''
Finally, consider the question actually asked of crabbers: ``Which
would you prefer--Processor Quotas, or nothing?'' This is an industry
that is in collapse, and the lives, the vessels and the homes of many
harvesters are right there on the line. Was this not really the
question of Solomon was asked to judge?
The Issue Papers claim that the baby was cut in thirds. Instead,
the NPFMC has given the whole baby to the processors, who declared they
had the power in Congress to block any programs, no matter the cost to
life or resource, to have their way.
Conclusion
We ask Congress to reject the Council's recommendation to
legalize Processor Quotas and by implication undo the antitrust
provisions of existing statutes.
After rejecting Processor Quotas, we hope Congress will use
its wisdom to judge the rest of the Council's program to end
the deadly race for crab, to improve safety and to improve
conservation of the resource, on its merits.
We ask that Congress allow and encourage the Council to
develop a rationalization plan for crab that doesn't include
Processor Quotas.
Senator Stevens. Our next witness is Frank Kelty, the
Natural Resources Manager of the City of Unalaska.
Mr. Kelty?
STATEMENT OF FRANK KELTY, RESOURCE ANALYST,
CITY OF UNALASKA
Mr. Kelty. Good afternoon, Mr. Chairman, Members of the
Committee. Thank you for inviting me to testify before the
Committee.
My name is Frank Kelty. I am the Resource Analyst for the
City of Unalaska. With me from Unalaska today--we have
traveled, I think, it must be about 7,000 miles to get here--is
the Honorable Mayor Pam Fitch, city council members Marcort,
Meeks, and Mr. Graves, and also our City Manager, Chris Latik.
Previously, I worked for 30 years in the Alaska seafood
industry as a crab plant manager for two operations in
Unalaska. I also served as the city's Mayor for 10 years. The
City of Unalaska strongly supports the Bering Sea crab-
management plan because it recognizes the investment of all
participants in the Bering Sea crab fishery, including
harvesters, processors, and communities. My oral testimony will
focus on the following issues--community dependence, safety,
and the lessons we have learned from AFA.
Mr. Chairman, we have a crab resource and industry in the
Bering Sea that is in trouble. That is why we are here today.
The fisheries in the Bering Sea have excess harvesting and
processing capacity, resource conservation and management
problems, lack of economic stability for harvesters,
processors, fishery-dependent communities, as well as major
safety concerns. The crab plan is strongly supported by the
Cities of Unalaska, Akutan, St. Paul, and St. George. These
communities process in excess of 85 percent of the king and
opilio snow crab of the Bering Sea. And I might like to point
out that City of St. Paul passed one of the first resolutions
over a year ago in support of this plan.
Mr. Chairman, I would like to emphasize the following
point. The communities that have been dependent on crab are, in
fact, supporting the plan. The plan, Mr. Chairman, guarantees
that the vast majority of the crab resource will stay onshore
in Alaska. I repeat, this plan is a shore-side plan that
guarantees the resource to Bering Sea communities. This has
long been your stated goal for the Alaska fisheries, and it is
implemented in this crab plan.
The North Pacific Council was extremely careful to
recognize historic dependence and to protect our stake in the
fisheries. Alaskan communities are protected through a
requirement that 90 percent of a processor's history stays in
the community within the region that they earned it and assures
us access to the resource during good times and bad.
The City of Unalaska has seen firsthand the benefits of a
rationalized fishery under the American Fisheries Act. The race
for pollock in the Bering Sea has stopped. We have seen
increased economic stability for pollock harvesters,
processors, and fishery-dependent communities, as well as
longer fishing seasons that benefit both employment in the
industry and in the community, improved resource management,
new product development, improved revenue streams for the
community, and increased safety.
As many of you know, the Bering Sea crab fishery is the
most dangerous in the world. As a crab plant manager, I cannot
count the number of times I have had vessels return to Unalaska
for crab unloading in January and February, March, that were
covered in a foot of ice or more. Sometimes the boats returned
to port having had two or three pilothouse windows blown out by
rogue waves that destroyed all their electronics. Fortunately,
the skipper can navigate blind for 2 days. I have had to make
phone calls to family members to tell them that their loved
ones were lost at sea during the crab season.
This crab plan will stop the race for fish by giving
harvesters the flexibility to pursue crab through longer
fishing seasons. Harvesters will be able to base their
decisions on accommodation of market, fishing, and weather
conditions, rather than face the pressure to fish in bad
weather because of such compressed seasons, as is certainly the
case. There is no question in my mind that this plan will help
save lives and reduce injuries.
As you know, there is a small, but vocal, opposition from
industry participants and community, most notably the City of
Kodiak. Their opposition is primarily based by the plan's
allowance for processing quota shares and on the years used to
determine processing and fishing history under the plan. These
opponents, by and large, reduced their participation in the
crab fishery some time ago rather than remain in the fishery,
like Unalaska and others that ride through the ups and downs.
They made their choice of their own free will and opted to
diversity into other fisheries, fisheries that they have been
successful at and have benefited from. Yet now they ask to be
held harmless at the expense of those who kept their commitment
to the crab fishery.
The Council has ensured that 10 percent of the fishery will
remain open access, so they will have an opportunity to
participate in the fishery. And the harvesting of processing
shares that they do hold from the late 1990s will have
significant value.
The Council, after careful deliberation, ratified the plan
by a unanimous vote--ratified the plan by a unanimous vote--and
followed up with the adoption of strong community-protection
measures. We believe that the final plan treats Kodiak and a
few other opposing communities fairly.
In summary, Mr. Chairman, we believe that this plan was
crafted by the Council in a deliberate, balanced, and open
fashion that has taken almost 4 years to complete. The plan has
measures that protect and enhance economic activity for both
harvesters and processors that have invested hundreds of
millions of dollars in their operations in Bering Sea
communities. So I would encourage you to move forward with
legislation to implement its provisions, and in a timely
manner.
Thank you for allowing me to testify, and I look forward to
answering any questions you may have.
Senator Stevens. Thank you very much.
[The prepared statement of Mr. Kelty follows:]
Prepared Statement of Frank Kelty, Resource Analyst, City of Unalaska
Mr. Chairman, my name is Frank Kelty, and I am the resource analyst
for the City of Unalaska. Unalaska strongly supports the Bering Sea
Crab Management Plan developed by the North Pacific Council. We have
submitted a written statement and supporting materials for the record.
My oral testimony will focus on the following issues: (1) community
dependence; (2) safety; and (3) the lessons we have learned from the
American Fisheries Act.
Mr. Chairman, we have a crab resource in the Bering Sea/Aleutian
Islands that is in trouble, and that is why we are here today. The
fisheries of the Bering Sea/Aleutian Islands have excess harvesting and
processing capacity; resource conservation and management problems;
lack of economic stability for harvesters, processors, and fishery-
dependant communities; as well as major safety concerns.
Community Dependence
The crab plan is strongly supported by the cities of Unalaska,
Akutan, St. Paul, and St. George. These communities process in excess
of 85 percent of the king and opilio crab in the Bering Sea. Opilio is
also referred to as snow crab.
Mr. Chairman, I would like to emphasize the following point: the
communities that have become dependent on crab are, in fact, supporting
the plan. The plan, Mr. Chairman, guarantees that the vast majority of
the crab resource will stay on-shore in Alaska. I repeat; this plan is
a shoreside plan that guarantees the resource to the Bering Sea
communities. It cannot be shifted to offshore catcher-processors, and
it cannot be taken out of state for primary processing. This has been
your long stated goal for the Alaskan fisheries, and it is implemented
by this crab plan.
The North Pacific Council was extremely careful to recognize
historic dependence and to protect our stake in the fisheries. Unalaska
believes that the plan itself is a community protective measure because
it assures that 90 percent of the processors' history stays in the
community within the region and assures us of the resource during good
times and bad.
Community opposition to the plan is coming from Kodiak. They feel
that the plan will impact their revenues, local processing plants, and
the 35 resident crab vessels. Kodiak was a major player in the crab
fisheries of the Bering Sea, but that was over 20 years ago, before the
development of the processing plants in the Bering Sea communities.
Between 1997 and 1999 (the qualifying years), Kodiak's historical
landings from the Bering Sea were very small, at less than 1 percent
for opilio snow crab and 6 percent for Bristol Bay red king crab.
Kodiak is 600 miles from Unalaska, and we are another 200 miles
from the king crab fish grounds and further yet from the major opilio
grounds. That distance means many days of running time to take crab
back to Kodiak. Harvesters have to worry not only about weather, but
deadloss of crab at unloading due to the length of time that the crab
has been on board the vessel.
Kodiak's vessels are protected under this plan. Because a good
harvesting history assures they will have great value, and Kodiak's
processors are granted their processing history. Kodiak's processing
companies will be able to compete for an additional ten percent of the
resource that has been set aside as open access. We believe that Kodiak
has been treated fairly by the North Pacific Council.
The City of Kodiak has requested changes to the plan to give their
community a greater stake in the fisheries. The crab allocation is a
zero sum game. Any increased allocation to Kodiak comes out of the fish
to be processed by Being Sea communities that have a far greater
dependence on the crab resources of the Bering Sea.
AFA
Unalaska has seen first hand the benefits of a rationalized
fishery. Under the American Fisheries Act, which rationalized the
Bering Sea Pollock fishery and is a far more restrictive plan than the
BSAI crab plan, the race for fish was stopped. We have seen increased
economic stability for harvesters, processors, and fishery dependant
communities, as well as longer fishing seasons that benefit both
employment in the industry and the community. We have seen improved
resource management, new product development, and increased safety.
With a rationalized Bering Sea/Aleutian Island crab fishery, we will
see many of the same benefits in all sectors of the Bering Sea/Aleutian
Island crab industry.
Safety
I have worked in the crab industry for over thirty years, and I
know that Bering Sea crab is the most dangerous fishery in the world. I
cannot count the number of times I have had vessels return to Unalaska
for unloading covered in a foot of ice in January or February.
Sometimes the boats have returned to port having had two or three
pilothouse windows taken out by rogue waves that destroyed all of their
electronics, forcing the skipper to navigate blind for two days and
barely making back to port. I have had to make phone calls to family
members to tell them their loved ones were lost at sea during the crab
fishery. This crab plan will help put a stop to that madness by making
the fishery safer. The rationalized Pollock fishery in the Bering Sea
has worked wonders for safety. Now, the Pollock fleet can wait out
major storms. The North Pacific Crab Plan affords this same safety net
to crab fishermen
In summary, Mr. Chairman, we support the public process that this
issue has gone through with the North Pacific Council, and we support
the council process. This plan has the support measures that protect
and enhance economic activity for crab industry and the communities of
the Bering Sea. The North Pacific Council's Crab Rationalization Plan
is vital to the long-term stability of the communities of the Bering
Sea that have become dependent on this resource.
Thank you for allowing me the opportunity to testify.
______
Prepared Statement of Frank Kelty, Resource Analyst,
City of Unalaska, Alaska
Mr. Chairman and Members of the Committee:
For the record, my name is Frank Kelty; I am the Resource Analyst
for the City of Unalaska. Previously, I worked in the Alaska Seafood
industry for 30 years in Unalaska, Alaska, as a manager for two
processing companies running crab operations. I also served the
community of Unalaska as and elected official for 19 years, until
December of 2000, when I resigned to work as Resource Analyst. The last
10 years of my time on Council, I served as the Mayor of Unalaska. I'm
here today to testify on behalf of the City of Unalaska in support of
the North Pacific Fishery Management Council's Bering Sea/Aleutian
Island Crab Rationalization Plan.
Included in your packet of information is the City of Unalaska
Resolution 2003-27 that was passed unanimously by the Unalaska City
Council in support of North Pacific Council's Bering Sea/Aleutian
Island (BSAI) Crab Rationalization plan. Also included are resolutions
and letters from other Bering Sea communities that support the plan.
Those communities include the City of Saint Paul, City of Saint George,
and the City of Akutan. These communities, as well as Unalaska, process
in excess of 85 percent of the King and Snow Crab harvested in the
Bering Sea Crab fisheries. The City of Unalaska is not only the largest
crab processing community in the State of Alaska, but the nations
number one commercial fishing port as well.
The BSAI rationalization plan has been developed in a very open
process. After three years of committee meetings and discussion by
harvesters, processors, community representatives, and members of the
public, came two years of intense discussion, analysis, and development
by the North Pacific Council, its staff, and the council's industry
advisory panel, with assistance from NMFS, ADFG, and independent
economists and fishery consultants. The plan was finally adopted by a
unanimous vote of the North Pacific Council 11-0 in June of 2002.
The current state of the Bering Sea crab fisheries is poor at best.
Included with the information in the packet, I have provided a
spreadsheet created by the Alaska Department of Fish and Game that that
reflects the recent history of the Bering Sea Fisheries covered under
this plan. As you can see, many of these fisheries have been closed for
years, and three of them have been under rebuilding plans for the many
years. The ones that have been open have had very small harvest quotas
and a very limited amount of fishing time, with a large number of
participants in these un-rationalized fisheries.
This situation has led to economic instability in the crab industry
as a whole and has caused major revenue shortfalls for many of the
fishery-dependent communities in the Bering Sea/Aleutian Island area,
especially smaller communities such as Saint Paul and Saint George that
depend almost exclusively on the harvesting and processing revenues
generated from the crab fisheries of the Bering Sea. With the continued
race for fish in the BSAI crab fisheries, we have witnessed the
continuous struggle that the State of Alaska fishery managers have in
managing an un-rationalized fishery. A good example is the fact that
the 2003 Opilio Snow Crab fishery might not have opened had a new
regulation from the Alaska Board of Fish not been granted to enact a
further reduction in the number of pots that harvesters could use
during the season. Without that reduction in fishing gear, we may not
have had a season, which would have been devastating to the harvesters,
processors, and fishery-dependent communities.
With the gear reduction regulation in place, we had a season that
lasted ten days, and a fleet of almost 200 vessels harvested 26 million
pounds of snow crab. This October, we will once again have another
derby-type crab opening on the Bristol Bay Red King Crab fishery. It
may last 4 or 5 days with a limited amount of harvest quota. This
fishery will attract well over 200 vessels. As a consequence, for the
2003 crab fishing season, the majority of the Bering Sea Crab fleet
will have a total of 14 or 15 fishing days. This type of scenario has
been ongoing on for years. I think this points out the crisis situation
this industry is in right now and why the Bering Sea Aleutian Island
Crab rationalization plan needs to move forward and be implemented as
soon as possible.
This plan, which the City of Unalaska supports first and foremost,
recognizes the investment of all participants in the Bering Sea crab
fisheries, including the harvesters, processors, and communities. My
community has seen first hand the benefits of a rationalized fishery
with Bering Sea Pollock fishery under the American Fisheries Act. Under
(AFA) the race for fish was stopped. We have seen a reduction in the
size of the fleet, as well as longer fishing and processing seasons
that benefit both employment in the industry and the community. It has
also provided economic stability to harvesters, processors, fishery-
dependent communities, and support sector business.
Under AFA, we have seen improved resource management, new product
development, and increased safety for the fleet. Included in the packet
of information provided, you will see some revenue graphs that point
out the increase in many of the City of Unalaska fishery-related
revenue streams since AFA was enacted. With a rationalized BSAI crab
fishery, we will see many of the same benefits in all sectors of the
industry. The following is an overview of some of the main components
of the plan that we support and that need to be pointed out.
Harvester Sector
Harvesters would be allocated quota shares (QS) in each fishery
rationalized by the program. Harvester shares will be allocated in two
classes of shares: Class A shares at 90 percent, and class B shares at
10 percent. Class A shares carry the regionalization tag and would have
to be delivered to the designated region. Class A shares of crab would
have to be sold to any processor in that region who has processor quota
shares. The 10 percent, class B shares do not have any regionalization
tag and can be sold to any processor in any region.
Quota shares and IFQs would both be transferable under the program,
subject to limits on the number of shares a person may own or use.
Leasing of quota shares may be prohibited, except within a cooperative,
after the first five years of the program. The transferability is
necessary to reduce fleet size and remove capital from the fisheries
under this plan. The limit on leasing of QS or sale of IFQs by persons
not in cooperatives is intended to create an incentive for cooperative
membership.
The Captains share allocation of 3 percent quota shares are open
shares for the first three years of the program and will be reviewed by
the North Pacific Council, which will decide whether to make the shares
either class A or a class B designation. A crewmember crew-loan program
will be developed to assist crewmembers' entry into the crab fisheries.
This program would permit harvesters to form voluntary cooperatives
with one or more processors holding processing quota shares (PQS).
Cooperatives are intended to facilitate efficiency in the harvest
sector by aiding harvesters in coordinating harvest activities among
members of the cooperatives and deliveries to processors. A minimum
membership of four unique QS holders would be required for cooperative
formation. Under AFA, Pollock cooperatives have so effectively
coordinated the harvest that less than 1 percent of the TAC is un-
harvested. We have also seen formation of profit sharing agreements
between harvesters and processors under AFA that have been very
successful in maximizing revenues for both sectors. I believe we would
see the same types of programs under a rationalized crab fishery.
Harvesters also have a binding arbitration program in place, which
will be used to settle price disputes between processors.
Processing Sector
The allocation of processor shares is easily the most controversial
part of the plan. However, their contribution and investments in the
BSAI crab fisheries cannot be ignored, as they are the economic anchor
for fishery-dependent communities in remote locations in the Bering Sea
area. Their investments in Unalaska are in the hundreds of million of
dollars. They provide employment and markets for a wide variety of
harvesters in different fisheries, and the importance of the revenue
they generate in the millions of dollars annually for both for local
governments and the State of Alaska cannot be understated.
The processing sector will be allocated PQSs in each fishery
rationalized by the program. These annual allocations of processing
privileges are referred to as Individual Processing Quotas (IPQs). IPQs
would be issued annually for 90 percent of the allocated harvests
corresponding to the 90 percent allocation of class A harvester shares.
Leaving the remaining 10 percent of processing unallocated and,
therefore, deliverable to any processor is intended to strike a balance
of bargaining power between the harvesting and processing sectors. In
addition, this 10 percent unallocated processing amount would allow for
new entry to that sector.
PQS allocations would be based on processing history during a
specified qualifying period for each fishery. A processor's allocation
in a fishery would be equal to the processor's share of all qualified
processing in the qualifying period. All allocations will be made to
the buyer of record on ADFG fish tickets, by the State of Alaska
Commercial Operators Annual report, and by fish tax records. This rule
reflects the intention to allocate shares to the entity that purchased
the crab and funded the processing activity.
Processor shares would be transferable, including leasing of PQS,
subject only to use and ownership caps, and restrictions that apply
during the two-year cooling-off period.
Ownership of PQS will be limited to 30 percent of the outstanding
PQS in a fishery. An exception to this would be in the Northern Region
where PQS shares in the Opilio Tanner fishery would be capped at 60
percent due to the limited number of processors in that region.
Community Protection
This is a very important part of the crab rationalization plan. I
maintain that it shows that the importance of the processors having
quota shares is their designation by region. The most important part of
the community protection provision is the regional designation of
shares, which will apply to processor allocations north of 56 20, N
Latitude. The South region will be all areas not included in the North
region, and all the corresponding 90 percent of the harvest
allocations, distributing landings and processing between specific
regions. Included in the handouts, are spreadsheets and graphs that
show landings, State of Alaska fish tax information for Opilio Snow
crab between the North and South regions, and community designations. I
have also included landings by community and percentages of landing
over a four-year period for Bristol Bay Red King Crab. These are
currently the two major Bering Sea fisheries. This information clearly
shows the connection between the communities of the Bering Sea and the
dependency of these communities on the crab fisheries of this area, and
it shows the importance of the processing plants in their communities.
It also points out the lack of dependency by some communities, such as
Kodiak, which is located in the Gulf of Alaska, over seven hundred
miles from the nearest Bering Sea crab fishery.
For the first two years of the plan, a cooling-off period is
currently in place. During this time, the processing quota earned in a
community may not be used outside of that community.
First Right of Refusal options apply for communities with at least
3 percent of the initial PQS allocation in any BSAI fishery, based on
history in a community. Exception will be made for those communities
that receive a direct allocation of any crab species; currently, the
only exception is Adak. This right allows qualified communities, or
community groups, and CDQ groups a first right of refusal to purchase
processing quota shares that are based on the original processing
history of the community.
Safety
This is one area under a rationalized fishery in which we will see
major improvements. Fishing in Bering Sea Crab fisheries, as many of
you know, has been labeled the most dangerous profession in the Nation.
Approximately 25 fishermen lost their lives in the Bering Sea during
the 1990s. This rationalization plan will provide a safer fishery in
much the same way as we have seen in the Halibut and Sablefish IFQ
fishery and the AFA rationalized Pollock fishery in the Bering Sea. In
my previous employment in the crab industry, I saw first hand the havoc
that the Bering Sea can wreck on people and on vessels in an un-
rationalized fishery. I cannot count the times I have had vessels
return to Unalaska for unloading covered in a foot of ice in January or
February. Sometimes they have returned to port having had two or three
pilot house windows taken out by rogue waves, and had all there
electronics and navigation gear destroyed and had to hand steer for two
days, barely able to make it back to town. I only had to deal with the
ultimate tragedy once, and that was enough for a lifetime. Having to
deal with the tragedy of a vessel lost with all hands, and having to
tell loved ones and friends that their family member or friend will not
be coming back was one of the toughest things I have ever had to deal
with in my life. The improvement that we will see in safety with a
rationalized fishery is something we cannot put a dollar value on, and
we should never forget that.
There are those who oppose this plan. Many of them do not have any
involvement with the crab fisheries of the Bering Sea, but are worried
that sometime in the future elements of this plan could impact their
fishery. We heard the same concerns during the implementation of the
American Fisheries Act (AFA) by the same folks that weren't involved in
the Pollock fishery, that it would cause problems for them in their
fisheries, but it has never happened. There are crab harvesters that
also do not support the plan. Many of their concerns might be due to
the qualifying years that are in the plan. Some vessels do not have
very much poundage or history during those years. These people might
hope to keep the derby fisheries going, thinking they may be better off
financially. We have heard the concerns that with processor shares,
their ex-vessels prices will be impacted. However, under AFA Pollock
cooperatives we have not seen that. What we have seen is ex-vessel
prices increase for the harvesters with profit sharing agreements and
other incentive programs.
We also have strong opposition to the plan from Kodiak. They feel
the plan will impact their revenues, their local processing plants, and
the 35 vessels that are home ported there. At one time, Kodiak was a
major player in the crab fisheries of the Bering Sea, but that was over
20 years ago, before the development of the processing plants in the
Bering Sea communities. Between 1997 and 1999, which were the
qualifying years in the plan, Kodiak historical landings from the
Bering Sea are very small, at less than 1 percent for Opilio Snow crab
and 6 percent for Bristol Bay Red King crab. That does not show much
dependency on the major crab fisheries in the Bering Sea, especially
when compared to the communities located in that area. This is due, in
part, to the distance they are from the crab grounds of the Bering Sea.
Kodiak is six hundred miles from Unalaska, which is another 200
miles from the King crab fishing grounds and further yet from the major
Opilio snow crab fishing grounds. That distance means many days of
running time to take crab back to Kodiak. Harvesters have to worry not
only about weather, but deadloss of crab at unloading due to length of
time that the crab has been on board the vessel. In addition, they have
to deal with the fishing gear that they may have left on the fishing
grounds, and they have increased cost of expenses just to make that
run.
I believe their current historical crab landings can be met under
this plan through the processor quotas shares that the local processing
plants will receive and through the open access 10 percent share of the
quota for the harvesters that can be delivered to any processor in any
community.
Unalaska supports this rationalization plan, as do other
communities located in the Bering Sea/Aleutian Islands, and the crab
processors and many of the harvesters that fish these fisheries. We
believe that the rationalization plan as proposed will address many of
the serious concerns facing all sectors involved in the crab industry
of the Bering Sea/Aleutian Islands. In Unalaska, we will see many of
the same benefits in a rationalized crab fishery that we have seen in
the rationalized AFA Pollock fishery. We support the public process
that this issue has gone through with the North Pacific Fishery
Management Council, and we support the council process. This plan has
the support measures that protect and enhance economic activity for all
sectors in the community of Unalaska. There is no doubt that everyone
realizes that our mainstay for continued success as a City is linked to
the fisheries from which we gain economic resources.
In closing, I would like to read a paragraph from North Pacific
Council Chairman Dave Benton's August 5, 2002 letter to Congress that
was attached to the North Pacific Council report to Congress on the
Bering Sea and Aleutian Islands Crab Rationalization Program:
``This program is certainly not with its controversy. The
adoption by the Council of processing quota shares as a
fundamental part of the program is probably the most
controversial aspect of the program. However, the Council
believes, as is reflected in its unanimous vote, that the crab
fisheries in the Bering Sea/Aleutian Islands require this
innovation, comprehensive management approach to adequately
recognize and protect the interests of all participants. It
recognizes all components of the fishery as a balanced,
inextricably linked system, rather than individual, competing
components. It may not be the appropriate model for other
fisheries. We do believe it is the appropriate management
approach for this fishery, and we respectfully submit that
Congress should allow for such regionally tailored approaches
in the management process. All Councils need such flexibility
as we consider development of rationalization programs for
other fisheries, for the benefit of all user groups, and to
sustain our precious fisheries resources for the Nation.''
Thank You.
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
Senator Stevens. Our last witness is Arni Thomson,
Executive Director of the Alaska Crab Coalition, Seattle,
Washington.
STATEMENT OF ARNI THOMSON, EXECUTIVE DIRECTOR, ALASKA CRAB
COALITION
Mr. Thomson. Good afternoon, Mr. Chairman and Distinguished
Members. On behalf of the Alaska Crab Coalition, I would like
to thank you for the opportunity to provide testimony on this
vitally important subject of crab rationalization. I am Arni
Thomson, Executive Director of the Alaska Crab Coalition.
Mr. Chairman, the ACC speaks on this subject from long
experience and deep conviction. Established in 1986, the ACC is
the longest-standing organization of Bering Sea crab fishing
vessel owners. I think it fair to say that among the several
organizations of Bering Sea crab harvesters, the ACC has been
the most consistent contributor to the efforts of Congress, the
Commerce Department, and the North Pacific Fishery Management
Council to provide for the improvement of our fisheries.
The ACC applauded your inclusion in the Consolidated
Appropriations Act of 2001 a directive that the North Pacific
Council consider rationalization of the Bering Sea crab
fisheries, including individual fishing quotas, processing
quotas, and community quotas, and to report back to Congress on
the outcome. This directive placed the Bering Sea crab
fisheries on the Council's priority list and guided the ensuing
analysis and debate in a direction ensured that all interested
parties a place at the table. Most importantly, this directive
implicitly recognized that all three major groups--harvesters,
processors, and communities--would have to be accommodated in
order to ensure the viability of any rationalization plan.
Mr. Chairman, the Council's plan represents a fair balance
of the effected interests, a compromise that promises to
achieve, for the first time, the fundamental goal of ending the
race for crab that has killed our fishermen, accelerated
bycatch and discards to damaging levels, and devastated the
economics of the fisheries for harvesters, processors, and
communities alike.
I would like to just stop briefly to point out that at the
beginning of our rationalization efforts, that the industry
took a real hard look at the AFA model in order to try to
emulate the AFA model in the crab fisheries. However, there
certain distinct characteristics between the fisheries, the
fisheries being unique, it just would not work. Basically, the
pollock fishery is a single-species fishery. There are long-
established one-on-one market relationships and contracts
between fishermen and specific processors. However, in the crab
fisheries, we have multi-species fisheries when the fishermen
oftentimes have multiple markets, and the fisherman very
expressly stated that they did not want a closed class of
processors.
So this gravitated toward a request from the fishermen to
the processors to produce a proposal of matching processor
quota shares for the fishermen's IFQs, and that is how we got
started down the road of the processor shares. And then we
added to that the regionalization concept which was proposed by
the City of St. Paul in the Pribilof Islands.
Going on, the following are six key elements of the plan.
Quotas for fishermen, processors, and communities that will end
the race for crab, reduce bycatch and waste, and promote
sustainability of the resources and economic stability for the
industry and communities, improve safety by ending the derby
fishery, preservation of the regional distribution of economic
activity through regional share designations of harvester and
processor quotas that will distribute landings and processing
between specific regions, plus additional protections.
Mr. Fraser mentioned to arbitration programs that were
presented to the Council, one by the harvesters and the other
by the processors. This is not entirely correct. The proposal
that was developed, primarily by the crab group, of whom he
represents, was a flawed proposal that did not commit the
fishermen to a binding arbitration program. So, in good faith,
the Alaska Crab Coalition could not support that program,
because we had initially proposed an arbitration program, and a
binding one. It was the only way that the arbitration would
work.
Having said that, a mandatory binding-arbitration program
to settle price disputes between harvesters and processors and
to ensure competitive market prices with a firm Council
commitment to change the program if it does not work. This
commitment is contained in the May 6th letter of the North
Pacific Council to the Congress.
Going on, initial harvest-share allocations to captains of
3 percent of the tack and the opportunity for skippers and crew
to purchase shares with a low-interest loan program, an
important precedent.
Comprehensive data collection and program review to assess
the success of the rationalization program and to provide
oversight of revenue share ration between harvesters and
processors.
Processor quotas are, as we all know, the main point of the
controversy. Some individuals and organizations would have the
Council or Congress throw out processor quotas and leave it at
that. But, Mr. Chairman, that would be repudiating the very
directive that Congress enacted into law and that Council
respected.
As we all know, Congress did not establish any priority
among the effected groups or among the various management
measures to be considered. Why not throw out harvester
community quotas and leave it at that? The question is
rhetorical, and the answer is obvious. Because harvesters,
communities, and processors all have the vital interests at
stake and the right not only to be considered, but also to be
accommodated in any rationalization plan.
In closing, Mr. Chairman, I ask that you take account of
the considered judgment of the ACC and other organizations that
believe that the plan developed by the Council family is the
only viable approach to restoring our Bering Sea crab fisheries
in a fair, balanced, and effective way. I respectfully urge
that you and your colleagues in Congress grant the authority
that remains necessary to bring this critically needed plan to
fruition.
Finally, at the suggestion of the North Pacific Council, in
its letter of May 6, 2003, we request that the Congress
explicitly authorize and direct the Council to take immediate
action to change the arbitration program if it does not work.
That concludes my comments, Mr. Chairman. Thank you very
much.
[The prepared statement of Mr. Thomson follows:]
Prepared Statement of Arni Thomson, Executive Director,
Alaska Crab Coalition
Senator Stevens, Senator Murray, Distinguished Committee Members:
The Alaska Crab Coalition (``ACC''), a trade association
representing the owners of Bering Sea crab fishing vessels, as well as
service and supply companies in the fishing industry, is grateful to
have been invited to testify at this important hearing on the Bering
Sea and Aleutian Islands (``BSAI'') crab rationalization plan
(``Plan'') adopted by the North Pacific Fishery Management Council
(``Council'') at its June 2002 meeting. The ACC strongly supports the
Plan.
Since its inception in 1986, the ACC has worked closely with the
Council, the Department of Commerce, and Congress in the development
and implementation on an array of statutes, regulations, and policies
aimed at the improvement of safety, conservation, efficiency, and
fairness in the BSAI crab fisheries. The challenges have been enormous:
the highest occupational fatality rate in the Nation, resources in
severe difficulty, the industry on its financial knees, and communities
at serious economic risk. What sets the Plan apart from all previous
management responses, and what promises to deliver the long-sought
after solutions, is its comprehensive approach to addressing the root
cause of the problems plaguing these fisheries--the race for crab.
Through implementation of the Plan, excess harvesting and processing
capacity will be removed from the BSAI crab fisheries in a way that
will be fair to harvesters and processors, alike, and will avoid
economic dislocation of dependent communities. Through a carefully
balanced system of harvester and processor quota shares and regional
delivery provisions, a sustainable equilibrium of production capacity
and resource availability will be achieved, markets will be stabilized,
safety will be improved, and communities will be protected.
Clouds of paper from one disgruntled special interest group or
another must not be allowed to obscure the fact that the Plan
represents the best collective judgment of the Council family, based on
the wise guidance of Congress. Blaring opposition must not be permitted
to drown out the chorus of support for this excellent Plan.
Background and Need for BSAI Crab Rationalization
The BSAI crab fisheries have long presented daunting challenges to
fisheries managers, our industry, and dependent communities. Safety
concerns have necessarily attended fishing operations in the extremely
harsh natural environment of the Bering Sea and Aleutian Islands area.
Conservation became an issue, as soon as major fishing fleets began to
exploit the resource. Allocation issues arose for our fishermen, when
we first sought to ``Americanize'' the fisheries, by wresting control
from the foreign fleets, and later, after such issues arose again, when
that goal was achieved and our domestic harvesting exceeded the
available resources.
As communities became dependent upon BSAI harvesting and
processing, the scope and complexity of economic and social issues
greatly increased. The full spectrum of these challenges became less
and less manageable, as BSAI crab resources suffered declines and
failures under enormous fishing pressures.
Following much debate and the rejection of a harvester-only
individual quota program, a license limitation program (``LLP'') was
adopted in 1995 and implemented in 1998, with the objective of slowing,
if not halting, increased harvesting capacity in the fisheries. Of
course, this was only a halfway measure, as it failed to prevent
``capital stuffing,'' that is, additional investments increasing the
efficiency of the limited number of vessels that were permitted to
operate in the fisheries. Limits on the number of pots per vessel and
various other management measures, including time and area closures,
also failed to solve the fundamental problem of excessive harvesting
capacity. The race for fish intensified.
In the superheated race for crab, these measures had perverse
safety, conservation, and economic effects. Crab pots are designed to
``soak'' for long enough to allow all the bait to be consumed, and for
the juveniles to leave, through escape panels, in search of other
forage. Fishing seasons comprised of a few days, coupled with pot
limits, led to a spiral of increased risk to the safety of fishermen
and to the sustainability of the resources, as frantic efforts were
made to maximize the numbers of pot lifts in short seasons. In these
circumstances, juvenile crab feeding on bait, would still be in the
pots at the time they were lifted, and a high percentage of juveniles
would perish, as a result of the changes in temperature, when they
ascended and descended the water column. The future of the crab
fisheries was dying with its juveniles. Many independent vessel owners
were left hanging precariously on the brink of bankruptcy. Worst of
all, the BSAI crab fishery remained the most dangerous occupation in
the United States.
In 1996, while the LLP was wending its way through the bureaucracy
toward implementation, the Sustainable Fisheries Act was enacted. It
included two measures first proposed by the ACC, new national standards
to limit and reduce bycatch, and to improve safety, and a third measure
supported by the organization, authority for the Federal Government to
conduct industry-funded fishing capacity buybacks. However, to the
disappointment of the ACC, the Act also included a four-year moratorium
on new individual fishing quotas.
Bering Sea pollock took center stage in the North Pacific, and in
October 1998, the American Fisheries Act established a unique system of
harvester/processor coops for that fishery, including a 90/10 formula
for mandatory deliveries to exclusive processors. Most of the Council's
time during the ensuing 18 months was consumed with resolving those
issues left to its jurisdiction by the new law.
During the year 2000, the crab industry considered various forms of
coops and a buyback. However, these potential management responses to
the crisis in the BSAI crab fisheries achieved a critical mass of
support.
At the close of the year 2000, the moratorium on individual fishing
quotas was extended for an additional two years. However, in the
Consolidated Appropriations Act of 2001 (P.L. 106-554), Congress also
enacted special legislation that served as a guidepost for future BSAI
crab management:
. . . The North Pacific Fishery Management Council shall
examine the fisheries under its jurisdiction, particularly the
Gulf of Alaska groundfish and Bering Sea crab fisheries, to
determine whether rationalization is needed. In particular, the
North Pacific Council shall analyze individual fishing quotas,
processor quotas, and quotas held by communities. The analysis
should include an economic analysis of the impact of all the
options on communities and processors as well as the fishing
fleets. The North Pacific Council shall present its analysis to
the appropriations and authorizing committees of the Senate and
House of Representative in a timely manner.
In January of the following year, the Council formally constituted
a 21-member Crab Rationalization Committee that represented all
affected interests, including the crab industry organizations,
dependent communities, and the environmental community. The work of
that committee culminated on March 23 of that same year with
endorsement, by a two-thirds vote, of a system that would provide
quotas for both fishermen and processors and regionalized landing
requirements. This served as the basis for the Council's eventual
adoption of a ``three-pie voluntary cooperative program.''
On June 10, 2002, the Council adopted the Plan by a unanimous vote
of 11-to-0. The very fact that the long public debate leading up to
this decision was spirited and even rancorous at times demonstrates
that the Council proceedings were a model of public participation, with
input received from every party who had a perspective to bring to the
table. There were countless hours of deliberation in the Council and
its committees, as well as within and among interested and affected
individuals and organizations over a period of more than two-and-a-half
years. Anyone who failed to offer his or her views cannot claim a lack
of opportunity to have participated in the process.
There was, it is true, a last-minute disagreement over a system of
arbitration designed to resolve price disputes. No organization was
more concerned than was the ACC, which withdrew support for the Plan,
pending the outcome of efforts to resolve the crisis. Fortunately, the
ACC was able to support the end-product, based on the expectation that
the Council and Congress would critically and continually review the
operation of the arbitration process, and that the Council would make
changes, if that proved necessary to assure fairness. This expectation
was proved correct, when the Council submitted its May 6, 2003, report
to the Congress, with the following statement concerning arbitration:
If the preferred arbitration program does not function as
intended, the Council is committed to using a different
arbitration structure to provide a fair price setting
environment. Because of the completed analyses of these
different structures, an alternative structure, such as the
``Steele Amendment,'' could be expeditiously adopted as part of
the binding arbitration program should Council review of the
program suggest that the arbitration program is not working as
intended. If Congress approves this program, such explicit
authority could be provided to the Council to ensure timely
action to address problems that might arise . . . We hope that
Congressional authorization of the program will provide
explicit direction to the Council concerning its obligation to
review and amend the program should any unanticipated negative
impacts arise.
The Plan
While there have been concerns that the Plan somehow establishes
unsuitable precedents for other fisheries, the fact is that it responds
in a tailored way to a unique combination of circumstances:
Horrendous weather and ice problems on the fishing grounds,
resulting in the highest occupational fatality rate in the
Nation.
Extreme over-capitalization in both the harvesting and the
processing sectors.
Heavy economic and social reliance of five communities,
located in two regions, on crab production.
Unstable and declining crab resources, and excessive bycatch
waste.
Foregone fishing opportunities, due to inability to manage
small resources.
The Plan responds, in a sustainable, fair, and balanced manner, to
the complex resource, environmental, economic, social, and safety
challenges confronting stakeholders in the major BSAI crab fisheries:
Vessel owners;
Skippers and crews;
Processors;
Communities;
The public at large.
To achieve this goal, the Plan contains the following primary
elements:
Harvest shares allocated to fishermen for 100 percent of the
total allowable catch (TAC), with 90 percent of those shares to
be delivered to processors holding processing shares, and the
remaining 10 percent to be deliverable to any processor.
Processing shares allocated to processors for 90 percent of
the TAC.
Regional share designations for processor allocations and
the corresponding 90 percent of the harvest allocations,
distributing landings and processing between specific regions,
plus additional community protections.
A mandatory binding arbitration program to settle price
disputes between harvesters and processors and to insure
competitive market prices.
Voluntary harvester cooperatives permitted to achieve
efficiencies through the coordination of harvest activities and
deliveries to processors.
Community Development Quota allocations of 10 percent of the
TAC.
Initial harvest share allocations to captains of 3 percent
of the TAC, and the opportunity for skippers and crew to
purchase shares.
Low-interest Federal loan program for captains and crew to
purchase harvest shares.
Comprehensive data collection and program review to assess
the success of the rationalization program and to provide
oversight on revenue share ratio between harvesters and
processors.
The Plan presents an impressive array of improvements over the
prevailing situation. These included:
Biological Benefits
Improved stock management through use of a TAC.
Reduced overharvests through individual allocations.
Reduced discards resulting longer soak times and better
sorting of undersized crab through escape mechanisms in gear.
Improved handling of discards by ending the race for crab.
Economic Benefits
Compensated reductions in capitalization through voluntary
share transactions.
Economic stability for the harvesting and processing sectors
and communities.
Social Benefits
Preservation of regional distribution of economic activity.
Facilitated entry to the fishery for crew.
Protection of historical interests of captains.
Safety Benefit
Improved safety by ending the derby fishery.
Conclusion
The ACC commends Senator Stevens for his foresight in authoring the
statutory guidance that was essential to the Councils' work in
developing the Plan. We likewise commend Senator Murray for her
contributions to strengthening the statutory framework for management
of our crab fisheries. We are deeply grateful for the impetus provided
by Congress for us to overcome the crisis in our fisheries, and we
respectfully request that the requisite, additional statutory authority
be provided as soon as possible for implementation of the crab
rationalization plan crafted by the Council family.
Thank you.
______
Addendum to the Statement of Arni Thomson, Executive Director, Alaska
Crab Coalition
I. The unique differences between the Bering Sea Pollock fishery
and the crab fishery, and the alternatives for rationalization that led
to development of the three-pie program with harvester, community, and
processor shares:
Rationalization solutions for processors, as well as
fishermen, begin with restricting entry. The pollock model
found in the American Fisheries Act of 1998 (AFA) was
specifically designed for the pollock industry, where stocks
were abundant, and long term single market contracts were in
effect, particularly within the onshore sector, when the
measure was enacted. The number of participants in both the
harvesting and processing sectors was stabilized, and the
startup costs for new entrants into the processing sector had
become almost prohibitive. Processing was, in 1998, and remains
today, at full capacity, so markets were not available to
accommodate new harvester entrants. During negotiations on the
AFA, processors insisted upon, and harvesters accepted, closed
classes for both sectors. In addition, the processors
requested, and the harvesters accepted, a statutory requirement
to contractually bind particular vessels to particular
processors, with very little practical flexibility for
harvesters to change processors. The AFA was a grand experiment
in fisheries management that everyone now agrees has worked
very well for the participants operating in the unique
circumstances of the pollock fishery.
In the initial crab industry discussions during the winter
and spring of 2000, a proposal was considered, utilizing key
elements of the shorebased AFA approach. This was a sincere
attempt to arrive at a prompt solution in critical
circumstances. However, there were several reasons the AFA
model did not work for Bering Sea crab. Most importantly, crab
fishermen preferred to avoid a closed class of processors and
restriction of markets, and believed that the unique
circumstances of their fishery justified and required a
different approach.
In the crab industry, there are multiple fisheries and
fishermen have multiple markets. With the exception of the
Aleutians Islands brown crab fishery, the crab stocks are low.
There is also gross overcapacity of fishing vessels and
floating processors. Using the AFA model for Bering Sea crab
would have eliminated small processors, as eligibility
criterion called for fishermen to match up with the processor
they had delivered the majority of their catch to in the
previous year. Most fishermen delivered the majority of their
product to the large companies, with the small companies being
minor markets for partial loads. The small companies, some of
which are price leaders, would have lost out on processing
privileges under the AFA model.
A buyback of crab processors was also discussed, but this
also meant closing the class of processors and limiting entry
in the future. As noted above, fishermen were opposed to a
closed class of processors.
Following failed attempts to adapt the AFA coops, crab
fishermen, themselves, asked processors for a processor shares
proposal. This was the point of departure for negotiations that
led to processors shares, and eventually, regional delivery
requirements to protect communities, as envisaged by provisions
of the Consolidated Appropriations Act of 2001.
The crab program, as structured with processor shares,
contrary to what opponents have to say, is less restrictive
than the AFA. There is no closed class of processors; new
processors can enter the industry at any time. Unlike the
situation for pollock harvesters under the AFA, crab fishermen
under the crab rationalization plan can move from coop to coop,
without a penalty year spent fishing in an open access fishery,
and fishermen can belong to more than one coop at a time.
During the public process of considering alternatives, the
industry also had discussions of a single-pie IFQ program,
combined with regionalization of deliveries of crab--without
processor shares or any AFA-style contractual relationship with
processors. The technology needed for the processing of crab is
much simpler and less costly than that required for pollock,
which facilitates the transition of fishermen into the
processing sector. This proposal continues to be very popular
amongst fishermen.
Due to the low status of the stocks, there is a significant amount
of excess capacity in floating processors that can be purchased
at rock bottom prices. Under this single-pie proposal,
harvesters could form large fishermen's cooperatives and either
purchase a floating processor, or contract with one or more
processors or catcher processors, to custom process their
product. Under custom processing agreements, fishermen retain
ownership of the product, and they can then market it
themselves, enabling them to extract more rents from the
fishery. In this scenario, harvesters would have the power to
vertically integrate the industry, bypass the existing
processing sector and disrupt the economics of the five
communities that are dependent on established processing
facilities. The reality of the outcome of this alternative
proposal's effects on efforts to maintain the historic revenue
sharing ratio between the harvesting and processing sectors,
and to maintain stability in coastal communities, contributed
heavily to the development and adoption of the crab
rationalization plan's three-pie quota share program.
The uniqueness of the crab fisheries cannot be
overemphasized. They have the only processor shares and have
the only regional delivery community protection requirement, as
well as the only harvester-processor arbitration process to
insure pricing fairness.
II. Transcript of the testimony of the CRAB Group, Linda Kozak,
Gordon Blue, Terry Cosgrove, Richard Powell and Jeffrey Stephan to the
North Pacific Fishery Management Council on October 8, 2000 in Sitka,
Alaska. At this time the CRAB Group requested (1) that the NPFMC begin
a full BSAI crab rationalization analysis; (2) formally establish a
BSAI crab rationalization committee; (3) direct the committee to assist
the Council staff to identify the elements and options of an analysis
for BSAI crab rationalization.
In response to a question from a Council member during this
testimony Ms Kozak stated on behalf of the CRAB Group that it was their
belief that the Council is the only forum for developing an analysis
for a program for crab IFQs, or crab co-ops.
See the attached notarized transcript in electronic format, pdf
file, from October 8, 2000.
Attachment
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
Senator Stevens. Well, thank you very much.
I think the issues are before the Congress very succinctly
here now. It would be my intention to call upon each of my
colleagues in the order in which they came to ask questions.
And I hope we are in agreement we will limit ourselves to 10
minutes on the first round.
I will recognize you, Senator Murray, for any questions you
might wish to ask.
Senator Murray. Well, thank you very much, Mr. Chairman,
and I know time is short, so I will limit myself to just a
couple of questions, and let me start with Mr. Thomson.
Can you explain to the Committee the differences between
the pollock fishery and the crab fisheries which led the
Council's crab committee to reject a plan that simply
duplicates the AFA?
Mr. Thomson. Senator Murray, that is a rather complex, but
very important, question. It will take me a minute, but I will
give it my best shot.
The American Fisheries Act, as I mentioned earlier, we did
take a very hard look at duplicating that model, since there
was a firm legislative precedent. However, the pollock fishery,
as I mentioned in my testimony, is a single-species fishery.
There are very close alignments between individual fishermen
and single processors, not multiple processors, very close,
established relationships, a number of which there were even
contracts in place. The fishermen and the processors agreed to
a closed class for both the fishermen and the processors. That
was point number one.
Second, they agreed, in the legislation, and this
particularly applies to the onshore sector, to statutorily bind
themselves, the fishermen, to a single processor for 90 percent
of his deliveries. So that was the basis for, particularly, the
onshore American Fisheries Act, the model that the crab
industry took a real hard look at.
However, on the crabbers' side, in looking at the AFA model
we quickly came to--start to pull away from it, because the
fishermen, a number of them, were very much opposed to closing
a class of processors and to that sort of market restriction.
We also, at that time, looked at the effects on the small
processors. In other words, most of the fishermen--or a number
of the fishermen had multiple markets, but their major markets
were the large crab-processing companies, and if you look at
the criterion which determines who your market will be, it
would be the one that you sold the majority of your product to.
The end result would have been the small processing companies,
of which a number of them are price leaders, being
disenfranchised in the program. They would virtually have not
been participants in the program. So we had to shift away from
the AFA model.
One of the other things we looked at that was proposed, and
discarded, was harvester IFQs with regionalization only. What
would have been the outcome of that? The outcome would likely
have been the pretty quick disappearance of several small
companies, because larger companies would have bought them up
or possibly bid the price up on crab to the point where smaller
companies would go away.
The other very clear outcome of the IFQ program with
regionalization-only would have been we would have seen the
harvesters form into some major fishermen's coops and then
contract with a floating processor or two floating processors
or some crab-catcher processes and essentially--although it
would have been a great deal for the fishermen that I
represent, the fact of the matter is we would have been able to
totally bypass the processing sector of the industry, and there
would have been devastating effects particularly on the
Pribilof Islands. So that was another option we looked at.
Then we looked at an option of buyback of the processing
sector, but any rationalization solution of processors or
fishermen requires restricting entry. And so to do a buyback,
you, first of all, would have to close the class of processors
and do a limited entry.
So I know I am kind of going on a bit here, but there were
some very important differences that made the AFA unworkable.
Senator Murray. OK, thank you.
Mr. Fraser, if Commerce decides against authorizing the
process of quotas, what specifically would you recommend that
the Council and that the Congress do?
Mr. Fraser. Mr. Chairman, I think--and Senator Murray--that
the Council has done a yeoman's task in developing this whole
program and 90 percent of the work they have done is pertinent
to moving forward, with a traditional IFQ program absent a PQ
element being contained within the program. So I think that is
the starting point.
As I mentioned in my testimony, I think there are some
processor protections that are inherent even in the non-
processor-quota portion of the program. The Crab Group does not
object to strengthening those. And I think that the AFA does
provide a model for moving in that direction.
Clearly, you know, no two fisheries are exactly the same.
One of the key differences between the two is that the amount
of capital on the processing side, in comparison to the harvest
side, in the pollock fisheries is much greater. The pollock
fishery is much more capital intensive in processing than crab
is, and thus the justification or more processor protection was
greater in the pollock fishery. And harvesters have not moved,
as Arni pointed out, near as much in the pollock fishery as
they do--because you do not have--and you do not have turnover
processors. You have got a small number of processors that were
essentially closed class even before the AFA was passed. And so
it was easier, I think, for harvesters in that fishery to
accept a reduced amount of competition that was inherent in
structuring the AFA, as long as there was some competition left
in that market arena.
In the crab fishery, we have seen 80 processors who
processed crab during the last 10 years during the period of
time the Council analyzed. Only 21 of the 80 would receive any
processing quota at this point, and most of the processing
quota is concentrated in the hands of just a half a dozen of
those, and most of those are the AFA processors.
I think you have to recognize that the AFA has, in and of
itself, provided a large measure of protection for AFA crab
processors and that the protection for the remaining processors
is where it needs to be focused as the Council moves forward
with the non-processor quota plan.
The sideboards that you, Senator Murray, and Chairman
Stevens, included in the AFA, the sideboards on crab
processing, gave the non-AFA processors more opportunity than
they will receive under this plan. This plan actually awards
the AFA processors more crab than the sideboard limit. I think
that if you move forward with a non-processor quota plan, you
ought to retain the sideboards that you have already included
within the AFA to protect those remaining non-AFA processors.
Senator Murray. Well, just a rough guess, how long would it
take for the Council to finalize another plan if we were not to
move forward legislatively?
Mr. Fraser. Mr. Chairman and Senator Murray, the Council
has an EIS that is prepared. At the request of the advisory
panel, a repeated request to the advisory panel, we asked that
they include in that EIS other alternatives than processing
quotas. There was quite a bit of resistance to that, but, in
the end, NOAA's general counsel advised the Council that they
must look at viable alternatives, including legal alternatives,
and the IFQ system is currently a legal alternative. So the
Council has a completed EIS for an IFQ system, as well as an
EIS for the processor quota system.
So I think they could move forward with a legal IFQ program
faster than by waiting for Congress to work through all the
issues relating to authorizing processor quotas. And I think,
as you have heard from Mayor Kelty and others--or it used to be
Mayor Kelty----
[Laughter.]
Mr. Fraser. But everyone in this room, I think, agrees for
the need to move forward.
Senator Murray. Right.
Well, Mr. Chairman, thank you. I know you have to move on,
and I will submit the rest of my questions for the record, if
that is OK.
Senator Stevens. Thank you very much.
Senator Murkowski?
Senator Murkowski. Thank you, Mr. Chairman.
There is obviously a lot of disagreement at the table, but
I think it would probably be fair to say that the reason we are
here, we are looking for, within the industry, to provide
safety within the industry, we are looking to sustain the
resource, and we are looking to provide the protection, the
economic stability, for the communities themselves, the coastal
communities. And we have heard from both Ms. Freed and Mr.
Fraser that your concerns are that the communities are not
protected.
What suggestions do you have? How can we address that
component?
Ms. Freed. I believe you have heard from Mr. Fraser that
the plan the Council has before it, without PQs, provides quite
a bit of protection. There is a regionalization component to
the plan, which provides protection for communities. And we
think that component does a very good job protecting
communities.
What is important to Kodiak is to provide the ability for
our plants to compete. Kodiak has processed Bering Sea crab for
over 30 years. The two years selected for the PQ portion of
this program selects 2 years in which people really could not
logistically bring their crab back to Kodiak, even if they
wanted to, because of fish and game regulations.
We want the opportunity to compete, to have crab come back
to Kodiak. And to do that, there needs to be a competitive
market for the harvesters. We believe that the plan, absent of
PQs, allows that to happen.
Senator Murkowski. Thank you.
Mr. Duffy, has there been an economic analysis or an
economic impact study done that has assessed or reviewed the
impact to the coastal communities on the regional Governments,
or, for that matter, the State tax revenues? And if there has
not been one done, will there be one conducted as part of the
EIS?
Mr. Duffy. Through the Chair, Senator Murkowski, there was
extensive economic analysis conducted, community profiles
developed, all kinds of information was generated through the
Council crab rationalization process. I am not sure of the
specific comment the representative from Kodiak made about the
state not analyzing something, but through the Council process
and leading to a Council conclusion in June, there were
extensive community profiles developed with all sorts of
economic analysis of the impacts of the different alternatives.
So the analysis has been done through the Council process,
Mr. Chairman.
Senator Murkowski. Go ahead, Mr. Chairman. Thank you.
Senator Stevens. Senator Cantwell, do you have any
questions?
Senator Cantwell. Yes, thank you, Mr. Chairman.
Nobody has mentioned the 100-million-loan buyout program,
and I know that it is in the rulemaking process. What effect do
you think that that will have on this plan? Should we wait
until we see that rulemaking? Do we think it is actually going
to reduce the capital investment that has been made? What are
the thoughts of the various panelists?
Mr. Thomson. A good question, too, Senator Cantwell.
The crab industry eagerly awaits the final implementing
regulations for the buyback program. And we are conservatively
estimating that we could reduce the over-capitalization
somewhere around 20 to 25 percent, or reduce capitalization 20
to 25 percent, going into this rationalization program. And the
low-cost loan would actually result in a redistribution of that
quota share in a proportional manner to all of the harvesters
that stay in the fishery. And so a lot of folks are really
looking forward to it.
Senator Cantwell. Do you think it is essential to making
this program work?
Mr. Thomson. I would not say it is essential, no. We think
it would definitely be a very positive benefit, though,
stepping into the program.
Mr. Fraser. Mr. Chairman, following up on that, the Crab
Group, as is implicit in our name, Crab Rationalization Buyback
Group, has long supported the buyback program and is very
grateful for the efforts of the Chairman and Senators getting
that program through, and there has been a lot of hard work to
make sure that it moves through a NMFS process and ends up
being implemented. But we are, as I understand it, on the verge
of implementation, and once that happens there is a very
expeditious process that is laid out in regulation that could
result in real benefits in next season, in fact, if all falls
into place on the time line, to the fleet.
So I think the buyback program is good. It should move
forward. The time line for implementation of rationalization
program is unknown at this point, whether or not Congress
authorizes processor quotas. So I think there are immediate
benefits there, and we should move forward with it.
Senator Cantwell. And, again, no effect on this decision?
No----
Mr. Fraser. I think they are standalone issues.
Senator Cantwell. OK.
Anybody else?
Mr. Fraser, you mentioned in your testimony that you
thought there was better protections for communities that might
be implemented, or better binding arbitration. What,
specifically, were you thinking of?
Mr. Fraser. Mr. Chairman and Senator Cantwell, on the
community-protection side, I think, as Ms. Freed alluded, the
community protections that were part of the trailing amendments
were intended to protect communities from the impacts of
processor quotas. If you do not go down that path, those sorts
of community protections are not needed. And, as she said--I
will leave her statement stand on the community-protection
side.
On binding arbitration, I have submitted a much more
lengthy comment in my written testimony, because it is largely
a technical kind of issue and there are a lot of nuances. But
Arni was correct, there were two basic models that came out of
the Committee. Most of the harvest groups supported a fleetwide
model, and that model would create a pre-season process where
one arbitrator would look at information from all processors
and all harvesters and establish a fair, minimum price which
would become the outside option for harvesters to then
negotiate their contracts with processors. And if they can
bring more to the table by delivering at specific times that
enhance the size of the pie, they may share those additional
benefits differently than that minimum price.
The last best offer arbitration process that the Council
adopted creates a segmented series of arbitrations, where the
arbitrators do not necessarily communicate with one another.
They happen in isolation. The arbitrator does not have the
ability to cross check the information from a processor whose
arbitration he is conducting with information about an
arbitration with a processor that is being arbitrated by
somebody else. And as a result, you do not have good
information. And I think that it is very critical that the
information base for any arbitration program be as good as
possible.
And right now, under the Magnuson-Stevens Act, NMFS is
prohibited from collecting economic data from processors. The
first thing you would need to do is make sure you change the
Act to mandate the collection of the relevant economic data and
make that data available to the arbitrator, because the Act
further restricts the ability of NMFS to provide economic data
that it does collect to any outside party. So it is essential
that the arbitrator have access to the full amount of
information.
But the key difference that--the model that Arni's group
supported and ended up being supported by the Crab Group and
others as an attempt to compromise was called the Steele
amendment. Mr. Steele is here. But it attempts to simulate the
situation we have now, where you have a fleetwide price-
formation process that establishes a price, and everybody has
to meet that minimum. A processor cannot come in and say, ``I
have got excessive costs, and you have to pay those costs for
me.'' In the status quo, that does not happen. Once somebody
establishes the price, all the other processors step up and
meet it. And we support that concept as being essential to an
arbitration process.
Senator Cantwell. So, Mr. Fraser, you are saying you are
not out-and-out against processor quotas if what you call a
``fleet model'' of an arbitration process would be in place?
Mr. Fraser. Mr. Chairman, Senator Cantwell, I think if we
go down the path of processing quotas, it is essential that we
have a good binding-arbitration process to help ameliorate the
fact that you are segmenting the market----
Senator Cantwell. I have that point, but would you support
a plan that had that model?
Mr. Fraser. I believe the concept of processor quotas is
fundamentally anti-competitive and bad public policy. What I
said in the advisory panel early on in the process is, if
somebody comes forward with such a model that really takes the
place of competition, and members of the Crab Group, members of
Tom Casey's group, (the Fisheries Conservation Group), the
Kodiak Association, (United Fishermen's Marketing Association),
the Alaska Fishery Market Association--(the group that does
crab marketing and bargaining)--if they support it, I would put
aside my personal objections to a processor quota and remove
myself from this process. But those groups are not endorsing
the package that the Council has put forward for binding
arbitration. Some members of those groups believe that a
different ratio of A and B shares (perhaps 50:50 or 70:30),
combined with a fleetwide binding arbitration model that
provided a minimum price as an outside option, would simulate a
competitive price. The position of the Crab Group is that
processor quota is bad public policy and that it will not be
possible to open Pandora's Box in just one fishery and then
prevent its spread to other fisheries.
Senator Cantwell. Well, if I could, Mr. Chairman, Mr.
Thomson, what is wrong with a fair minimum price set by an
arbitrator as a beginning point?
Mr. Thomson. Senator Cantwell, you say what is wrong with
the----
Senator Cantwell. What is wrong with the concept that Mr.
Fraser said in having one arbitrator have analysis--basically
having an information database that has all the information and
setting what is a fair minimum price so that there can be a
starting point as a way to protect them. We often run into this
as we are trying to sell our agricultural products abroad for
countries who do not want to open up to our product. They just
want to know what the minimum price is going to be so that they
know that they are not going to get taken. Then everybody else
comes in to negotiate the price.
Mr. Thomson. We do support that. We fully support it. And
that concept is implicit in the Steel amendment. And the
Council, in its recent letter to your distinguished Members,
states that if there is a problem with the process, the
arbitration process that they have approved as of April of this
year, that they will immediately address that, take a look at
that process, and implement this minimum fleet-wide arbitration
process. They will replace the existing one with the minimum,
fleet-wide.
Senator Cantwell. So that is a back-up and----
Mr. Thomson. It is a back-up----
Senator Cantwell. I do not want to--I know that I am--I do
not want to--I want to give time to the Chairman, yield my time
back, but I--Mr. Fraser, you are shaking your head that that is
acceptable process, or no?
Mr. Fraser. It is not an acceptable process. I think Arni
and I agree that the Steele amendment is better than what the
Council adopted but it needs to be in place on day one.
Senator Cantwell. OK. Thank you.
Thank you, Mr. Chairman.
Senator Stevens. Well, thank you very much.
I think we ought to put our eyes back on the goal that we
sought to achieve with the original Magnuson Act, and that was
to protect the species, to assure the reproductive capability
of the North Pacific was to be the primary source of fisheries
for the United States, that we wanted to assure the
survivability of the species. We really did not envision
protecting fishermen--you will not like that--or protecting
communities. We were protecting the fish.
Now, I still think we have got some ways to go to assure
ourselves that we are still on that line. I think it can be
achieved under the recommendations of the Council, but I am
worried a little bit about some of our powers. It is my
understanding the Commission will still establish seasons,
right? They will establish annual quotas, right? And this
formula will apply to their annual actions.
One of the things we have to be sure, that we do not get
ourselves in the position that some Federal court somewhere is
going to start regulating the fisheries off the Alaska coast. I
think that that would be--a delay of a year or two in
litigation over this plan would be destructive to everybody,
including the fishery, I think, because of the problems.
Do you all agree that--not agree, but would you care to
comment upon a follow-up to Senator Cantwell's question? Do you
envision that we have the authority now to condition the
approval of this change in the law upon compliance with the
recent letter that was received by us concerning the
arbitration procedure? Do you understand?
Mr. Fraser. Mr. Chairman, I will rush in, I guess.
Senator Stevens. There is a letter on file, as Mr. Thomson
said, saying that the Council would consider this change. And,
in effect, it is a tacit admission that perhaps the procedure
for arbitration is not sufficient to carry out their original
intention, and they indicated they would revisit it.
Mr. Duffy, do you believe we----
Mr. Duffy. Thank you, Mr. Chairman.
Senator Stevens.--should consider making our action here
conditioned upon the follow-up to that letter?
Mr. Duffy. Mr. Chairman, the discussion was about binding
arbitration. It was, indeed, very complex. Counsel, as well as
the industry, spent 6 months trying to sort this one out. It
was very difficult. In the end, it a 6-5 vote was made by the
Council, that is the binding-arbitration process. It is a pre-
season non-binding price formula. You go through the season,
and if you cannot reach agreement between the harvester and
processor, there are provisions for binding arbitration.
This concept of the Steele amendment is incorporated into
the Council motion in that what the proponents were trying to
do in the Steele amendment can be looked at by the nonbinding
arbitration process prior to the start of the next season to
get a feel for how that works.
If the question is, ``Can Congress change what the Council
has recommended on binding arbitration,'' the answer is yes.
Senator Stevens. We never have. We have never done that
before, because we created the Council with the idea that
regional management was better than management from Washington.
But, strangely, I hear some of you saying you would rather have
some management from Washington now.
[Laughter.]
Mr. Duffy. Mr. Chairman, if I could respond to that. I do
not want that to be attributed to me because I am a big
supporter of the Council process and I think we have done a
very good job shaping a very difficult program.
What I was suggesting is the binding-arbitration
discussions and the vote were very, very close, and there were
a lot of very difficult concepts for the Council to wrestle
with. And we would not be the first to admit--or the last to
admit, that we were, you know, experts in binding arbitration.
It is a subject where a lot of attorneys make a living. So it
was a complex issue.
Mr. Fraser. Mr. Chairman, to follow up on that, the
Council--actually NOAA General Counsel has sent a letter to
Department of Justice, the Antitrust Division, requesting some
bounds, I think, on what constraints they will bump up against
in setting up different binding-arbitration programs. I concur
with Mr. Duffy, the M-S Act process under which you delegate
authority to the Council for managing fish. However, we would
be stepping into new territory here if you start delegating
authorities to councils to manage markets. And it is unclear at
this point how broad of authority you would have to grant to
the Council to give them the ability to change horses from one
system of binding arbitration to another if, in fact, the
system that they have identified at this point does not work
out.
And so any Council action to deal with a flawed arbitration
program will not be final. Prior to review they will wait a
year until they see the results. Fishermen will be described as
coming to the Council whining for a change, so they might want
to see 2 years of results. And then there is an analysis
process, and it takes a year for the Council to work through
its process and then 6 months for NMFS to implement things,
assuming the fix is within the scope of authority Congress
might grant. So we are looking a number of years down the road.
If the effects are seriously negative, they may be
irreversible. That is the concern.
So from our perspective, as harvesters, we would like to
see a binding-arbitration program in which we have confidence
clearly built in ahead of time before you go down any further
on the path of authorizing the processor quotas.
Senator Stevens. Well, the difficulty I have is I think
that the Council has responded to an act of Congress that has
already been passed. And you are requesting, Mr. Fraser, that
we amend our own action in seeking to have the Council make
this determination.
I am interested in your comment concerning the antitrust
provisions. That has been one of the worries that has been
raised here in this Committee. And I am fearful, again, that
should this be taken to court on the antitrust concept, we are
liable to find a court managing the fisheries off our coasts. I
want to make sure that does not happen. And if the antitrust
provisions, if there is a decision that this is in violation of
antitrust, I want to make sure it is returned to the Council to
manage this fishery and not kept in some Federal court.
But, beyond that, I am reminded of the time I spent with
former Secretary of Commerce Mack Baldrige, when he was still
with us, and he was very interested in the king crab, and he
taught me about the way king crab travel across the ocean
floor. He had a report that one of you issued concerning the
pods, that they are 40 and 50 feet high, and how they travel so
far.
It leads me to this question. What assurance is there that
when a harvester comes to a processor, that the processor will
be open under this proposal? We have a processor share, and you
must sell to a processor that has part of that share. But if
you are the harvester, and you come in when the processor that
you are obligated to deliver to is closed, where do you go?
Where do you go, Mr. Thomson?
Mr. Thomson. The way the program is designed is there is--
90 percent of the processing shares are allocated to certain
processing companies. But you have to match up with one of the
companies. If a company is closed or its quota share is filled
up with other harvesters, he has to go someplace else. But
there will be--there is somebody there to match up with.
Senator Stevens. But it has to be someone within the 90
percent, right?
Mr. Thomson. It has to be someone within the 90 percent.
But we envision that this thing will operate similar to the
AFA, that there are so many advantages to the fishermen joining
coops that months ahead of time you are going to pair up into a
coop, say, with another 10, 20, 30 boats. Fishermen will have
to make decisions who is going to fish, who is going to tie up
their boats, so that they can save money in order to make
money.
I do not know whether I am answering your question
correctly, but we do not envision this becoming a problem.
Senator Stevens. Do you believe that the processor holding
part of the 90 percent shares has an obligation to remain open
to receive the harvesters' product when it is brought to the
shore?
Mr. Thomson. Yes, I believe he has an obligation. I mean,
he is there to be in business and to make money. In order to do
that, he has to contract and receive product.
Senator Stevens. Let me tell you, I am sure you remember
that I had an unfortunate invasion into this process when we
commissioned the Lady Anne back in 1980, and that was one
investment that I had--the only investment that I can recall
that I had to buy my way out of.
[Laughter.]
Senator Stevens. This is not a business without great
financial risk, and I am worried about what happens to
processors under this proposal when there is such an enormous
share allocated to certain processors with no chance of entry
into that circle at all in the future.
Now, I want to make sure I understand this, now. Annually,
the allowable catch is determined. If it is increased next year
by 20 percent, the same processors get 90 percent. Is that
right?
Mr. Thomson. That is correct, sir.
Senator Stevens. Mr. Duffy, is there any chance for a new
processor to enter into this without another act of Congress?
Mr. Duffy. Yes, Mr. Chairman, there is 10 percent B shares.
That is one option. So any harvester can process their B
shares. In addition to that, any processor without processor
shares can purchase processing quota from those that currently
hold it if they strike an agreement.
Senator Stevens. I saw that, but does that apply when there
is an increase in the quota?
Mr. Kelty. Mr. Chairman?
Senator Stevens. Yes?
Mr. Kelty. There are caps in place. If the quota goes up,
then there are caps on the amount that the processors can
process, and that would mean more poundage into the open-access
B-share category.
Senator Stevens. The 10 percent is not 10 percent in that--
do you agree, Mr. Duffy?
Mr. Duffy. Mr. Chairman, the way it works with these caps
is--let us say, for example, opilio crab stock goes up
dramatically. At some point, the 90:10 is diminished to 80:20,
depending on the total amount of the stock there. With those
caps in place, the amount of B share that anyone can process is
increased when you hit those caps.
Senator Stevens. And does the Commission make that
determination?
Mr. Duffy. Yes, that is part of the comprehensive Council
plan. And on an annual basis, the Department of Fish and Game
will be the total allowable catch, by species.
Ms. Freed. Mr. Chairman, if I could respond to that, as
well. We have looked very carefully at those caps set on the
community protection committee, and we are very disappointed at
how high those caps are set. We believe that they offer no
respite from the 90:10, not, certainly, in our lifetimes.
We would also like to mention that the 10 percent B share,
open share, we do not believe that provides any leverage for
the fishermen at all, and provides no leverage for communities.
Ten percent, frankly--even in open access right now, fishermen
are told if they do not bring their crab to a particular plant,
they do not have a tender contract. Or if they do not bring
their crab to a plant, they are not going to have other
opportunities with a particular processor.
We have had public hearings where fishermen have testified
before the city council saying, ``We do not believe that there
is any chance that that 10 percent will ever be able to be
brought back to Kodiak, because the processors that we deliver
our 90 percent to will hold us hostage.''
Senator Stevens. We can take care of that, Ms. Freed. I do
believe most of the boats you are talking about are home-ported
in Kodiak, and they certainly could bring back their last catch
without any problem. I do not think--that is another antitrust
problem. If you think that, for 1 minute, that a processor can
demand that they get part of the 10 percent in exchange for an
agreement to buy part of the 90 percent, you have got yourself
an antitrust case right there. Now, that is not going to happen
twice, I will tell you, because that is an antitrust violation.
Ms. Freed. Well, Mr. Chairman, also the fact is, is that
there is so little quota that comes to Kodiak because of the
years that were selected that boats will not be able to bring
their last load home as they have traditionally done. That is
one of the things that we took to the North Pacific Council and
asked them to consider, and they dismissed that out of hand.
Boats will not be able to bring their last load home to
Kodiak, even boats that are home-ported in Kodiak, because
under the current plan, Kodiak gets roughly 3 percent of the
quota. Traditionally, what comes home in a last load when there
is open opportunity for the fishermen to bring their crab back,
is 10 percent or more.
Senator Stevens. Well, what about that 10 percent? Why
could not you get part of the 10 percent?
Ms. Freed. The way the last-load home works is, most of
them would have more onboard, potentially under these low tax
right now, that is more than just the 10 percent. But what I am
saying is, that is not equivalent to what the processing quota
is and their ability to bring that product home now, because
they have to deliver 90 percent of their product to plants that
are not in Kodiak.
Mr. Kelty. Mr. Chairman?
Senator Stevens. Yes, Mr. Kelty.
Mr. Kelty. Yes, I have to take exception to some of Ms.
Freed's comments. You know, the Kodiak history through 1997
through 1999, it is a little bit under 4 percent on red crab.
It is not even 1 percent on opilio. Part of my testimony that I
turned in, I went back trying to find some history for Kodiak,
and I went back all the way to 1990. 1990 to 2003 for opilio,
1.9 billion pounds were processed. Kodiak got 6.6, only 6.6
million out of 1.9 billion pounds of opilio went back to
Kodiak. Years that--we had 300-million-pound quotas in the
early 1990s. Less than a million pounds went back to Kodiak. On
red king crab, in that same timeframe, from 1990 to 2002, there
was 126 million pounds of red king crab processed; 5.5 million
pounds went back to Kodiak, about 3.8 percent, during that
whole 10-year period.
It just shows you there is a lack of dependency on the
Bering Sea resource going back to that community. This last
opilio season, in 2003, there was 197 boats fished. Only one
vessel--it was a one-trip season, 26 million pounds of opilio
caught--only one vessel decided to go back to Kodiak, with less
than 100,000 pounds. On the king-crab season the year before, I
think seven vessels went back. And these are one-trip seasons.
Senator Stevens. We have got to move on here.
Senator Murray, do you have any further questions?
Senator Murray. No, thank you, Mr. Chairman.
Senator Stevens. Senator Murkowski, do you have any further
questions?
Senator Murkowski. I want to ask, really quickly, on this
10 percent, the undesignated shares. You know, I understand the
purpose and the intent and the leverage to the fishermen and
hopefully to encourage new processors to come into the industry
so we do not have the antitrust concerns. But in terms of how
we got to the 10 percent--again, I guess, Mr. Duffy, this would
be directed to you--what kind of an analysis went into that to
determine it is a 90:10? How did we arrive here? Just very
briefly, because I am sure there is more history than I want to
know.
Mr. Duffy. Very briefly, through the Chair, Senator
Murkowski, the concept of a 90:10 split was within a range of
discussions that happened in the industry committee. That
industry committee met for over a two year period. So that was
really where you had a range of--I think it was 70 to 90
percent was discussed, if you went down the processor-quota
idea.
So in terms of the 90 percent and specifically where was it
analyzed, I think the concept, generally, of processor-quota
shares, their benefits to communities, the potential
relationship impacts it might have between harvesters and
processors, was all in the Council analysis. But I do not think
you can take your finger and point to a specific point or place
in the analysis where it says 90 percent or 85 or 95 or 100.
The Council had to listen to the industry committee, the
advisory panel, the science committee, and had to come to a
decision about what they thought an appropriate share of the
rents would be through a program that protects communities,
harvesters, and processors. That is where the 90 percent
processor quota, 10 percent B share came from.
Mr. Fraser. Neither the Advisory Panel or Scientific and
Statistical Committee recommended 90:10. Nor did the crab
industry committee appointed by the Council make a specific
recommendation on the A:B ratio. However, ACC which supported
the Committee report to the Council came to the June 2002
meeting recommending 80:20. The Council did not debate any
alternative ratios, and so the record fails to identify why
90:10 is the preferred option. If I could just add, Mr.
Chairman, that the one place that the analysis did deal with it
was the two economists who were contracted by the Council,
Professors Milan and Hamilton, and that analysis was
qualitative. And the relevant quote was that, ``as the ratio
increases of processor quota to independent market quota, it is
going to favor processors over harvesters.'' And I think that
is just--it is almost intuitively obvious. But we are at a
ratio that is clearly at the high end of favoring processors.
Senator Murkowski. And if I recall your testimony
previously, you had said that that summation had been taken out
of the----
Mr. Fraser. It had been removed from the main document
and----
Senator Murkowski. OK.
Mr. Fraser.--put back in as an Appendix.
Senator Murkowski. Thanks.
Thank you, Mr. Chairman.
Senator Stevens. Mr. Duffy, I have some other questions to
ask, if I may. We are obligated to get out of this room by
4:30.
Is there a commitment by the Council to monitor the crab
plan and make any needed adjustments if one of the three pies
or partners--harvesters, processors, or communities--are
disadvantaged by some unintended consequence?
Mr. Duffy. Mr. Chairman, I think I would be appropriate for
that.
The Council, in their problem statement, talked about--and
goal statement--talked about trying to achieve equity between
the harvesters and processors. In developing this program, we
put a number of measures in place we hope will do that. The
program-review portion of it is review by the Council 3 years
after implementation and then comprehensive review of the
program 5 years after implementation and a commitment by the
Council to modify the program, if needed or as appropriate, to
ensure equity between the harvesters and processors. The
Council is on record as taking that approach.
Thank you, Mr. Chairman.
Senator Stevens. Mr. Thomson, there has been, implicit in
some of the answers, a fear that the processors are going to
use the leverage of their 90 percent contracts to control the
10 percent. I have in mind some sort of an amendment that would
say that anyone that did that would forfeit forever their
interest in the 90 percent. Would you object to that?
Mr. Thomson. No, Mr. Chairman.
Just a quick comment, if I might, Mr. Chairman----
Senator Stevens. Yes, sir.
Mr. Thomson.--to a question that you asked previously about
the Council, this Council-approved program and whether or not
it should be changed or altered or if we should just authorize
the Council-approved program. The ACC has long supported the
North Pacific Council process. We live with the decisions of
the North Pacific Council, and so we support the program that
the Council has approved. We agreed to that going into this
process, and I thought other folks that were involved in this
process similarly asked for the Council process to develop this
program, and I thought that they had similarly agreed that what
came out of this process at the North Pacific Council on crab
rationalization was going to become the program that we would
live with.
Senator Stevens. I agree with that. It does seem, however,
that when we pass this bill, if we do, we have established a
new paradigm for the king crab as far as the Council's
authority for the future. And as I understand it, they would
not have the authority to modify substantially the whatever it
would be, the plan we approved, without the approval of the
Congress. We could change that, of course, and say that it
could be modified with the approval of the Secretary. I think
you all should look at that to see, because I do believe, in a
dynamic situation like we have got now, that the possibility of
unforeseen circumstances is great.
We have just had a report over the weekend, as you know,
from a global study, of the status of fisheries in the world. I
was interested to see that there were not any specifics
mentioned about the North Pacific. I think the North Pacific
Council has the best track record of all the councils in our
country, and I think that the United States has the only track
record of putting forth a management plan that protects the
fish.
Now, we have got to emphasize that even more for the
future, and I think that if we are to avoid the unforeseen
consequence of having a court intervene in this management
process because of violation of some concept that is developed
by one of these studies, and that is not--when we look at
endangered habitats and the whole concept now, that is not an
unforeseen possibility, we have a definite possibility of
intervention by litigation in the management of the North
Pacific if we do not really tow the line as far as protection
of the species. So I would urge you to keep that in mind.
My last question, really, is the problem of Kodiak. And it
is been very well expressed here, but I want to ask you all. Do
you see any chance for Kodiak to restore any portion of its
previous fishery under the plan that is before us now? What is
the answer to Kodiak's objection?
Mr. Duffy?
Mr. Duffy. Mr. Chairman, if I could, we, as a Council,
tried to reflect, in the amount of processing quota, historical
dependence on that fishery. As you know, under National
Standards and under advice from the National Marine Fishery
Service, recency is an issue we often look at. The years
selected for processing quota for Kodiak, 1997 through 1999,
for opilio and Bristol Bay red king crab are within the range
of the 1991 to 2000 season and what Kodiak has processed on
crab.
Additional measures Kodiak can use is, they can use sweep-
up provisions to get crab in the Gulf of Alaska that is just
bits and pieces in smaller communities' processing quota. They
can purchase IPQ to expand their processing base in the
community of Kodiak. And they have some first right of refusal
processor quota if certain processors choose to leave that
community. So they do have some options available to them upon
approval of this program and implementation already.
Senator Stevens. Is there any future with regard to that
harvester-processor portion of the industry as far as Kodiak is
concerned?
Mr. Duffy. Mr. Chairman, if you are talking about the
catcher-processor----
Senator Stevens. Catcher-processor, yes.
Mr. Duffy. Yes, the catcher-processor sector. What the
Council has done with the catcher-processor sector is given
each catcher-processor CP shares that reflect what they harvest
and what they processed on board. We have also, in addition to
that, given them harvester quota for quota that was previously
delivered to the shore. So they are made whole under this
program. Now----
Senator Stevens. But can any of that allocation to
processors, the catcher-processor, be brought ashore in Kodiak?
Mr. Duffy. Yes, the catcher-processor shares can be
transferred to shore-based processing through the Council----
Senator Stevens. Must they be allocated to those that hold
the 90 percent?
Mr. Duffy. The CPs, if they change to shore-based, they
would be subject to the 90:10 rule----
Senator Stevens. Why?
Mr. Duffy. Well, because they would be moved to shore-based
processors who are a part of the processor-quota program.
Senator Stevens. I do not understand that. I should have
asked that question sooner. That was in the back of my mind
when I came here. Why should the catcher-processor quota, if
they decide to go ashore, automatically go to the 90 percent?
Mr. Duffy. Remembering Council discussions on that issue--
and I will let others chime in--but remembering Council
discussions, our intent was to make the catcher-processor fleet
whole, but to not let them grow at the expense of coastal
communities. So what we set up was a one-way transfer of CP
shares to the shore-based----
Senator Stevens. Could a community purchase the shares of a
catcher-processor?
Mr. Duffy. Can a community purchase catcher-processor
shares? Yes, I believe.
Mr. Kelty. Yes, because they can--communities can also
purchase the harvester shares.
Mr. Duffy. Right.
Mr. Kelty. So the CP is a harvester, so they could purchase
shares that way.
Senator Stevens. Ms. Freed, why is not there an avenue for
Kodiak to increase its shore-side processing by purchasing some
of the shares of the catcher-processors?
Ms. Freed. Our contention is that other communities are
getting their historical share under this program; Kodiak is
being forced to purchase it. The Kodiak community can ill
afford to purchase quota that has been historically coming to
Kodiak.
I have got some economic information attached to my
testimony, but in 2000, 10.93 percent of the Bering Sea red
king crab came to Kodiak. In 2001, it was 10.5 percent. I
cannot tell you what it was in 2002 and 2003, because Fish and
Game does not have the statistics, but I can tell you I looked
out of my office window and saw 14 boats that were home-ported
in Kodiak sitting in Kodiak waiting to unload. That is well
more than 3.8 percent that is available.
The sweep-up provision for the coastal communities in the
Gulf of Alaska that would provide us with an opportunity to add
to our quota is, again, a purchase program. It would force the
community of Kodiak to spend money that we might need for law
enforcement, for our ambulance service, to buy quota to make us
whole, and that is an unacceptable situation for Kodiak. It is
not happening for many other communities. It is happening to
some other communities in the Gulf of Alaska.
Senator Stevens. Well, I do thank you all. It is a very
perplexing question for us. We are going to have to confer
further on it. There is no question about that.
We will accept additions to your comments. Those persons
that want to add to this record would do so through these five
witnesses. I was limited to the five witnesses by the agreement
we had to hold this hearing.
And, again, I thank you all for coming. You know, we get to
the point sometime that we forget about the objectives of what
we fought for back in the 200-mile limit bill and the whole
concept of being able to protect our fishery against, primarily
against, foreign intrusion. And I do think that we have to get
back to that and keep foremost in our mind that we, as
particularly the Northwestern part of the United States and
Alaska, are the stewards of that offshore potential and the
great species that exist. And the history of the AFA is that we
have improved the condition of pollock. The history of this
agreement can only be tested in terms of whether it increases
the stability of the production of crab in the North Pacific. I
hope we keep that in mind.
Thank you very much.
[Whereupon, at 4:30 p.m., the hearing was adjourned.]
A P P E N D I X
Prepared Statement of Coleman A. Anderson
Senator John McCain, Chairman
Mr. Chairman, members of this committee: I do not support Processor
quota in any form. It seems that all common sense has been thrown out
the window on this issue of PQS. As an operator of a crab catcher
vessel and minority owner in a crab catcher processor I would be left
with only four companies to sell to, or compete against for the rest of
my life in the industry. I would like the committee to consider the
following points:
The ACC does not represent independent fisherman and is now
the smallest group of vessel owners in the industry.
The NPFMC council's recommendations do not allow for
vertical integration by the remaining processors in the
industry nor the fact that CIP vessels would no longer be
allowed to purchase crab at the end of the seasons. This will
skewer the 90:10 split to 95:05 in reality and that would not
be a free market in any sense of the term.
This entire process is about harvester reduction, the
processing sector has downsized it's crab operations years ago
with the reduced GHL's that we all live with. All of the crab
processing equipment in Alaska is fully depreciated or the
former owners are out of business. There is no reason to
guarantee the four remaining companies a free ride for the rest
of their corporate existence.
Why has Binding Arbitration become such a hot issue? Because
no amount of talk, meetings, and proposed full time positions
for legal advisors can create the compelling need to arbitrate
by the processing sector if the are guaranteed the crab in the
end by law.
The councils proposals will result in the harvesting and
labor sectors of this industry being absorbed by the four
remaining processing corporations with no place left to go in
an industry that most of us have dedicated our lives to.
As a life long Alaskan and 30 year participant in it's fisheries I
have watch all of Alaska's recourses sold and bartered as if Alaska
were a third world nation: Please don't allow this to continue here.
Thank You,
Coleman A. Anderson
______
Prepared Statement of Michael F. Burns, President and Co-Owner,
Blue North Fisheries
Thank you for the opportunity to tell you a little about our
company, Blue North Fisheries, and the likely adverse effects on Blue
North of the Bering Sea/Aleutian Islands Crab Rationalization plan
recommended by the North Pacific Fishery Management Council (NPFMC).
Generally we are supportive of this plan, but have concerns about the
design of the processing quota element of the Council's
recommendations. First let me explain the background of our
participation, the problem with the Council's recommendation for
processing quotas as we see it, and a proposed solution to correct the
problem.
Blue North Fisheries participated in the Bering Sea crab fisheries
for over fifteen years as crab harvesters. As a result of the enactment
of the American Fisheries Act (AFA) in October 1998, a limited group of
large, so-called ``AFA processors'' was given a ``closed class'' for
the right to process pollock, and to be fair to processors not allowed
to process pollock, the AFA ``sideboards'' limited the amount of crab
that these AFA processors could process. In reliance on the AFA
sideboards, and concerned about a restricted processing marketplace,
Blue North Fisheries invested more than $2.5 million in acquiring a
processing vessel to process crab caught by its commonly owned catcher
vessels. Starting in October 1999 and continuing to the present, Blue
North processed more than one million pounds of raw, delivered crab,
all of which was caught by Blue North catcher vessels in the Bristol
Bay Red king crab fishery and the Bering Sea C. opilio (snow crab)
fishery. To our knowledge, Blue North is the only processing entity
with significant recent participation that entered the processing
sector subsequent to the AFA.
Additionally, Blue North also engaged in joint venture processing
operations with Western Alaskan communities since 2001, providing
processing markets in coastal communities traditionally deprived of the
opportunity to participate in the crab fisheries adjacent to their
villages. These operations culminated in the company's recent
investment partnership with the CDQ organization Coastal Villages
Region Fund (CVRF), now fifty-percent investment partners in Blue
North's four crab vessels including the processor Blue Dutch.
In June 2002, the NPFMC decided to recommend the establishment of
processing quota shares, for the first time ever. Given the extent of
our crab processing in 1999-2002 (continuing in 2003), we reasonably
expected to receive processing quota shares equivalent to our present
effort. However, the NPFMC omitted the most recent three years and
picked 1997-1999 (years that are most advantageous for the AFA
processors) for the Bristol Bay Red king crab and Bering Sea C. opilio
(snow crab) fisheries. By contrast, the AFA used for historical
participation the three years preceding the year of enactment, which we
believe is the correct approach. The result of the NPFMC's selection of
less-than-current history is that Blue North is extremely disadvantaged
by receiving no processing quota for opilio and only a negligible
amount for king crab, despite active current processing--processors
with less recent participation will receive more processing quota
shares. Additionally, the Western Alaska communities of CVRF that have
been deprived of opportunities in fisheries off their doorstep will be
similarly disadvantaged by this lack of processing opportunity.
``Present participation in the fishery'' is the very first
criterion listed in the Magnuson-Stevens Act (section 303) for
development of limited access programs. In determining crab processing
quota shares, the NPFMC failed to adequately characterize present
participation in its analysis, disregarding the most current three
years of participation. The net result of this omission is to award
processing privileges to AFA processors in excess of the processing
sideboard caps stipulated as a condition of the benefits conferred by
the AFA. This windfall to the large, often multi-national, AFA
processors comes at the expense of our small, domestic processing
operation.
In conclusion, we request that any implementing legislation for the
North Pacific Council's crab rationalization plan include a
``grandfather provision'' to reflect Blue North Fisheries' present
participation in crab processing. Without such a grandfather provision,
the NPFMC plan has the effect of penalizing Blue North for reasonable
reliance on the AFA sideboards. Specifically, Blue North would like to
have sufficient processing quota shares to enable its processing vessel
to process the catch of its commonly owned catcher vessels. This
solution will correct the unfairness resulting from the Council's
approach and compensate for the Council's failure to adequately address
the requirements of the Magnuson-Stevens Act regarding present
participation. The amount provided to Blue North under an amendment
like this would be less than 1.5 percent of the total amount of the
crab processing quota shares, most of which is allocated to large,
economically advantaged shoreside processors. With this adjustment,
Blue North Fisheries would fully support the crab rationalization
program recommended by the North Pacific Fishery Management Council.
Thank you for the opportunity to submit this statement for the
hearing record.
______
Prepared Statement of Robert L. Chevalier, Sitka, AK
U.S. Senate Committee on Commerce, Science, and Transportation
Senator John McCain, Chairman
Mr. Chairman:
I believe that there are grave dangers to the United States
inherent in the Crab Rationalization Plan being considered by the North
Pacific Fishery Management Council. The domestic issues of
socioeconomic violence and antitrust law violations are bad enough; the
international implications are truly frightening, and the U.S. may be
in danger of giving away control of its fisheries and oceans simply by
adapting the processor shares provisions of the Crab Rationalization
Plan.
Under the Individual Fishing Quota plan, at present used to manage
longline fisheries in the Gulf of Alaska and the Bering Sea, anyone
wishing to purchase quota and prosecute a fishery must be an American
citizen, and MUST, under pain of ultimately forfeiting all fishing
rights, obey all regulations set forth by managing agencies designated
by the Magnuson-Stevens Act. However, quota shares given to processors
will become simply a business asset controlled by the owner of the
processing facility, many of whom are foreign or multinational
corporations. Under the provisions of GATT the holder of resource
assets may have rights which supercede those of the state or Federal
governments, and the U.S. may very well lose the ability to manage the
fisheries in question: the provision of so-called free trade will
supercede all matters of conservation or American utilization of marine
resources.
In short, the processor share provisions of the Rationalization
Plan may destroy the Magnuson Steven Act, and all American fisheries.
Respectfully Submitted,
Robert L. Chevalier
______
Prepared Statement of Dorothy Childers, Executive Director,
Alaska Marine Conservation Council
Chairman McCain, Senator Stevens and members of the Committee:
Thank you for holding today's hearing on processing quota and the
proposed Bering Sea crab rationalization plan. A decision by Congress
regarding authorization of processing quota is clearly of major
significance in terms of future U.S. fisheries management policy.
The Alaska Marine Conservation Council (AMCC) is a community-based
organization of fishermen, traditional subsistence harvesters,
scientists, small business owners, families and others concerned about
the health and diversity of our marine ecosystems.
AMCC supports rationalization of the Bering Sea crab fisheries. We
are also working at the North Pacific Fishery Management Council to
advance conservation and community benefits of groundfish
rationalization plans that are currently under development in Alaska.
We do not support processing quota as an element of rationalization
because of the controlling effect this system would have on markets,
fishing families, communities and the public process for managing the
public's fishery resources.
Under the proposed Bering sea crab plan, the fishery would be
managed under a quota system in which vessel owners and fishermen would
be allocated individual fishing quotas (IFQs) and seafood processors
would be awarded processing quotas to buy and process the catch. The
proposed ``two-pie'' plan would require fishermen to sell at least 90
percent of their catch only to processing companies holding processing
quota.
Processing quota is an anti-competitive system in which the
government would allocate market share to certain corporations.
Processing quota will limit who can do business to a select pool of
quota holders. About 90 percent of the shares would be allocated to 12
out of the 30 eligible companies and will allow for significant quota
consolidation through transfers, leasing or buying out other companies
holding quota. There are no effective limits on how much processing
quota a single corporation could control.
As a conservation organization, AMCC is involved in the processing
quota debate as a matter of public policy because it speaks to
governance of the public's fishery resources and the future face of the
seafood industry. Our objective is to promote fishery management
systems that ensure wise conservation management for long-term
sustainability of fish, ocean ecosystems and our communities--healthy
fisheries and healthy communities go hand-in-hand. The allocation of
processing quota goes far beyond the kinds of allocation decisions
regularly made by regional councils. We believe processing quota will
lead to a concentration of control by processing corporations that will
tip the scale, diminishing the role of community voices in the
management of our resources. For fishery-dependent communities,
processing quota does not offer a safety net but is rather a political
accommodation that over time will create more problems than it solves.
Certainly concerns raised by community entities regarding inadequate
protections in the Bering Sea crab plan bear this out.
The NPFMC has a huge and time-consuming responsibility for the
conservation management of our fisheries. The Magnuson-Stevens Act
authorizes regional councils to manage the harvesting sector in
accordance with complex rules associated with the biological needs of
the fish and ocean ecosystems along with socio-economic considerations
related to limited access and allocation between harvesters. Fishery
managers should not also be tasked with designing and managing market
systems for fish after they are caught. Although many hours have been
spent developing the Bering Sea crab plan, we remain concerned about
the degree of scrutiny applied to the processing quota component:
The North Pacific fishery managers' deliberations on the
crab plan have not included a discussion on what problems
processing quota is expected to resolve or what alternatives
might be suitable to address them. For example, do the
relatively small number of processing companies that will
receive quota require a perpetual endowment of market share? Or
are their problems related to being able to weather a short-
term transition from an open access fishery to a slower-paced
quota system for harvesters?
The Senate Subcommittee on Oceans and Fisheries requested a
report from the Government Accounting Office (GAO) reviewing
the ``economic effect of the IFQ program on Alaskan halibut and
sablefish processors.'' \1\ The report revealed clear
weaknesses in the one economic study that has been used as the
rationale for processing quota. This study concluded that IFQs
gave fishermen more bargaining power and caused processors to
lose market share, profit margin or go out of business.
---------------------------------------------------------------------------
\1\ GAO. 2002. Individual Fishing Quotas, Better Information Could
Improve Program Management. The full report is available at
www.gao.gov.
The GAO reported that they could not determine what the effects
of halibut/sablefish IFQs were on processors and to what degree
---------------------------------------------------------------------------
IFQs caused changes in the processing sector.
``. . . we identified a number of problems with the
study's methodology and scope that brings into question
the reliability of the study's estimates. These
problems include (1) the pre- and post-IFQ time periods
do not provide an accurate measure of processors'
economic welfare, (2) the study's results may not be
representative of the industry as a whole, and (3) the
document requesting economic information from
processors may have biased participant response. The
study's authors acknowledged that examining the pre-
and post-IFQ impacts on the processing sector does not
necessarily imply that the IFQ program alone caused
these effects.'' GAO p. 25.
``. . . the document requesting economic information
from processors may have had biased participant
responses. In the preamble to the survey document,
participants were told, among other things, that the
purpose of the study was to test the theory that a
harvester-only quota allocation transfers wealth from
processors to harvesters and that the survey's results
would be used to assist in designing future IFQ or
other fishery rationalization programs. Such statements
leave little doubt as to how responses could benefit or
harm processors with economic interests in other
fisheries. According to standard economic research
practice, these types of statements are to be avoided
when designing a survey as they can influence
results.'' GAO p. 27.
``Along with an increase in buyer-broker halibut
purchases, there was a decrease in the number of
individual shore-based plants that processed halibut
and sablefish. While some plants stopped processing
halibut and sablefish, others decided it was beneficial
to start. Between 1995 and 2001 . . . 68 plants stopped
processing halibut and 56 started, resulting in a net
decrease of 12 plants. Similarly, 54 plants stopped
processing sablefish and 40 started, resulting in a net
decline of 14 plants.'' GAO p. 23.
``The IFQ program, however, did not necessarily cause a
plant to stop processing halibut and sablefish,
According to industry and government officials, some
plants stopped processing halibut or sablefish because
the plant was sold to another processor, the plant
closed for personal reasons, plant management made poor
business decisions, or the plant burned down.'' GAO p.
23.
The North Pacific Fishery Management Council recommendation to
Congress has not taken the GAO report into account.
These concerns notwithstanding, we very much recognize the
important role that the processing sector fills in our communities. The
question of whether or not Congress should legalize processing quota
should not be reduced to a feud between processors and fishermen with
communities caught in the middle. Rather than asking what it will take
to get fishermen and communities to accept processing quota, Congress
should ask a completely different question: What is the appropriate way
to support each sector of the seafood industry in order to better serve
(1) fisheries conservation, (2) stability of our communities, and (3)
preservation of healthy markets and free enterprise?
To that end, we urge Congress to look closely at the issues facing
the processing sector and to pursue other ways to address them in. a
manner that preserves competition, encourages innovation and protects
communities. A decision to authorize processing quota in Bering Sea
crab fisheries will establish a precedent that will be hard to contain
in other Alaskan fisheries or beyond.
Thank you for the opportunity to share our perspective on
processing quota with the Committee.
______
Prepared Statement of Kale Garcia, Owner-Operator F/V Aquila
Submitted to U.S. Senate Committee on Commerce, Science, and
Transportation
Senator John McCain, Chairman
Senator Ted Stevens, Presiding
May 20, 2003
Senators,
I have been participating in the Bering Sea, Bristol Bay crab
fisheries since 1980. I just wanted to go explore Alaska and getting
there was on a Bering Sea bound, crab catcher-processor, hired on as a
processor. In that same season, I was able to fill in as a deck hand on
a crab vessel. I got off and swore I'd never do that again! I began
working my way up to Captain, as it didn't take long to figure out
where I wanted to work on a crabber. Since then I've rarely missed a
season.
As an owner operator, I have now come around in full support of
crab rationalization via the industry funded buyback and IFQ's. I
believe this will dramatically improve the safety of the lives at sea
and will help the crab stocks as the fishery will be approached with a
different mind set. It will change the way I fish by allowing me to
time the weather and tides for safer working conditions. This will also
allow more soak time on the gear which will reduce and possibly
eliminate the incidental by-catch of smaller juvenile crabs. This will
definitely have a positive effect on the fishery and most everyone
associated with the crab industry.
I am adamantly opposed to any type of processor quota share. If it
was decided to go along with the 90:10 two pie system, this would
possibly take away the above mentioned positive impacts on the industry
as I would be told when the processor would be buying crab, or when I
could go fishing. It is currently illegal under Alaska state law for me
to bring in live crab to one port and leave to take my ten percent
somewhere else. It would probably not be economically feasible. If I'm
the owner of the boat and permit to go catch the crab, then I should
own the crab up until the time they are sold. However, if I have to
sell 90 percent to a predetermined market then who actually owns the
product? How am I to get a fair price? Processors have the opportunity
to own catcher vessels and harvester shares now. In fact there is a
significant amount of vessels that are processor owned.
I am not interested in remaining in the industry under a two pie
system. I have 23 years of hard work and everything financially at
stake in the crab industry. There would be a huge devaluation in
vessels and IFQs under the current proposed plan. When the boat tenders
salmon in the summertime it is a full on family job for myself, my wife
and kids. I am not sure where I would start over or what I would do. I
do think there are many other alternatives to this absurdly proposed
idea.
Thank You,
Kale Garcia,
Owner-Operator F/V Aquila.
______
Statement of Erin D. Harrington, Kodiak, AK
Senator John McCain,
Committee on Commerce, Science, and Transportation,
U.S. Senate,
Washington, DC.
Dear Senator McCain and Committee Members;
I'm writing to you to express my concerns about the proposed Bering
Sea/Aleutian Islands crab rationalization plan. In particular, I feel
it is important that I express my complete opposition to processor
quota as a tool for fishery rationalization.
I believe that the award of processing quota to Alaskan processors
through this plan will result in numerous side effects that will be
deleterious to the commercial fishing industry in Alaska. In my
opinion, the most important of these effects will be
the restriction of future development of entrepreneurial
seafood marketing businesses by fishermen or next-generation
processors;
the establishment of a precedent that will be carried
through to rationalization plans in other fisheries, where
processor quota is entirely inappropriate and would result in
the degradation of small communities and forward-thinking
businesses; and
a protectionist system that locks a set number of
corporations (primarily foreign-owned) into the Alaskan fishing
industry, and renders them bullet-proof from the natural
fluctuations of the market-based system that would otherwise
compel them to innovate and improve in order to succeed.
My family has been involved in fisheries in Alaska and
Massachusetts for two generations. I personally have fished since I was
a little girl. In recent years, I have broadened my involvement in the
fishing industry as a reporter for several trade publications, as the
project leader for a large fishermen's organization in Alaska, and now
as a graduate student in Seafood Marketing and Economics at the
University of Alaska. I am writing this letter on behalf of my parents
and my brother, all of whom are directly involved in the fishing
industry.
Senator, I'm sure that you and your colleagues are well-versed on
the impact of corporate consolidation in the world food systems on
family farms. You also have likely become aware of the development of
farmers' cooperatives, and the
[The end of this letter was unavailable at time of printing.]
______
Prepared Statement of Robert Halvorsen, Professor of Economics,
Department of Economics, University of Washington
The Economic Analysis of Individual Processor Quotas
On behalf of the City of Kodiak, I am submitting this written
testimony to the record of the May 20, 2003 Hearing on Bering Sea/
Aleutian Island Crab Rationalization before the Committee on Commerce,
Science, and Transportation of the United States Senate. My testimony
first addresses the general issue of whether fishery rationalization
programs should include explicit mechanisms to protect processors from
possible adverse effects. It then discusses one specific mechanism,
Individual Processor Quota (IPQ), which has been recommended for use in
the Bering Sea/Aleutian Islands (BSAI) crab fisheries. I previously
presented testimony on these topics to the February 13, 2002 Oversight
Hearing on Individual Fishing Quotas before the House Subcommittee on
Fisheries Conservation, Wildlife and Oceans. My testimony in the
present statement has been revised and expanded in the light of
information and analyses that have become available since that time.
The City of Kodiak has also asked me to submit testimony for the
record on the experience to date with rationalization of the BSAI
inshore pollock fishery under the American Fisheries Act of 1998, and
the implications that can be appropriately drawn regarding the proposed
rationalization program for the BSAI crab fisheries. My testimony on
this topic is being submitted separately in an additional statement for
the record.
Executive Summary
My analysis of the likelihood of fishery rationalization programs
having adverse effects on processors begins by considering the
importance of market structure in determining the outcomes of a
rationalization program. To illustrate the importance of market
structure and the balance of bargaining power on the outcomes of
rationalization programs, I review a recent analysis of the BSAI
inshore pollock fishery that was prepared for the North Pacific Fishery
Management Council (Halvorsen, Khalil, and Lawarree 2000). The analysis
demonstrates that market structure is critical in determining the
distributive outcomes of rationalization programs.
I then review the available evidence on whether processors have in
fact incurred losses as a result of fishery rationalization programs.
There has been surprisingly little empirical research on this issue.
The strongest conclusion that appears to be supported by the existing
research is that some processors are made worse off and some are made
better off by the implementation of an Individual Fishing Quota (IFQ)
program. A recent unpublished study of the North Pacific Halibut and
Sablefish IFQ program (Matulich and Clark 2002) does claim that
processors in that fishery experienced large losses. However, a review
of this study by the General Accounting Office concluded that problems
with the study's methodology and scope bring into question the
reliability of its results. My own review of this study leads to the
even more negative conclusion that critical defects in its theoretical
and empirical methodology invalidate its results.
Next I discuss the rationales that have been advanced for
compensating processors for any losses that they may incur as the
result of a rationalization program. My primary emphasis is on the
argument that if processors are not compensated they may block the
implementation of a rationalization program, with the result that the
potential efficiency gains from the program cannot be realized. I note
that there are several problems with this argument. First, attempts to
block a program unless distributive outcomes are altered may simply
reflect an attempt to increase the size of already positive net
benefits, rather than to avoid losses. Second, if harvesters become
concerned that the attempt to keep processors safe from harm will
result in losses for harvesters, they may also try to block
implementation. Lastly, when efforts to hinder implementation are
rewarded, an incentive is created for increased obstructive behavior in
the future.
Following this general discussion of the effects of rationalization
programs on processors, I consider two recently proposed concepts that
have received a considerable amount of discussion in the context of
rationalization programs in North Pacific fisheries. One is that
rationalization programs should satisfy the criterion of being ``Pareto
safe,'' which requires that no fishery entities be made worse as a
result of rationalization. The other is that an IFQ program should also
involve the allocation of Individual Processor Quotas (IPQs), as in the
proposed program for the BSAI crab fisheries.
The inclusion of both IFQs and IPQs in a rationalization program
was originally referred to as a ``two-pie'' system. The concepts of
Pareto safety and two-pie programs are linked in that the two-pie
approach has been advocated by Matulich and Sever (1999) as a feasible
way of achieving Pareto-safe rationalization in at least some policy-
relevant situations. In particular, Matulich and Sever claimed to have
shown that a two-pie allocation would be Pareto safe in a bilateral
monopoly, that is, a fishery with only one harvester and one processor,
and asserted that their analysis of this case would be applicable to
the BSAI inshore pollock fishery because it ``emulated'' a bilateral
monopoly.
Neither of these claims is correct. First, as discussed in
Halvorsen, Khalil, and Lawarree (2000), the characterization of the
BSAI inshore pollock fishery as a bilateral monopoly was highly
misleading because it ignored critical elements of the inshore
fishery's market structure. Second, and more importantly, my testimony
shows in section 6 that the claim that a two-pie allocation is
guaranteed to be Pareto safe in a bilateral monopoly is incorrect.
Therefore, even if a real-world fishery could be found that was a
bilateral monopoly, there is no reason to believe that a two-pie
allocation would be Pareto safe.
Since there are no other market structures for which the Pareto
safety of a policy feasible two-pie system has even been asserted, no
credence should be given to claims that a two-pie system is a ``policy-
superior initial allocation of rights'' (Matulich, Mittelhammer, and
Reberte 1996, page 112). Instead, the inclusion of IPQs in a fishery
rationalization program should be viewed as simply one possible
mechanism for enhancing outcomes for processors, bearing in mind that
the possible outcomes under IPQs have received very little credible
economic analysis and are untested by experience in any real-world
fishery.
I conclude that evaluations of the appropriateness of allocating
IPQs as part of a specific rationalization program should include (i)
an assessment of whether compensation for processors is desirable,
given the characteristics of the specific fishery, in particular the
balance of bargaining power, and (ii) the relative merits of IPQs
versus other possible compensation programs, given the characteristics
of the specific fishery.
1. Introduction
My testimony first addresses the general issue of whether fishery
rationalization programs should include explicit mechanisms to protect
processors from possible adverse effects. It then discusses one
specific mechanism, Individual Processor Quota (IPQ), which has been
recommended for use in the Bering Sea/Aleutian Islands (BSAI) crab
fisheries. I will discuss in turn the conditions determining the
probability, extent, and incidence of possible adverse effects of
rationalization programs on processors, and the efficiency and equity
rationales that have been advanced in favor of ensuring that processors
are protected from any such effects.
I will pay particular attention to two recently developed concepts
that have received a considerable amount of discussion in the context
of rationalization programs in North Pacific fisheries. One is that
rationalization programs should satisfy the criterion of being ``Pareto
safe,'' which requires that no fishery entities be made worse as a
result of rationalization. The other is that an IFQ program should also
involve the allocation of Individual Processor Quotas (IPQs) in what
has come to be known as the ``two-pie'' approach. The two concepts are
linked in that the two-pie system has been put forward as a feasible
way of achieving Pareto-safe rationalization in at least some policy-
relevant situations by Professor Scott Matulich and his co-authors
(Matulich, Mitte1harnmer, and Reberte 1996, Matulich and Sever 1999).
2. Theoretical Analysis of the Effects of IFQ Programs on Processors
Unless specified otherwise, the phrase ``IFQ program'' will refer
to a rationalization program based on the allocation of quota shares
for harvesting only, and in which the lFQs are allocated only to
harvesters. This is in contrast to possible rationalization programs
that include IFQs but also make explicit provision for compensating
processors for possible negative effects on them, either by allocating
part of the total harvesting quota shares to them or by incorporating
processing quotas as well, as in a ``two-pie'' system.
Also, the phrase ``processors'' will refer to the processors that
have been operating in the fishery prior to the time that it is
rationalized. One common result of rationalization is the entry of new
processors, who are obviously made better off by the opportunities
created by rationalization. An appraisal of the overall effect of a
rationalization program on the processing sector should clearly take
the positive effects on new processors into account as well. However,
my review of the theoretical analysis of the effects of IFQ programs on
processors will concentrate on the effects on previously operating
processors only.
In analyzing and predicting the effects of an IFQ program on the
well-being of incumbent processors, it is critical to take into account
the specific conditions of the fishery being considered. One important
set of conditions concerns the market structure of the fishery.
Implications of alternative models of market structure
The first analyses to emphasize the possibility of processors
incurring losses as a result of the introduction of an IFQ program
(Plesha and Riley 1992, Matulich, Mittelhammer and Reberte 1996)
assumed that the fishery was perfectly competitive, the end of the race
for fish created excess processing capacity with no alternative uses,
and the firms in the industry were not vertically integrated (that is,
processors did not own harvesters or vice versa). Given these
assumptions, they conclude that processors would be made worse off by
an IFQ program because they would fail to obtain any of the rents from
fish and would also lose part of the value of their capital.
However, if all other circumstances were the same, but processors
and harvesters were vertically integrated (as for example in a fishery
comprising only factory trawlers), then processors could not be made
worse off because they would receive the full benefits of the
rationalization program (Matulich and Sever 1999). In a mixed case,
with some processors vertically integrated and others not, the
incidence of gains and losses might differ by type of entity, with non-
integrated processors being more susceptible to suffering losses than
integrated ones (Halvorsen, Khalil, and Lawarree 2000).
Perfect competition is one of the standard models used in economic
analyses, in large part because of its analytical simplicity. Examples
of other standard models familiar from economic theory include monopoly
(a single harvester facing perfectly competitive processors), a
monopsony (a single processor facing perfectly competitive harvesters),
and a bilateral monopoly (a single harvester facing a single
processor). In the first case, the monopolist would obtain all the net
benefits of the fishery, in the second case the monopsonist would, and
in the third case the division of net benefits would depend, among
other things, on the alternative opportunities available to the
participants.
These three standard models also have the advantage of analytical
simplicity, but are not in general directly applicable to the analysis
of the effects of IFQ programs for two reasons. First, the
characteristics of the market structures of real-world fisheries are
more complex than such simple theoretical models imply. And second, if
a fishery did conform to one of these model specifications, then it
would be expected to be capable of maximizing aggregate net benefits on
its own, which would preclude the development of a race for fish. For
instance, a monopolist harvester would optimally allocate its fleet
over time rather than engaging in a race to fish between its own
vessels. Accordingly, rationalization programs such as an IFQ program
would be redundant.
However, consideration of these standard models does illustrate the
wide range of results possible with respect to the division of the net
benefits of a fishery, and therefore the need to take market structure
into account when assessing the effects of an IFQ program on the
participants in the fishery. Also, to the extent that a fishery being
considered for an IFQ program has characteristics similar to a standard
model, some inferences may be drawn about the probability that
processors could be adversely affected by the implementation of the
program. For example, other things equal, implementing an IFQ program
in a fishery with very few processors and many harvesters is less
likely to result in processor losses than in a fishery with many
processors and harvesters.
More generally, these examples suggest the importance of bargaining
power in determining the distributive effects of an IFQ program, and
therefore the need to use the tools of game theory to assess the
possible outcomes of a particular IFQ program. These tools include
cooperative bargaining theory (e.g., Nash 1953) and non-cooperative
bargaining theory (e.g., Osborne and Rubinstein 1990). A recent example
of an analysis of a fishery using cooperative and non-cooperative
bargaining theory is Halvorsen, Khalil, and Lawarree (2000).
Halvorsen, Khalil. and Lawarree Study
The analysis by Halvorsen, Khalil, and Lawarree was conducted on
behalf of the North Pacific Fishery Management Council and considered
the prospective distribution of net benefits from rationalization of
the inshore sector of the Bering Sea/Aleutian Islands (BSAI) fishery
under the American Fisheries Act (AFA). Although most of the specific
results of the analysis are directly applicable only to that particular
fishery, a brief review of the main elements of the analysis is useful
to illustrate the issues involved. The review also will be useful as
background for the evaluation of the two-pie allocation, which was
initially discussed in the context of the inshore pollock fishery.
Applicability of standard economic models:
Halvorsen, Khalil, and Lawarree evaluated, and rejected, the
suitability of several standard economic models that had been proposed
for application in the inshore pollock fishery. For example, Wilen
(1998) had argued that the inshore fishery was best characterized as a
single monopsony, in part because of the dominant position of two firms
in the main market for surimi products. Halvorsen, Khalil, and Lawarree
concluded that Wilen's analysis substantiated the hypothesis that
processors had significant market power, but that the fishery was not a
monopsony.
One reason given for rejecting the conclusion of monopsony was that
for the processors to behave as a monopsony they would have to overcome
serious economic and legal difficulties associated with being a
successful cartel. Also, there was evidence that the processors had not
always acted in a united way, as they would have if they were a
monopsony. For example, when the Bering Sea Marketing Association
(BSMA) went on strike against several processors in 1999, the largest
processor in the fishery, which was not a party to the negotiations,
had its fleet continue to fish, making prolongation of the strike too
costly to both the members of the BSMA and their processors. The
existence of the BSMA also argued against the conclusion that the
inshore sector was a monopsony, because its collective bargaining is
not consistent with harvesters acting as passive price takers. Lastly,
as noted above, an effective monopsony would have been capable of
substantially rationalizing the fishery without the intervention of the
AFA.
The existence of the BSMA was considered especially important by
Matulich and Sever (1999), who argued that it implied that the inshore
sector was a single bilateral monopoly. They claimed that the
dissemination of price information to each processor by the marketing
association during the course of negotiations allowed the processors to
unity even though they were not sharing information among themselves.
In other words, Matulich and Sever were claiming that the BSMA, acting
as the representative of independent catcher vessels, unwittingly made
it possible for the processors to unite against its own clients.
One serious factual problem with Matulich and Sever's analysis is
that the BSMA did not represent all of the independent catcher vessels,
and the largest processor was not a party to the negotiations. Also,
the theoretical analysis left two critical questions unanswered. First,
why would the marketing association not take advantage of the
processors' lack of communication and play one against the other by
misrepresenting received price offers? Second, even if it did not do
so, why would information on prices be sufficient to allow the
processors to overcome the other economic and legal difficulties
hindering their behavior as a single agent?
Another critical factual problem with Matulich and Sever's analysis
is that it ignored the existence of substantial vertical integration in
the fishery. Based on National Marine Fishery Service data, processor
controlled vessels harvested approximately half the total allocation of
catch to the inshore sector. This makes the existence of a united
harvesting sector implausible, because processor controlled vessels
would be subject to conflict of interest issues and could not be
expected to consider only the effects on harvesters of the results of
negotiations with processors.
Furthermore, the degree of vertical integration was not uniform
across processors. For example, two of the largest processors, which
together accounted for about two-fifths of the total inshore catch,
were estimated to obtain more than eighty percent of their fish from
their own processor controlled vessels, whereas another large
processor, with about one-fourth of the total inshore allocation,
obtained virtually all of its fish from independent catcher vessels.
The differences in the degree of vertical integration implied
differences in the effects of a given negotiated outcome, complicating
any effort of the processors or harvesters to act in unison.
Based on their assumption that the inshore sector was a bilateral
monopoly, Matulich and Sever (1999) recommended that a two-pie
rationalization approach be implemented, and claimed that it would
result in a Pareto-safe distribution of net benefits. However, as
discussed in section 6 below, Matulich and Sever's theoretical analysis
of the two-pie system under bilateral monopoly is fundamentally flawed,
and their conclusion that it would guarantee a Pareto safe outcome is
simply incorrect. Furthermore, even if their analysis of a two-pie
program under bilateral monopoly had been correct in theory, advocacy
of this particular policy approach for this specific fishery was based
on a highly misleading characterization of the fishery's market
structure.
Bargaining power:
Halvorsen, Khalil, and Lawarree (2000) used concepts from game
theory to evaluate the nature of competition in the industry, and the
resulting balance of bargaining power. They concluded that the
processors had a number of important bargaining advantages. The large
portion of the harvest caught by processor controlled vessels reduced
the reliance of the vertically integrated processors on supply from
independent catcher vessels, while also providing processors an
informational advantage because the independent catcher vessels they
bargained with did not own inshore processing plants. Also, because the
processing sector was highly concentrated and new entry was prohibited
under the AFA, processors would be expected to realize that aggressive
tactics yielding short-term gains were unlikely to be profitable in the
long-run. Independent catcher vessels did have one bargaining advantage
in that they were able to legally bargain as a group. However, it was
concluded that on balance the processors had substantially more
bargaining power than independent catcher vessels.
The Halvorsen, Khalil, and Lawarree analysis noted that
rationalization of the inshore pollock fishery was expected to result
in a large increase in the effective amount of processing capacity,
which would provide more opportunities for processors to engage in
aggressive competition, but the long-term incentives for refraining
from doing so would remain. Therefore they concluded that the
rationalized fishery would be characterized by ``moderate but not
cutthroat competition'' among processors.
These conclusions concerning bargaining power were then applied to
analyze two alternative rationalization programs being considered by
the Council: processor-specific cooperatives (an implicit processor
compensation mechanism) and the Dooley-Hall proposal for non-processor-
specific cooperatives (an approximation to IFQs). Halvorsen, Khalil,
and Lawarree concluded that there was a significant probability that
some independent catcher vessels would be adversely affected by the
requirement of processor specific cooperatives, but the positive net
benefits from the reallocation aspect of the AFA, as well as the
potential net economic benefits from rationalization of the fishery,
decreased the likelihood that they would be adversely affected relative
to the situation pre-AFA. They also concluded that the Dooley-Hall
proposal would be more favorable to independent catcher vessels, and
less favorable to processors, than the processor-specific cooperatives.
Their conclusions concerning the relative bargaining power of
harvesters and processors in the inshore BSAI pollock fishery would
also have been relevant to the analysis of the effects on processors of
alternative rationalization programs including IFQs. However, it is
important to note that the conclusions were based on the conditions in
this specific fishery. Because market structure is critical in
determining the distributive outcomes of IFQ programs, and the
characteristics of market structure differ greatly across fisheries,
the distributive effects of rationalization policies require fishery
specific analysis.
The requirement of rigorous, fishery-specific, analysis of the
probable effects of IFQ programs is especially important because very
little information is available on the actual effects of previous
programs. The following section reviews several recent studies that
have evaluated the actual effects on processors of previous IFQ
programs.
3. Empirical Analysis of the Effects of IFQ Programs on Processors
There has been surprisingly little empirical research on the actual
effects of IFQ programs on the processing sector. However, there are
three studies that have examined the effects of IFQ programs in the
United States and British Columbia halibut fisheries. All three studies
document the fact that there were many winners from the programs in the
form of new processors who entered the fisheries in order to benefit
from the opportunities created by the IFQ programs. However, one study
differs from the others with respect to its conclusions concerning the
effects of the IFQ programs on processors that had been operating in
the fishery before the programs were adopted.
British Columbia Study
The British Columbia halibut fishery was studied by Casey et al.,
(1995). The IFQ program was implemented in May 1991 and the study is
based on the results of in-person interviews with processors in
September 1993 and a mail survey of harvesters in May 1994. The study
documents large net economic benefits from rationalization of this
fishery, including the ability to switch from mostly frozen products to
more highly valued fresh fish. Ex vessel price is estimated to have
increased by more than half. The survey of processors indicated that
some of the processors that had been operating prior to the IFQ program
felt that it had made them better off while some felt it had made them
worse off.
General Accounting Office Study
The GAO study (United States General Accounting Office 2002) based
its determination of the effect of IFQ programs on processors on an
assessment of the economic effects of the North Pacific halibut and
sablefish IFQ program. The study methodology included analysis of data
from the National Marine Fisheries Service and the Alaskan Department
of Fish and Game, interviews with fishery participants, and a review of
a study commissioned by the Alaskan Department of Fish and Game
(Matulich and Clark 2002).
The GAO's summary of its main conclusions was (page 20): ``Some
processors were adversely affected by the implementation of the halibut
and sablefish IFQ program while others benefited. However, quantifying
the economic effects of the IFQ program on processors is difficult
because much of the data needed to measure changes in profitability are
proprietary. Furthermore, other factors besides the IFQ program may
lead to changes in processors' economic situation.''
The GAO's review noted that the Matulich and Clark study of the
halibut and sablefish IFQ program had concluded that processors were
hurt significantly by the IFQ program. However, the GAO also noted that
it could not validate or replicate the study's results, and that it had
identified a number of problems with the study's methodology and scope
that brought into question the reliability of the study's estimates.
Among the problems identified by the GAO were: the study's assumption
that all costs except labor and material inputs were constant over a
seven year period was clearly unjustified, the choice of base year
exaggerated the size of any negative effect, the results might not be
representative of the industry as a whole, and the document requesting
economic information from processors might have biased participant
responses.
In April 2002 I presented a critique of the Matulich and Clark
study in public testimony before the North Pacific Fishery Management
Council. The conclusions I reached concerning the reliability of the
study's results were even more negative than those reached by the GAO.
My analysis of the study is summarized in the following sub section.
Matulich and Clark Study
The Matulich and Clark study assumes that the effect of the IFQ
program on processors' economic welfare can be measured as the change
in ``quasi rents'' retained by processors, which they define as the
change in revenues in excess of all variable processing costs. However,
the use of this measure to evaluate welfare changes is not consistent
with economic theory and would not provide reliable estimates of
changes in welfare even if it were estimated accurately. Furthermore,
their empirical methodology is deeply flawed and would be incapable of
providing reliable estimates of welfare change even if a theoretically
correct measure were being used. In short, the study measures the wrong
thing, and measures it poorly.
Theoretical assumptions:
Quasi-rents are fundamentally a short-run concept. The short-run is
defined as the period of time during which at least one of the firm's
input quantities cannot be changed. It should be noted that, although
it is customary for expository reasons to use capital inputs as
examples of fixed inputs, and labor and materials inputs as examples of
variable inputs, some capital inputs may in fact be variable in the
short-run (e.g., motor vehicles, personal computers), and some labor
and materials inputs may be fixed (e.g., because of transportation
costs, job-specific human capital, or contractual commitments).
The difference in the short-run between the firm's total revenue
and total expenditures on variable inputs is defined as the quasi-rent
to the fixed inputs. That is, the amount in excess of the amount
required to keep the fixed inputs in their current use. Thus a decrease
in quasi-rents would indicate a decrease in the firms' welfare in the
short-run.
In the long-run, by definition, all inputs are variable, and the
amount required to keep inputs in their current use is equal to what
they could earn elsewhere, including a normal rate of return to
capital. Therefore the relevant concept for measuring a firm's welfare
change in the long-run is that of economic profit. Quasi-rents would
have no operational meaning, being simply equal to economic profit if
correctly measured.
There is no direct connection between the economic concepts of the
short-and long-run and calendar time. Instead, the amount of time
required before all inputs can be considered variable will vary across
industries, although it is plausible that in any given industry the
number of inputs that are fixed will decrease with the length of time
being considered.
Thus the first-step in attempting to use quasi-rent data to measure
changes in a firm's welfare should be a careful evaluation of which, if
any, inputs are fixed, given the period of time over which changes are
being evaluated. Simply assuming that labor and material inputs are
variable and all other inputs are fixed, as done by Matulich and Clark,
is not adequate even for a period as short as a year, and is clearly
unjustified for the seven-year period over which they evaluated
changes.
To illustrate the type of error their assumption can introduce in
the measurement of welfare change, consider the results of applying
Matulich and Clark's definition of quasi-rents to evaluate changes that
are distant enough in time for all inputs to be variable. Further
suppose for ease of exposition that in both periods the price of
processed fish is $1.00, the cost of raw fish is $0.40, and average
cost is $0.50. Thus economic profit per unit would be equal to $0.10 in
both periods and the firm's welfare would be unchanged. Nevertheless,
if the firm had become more labor intensive over time, the unit quasi-
rent as calculated by Matulich and Clark would have indicated a
decrease in welfare. For example, if average costs were split equally
between capital and labor costs in the first period, but labor costs
accounted for 80 percent of average costs in the second period, the
unit quasi-rent as calculated by Matulich and Clark would have
decreased from $0.35 to $0.20, a decrease of 43 percent.
While this example is hypothetical, it does illustrate that quasi-
rent as evaluated by Matulich and Clark does not provide reliable
estimates of changes in welfare over longer periods of time. More
specifically, increases in labor intensity, other things equal, will
result in decreases in welfare as evaluated by their measure.
Lastly, even if reliable estimates of welfare changes were
obtained, their normative significance would depend in part on the
benchmark on which they were based. Matulich and Clark choose as their
benchmark the welfare of processors in 1992-1993, asserting that this
period represented an open-access long-run equilibrium. One reason for
doubting this assumption is that it is not clear that the fishery would
have stabilized at the 1992 levels in the absence of an IFQ program. If
not, then a more appropriate benchmark would be the counterfactual case
of how the fishery would have developed in the absence of a program.
More directly, the 1992-1993 experience reflected the fact that
fishery participants expected an IFQ program to be implemented, and
this provided incentives for different behavior than would have
occurred in an open-access equilibrium. For example, harvesters might
have considered it worthwhile to fish at a loss in order to try to
capture or protect catch history.
Empirical methodology:
Matulich and Clark obtained the data used to estimate changes in
quasi-rents from a questionnaire distributed to a sample of processors.
The principal types of data requested were total revenue, total raw
fish cost, and total variable processing costs, defined as the
aggregate of several specific cost elements, including custom
processing fees, wage and housing costs for direct labor, and packaging
and freight costs. Data on total revenue and total raw fish cost were
verifiable from Alaska Department of Fish and Game data, the data for
variable processing costs were not.
Economists using survey research techniques have noted that the
design of a questionnaire can result in a number of different types of
biases. In particular, respondents may engage in strategic
misrepresentation of the data if it is clearly in their economic
interests to do so. Therefore, one of the most important protocols for
survey design is to avoid providing material that establishes a clear
link between a participant's responses and his or her economic
interests.
The survey design used by Matulich and Clark clearly violates this
protocol. The material accompanying the questionnaire noted that
Professor Matulich was the principal investigator, that he had written
an article showing that the type of program used for halibut and
sablefish transfers wealth from processors to harvesters, and that the
purpose of this survey was to see if that was true empirically. It was
further noted that the purpose of the study was to obtain information
for use in evaluating future rationalization programs; in particular to
help policy makers to avoid unintended distributive effects, and that
distributional impacts would be based on measuring changes in
processors' total revenue minus various processing costs.
This material would have made it clear to the processors how their
responses could benefit or harm them in the future when other fisheries
in which they participated were rationalized, and this would have
provided an incentive for strategic misrepresentation. At a minimum,
processors that had benefited from the IFQ program would realize that
reporting this might be harmful to their future interests, and
therefore have simply not participated in the survey.
An important difficulty in assessing the treatment of a number of
empirical issues is that the discussion in the report is often
qualitative where it would normally be expected to be quantitative.
Examples include the section on data problems where it is noted that it
was ``not uncommon'' for aggregation problems to prevent accurately
measuring variable processing costs, and that ``some'' firms were
unable to access historical data. No information is provided on the
number of firms that were eliminated from the sample for these reasons.
Similarly, they report that there were a ``few'' instances'' where
inventory issues were ``problematic''.
In addition, Matulich and Clark report that some firms were
considered to be outliers, usually by exhibiting unrealistically high
quasi-rents. These firms were contacted for an explanation, and if it
was not satisfactory, the firm was dropped from the sample or its data
replaced by the sample average. The number of firms considered
outliers, how many were considered to report too high quasi-rents, how
many justified their data, were dropped, or had their data replaced by
sample averages, is not specified.
Lastly, and most surprisingly, Matulich and Clark do not report the
number of participants included in the final data. They report that the
number of buyers/processors asked to participate in the survey was 53
for halibut and 46 for sablefish, accounting for 88 percent to 96
percent of all fish purchased, and that the retained survey data
accounts for 52 percent to 61 percent of fish purchased. Given the
degree of concentration in these fisheries, this may or may not
represent a substantial percentage of the number of total firms.
Matulich and Clark do note that the final sample does not include
data for any firms that exited the fisheries, which accounted for one-
fifth of the total 1992-1993 catch in both fisheries. Although these
firms might be expected to have been less profitable than the surviving
firms, they are assumed to have had the same quasi-rent share in 1992-
1993 as the surviving firms. Similarly, surviving firms that lost
market share are assumed to have had the same quasi-rent share as the
firms increasing market share.
Matulich and Clark conclude from their analysis that 82 percent of
the halibut processing sector (as measured by raw fish weight rather
than number of firms) lost quasi-rents relative to the pre-IFQ period,
with the average loss being 56 percent. Even more dramatic results are
reported for the sablefish processing sector. However these results
cannot be considered to be reflective of the actual effects on the
economic welfare of processors. The basic problem with their approach
is that the results depend on the estimates of total variable
processing costs, which in turn depend on arbitrary, and unrealistic,
assumptions concerning which inputs are variable over a seven-year time
span. In addition, estimation of total variable costs conditional on
these assumptions depends on survey data from processors, who can be
expected to clearly realize that there are incentives for strategic
misrepresentation.
4. Rationales for Compensation
As discussed in the previous two sections, there are no general
theoretical or empirical grounds for concluding that processors as a
whole are likely to be adversely affected by the implementation of an
IFQ program. However, it is possible that situations could arise where
processors would be adversely affected. Therefore it is useful to
consider whether it would be desirable to incorporate measures to
prevent such adverse effects from occurring, or to compensate
processors if they do occur. Arguments have been put forward for
preventing processors from being harmed by rationalization programs
based on considerations of both efficiency and equity. Each type of
rationale will be considered in turn.
Efficiency
One rationale advanced for compensating processors for possible
losses is that not doing so could have adverse consequences for
economic efficiency by creating impediments to the implementation of
efficiency-enhancing rationalization programs. This possible source of
inefficiency is emphasized by Matulich, Mittelhammer, and Reberte
(1996). Having concluded that processors could suffer losses as the
result of the introduction of IFQs in a perfectly competitive fishery,
they note (page 112), ``These losses could promote political gridlock
and jeopardize adoption of an ITQ policy unless they are fully
compensated or redistribution is avoided by a policy-superior initial
allocation of rights to both harvesters and processors.''
This argument assumes that processors do not have enough economic
bargaining power in rationalized fisheries to avoid losses, but do have
enough political bargaining power to block efficiency-enhancing
rationalization programs. However, as the Halvorsen, Khalil, and
Lawarree analysis of the BSAI pollock fishery indicated, processors may
in fact have more bargaining power than harvesters in real-world
fisheries.
Therefore, a situation in which processors seek rent-enhancing
mechanisms as the price of agreeing to rationalization programs may
simply reflect the desire of processors to obtain a larger share of the
net benefits the program would create, rather than that they are
seeking to protect themselves from suffering losses. Under these
circumstances, utilizing mechanisms to enhance the processors' share of
the net benefits could actually create new impediments to the
implementation of rationalization programs by causing harvesters to
fear that they would lose out from the implementation of the program.
The history of the American Fisheries Act is instructive in this
regard. Rationalization of the BSAI pollock fishery was based on the
creation of harvesting cooperatives. Processors in the inshore sector
expressed concern that cooperatives might put them at a bargaining
disadvantage. In response, the AFA rules for cooperatives required that
they be processor-specific, and that membership in the cooperative for
each processor was limited to vessels that were qualified for that
processor, as determined by where a catcher vessel had previously
delivered the largest share of its total catch.
In response, an association of independent catcher vessel owners
expressed concern that the AFA rules for inshore cooperatives would
harm them because of the restrictions placed on where they could market
their fish, and proposed an alternative set of rules known as the
Dooley-Hall proposal. Resolution of this conflict required extensive
hearings before the North Pacific Fisheries Management Council. In
addition, concerns were raised about the effects of processor-specific
cooperatives on small entities as defined in the Regulatory Flexibility
Act.
Another possible disadvantage of responding to processors'
resistance to the adoption of a rationalization program by
incorporating mechanisms to enhance their position is that doing so
might have adverse efficiency consequences in the longer run. If it
appears that policy makers are willing to appease opponents of
rationalization by enhancing their rewards, this will provide
incentives for increased obstructive behavior in the future, and
thereby imperil the implementation of rationalization programs in other
fisheries.
Equity
The other principle rationale for compensating processors against
possible losses is that it would be inequitable not to do so. For
example, Plesha and Riley (1992) and Matulich, Mittelhammer, and
Reberte (1996) argue that there is a Fifth Amendment ``taking'' issue
if the rationalization of a fishery results in losses for processors.
Without attempting to address the implied legal issues, some
observations can be made on whether investment losses from
rationalization are inequitable from an economic point of few. It seems
unlikely that the investments assumed to be at risk from
rationalization were made with the anticipation that the race for fish
was certain to be the long-run equilibrium outcome for the fishery in
question. Faced with an uncertain future, processors' investment
decisions can be assumed to have taken into account the possibility of
various alternative scenarios, including regulatory policies to end the
race for fish. Accordingly, investment decisions would be optimized to
reflect trade-offs between the various possible future outcomes. For
example, there might be a trade-off between further increasing
investment in capacity in order to secure competitive advantages by,
for example, deterring the entry of new processing firms into the
fishery, versus the advantage of having the smaller amount of capacity
that would be optimal if the race for fish were ended. It is not clear
why losses that had been anticipated to occur under a particular
scenario should instead be compensated on equity grounds when that
scenario turns out to be the actual outcome.
Another equity issue concerns the distribution of net benefits
within the processing sector. For example, as noted above, in a
processing sector comprising some firms that are vertically-integrated
and some that are not, the non-integrated processors would be more
susceptible to suffering losses from rationalization than would the
integrated processors. But the choice to not be vertically-integrated
presumably reflects a judgement by these firms that they obtained
enough economic advantages by refraining from acquiring harvesting
capacity to compensate for the increased risk of losses if the fishery
were rationalized. Adopting a policy to compensate all processing firms
for possible losses would change the anticipated benefits and costs of
these business decisions after the fact and thereby effectively
discriminate in favor of the non-integrated firms, partly at the cost
of harvesters.
Matulich and Sever (1999) use the term ``Pareto safe'' to refer to
the concept of a rationalization program that is ``equitable in the
sense of not redistributing status quo ante wealth of historical
participants'' (page 204). They then argue that if a rationalization
program is not Pareto safe, ``politically powerful interest groups may
form to block a switch to ITQ management, jeopardizing the efficiency
benefits of rights-based fishing (page 215). The desirability, and
feasibility, of relying on the concept of Pareto safety in designing
and evaluating fishery rationalization polices is discussed in the
following section.
5. Pareto Safe Rationalization
Although the term ``Pareto safe'' appears to have originated in the
writings of Matulich and his co-authors, essentially the same concept
has been long known in the economic policy analysis literature as the
Pareto criterion. ``The logical criterion for proving that a policy
change, or any other change, is beneficial was first stated by a
nineteenth century Italian social scientist, Vilfredo Pareto. Pareto's
rule is very simple: Program X improves the welfare of the society if
it makes at least one person better off and no one worse off.''
(Gramlich 1990).
The recommendation by Matulich and his co-authors that fishery
rationalization policies should be required to be Pareto safe is
equivalent to saying that they should satisfy the Pareto criterion.
However, the Pareto criterion only provides information on a policy's
effect on economic welfare when the policy would result in no
individual being made worse off. A policy that involved small losses to
one individual, and large gains to many others, would fail the Pareto
criterion, even though it might have a large positive effect on
economic welfare. And virtually all feasible public policies result in
at least one individual being made worse off.
This has led to the general rejection of the Pareto criterion as a
practical basis for evaluating public policies. As Ng (1984, page 1033)
summarizes, ``The Pareto criterion is widely accepted as a sufficient
condition for an improvement in social welfare. . . . However, most, if
not all, changes in the real world involve making some better off and
some (no matter how small the number) worse off. Thus the Pareto
criterion in itself is of little practical use.''
The practical difficulties of trying to implement Pareto safe
fishery rationalization policies can be illustrated by considering the
effects on individual harvesters of implementing an IFQ program.
Matulich, Mittelhammer and Reberte (1996, page 112) indicate that an
IFQ policy would be Pareto safe within the harvesting sector, because
``endowing individual harvesters with fully transferable, permanent,
and exclusive fishing rights is tantamount to assigning property rights
over the fish stock . . . [an important benefit] . . . arises out of
gains from free trade in which more efficient users of the resource are
able to purchase rights from less efficient users. Such trade fully
compensates the sellers.'' While this is a reasonable summary of the
efficiency arguments in favor of IFQs, it does not provide a basis for
concluding that no individual harvesters are made worse off.
There are at least two ways in which individual harvesters can be
made worse off under an IFQ program. First, it is not feasible to
ensure that the original distribution of quota among harvesters matches
their actual participation in the fishery. For example, a standard
procedure is to base quota share allocations on catch history over some
historic period. If a participant's harvest was unusually low during
all or part of that period he may not receive sufficient quota to leave
him as well off as before. Similarly, if the catch history period is
not fairly recent, a large proportion of the quota shares may go to
individuals no longer active in the fishery rather than to those
currently active (see, for example, North Pacific Fishery Management
Council 2002, Appendix 2-7, page 8). Second, the assumption that the
price of quota will fully compensate the sellers depends on the
implicit assumption that the market for quotas is perfectly
competitive, which need not be the case (Anderson 1991).
It should be noted that similar issues could arise in a program
involving the allocation of individual processor quotas. The allocation
of the quotas might not reflect an individual processor's actual
participation in the fishery, for example if a facility was
incapacitated during part of the historic period used to determine
shares. And fisheries with a small number of processing firms, or a few
large and many small firms, are particularly susceptible to market
imperfections that might prevent the price of quota from fully
compensating the seller.
Thus the Pareto safe concept is not of much practical help in
evaluating the effects of fishery rationalization programs at the
individual participant level. Matulich and his co-authors in fact
rarely refer to applying the Pareto safe concept at this level, but
instead focus on Pareto safety at the level of the aggregate harvesting
and processing sectors. In particular, as noted above in section 4,
they suggest that a rationalization policy is unlikely to be adopted if
it would create uncompensated losses for the processing sector.
Matulich, Mittelhammer and Reberte (1996, page 126) speculate that
a Pareto safe allocation might be obtained under a ``symmetrical rights
distribution'' and suggest, ``Candidates worthy of consideration
include (i) a split of harvest quota shares between fishers and
processors; (ii) a ``two-pie'' allocation, in which catching rights are
awarded to fishers and processing rights are awarded to processors; and
(iii) full-utilization quota shares. . . .''
Matulich and Sever (1999) investigate the properties of the first
two of these proposals, referring to the first one as a ``one-pie split
allocation.'' They first consider the application of the one-and two-
pie allocations to a fishery that is perfectly competitive and conclude
that neither type of allocation is capable of providing policy feasible
Pareto safe outcomes. They then consider the application of these
allocations to a bilateral monopoly. The one-pie allocation is again
concluded to not be capable of providing policy feasible Pareto safe
outcomes. However, they claim to show that the two-pie system is
guaranteed to be Pareto safe not only at the aggregate level but also
at the level of individual participants. The validity of this
remarkable claim is discussed in the following section.
6. Two-Pie Allocations and Pareto Safety
Matulich and Sever's attempt to demonstrate that a two-pie
allocation would be Pareto safe in a bilateral monopoly is based on a
series of dubious assumptions. The first is their assumption that the
bilateral monopoly would be able to negotiate an ex vessel price that
maximized joint profits under conditions of a race for fish, but would
be unable to negotiate rationalization measures that would end the race
for fish and thereby increase the potential joint profits. No
explanation is given for this assumed constraint on the bilateral
monopoly's ability to maximize joint profits. Instead it is simply
implicitly assumed that the race for fish can be ended only by an
externally imposed rationalization program.
In their analysis of the two-pie allocation, efficiency is assumed
to be attained through quota trading, and to be independent of the
bargained ex vessel price. In particular, they note that the ex vessel
price might be outside of the Pareto safe range. However, they argue
that the actual price will fall within the Pareto safe range because
(page 214):
``While the efficient price does not guarantee Pareto safety,
intrinsic bargaining behavior should, provided the bargaining
association is responsive to the well being of its entire
membership. Bargaining agents have internal incentives to
negotiate a price that not only maximizes joint profits
(efficiency) but also leaves no member worse off . . . at least
one Pareto-safe price exists-the open access ex vessel price,
P\0\ . . . As long as the parties desire to reach a Pareto-safe
agreement, they can do so by settling on a rent share that
implies P\0\ as the ex vessel price. Thus, there are no
functional impediments to achieving an efficient price that is
also Pareto safe.''
Matulich and Sever then use the Nash (1953) bargaining solution
concept to indicate how the rent shares might be determined, given that
``the bargaining agents are assumed to act so as to leave no member
worse off under ITQs relative to open access'' (page 214). The solution
of the Nash model does not strengthen the claim that the two pie
allocation is Pareto safe, but instead is performed under the
assumption that the price must fall with the Pareto safe range.
In short, Matulich and Sever's claim that ``[e]ach sector is
guaranteed sufficient quasi-rents to make all members Pareto safe''
(page 214) is based simply on the assumption that bargaining agents
will want and be able to set prices that are Pareto safe for all their
members. This assumption is merely asserted, rather than derived from
economic theory, and is unlikely to be satisfied in a real-world
fishery, in which each side would contain heterogeneous participants.
It is not obvious, and Matulich and Sever do not suggest, how such a
difficult principal-agent problem in each sector could be structured so
that the agent is constrained to leave no member worse off.
Even if it is assumed for the sake of argument that both sides of
the bilateral monopoly desire agreements that are Pareto safe as among
their own members, a Pareto safe price need not be the outcome of the
bilateral monopoly negotiation. This can be demonstrated using a Nash
bargaining model with the outside options for both sides correctly
specified.
To determine the outside option for the harvester sector of the
bilateral monopoly, consider what its alternative would be if it did
not reach an agreement with the processor sector. Because it would have
IFQs it could harvest the fish, but the processor sector could simply
refuse to process the harvest. Therefore the outside option for the
harvester sector is zero rent. Similarly, the harvester sector could
threaten to not fish, so that the outside option for the processor
sector is also zero rent, assuming that it has no processor controlled
vessels. With these outside options, there is no reason to assume that
the bargaining outcome would be Pareto safe. And if the processor
sector also owns some catcher vessels, the outcome could be very
unfavorable for harvesters, as shown in Halvorsen, Khalil, and Lawarree
(2000).
To summarize, Matulich and Sever's claim that they have shown that
a two-pie allocation is guaranteed to be Pareto safe under bilateral
monopoly is incorrect, and there is no other market structure for which
this claim has even been made. Therefore, no credence should be given
to claims that a two-pie system is a ``policy-superior initial
allocation of rights'' (Matulich, Mittelhammer and Reberte 1996, page
112). Instead, the inclusion of IPQs in a fishery rationalization
program should be viewed as simply one possible mechanism for enhancing
outcomes for processors, bearing in mind that the possible outcomes
under IPQs have received very little credible economic analysis and are
untested by experience in any real-world fishery.
References
Anderson, Lee G. (1991), ``A Note on Market Power in ITQ
Fisheries,'' Journal of Environmental Economics and Management, 21 (3):
291-296.
Casey, Keith E., Christopher M. Dewees, Bruce R. Turris, and James
E. Wilen (1995), ``The Effects of Individual Vessel Quotas in the
British Columbia Halibut Industry,'' Marine Resource Economics, 10 (5):
211-230.
Gramlich, Edward M. (1990), A Guide to Benefit-Cost Analysis,
Second Edition, Prentice Hall.
Halvorsen, Robert, Fahad Khalil, and Jacques Lawarree (2000),
``Inshore Sector Catcher Vessel Cooperatives in the Bering Sea/Aleutian
Islands Pollock Fisheries,'' discussion paper prepared for the North
Pacific Fishery Management Council, February 7.
Matulich, Scott C. and Michael Clark (2002), ``Efficiency and
Equity Choices in Fishery Rationalization Policy Design: An Examination
of the North Pacific Halibut and Sablefish IFQ Policy Impacts on
Processors,'' Regional Information Report No. 5J02-02, Alaska
Department of Fish and Game, January 24.
Matulich, Scott C., Ron C. Mittelhammer, and Carlos Reberte (1996),
``Toward a More Complete Model of Individual Transferable Fishing
Quotas: Implications of Incorporating the Processing Sector,'' Journal
of Environmental Economics and Management, 31 (1): 112-28.
Matulich, Scott C. and Murat Sever (1999), ``Reconsidering the
Initial Allocation of ITQs: The Search for a Pareto-Safe Allocation
Between Fishing and Processing Sectors,'' Land Economics, 75 (2): 203-
19.
Nash, John F. (1953), ``Two-Person Cooperative Games,''
Econometrica, 21 (1): 128-40. Ng, Yew-Kwang (1984), ``Quasi-Pareto
Social Improvements,'' American Economic Review, 74 (5): 1033-1050.
North Pacific Fishery Management Council (2002), ``BSAI Crab
Rationalization Program Alternatives,'' Initial Council Review Draft,
January 22.
Osborne, Martin and Ariel Rubinstein, Bargaining and Markets,
Academic Press, 1990. Plesha, Joseph T. and Christopher C. Riley
(1992), ``The Allocation of Individual Transferable Quotas to Investors
in the Seafood Industry of the North Pacific,'' photocopied.
United States General Accounting Office (2002), ``Individual
Fishing Quotas: Better Information Could Improve Program Management,''
December.
Wilen, James E. (1998), ``Testimony to be Presented to the North
Pacific Fishery Management Council Regarding Inshore/Offshore III,
Bering Sea--Aleutian Islands Pollock Allocation,'' photocopied.
Appendix PMajor Differences Between AFA Inshore Cooperatives and Council's Crab Rationalization Program
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
AFA Inshore Pollock Cooperatives Proposed BSAI Crab Rationalization
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Allocation of processing history:
Initial allocation 90% of previous year's deliveries 90% of historic share of deliveries
Guaranteed percent Zero 100%
Individual vessels' ability to reallocate 10% per year, cumulative Zero
Percent of vessels' approval required to retain 80% Zero
Regionalization:
Decreased efficiency gains No Yes
Increased processor concentration No Yes
Complexity:
Number of markets 7 cooperatives 15 regional fisheries
Number of prices Maximum of 14 Maximum of 75
Need for price arbitration No Yes
Monitoring and enforcement costs Much less than in IFQ program Much more than in IFQ program
Large fluctuations in TAC No Yes
Total benefits from rationalization:
Increase in total TAC Yes No
Buyback program No Yes
Increase in value of product Large Moderate
Efficiency benefits from cooperatives Scheduling, avoidance of under harvesting None if cooperatives do not form
Information base for program evaluation
Experience in other fisheries Whiting and offshore pollock cooperatives IPQs never before used in any fishery
Experience in other markets Agricultural cooperatives None
Previous theoretical analysis Extensive for IFQs and cooperatives Very little
Previous empirical analysis Extensive for IFQs and cooperatives None
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
______
Statement of Robert Halvorsen, Professor of Economics, Department of
Economics, University of Washington
On Implications of the American Fisheries Act for Bering Sea/Aleutian
Islands Crab Rationalization
On behalf of the City of Kodiak, I am submitting this written
testimony to the record of the May 20, 2003 Hearing on Bering Sea/
Aleutian Island Crab Rationalization before the Committee on Commerce,
Science, and Transportation of the United States Senate. My testimony
discusses the rationalization program for the Bering Sea/Aleutian
Islands inshore pollock fishery that was implemented under the American
Fisheries Act of 1998, and the implications that can be appropriately
drawn from this experience regarding the proposed rationalization
program for the BSAI crab fisheries.
The City of Kodiak also asked me to submit written testimony on the
general issue of whether fishery rationalization programs should
include explicit mechanisms to protect processors from possible adverse
effects, with specific consideration of Individual Processor Quotas
(IPQs) as one such mechanism. My testimony on this topic has been
submitted separately and will be referred to below as Halvorsen (2003).
Executive Summary
The American Fisheries Act (AFA) included both a reallocation of
total allowable catch from the offshore sector to the inshore sector
and a rationalization program for the inshore sector based on
cooperatives that were tied to individual processing facilities. In
response to controversy over the AFA rules for cooperatives, the North
Pacific Fishery Management Council commissioned studies of the effects
of the proposed rules as well as of several alternatives (Halvorsen,
Khalil and Lawarree, 1999 and 2000).
The studies concluded that the qualification requirements for
membership in a cooperative disadvantaged independent catcher vessels.
However, the rule that a cooperative could deliver 10 percent of its
harvest to an alternative eligible processing facility, and, more
importantly, that continued existence of a cooperative required
approval by 80 percent of its members, decreased the probability that
independent catcher vessel owners would be adversely affected by the
AFA provisions for cooperatives. Also, the reallocation aspect of the
AFA, as well as the potential efficiency benefits from rationalization
of the fishery, would result in large net economic benefits for the
inshore fishery as a whole.
The principal alternative to the AFA cooperatives that was
considered in the studies was the ``Dooley-Hall proposal'' under which
cooperatives would be allowed to deliver to any eligible processing
facility and any eligible catcher vessel would be allowed to join any
cooperative. The studies concluded that, for the purpose of evaluating
the effects on the distribution of the net benefits of the fishery, the
Dooley Hall proposal could be analyzed as if it were an IFQ program.
Two previous theoretical analyses of the distributional effects of
an IFQ program (Plesha and Riley, 1992 and Matulich, Mittelhammer, and
Reberte, 1996) had concluded that an IFQ program would have highly
adverse effects on processors, and Professor Matulich argued that this
would be the result of implementing the Dooley-Hall proposal in the
inshore pollock fishery. However, the theoretical analyses had little
relevance for the inshore pollock fishery. They assumed cutthroat
competition among processors, whereas this was highly unlikely given
the degree of concentration and other market structure characteristics.
Also, the models assumed that processors received no IFQs but in the
pollock fishery processor-controlled vessels would receive at least
half of the total harvest allocation. The models also ignored the
existence of differentiated capital, informational rents, and
alternative uses for capital.
The Halvorsen, Khalil and Lawarree studies included a consideration
of the results of the Dooley-Hall proposal under the extreme
alternative assumption that the fishery was a bilateral monopoly, with
processors behaving as if they were one entity (a monopsony), and a
Dooley-Hall cooperative containing all independent catcher vessels
acting as a monopoly. It was noted that the bilateral monopoly model
was highly unlikely to be an appropriate characterization of the
fishery. One reason for considering this model was that it was the
market structure for the inshore pollock fishery that Matulich and
Sever (1999) had assumed in their study of a ``two-pie'' allocation.
Halvorsen, Khalil and Lawarree concluded that under a bilateral
monopoly processors would attain at least three-fourths of the total
net benefits of the fishery.
The analysis of the Dooley-Hall proposal under the assumption of
either cutthroat competition or bilateral monopoly did not provide a
realistic prediction of actual outcomes in the inshore pollock fishery.
Under more realistic assumptions concerning the fishery the most
reasonable outcome was that independent catcher vessel owners would
fare somewhat better, and processors would fare somewhat worse, under
the Dooley-Hall proposal than under the AFA, but this did not imply
that a rationalization program based on the Dooley-Hall proposal would
have adverse effects on processors.
The Council decided to retain the original AFA provisions for
cooperatives in implementing the rationalization program for the
inshore pollock fishery. The actual experience of the fishery
subsequent to rationalization has been consistent with the overall
economic analysis of Halvorsen, Khalil, and Lawarree. Total net
economic benefits in the inshore sector were increased because of the
increase in the sector's share of the total allowable catch (TAC), as
well as substantial increases in the TAC itself. In addition,
implementation of the cooperatives has resulted in increased
efficiency, including increases in the value of output and decreases in
harvesting costs. As a result, the net effects of the AFA on both
harvesters and processors is generally agreed to have been positive,
although the relative size of the benefits remains a matter of
contention.
In its deliberations concerning rationalization of the BSAI crab
fisheries, the Council considered the use of a program similar to the
AFA cooperatives in the inshore pollock fishery, as well as an entirely
different type of rationalization program based on the use of both IFQs
for harvesters and individual processing quotas (IPQs) for processors.
As in the case of the inshore pollock fishery, the Council commissioned
an independent economic analysis of the proposed alternatives. The
study, Milon and Hamilton (2002), concluded that a two-pie system in
which processors were allocated IPQs for the entire harvest would
result in the processors capturing all of the net economic benefits
from the fishery.
The study also considered a program with both Class A harvesting
quotas that must be delivered to processors that hold IPQs, and Class B
harvesting quotas that may be delivered to any processor. As the ratio
of B to A quotas increases, harvesters fare better. If all harvesting
quotas are Class B, the outcome is the same as for an IFQ only program,
with the likelihood that both harvesters and processors would benefit
from rationalization.
The Milon and Hamilton study also considers the effects of
regionalization of the fisheries. It concludes that regionalization
reduces cost efficiency by imposing constraints that prevent harvest
within a region being transferable to other regions. In addition,
regionalization increases the market power of processors, resulting in
lower ex vessel prices.
Milon and Hamilton's conclusion that both harvesters and processors
are able to gain under an IFQ program that did not include IPQs is
consistent with Halvorsen, Khalil, and Lawarree's analysis of the
Dooley-Hall proposal under the AFA. As in other individual harvesting
quota programs, the implementation of an IFQ program in the BSAI crab
fisheries would allow the creation of rents and facilitate
rationalization in both the harvesting and processing sectors. The
realization of significant economic benefits from rationalization of
the fisheries would make it possible for both processors and harvesters
to benefit, and both the Halvorsen, Khalil and Lawarree study and the
Milon and Hamilton study indicate that this is the most likely outcome.
The research program by Professor Matulich that has attempted to
establish the presumption that harvesters would gain and processors
would lose under an IFQ program is simply not credible, either on
theoretical or empirical grounds.
The ``three-pie voluntary cooperative program'' proposed by the
Council for the BSAI crab fisheries includes both regionalization and
the requirement that 90 percent of total harvesting quotas be Class A
shares that can only be delivered to processors holding IPQs. Class B
shares, which can be delivered to processors that do not hold IPQs,
account for I 0 percent of the harvest allocation. As opposed to the
likely outcome under a standard IFQ program that both processors and
harvesters would benefit, the Milon and Hamilton study concludes that a
program in which processors receive IPQs equal to the total harvest
would benefit only processors, with the value of harvester IFQs being
driven to zero. They also conclude that as the ratio of IPQs to IFQs
decreases, harvesters will fare better. However, there is no reason to
believe that reducing the IPQ share from 100 percent to 90 percent is
sufficient to allow independent catcher vessels to avoid adverse
effects from the rationalization program.
The basis for the Council's conclusion that a 90 percent-10 percent
program would balance the interests of both processors and harvesters
is not clear. Some proponents of the IPQ program have based their
arguments in its favor on an analogy with the AFA inshore pollock
cooperatives. It is now generally agreed that outcomes under the AFA
cooperatives benefited both processors and harvesters. However, the 90
percent-10 percent IPQ program is fundamentally different from the AFA.
The only apparent similarity is that a cooperative can deliver 10
percent of its harvest to an eligible processing facility other than
the one it is qualified for under the AFA, and harvesters could deliver
10 percent of their harvests to processors without IPQs under the
proposed program for the BSAI crab fisheries. However this apparent
similarity is purely superficial and does not imply, as some proponents
have argued, that processors are guaranteed 90 percent of their
historic processing shares under the AFA program as well as under the
three-pie program.
There is in fact no basis for believing that the rationalization
program proposed for the BSAI crab fisheries will have similar outcomes
to those obtained under the AFA cooperatives for the inshore pollock
fishery. Instead, the crab rationalization program as currently
structured is much more advantageous for processors both because of the
guarantee of 90 percent of their processing history and because of the
regionalization restrictions. The program is also far more complex,
making it problematic that market processes can be relied on for the
determination of appropriate prices. And there is less of a margin for
error in case of unintended distributional effects because the net
benefits to the fishery as a whole are expected to be substantially
smaller. Some of the major differences between the two programs are
summarized in the Appendix.
The generally favorable outcomes under the AFA provide no assurance
of favorable outcomes under the proposed rationalization program for
the BSAI crab fisheries. The 90 percent-10 percent proposal appears to
be a step in the direction of a compromise that less overwhelmingly
favors processors than a 100 percent IPQ program would. However, it is
not clear why the 100 percent program should have been the apparent
starting point.
A more obvious approach would be to begin by considering the
effects of a standard IFQ program. Both the Halvorsen, Khalil and
Lawarree study and the Milon and Hamilton study indicate that a
straight IFQ program would benefit both harvesters and processors.
Given the existence of concerns that an IFQ program might instead
result in adverse outcomes for processors, consideration of
modifications to the program to rebalance the outcomes would be
reasonable. Regionalization is one such modification, and whatever its
primary purpose, one effect of regionalization in the BSAI crab
rationalization program is to increase processors' bargaining power.
The allocation of IPQs would be another possible modification.
Restricting deliveries of some portion of the harvest under IFQs to
processors holding IPQs would tilt the distribution of the benefits of
the program in favor of processors, with processors faring better the
larger the share of IPQs in the total harvest. Given that it is not
clear that processors would fare badly without IPQs, as well as the
total lack of practical experience with IPQs, it would be reasonable to
limit the original allocation of IPQs to a modest share of the total
harvest. And if the goal is to have an effect similar to the overall
provisions of the AFA inshore cooperatives, which allow for the
possibility that a processor's claim on its processing share can be
eroded over time or even be completely lost, then this share should
most likely be less than 50 percent.
The Council has commendably committed itself to closely monitoring
the outcomes under the BSAI crab rationalization program and making
modifications to the program if the outcomes are judged to be
unsatisfactory. Therefore, if it turned out that the initial allocation
of IPQs was seen to be causing unsatisfactory results, the Council
could modify the allocation accordingly. Because IPQs do not contribute
to the realization of the efficiency benefits of rationalization, such
changes would have primarily only distributional results. Such a
flexible approach would appear to be clearly superior to one in which
the Council commits itself to a 90 percent-10 percent program with the
intention of making unspecified modifications to other aspects of the
program if the outcomes are considered unsatisfactory.
American Fisheries Act
The AFA increased total net economic benefits in the inshore sector
of the BSAI pollock fishery in two major ways; it increased the inshore
sector's share of the total allowable catch from 35 percent to 50
percent, and it made rationalization of the fishery possible through
the formation of harvesting cooperatives. Each cooperative was to be
allocated a share of the inshore sector's TAC based on the catch
history of the vessels belonging to the cooperative. Therefore, the AFA
made it possible for catcher vessels that joined a cooperative to
obtain pollock harvest allocations despite the moratorium on new IFQ
programs. Because inshore processors had expressed concern that this
might put them at a bargaining disadvantage, the AFA required that each
cooperative be tied to a specific processing facility.
AFA cooperatives
Under the AFA cooperative provisions, qualification for membership
in a cooperative is based on the facility to which a catcher vessel
delivered the largest share of its total catch in the previous year. If
a catcher vessel does not join a cooperative it can fish for the share
of the total inshore allocation not apportioned to the cooperatives.
The term ``open access fishery'' is used to refer to this part of the
inshore pollock fishery, with the understanding that access of catcher
vessels and processing facilities is limited by the AFA. A catcher
vessel fishing in open access can chose to remain in open access and
deliver its fish to the eligible processing facility, or facilities,
offering the best price. Alternatively, it can chose in any year to
join the cooperative associated with the processing facility to which
it delivered the largest share of its fish in the previous year.
During the Council's deliberations concerning the implementation of
the AFA, an association of independent catcher vessel owners protested
to the Council that the AFA rules for inshore cooperatives would be
harmful to them because of the restrictions on where they could market
their fish. The association proposed an alternative set of rules, which
became known as the Dooley-Hall proposal. The most important proposed
change was to eliminate the qualification requirements so that a
cooperative could deliver to any eligible processing facility and any
eligible catcher vessel could join any cooperative.
In response, the Council commissioned a study by myself and two
University of Washington colleagues to help determine if the
implementation of the inshore sector cooperatives as provided for in
the AFA would have beneficial or adverse effects on independent vessel
owners and to evaluate the effects of alternative proposals (Halvorsen,
Khalil and Lawarree, 1999). Subsequently, the Council commissioned a
second study that included the evaluation of an expanded set of
alternative proposals (Halvorsen, Khalil and Lawarree, 2000). The
studies were extensive and included interviews with many participants
in the pollock fishery as well as theoretical and empirical economic
analysis.
The interviews with participants were consistent with respect to
two issues; that the AFA provisions for cooperatives were designed to
avoid any possible adverse effects of rationalization on processors,
and that processors would fare worse, and independent catcher vessels
would fare better, under the Dooley-Hall proposal than under the AFA
cooperatives. However, there was sharp disagreement concerning the
magnitude of the distributional effects. We had representatives of
processors tell us that both sides would gain under the AFA rules,
whereas the processors would lose disastrously under the Dooley-Hall
proposal, while representatives of independent catcher vessel owners
said that both sides would gain under the Dooley-Hall proposal, whereas
they would lose disastrously under the AFA rules.
In addition, one academic economist, Professor Scott C. Matulich,
whose research was funded by some of the processors, claimed in public
testimony at the October 1999 Council Meeting that his research proved
that everyone would gain under the AFA cooperative provisions whereas
processors would lose disastrously under the Dooley-Hall proposals.
[The research upon which Professor Matulich based these conclusions is
critically reviewed in the written testimony that I submitted on May
20, 2003 (Halvorsen 2003)]. He subsequently amended his position and
testified at the June 2000 Council meeting that the AFA had introduced
a new market failure that ``encourages vertically integrated processors
to impose higher average raw fish costs on non-integrated or less
integrated processors'' which could drive them out of business, even if
the non-integrated processors were more efficient. Furthermore, the AFA
``may be the death knell of the independent fisherman'' because all
processors would want to become vertically integrated (Matulich 2000,
page 15).
Our studies reached much more moderate conclusions than were
expressed either in the participant interviews or Professor Matulich's
testimony. We concluded that the balance of bargaining power in the
fishery under the AFA provisions for cooperatives favored processors,
because the apparent necessity of spending a year in open access in
order to change processing facilities could be very costly for a
catcher vessel. Therefore there was a significant probability that
independent catcher vessels would be adversely affected by the AFA's
provisions for cooperatives. However, we also noted that the positive
net benefits from the reallocation aspect of the AFA, as well as the
potential net economic benefits from rationalization of the fishery,
decreased the likelihood that catcher vessels would be adversely
affected relative to the situation pre-AFA.
We also concluded that two other provisions of the AFA would
decrease the probability that independent catcher vessel owners would
be adversely affected. One was the rule that the cooperative contract
could provide for up to 10 percent of a cooperative's harvest to be
processed by a different processing facility. In our study we noted
that cooperatives were expected to be able to negotiate a higher price
for the 10 percent of their deliveries that could be delivered to any
eligible processing facility. However, it turned out that there was a
more important way in which the 10 percent rule improved an independent
catcher vessel's bargaining power, which we did not recognize at the
time of our study.
The principal reason for expecting that independent catcher vessels
might be adversely affected by the AFA provisions for cooperatives was
that the apparent necessity of spending a year in open access in order
to change processing facilities could be very costly. What we did not
anticipate was that, by utilizing the 10 percent of the cooperative's
harvest that could be delivered to another eligible processing
facility, catcher vessels have been able to qualify for a new
cooperative without having to spend a year in open access. This has
been possible because the 10 percent applies to the cooperative's
collective deliveries, not to each individual vessel's deliveries.
Therefore the 10 percent rule freed up enough fish to make it feasible
for even a large vessel to deliver more of its harvest to a new
processing facility than to its existing facility and therefore to be
able to switch processors without incurring the cost of going through
open access.
This has improved the bargaining power of independent catcher
vessels more than we expected, because if they are not satisfied with
the outcome of price negotiations they can credibly threaten to switch
processors in the following year. If switching did occur, a processing
facility's claim on its cooperative's original total harvest could
decrease cumulatively over time at a rate of up to 10 percent per year,
so that by the end of the original five year life of the AFA it might
have lost more than 40 percent of its original processing share. And
the process of erosion could continue into subsequent years if, as
happened, the life of the AFA was extended. From the point of view of
the affected processing facility, the experience in the first five
years would be the equivalent in the proposed BSAI crab rationalization
plan to having its individual processing quota decrease from 90 percent
of its processing history to less than 50 percent.
The other provision that was favorable for independent catcher
vessels was the rule that a cooperative had to be approved by 80
percent of the catching vessels qualified for it. This rule was even
more important for the bargaining power of independent catcher vessels
than the 10 percent rule, because it involved the possibility of a
processing facility losing its entire claim on its cooperative's
original total harvest. The rule made it possible for independent
catcher vessels to credibly threaten to dissolve their cooperative by
voting against its continued existence. This would not only deprive the
processor of any claim on the deliveries of the independent catcher
vessels, it would also cause the processor to lose the catch
allocations of any catcher vessels that it owned, because catch
allocations could only be obtained through membership in a cooperative.
The 80 percent rule makes the continued existence of a cooperative
from year to year quite uncertain. As a result, the National Marine
Fisheries service required that each in shore cooperative apply for its
allocation on an annual basis. Each cooperative must apply for its
allocation annually and must certify annually that the cooperative
meets all the requirements in the AFA and its associated regulations.
If it is not able to do so, the processing facility will lose its
entire claim on its processing share.
Dooley-Hall cooperatives
The Dooley-Hall proposal would have modified or eliminated several
of the AFA rules for inshore cooperatives. The most important proposed
change was to eliminate the qualification requirements, so that
cooperatives would not be tied to specific processing facilities.
Instead, under the Dooley-Hall proposal a cooperative could deliver to
any eligible processing facility, and any eligible catcher vessel could
join any cooperative. Elimination of the qualification requirements
would make the 80 percent rule inoperable, and it would be replaced by
a rule requiring that the cooperative contract be signed by the owners
of five or more catcher vessels.
Except for the requirement that catcher vessels would have to
belong to a cooperative in order to obtain the advantages of pollock
harvest allocations, the Dooley Hall proposal would be essentially
equivalent to an IFQ program. The cooperative requirement had some
important practical implications for the management of the fishery, for
example, with respect to monitoring and enforcement issues. However,
for the purpose of evaluating the principal effects of the Dooley-Hall
alternative on the different types of participants in the fishery, it
could be analyzed as if it were an IFQ program. This also implies that
our analysis of the Dooley-Hall proposal is relevant for the analysis
of implementing an IFQ program in a fishery with similar
characteristics to the inshore pollock fishery.
At the time of our analysis, there already existed a considerable
amount of theoretical and empirical information on fishery management
programs involving IFQs, including the National Research Council (1999)
study mandated by the Sustainable Fisheries Act of 1996. A consistent
finding in these studies was that IFQ programs score highly on
efficiency grounds, allowing the creation of rents and facilitating
rationalization in both the harvesting and processing sectors. However,
some of the studies also note that implementation of these programs
could result in adverse effects on some participants in the fishery,
and that the distributional effects of IFQ programs have been
controversial.
As noted above, the principal controversy with respect to the AFA
and Dooley Hall proposals was their effects on the relative well-being
of processors and harvesters. Our analysis showed that the degree of
competitive behavior among processors would be the most critical factor
in determining outcomes under the Dooley-Hall proposal. Processors had
a number of important bargaining advantages. Their ownership of
catching vessels reduced their reliance on supply from independent
catcher vessels, while also providing them an informational advantage
given that independent catcher vessels did not own inshore processing
plants. Independent catcher vessels did have the advantage of being
able to legally bargain as a group through the Bering Sea Marketing
Association (BSMA), but the BSMA did not represent all of the catcher
vessels and the largest processor was not a party to the negotiations.
The basic change that the AFA made in the economic circumstances of
the fishery was that catcher vessels would have a claim on the
available harvest allocation based on their catch history.
Historically, the lack of such a claim had been the primary incentive
for the race to fish, and therefore the AFA was expected to decrease
the daily rate of harvest and thereby extend harvesting periods. This
in turn would create a large amount of extra capacity in both the
harvesting and processing sectors, which might affect the degree of
competitive behavior among processors.
However, the processing sector was very concentrated, entry was
legally blocked by the AFA, and processors should anticipate that
aggressive tactics that gave them short term gains would not be
profitable in the long run because they can all engage in such tactics.
We did not believe that even the large amount of excess capacity that
was anticipated would make the processors blind to the benefits of
moderate competition. Their investments in capacity were sunk costs and
cutthroat competition would not make them profitable. Therefore, it was
highly improbable that the situation would deteriorate, from the
processors' point of view, into one of cutthroat competition. However,
because this was not a logical impossibility, the outcomes that would
be produced by Dooley-Hall proposal under cutthroat competition were
examined.
This case had previously been considered in the context of a race
for fish that is ended by the introduction of IFQs, first in a
theoretical analysis conducted by two employees of Trident Seafood
(Plesha and Riley, 1992), and subsequently in a paper by Matulich,
Mittelhammer, and Reberte (1996). The latter paper was the basis for
Professor Matulich's Council testimony concerning the effects of the
Dooley-Hall proposal. The paper concluded that if processing capital
had no alternative uses or salvage value, then the excess capacity
created by an IFQ program would cause the ex vessel price of fish to
increase to the point where it was equal to the difference between
final product price and short-run average variable processing cost.
Processors would leave the industry until excess capacity no longer
existed. During the transitional period, catcher vessels not only would
gain all the rents from the fish, but also the ``quasi-rents'' from the
processors' capital. Processors that survived would realize gains in
efficiency and market shares, but might have to pay a higher ex vessel
price for fish.
However, this theoretical analysis had little relevance for the
inshore pollock fishery. First, it was extremely unlikely that
processors in such a highly concentrated fishery would not be able to
do better than the perfectly competitive market outcomes. Second, the
model assumes that processors receive no IFQs, but in the pollock
fishery processors owned catcher vessels that were being given harvest
allocations on the same terms as the independent catcher vessels. Data
provided by the National Marine Fisheries Service indicated that
processor-controlled vessels would receive at least half of the total
harvest allocation. Therefore, for at least half the total allocation,
any increase in price paid by processors would, except for the effect
on crew share, be merely an internal transfer, not an economic loss.
Third, the model assumes that there is only one basic type of
processing capital, so that excess aggregate capacity implies that all
capital is in excess supply. However, pollock processors produce two
main types of primary product, surimi and fillets. Under then current
market conditions, fillet capital would not be in excess supply even
with a large amount of total excess capacity, and therefore would
continue to earn quasi-rents. The model's assumptions that processors
could not earn informational rents, and that pollock processing capital
had no alternative uses, were also too pessimistic.
The extreme alternative to cutthroat competition among processors
is that they act as a monopsony. Monopsony implies that the processors
fully account for their mutual interdependence and behave as if they
were one entity. If independent catcher vessels also were able to
unite, as could occur under the Dooley-Hall proposal by the formation
of a single cooperative containing all independent catcher vessels, the
monopsony case would become equivalent to that of a bilateral monopoly.
This is the market structure that Matulich and Sever (1999) assumed for
the inshore pollock fishery in their paper claiming that a two-pie
allocation would make everyone better off. As discussed in Halvorsen
(2003), this was a highly misleading characterization of the actual
fishery. However, because such an outcome was not a logical
impossibility under the Dooley Hall proposal, the outcomes that would
be produced by the Dooley-Hall proposal under bilateral monopoly were
also examined.
In this scenario, the independent catcher vessels' cooperative
would give them an exclusive right to approximately half of the total
inshore allocation. If they could not negotiate an agreement with the
processors, they would not fish. Similarly, the most severe threat of
the processors would be to not process the fish. In both of these
cases, the processors and the independent catcher vessels would receive
nothing from the independent catcher vessels' share of the allowable
harvest. The standard game theory outcome, known as the Nash bargaining
solution, would be for them to split the benefits from the independent
catcher vessels' fish equally. The processors would also obtain all of
the rent from the half of the total allocation that they controlled.
The end result would be that processors would obtain about three-
fourths of the total net benefits from the fishery (one-half from the
processor-controlled vessels and one-fourth from the independent
catcher vessels). Thus the assumption that the inshore pollock fishery
is a bilateral monopoly is inconsistent with a conclusion that the
Dooley-Hall proposal would have had seriously adverse results for
processors.
The analysis of the Dooley-Hall proposal under the assumptions of
cutthroat competition and bilateral monopoly is instructive, but would
provide a realistic prediction of actual outcomes only in a fishery
that fit one of these extreme cases. For the inshore pollock fishery,
the most probable outcome was that processors would be moderately
competitive. Analysis of the outcomes in this case was necessarily more
speculative, because it could not rely on the results of the standard
simple models of cutthroat competition and bilateral monopoly. However,
the same key trade-offs would be at work. Independent catcher vessels'
would benefit under the Dooley-Hall proposal from the elimination of
the costs of open access and from the ability to form cooperatives
independent of the processing facilities, but would lose the bargaining
leverage provided by the 80 percent rule. We concluded that the most
reasonable outcome was that independent catcher vessels would have been
somewhat better off, and processors somewhat worse off, under the
Dooley-Hall proposal than under the AFA rules. This did not imply that
processors were expected to be harmed by a rationalization program
based on the Dooley-Hall proposal, only that they would not be as well
off as under the AFA provisions for cooperatives.
Outcomes under the AFA
The analysis of the potential effects on harvesters and processors
of rationalization programs under the AFA was facilitated by the
existence of previous economic analysis of, and practical experience
with, its principal features. For example, there is a large body of
relevant literature on agricultural cooperatives that was surveyed in
an appendix to the Halvorsen, Khalil, and Lawarree studies written by
Professor Steven T. Buccola of Oregon State University. In addition,
the rationalization benefits attainable by fish harvesting cooperatives
had been demonstrated by the Pacific whiting cooperative established in
1997 and the AFA cooperatives in the offshore sector that were
implemented in 1999. Similarly, the analysis of probable outcomes under
the Dooley-Hall proposal could draw on the existing literature on IFQ
programs.
The rationalization program that was implemented in the inshore
pollock fishery retained the original AFA provisions for cooperatives.
The actual experience of the fishery subsequent to rationalization has
been consistent with the overall economic analysis of Halvorsen,
Khalil, and Lawarree. Total net economic benefits in the inshore sector
have been increased as a result of the increase in the sector's share
of the TAC, as well as substantial increases in the TAC itself. In
addition, implementation of the cooperatives has resulted in increased
efficiency. Improved targeting of pollock during the peak roe season
has contributed to greatly increased ex vessel prices during this
season. The value of output has also increased because slowing the race
for fish permitted an increase in the recovery rate and a shift to
higher valued products. Harvesting costs have been reduced by the
transfer of quota shares from less efficient vessels to more efficient
vessels, and easy transferability of allocation within a cooperative
has facilitated the full harvesting of the available allocations. As a
result, the net effects of the AFA on both harvesters and processors is
generally agreed to have been positive, although the relative size of
the benefits remains a matter of contention.
On the other hand, the fishery has not demonstrated the outcomes
suggested by Professor Matulich. There has been no apparent tendency
for vertically-integrated processors to have gained at the expense of
less-integrated processors, much less to have been able to drive them
out of business. And the AFA does not appear to have had significant
negative effects on the share of independent catcher vessels in the
fishery, much less to have been the ``death knell of the independent
fisherman.''
Because the Dooley-Hall proposal was not adopted by the Council, it
is not possible to compare actual outcomes with the conclusions
concerning the probable outcomes that were reached by Halvorsen,
Khalil, and Lawarree. (Three studies of the economic effects of IFQ
programs in halibut fisheries are discussed in the written testimony I
submitted on May 20, 2003). However, the analysis sheds light on the
possible outcomes of IFQ programs in fisheries with similar
characteristics to the inshore pollock fishery. In particular, it
indicates that implementation of an IFQ program in a fishery with a
similar market structure would be highly unlikely to have the adverse
effects for processors that are implied by Matulich, Mittelhammer, and
Reberte's (1996) analysis.
BSAI Crab Rationalization Plan
In its deliberations concerning rationalization of the BSAI crab
fisheries, the Council considered the use of a program similar to the
AFA cooperatives in the inshore pollock fishery, as well as an entirely
different type of rationalization program based on the use of both IFQs
for harvesters and individual processing quotas (IPQs) for processors.
As in the case of the inshore pollock fishery, the Council commissioned
an independent economic analysis of the proposed alternatives.
Milon and Hamilton study
The study, Milon and Hamilton (2002), draws conclusions for
economic performance under the assumptions that the harvesting sector
is perfectly competitive and the processing sector contains few enough
firms that it might be able to exercise market power. One of the
rationalization programs analyzed is a two-pie system in which
processors are allocated IPQs for the entire harvest. Milon and
Hamilton conclude that each processor would maximize profits by paying
the lowest ex vessel price that is required to support harvester
delivery. The result would be that processors capture all of the net
economic benefits from the fishery. The allocation of IFQs in such a
program would be redundant, and the value of harvester IFQs would be
driven to zero.
They also consider a program with both Class A harvesting quotas
that must be delivered to processors that hold IPQs, and Class B
harvesting quotas that may be delivered to any processor. If all
harvesting quotas are Class B, the outcome is the same as for an IFQ
only program, with the likelihood that both harvesters and processors
would benefit from rationalization. As the ratio of A to B quotas
increase, processors are expected to fare better, and harvesters to
fare worse. And when all the quotas are Class A, processors capture all
of the net economic benefits from the fishery.
The study also considers the effects of regionalization of the
fisheries. It concludes that regionalization reduces cost efficiency by
imposing constraints that prevent harvest within a region being
transferable to other regions. In addition, regionalization increases
the market power of processors, resulting in lower ex vessel prices.
Milon and Hamilton's assumption that the harvesting sector is
perfectly competitive is not completely accurate, because catcher
vessels have negotiated prices collectively through the Alaska
Marketing Association, and have engaged in strikes. However, the
effectiveness of strikes has been undermined because catcher processors
have continued to fish during them. Also, the lengthening of fishing
seasons under rationalization is expected to decrease the effectiveness
of collective bargaining (North Pacific Fishery Management Council,
2002, page 376). Therefore, the assumption that the harvesting sector
is competitive does not invalidate their conclusions concerning the
qualitative effects of IPQs and regionalization.
Implications of the AFA study for an IFQ program
Milon and Hamilton's conclusion that both harvesters and processors
are able to gain under an IFQ program that did not include IPQs is
consistent with Halvorsen, Khalil, and Lawarree's analysis of the
Dooley-Hall proposal under the AFA. As noted above, the Dooley-Hall
proposal was essentially equivalent to an IFQ program, and therefore
the principal factors that affected outcomes under it would also affect
outcomes under an IFQ. Because some of the most important of those
factors are also present in the BSAI crab fisheries, our analysis of
the Dooley-Hall proposal can shed light on the economic effects of
implementing an IFQ program in the crab fisheries.
The market structure of the BSAI crab fisheries is similar to that
of the inshore pollock fishery in several important respects. Most
importantly, the processing sector is highly concentrated, especially
on a regional basis. Processors can be expected to realize that
aggressive tactics yielding short-term gains are unlikely to be
profitable in the long run. Entry is not prohibited but there would be
some barriers to entry, even if existing processors were not given any
rights to their processing history. Processors own catching vessels,
which reduces their reliance on supply from independent catcher vessels
and also provides them an informational advantage. Independent catcher
vessels do have the advantage of being legally able to bargain as a
group, but the effectiveness of collective bargaining is expected to
decrease under rationalization.
Implementation of an IFQ program would give catcher vessels a claim
on the available harvest allocation based on their catch history. This
would eliminate the primary incentive for the race to fish, and
therefore can be expected to result in longer harvesting periods. This
in turn would create extra capacity in both the harvesting and
processing sectors, which might affect the degree of competitive
behavior among processors.
However, because the processing sector is very concentrated and
there are barriers to entry, the situation should be similar to that
analyzed in the AFA study. Processors should anticipate that aggressive
tactics that gave them short-term gains would not be profitable in the
long run because they can all engage in such tactics. Not even a large
amount of excess capacity would make the processors blind to the
benefits of moderate competition. Their investments in capacity are
sunk costs and cutthroat competition would not make them profitable.
Therefore, it is highly improbable that the introduction of IFQs would
result in cutthroat competition on the part of processors.
Therefore, as in the inshore pollock fishery, the cutthroat
competition among processors that is required for an IFQ program to
lead to the adverse results for processors predicted by Matulich,
Mittelhammer, and Reberte (1996) is highly unlikely. During the
Council's deliberations on the BSAI crab rationalization, Professor
Matulich attempted to buttress his theoretical arguments for the
adverse effects of IFQs by citing an empirical study of the halibut and
sablefish IFQ program (Matulich and Clark, 2002). The theoretical and
empirical problems with this study were discussed in my previous
written testimony to this Committee (Halvorsen 2003), which also
summarizes the problems that the United States General Accounting
Office (2002) noted concerning the study's methodology and scope.
The extreme alternative to cutthroat competition among processors
is that they act as a monopsony. A pure monopsony would capture all of
the net benefits in the fishery. A less extreme case is one in which
the independent catcher vessels are also able to unite effectively
under an organization such as the Alaska Marketing Association. With
independent catcher vessels, as well as processors, united, the
analysis becomes one of bilateral monopoly for the entire industry. The
situation is equivalent to there being only one processing firm, which
also owns some catcher vessels, and one entity that owns all the
independent catcher vessels.
The first step in the analysis is to understand the outside options
(threat points) of the two players. Since there would be effectively
only one processor, the independent catcher vessel owner's only threat
would be to refuse to fish. The threat of the processor would be to
refuse to process the fish of the independent catcher vessel owner. If
the independent catcher vessel did not fish, its net benefit would be
zero, whereas if the processor refused to process fish from the
independent catcher vessels, it would still have available to it the
fish caught by its own catcher vessels. Therefore, the Nash bargaining
solution would imply a distribution of net benefits in which the
processor would obtain more than half of the total net economic benefit
of the fishery.
The assumptions of cutthroat competition, monopsony, or bilateral
monopoly are not plausible characterizations of market structure in the
BSAI crab fisheries under a standard IFQ program. The most probable
outcome would be that processors would be moderately competitive. They
would have a bargaining advantage in that their ownership of catching
vessels reduces their reliance on supply from independent catcher
vessels, while also providing them an informational advantage.
Regionalization of the crab fisheries would increase their bargaining
power by increasing the effective degree of concentration. Independent
catcher vessels would have the advantage of being able to legally
bargain as a group, but the effectiveness of collective action is
expected to decrease because of elongation of the fishing seasons. The
balance of bargaining power would be expected to be in favor of
processors, but not overwhelmingly so.
As in other individual harvesting quota programs, the
implementation of an IFQ program in the BSAI crab fisheries would allow
the creation of rents and facilitate rationalization in both the
harvesting and processing sectors. The realization of significant
economic benefits from rationalization of the fisheries would make it
possible for both processors and harvesters to benefit, and both the
Halvorsen, Khalil and Lawarree study and the Milon and Hamilton study
indicate that this is the most likely outcome. The research program by
Professor Matulich that has attempted to establish the presumption that
harvesters would gain and processors would lose under an IFQ program is
simply not credible, either on theoretical or empirical grounds.
The Council's ``three-pie'' proposal
The ``three-pie voluntary cooperative program'' proposed by the
Council for the BSAI crab fisheries includes both regionalization and
the requirement that 90 percent of total harvesting quotas be Class A
shares that can only be delivered to processors holding IPQs. Class B
shares, which can be delivered to processors that do not hold IPQs,
account for 10 percent of the harvest allocation. As opposed to the
likely outcome under a standard IFQ program that both processors and
harvesters would benefit, the Milon and Hamilton study concludes that a
program in which processors receive IPQs equal to the total harvest
would benefit only processors, with the value of harvester IFQs being
driven to zero. They also conclude that as the ratio of IPQs to IFQs
decreases, harvesters will fare better. However, there is no reason to
believe that reducing the IPQ share from 100 percent to 90 percent is
sufficient to allow independent catcher vessels to avoid adverse
effects from the rationalization program.
The basis for the Council's conclusion that a 90 percent-10 percent
program would balance the interests of both processors and harvesters
is not clear. Some proponents of the IPQ program have based their
arguments in its favor on an analogy with the AFA inshore pollock
cooperatives. It is now generally agreed that outcomes under the AFA
cooperatives benefited both processors and harvesters. However, the 90
percent-10 percent IPQ program is fundamentally different from the AFA.
The only apparent similarity is that a cooperative can deliver 10
percent of its harvest to an eligible processing facility other than
the one it is qualified for under the AFA, and harvesters could deliver
10 percent of their harvests to processors without IPQs under the
proposed program for the BSAI crab fisheries.
However this apparent similarity is purely superficial. Under the
three-pie program processing companies are guaranteed 90 percent of
their historic processing shares, whereas under the AFA program a
processing facility is not guaranteed to retain any of its claim on its
cooperative's original total harvest. In addition, other differences
between the AFA and the proposed crab rationalization program
contradict the supposed analogy. The following section summarizes some
of the major differences between the two programs.
Differences between the AFA and Crab Rationalization Programs
Protection of processors' market shares
Under the proposed rationalization program for the crab fisheries,
a processor would receive processor quota share equal to 90 percent of
its historic processing share. This amount would be guaranteed to the
processor because harvesters could deliver their Class A allocation
only to processors holding processing share. Therefore the only way
that a processor could lose more than 10 percent of its historic market
share would be if it set the ex vessel price so low that vessels would
prefer to forego fishing rather than deliver fish to it, because that
would be their only alternative.
Under the AFA, each processing facility has the right to process 90
percent of its cooperative's total harvest, but this does not guarantee
that it will receive 90 percent of its historic processing share,
because vessels have alternatives to remaining in the cooperative. One
alternative for a vessel is to fish in the open access portion of the
fishery and deliver its fish to another eligible processing facility.
It could then either remain in open access or join the cooperative of
the new processing facility. Another option is to qualify for another
cooperative without going through open access by delivering its fish to
the alternative processing facility as part of the 10 percent of the
cooperative's total harvest that can be delivered to any eligible
facility. And the existence of a cooperative requires the approval of
at least 80 percent of the vessels qualified for it. As a result, a
processing facility is faced with the possibility of losing all of its
processing rights if more than 20 percent of the vessels in the
cooperative decide that it should be dissolved.
Therefore protection of processors' market shares would be much
greater under the proposed program for the crab fisheries than under
the AFA, thereby giving the processors much greater bargaining power.
In particular, under the AFA a processor will retain market share only
if it offers an ex vessel price that is competitive with the price
being offered by other processors, whereas under the proposed crab
rationalization program a processor could retain 90 percent of its
market share even if it offered a price so low that the catcher vessel
was barely able to cover the average variable costs of harvesting the
fish.
Regionalization
Under the proposed rationalization program for the crab fisheries,
Class A harvest shares and processor shares for each crab fishery would
be regionally designated, whereas under the AFA the entire inshore
sector is treated as a single region. This difference has important
implications both for the net economic benefits that can be realized
from rationalization and for the distributional consequences of
rationalization.
The Council itself recognizes that regionalization reduces net
economic benefits by restricting consolidation of activities that are
desirable for reducing capacity and gaining efficiency in both the
harvesting and processing sectors under rationalization (Report to
Congress, August 2002, page 18). The lack of such constraints under the
AFA increased the total net economic benefits that were available to be
shared by harvesters and processors.
Regionalization also has implications for the distributional
consequences of rationalization because it subdivides the markets for
crab and thereby increases the already high degree of concentration
among processors. It also creates an incentive for processors to
consolidate their market shares on a regional basis, which would
increase the degree of concentration still more. The greater bargaining
power attained by processors can be expected to adversely affect the
price received by harvesters for Class B allocations as well as for
Class A allocations, both because it might be difficult logistically to
deliver to different markets and because processors might be able to
require bundling deliveries of the two classes of fish.
Complexity
The ``three-pie voluntary cooperative program'' being recommended
for the crab fisheries is much more complex than the rationalization
program implemented under the AFA. The greater complexity can be
expected to have serious negative consequences both with respect to the
cost of management and with respect to the functioning of the market
for fish and for quota shares.
Implementation of the proposed rationalization program for the crab
fisheries would require the determination of share allocations in each
region of each fishery for each individual vessel and processor.
Ongoing management measures would include annual monitoring and
enforcement measures at the same level of detail. Eventual formation of
voluntary cooperatives might reduce some of the management costs with
respect to harvesting, but the extra costs of managing processing
activities would continue.
More importantly, the increased complexity of the system might make
the determination of prices through a decentralized market structure
impracticable. For each regional market in each fishery the prices that
would have to be determined include the ex vessel price of Class A
fish, the ex vessel price of Class B fish, the price of Class A
harvesting quota, the price of Class B harvesting quota, and the price
of processing quota.
Attaining equilibrium prices in such a complex system would be
difficult even in large, well-functioning, markets, and the markets in
the crab fisheries would be both thin and imperfectly competitive. In
addition, the large fluctuations in total allowable catch would
complicate the determination of equilibrium prices and hinder the
ability of the system to converge to stable values. In recognition of
the possibility of the price system breaking down, the rationalization
plan includes a binding arbitration program. However, the necessity of
such a procedure increases the cost of managing the fisheries under the
proposed rationalization plan, and even if the arbitration procedure
were well designed, it would not be an adequate substitute for a well-
functioning market.
Net benefits from rationalization
Rationalization of the pollock fishery under the AFA created large
net economic benefits for the inshore sector, which facilitated
outcomes benefiting both the harvesting and processing sectors. As
already noted, the regionalization requirement under the proposed plan
for the crab fisheries would decrease the potential net economic
benefits to be obtained by rationalization. But even if this provision
did not exist, the total net economic benefits of rationalization in
the crab fisheries could not be expected to be as large as they were
under the AFA, because participants in the inshore pollock fishery
benefited both from a large increase in the sector's total allocation
and from large rationalization benefits from the formation of
cooperatives.
The sector's total allocation was increased first by an increase in
its share of the total directed pollock fishery from 35 percent to 50
percent, and subsequently by an increase in the total allowable catch
for the pollock fishery. The combined result was that the inshore
sector's total allocation has increased by 80 percent from the pre-AFA
level in 1998 to the present.
Large efficiency benefits were realized from the formation of the
AFA cooperatives. Rationalization under the AFA permitted improved
targeting of pollock during the peak roe season, resulting in greatly
increased ex vessel prices during this season. The value of output was
also increased because slowing the race for fish permitted an increase
in the recovery rate and a shift to higher valued products. Harvesting
costs have been reduced by the transfer of quota shares from less
efficient vessels to more efficient vessels, and easy transferability
of allocation within a cooperative has facilitated the full harvesting
of the available allocations.
In contrast, the proposed rationalization program for the crab
fisheries does not include an increase in the total allocations
available to these fisheries. It does incorporate a buyback program,
but the efficacy of the buyback program has yet to be determined, and
in any case could not result in benefits equivalent to the 80 percent
increase in total allocation experienced by the inshore pollock
fishery. Similarly, increases in the value of output due to
rationalization are not anticipated to be as large for the crab
fisheries, and increases in harvesting efficiency are likely to be
hindered by the restrictions imposed by the proposed program.
Implications
There is no basis for believing that the rationalization program
proposed for the BSAI crab fisheries will have similar outcomes to
those obtained under the AFA cooperatives for the inshore pollock
fishery. Instead, the crab rationalization program as currently
structured is much more advantageous for processors both because of the
guarantee of 90 percent of each processing company's processing history
and because of regionalization. The program is also far more complex,
making it problematic whether market processes can be relied on for the
determination of appropriate prices. And there is less of a margin for
error in case of unintended distributional effects because the net
benefits to the fishery as a whole are expected to be substantially
smaller.
Conclusions
The generally favorable outcomes under the AFA provide no assurance
of favorable outcomes under the proposed rationalization program for
the BSAI crab fisheries. To the contrary, the economic analyses
commissioned by the Council for both the AFA and the proposed crab
rationalization programs indicate serious problems with the crab
program. In particular, the choice of a 90 percent-10 percent IPQ
program is highly unlikely to provide a reasonable balance between the
interests of processors and harvesters.
An IPQ program in which processors are guaranteed 100 percent of
their harvesting history would almost certainly result in most, if not
all, of the benefits of the rationalization program accruing to
processors. The 90 percent-10 percent proposal appears to be a step in
the direction of a compromise that less overwhelmingly favored
processors. However, it is not clear why the 100 percent program should
have been the apparent starting point.
A more obvious approach would be to begin by considering the
effects of a standard IFQ program. Both the Halvorsen, Khalil and
Lawarree study and the Milon and Hamilton study indicate that a
straight IFQ program would benefit both harvesters and processors.
Given the existence of concerns that an IFQ program might instead
result in adverse outcomes for processors, consideration of
modifications to the program to rebalance the outcomes would be
reasonable. Regionalization is one such modification, and whatever its
primary purpose, one effect of regionalization in the BSAI crab
rationalization program is to increase processors' bargaining power.
The allocation of IPQs would be another possible modification.
Restricting deliveries of some portion of the harvest under IFQs to
processors holding IPQs would tilt the distribution of the benefits of
the program in favor of processors, with processors faring better the
larger the share of IPQs in the total harvest. Given that it is not
clear that processors would fare badly without IPQs, as well as the
total lack of practical experience with IPQs, it would be reasonable to
limit the original allocation of IPQs to a modest share of the total
harvest. And if the goal is to have an effect similar to the overall
provisions of the AFA inshore cooperatives, which allow for the
possibility that a processor's claim on its processing share can be
eroded over time or even completely lost, then this share should most
likely be less than 50 percent.
The Council has commendably committed itself to closely monitoring
the outcomes under the BSAI crab rationalization program and making
modifications to the program if the outcomes are judged to be
unsatisfactory. Therefore if it turned out that the initial allocation
of IPQs was seen to be causing unsatisfactory results, the Council
could modify the allocation accordingly. Because IPQs do not contribute
to the realization of the efficiency benefits of rationalization, such
changes would have primarily only distributional results. Such a
flexible approach would appear to be clearly superior to one in which
the Council commits itself to a 90 percent-10 percent program with the
intention of making unspecified modifications to other aspects of the
program if the outcomes are considered unsatisfactory.
References
Halvorsen, Robert (2003), ``Written Testimony Submitted to the
Record of the May 20, 2003 Hearing on Bering Sea/Aleutian Island Crab
Rationalization before the Committee on Commerce, Science, and
Transportation of the United States Senate.''
Halvorsen, Robert, Fahad Khalil, and Jacques Lawarree (1999),
``Inshore Sector Catcher Vessel Cooperatives in the Bering Sea/Aleutian
Islands Pollock Fisheries,'' discussion paper prepared for the North
Pacific Fishery Management Council, September 13.
Halvorsen, Robert, Fahad Khalil, and Jacques Lawarree (2000),
``Inshore Sector Catcher Vessel Cooperatives in the Bering Sea/Aleutian
Islands Pollock Fisheries,'' discussion paper prepared for the North
Pacific Fishery Management Council, February 7.
Matulich, Scott C. (2000), ``Unintended Consequences of the
American Fisheries Act,'' report to the North Pacific Fisheries
Management Council, June.
Matulich, Scott C. and Michael Clark (2002), ``Efficiency and
Equity Choices in Fishery Rationalization Policy Design: An Examination
of the North Pacific Halibut and Sablefish IFQ Policy Impacts on
Processors,'' Regional Information Report No. 5J02-02, Alaska
Department of Fish and Game, January 24.
Matulich, Scott C., Ron C. Mittelhammer, and Carlos Reberte (1996),
``Toward a More Complete Model of Individual Transferable Fishing
Quotas: Implications of Incorporating the Processing Sector,'' Journal
of Environmental Economics and Management, 31 (1): 112-28.
Matulich, Scott C. and Murat Sever (1999), ``Reconsidering the
Initial Allocation of ITQs: The Search for a Pareto-Safe Allocation
Between Fishing and Processing Sectors,'' Land Economics, 75 (2): 203-
19.
Milon, J. Walter and Stephen F. Hamilton (2002), ``A Comparative
Analysis of Alternative Rationalization Models for the Bering Sea/
Aleutian Islands (BSAI) Crab Fisheries,'' discussion paper prepared for
the North Pacific Fishery Management Council, March.
National Research Council (1999), Sharing the Fish: Toward a
National Policy on Individual Fishing Quotas, National Academy Press.
North Pacific Fishery Management Council (2002), ``BSAI Crab
Rationalization Program Alternatives,'' March.
North Pacific Fishery Management Council (2002), ``Bering Sea and
Aleutian Islands Crab Rationalization Program,'' Report to Congress,
August.
Plesha, Joseph T. and Christopher C. Riley (1992), ``The Allocation
of Individual Transferable Quotas to Investors in the Seafood Industry
of the North Pacific,'' photocopied.
United States General Accounting Office (2002), ``Individual
Fishing Quotas: Better Information Could Improve Program Management,''
December.
______
Community Impacts Related to Crab Rationalization
Utilizing Processor Quotas
This paper condenses information concerning crab rationalization
affects on communities as presented in the North Pacific Fishery
Management Council's (Council) analyses. The emphasis of this
examination is on the ``two-pie'' or processor quota aspects of the
proposed program. Also, the paper focuses on the community of Kodiak.
However, the general type of information presented in the Council
analyses for Kodiak is similar for Dutch Harbor and, to a lesser
extent, a few other communities. Most of the information in this paper
is taken directly from Council documents and is referenced as such.
This paper presents a more coherent picture of the community impacts
expected under crab rationalization. All of this information was
available to Council decision makers and, according to their official
record, comprised a basis for their decisions.
A thorough review of the Council's crab rationalization analysis
documents reveals absolutely no quantitative analysis of community
impacts.\1\ What quantitative impact analysis exists focuses on
allocations strategies to harvesters and processors. There is no
quantitative analysis or discussion of the impacts to communities,
regions, states, or the Nation as a whole.
---------------------------------------------------------------------------
\1\ Documents reviewed for this paper include: (1) North Pacific
Fishery Management Council (NPFMC). 2002. Bering Sea Crab
Rationalization Program Alternatives: Public Review Draft. May, 2002.
Including appendices; (2) EDAW (SIA). 2002. Social Impact Assessment,
BSAI Crab Rationalization: Overview and Community Profiles. Appendix 2-
6 to NPFMC 2002.; (3) NPFMC AP/CQS/S. 2002. BSAI Crab Rationalization
Program Trailing Amendments: Additional Provisions, Captain's Quota
Shares, and Sideboards. December 2002.; (4) NPFMC CP/BA/DC. 2003. BSAI
Crab Rationalization Program Trailing Amendments: Community Protection,
Binding Arbitration, and Data Collection. February 2003.; and, (5)
NPFMC CP/BA. 2003. BSAI Crab Rationalization Program Trailing
Amendments: Community Protection and Binding Arbitration. April 2003.
---------------------------------------------------------------------------
The most concise, and perhaps the only qualitative summary of
expected impacts of a ``two-pie'' allocation on the community of Kodiak
is found in the Social Impact Assessment.
Kodiak processors at present do not have a substantial
established history in the Bering Sea crab fisheries for the
qualifying periods being considered. While important to BSAI
crab processing in the more distant past, local processors have
minimal contemporary involvement, being in some cases
effectively restricted to short season's ``last load''
deliveries of locally based vessels. In a rationalized Bering
Sea crab fishery with only harvester rights allocated, Kodiak
processors generally feel that they could compete for more than
their historical percentage of the Bering Sea crab processed in
Kodiak. That is, the thought is that in a system free from a
race for crab (allocated harvester rights) but unconstrained as
to where crab can be delivered (no allocated processor rights,
cooperatives, or regionalization), Kodiak processors could
compete by offering higher prices to compensate for their
relatively greater distance from the resource. The co-op option
featuring assignment of quota to processor involved co-ops
based upon where the majority of catch has historically been
delivered would effectively serve to exclude Kodiak processors
from the BSAI crab fisheries. Similarly, direct processor
allocations would serve to diminish the role of, if not
completely exclude, Kodiak processors from the fishery, with
the degree of the effect dependent upon the percentage of the
processors' history which if firmly allocated-the higher the
percentage more firmly allocated by past processing history
(away from Kodiak processors, from their perspective), the more
potentially harmful the effect. For instance, a 100 percent
allocation of processor history would constrain Kodiak
processors to one to three percent of Bering Sea crab, unless
they could somehow buy a greater share from a willing
seller.\2\ (emphasis added)
---------------------------------------------------------------------------
\2\ SIA, page 127.
The analysis presented above is unequivocal in its prediction of
the negative affect on Kodiak of a two-pie system with processor
shares. Therefore, if the Council and the State of Alaska base their
support of a two-pie system, with 90 percent of the processing
privileges directly assigned to specific processors, on the analysis
documents they are explicitly attempting to legislatively depress
Kodiak's (and other smaller communities) future benefits from
rationalized BSAI crab fisheries.
Summary of Council Analysis Statements Concerning Community Impacts
There are four general sections of the analysis within which
community impacts would reasonably be described and analyzed:
(1) Section 2.6: Community and Social Impacts (pages 105-137) and
specifically Section 2.6.4: Detailed Community Level Impacts.
(2) Section 3.6: Regionalization (from the May 2002 document and
modified by trailing amendments)
(3) Section 3.16: Economic Effects (pages 395-408).
(4) Appendix 2-6: Community Profiles and Social Impact Assessment
(SIA)
Section 2.6.4 is presented below in its entirety:
As noted in the introduction to this section, community and
social impacts of crab rationalization approaches are discussed
both in this section and in an appendix to this volume, and
these two discussions, taken together, comprise the SIA for
crab rationalization. Appendix 2-6 (Social Impact Assessment:
BSAI Crab Rationalization Overview and Community Profiles)
details the localized nature and intensity of engagement with
and dependency on the crab fishery at the community level, and
presents an analysis of the direction and magnitude of the
social impacts likely to result from crab rationalization for
the series of communities profiled as well as for the CDQ
region.
Section 3 of the document presents an analysis of the alternatives
via a variety of types of impacts. Communities are not mentioned in the
analysis including environmental impacts, endangered species, marine
mammals, or safety.
Scattered throughout the SIA and analysis sections are various
suggestive or summary statements about the ability of estimating or
forecasting community impacts.
Some communities could also be affected by rationalization. In
the current derby fishery, processing activity is likely to be
located to facilitate success in the race for fish. In a slower
rationalized fishery, processing activity could relocate to
different communities to realize cost efficiencies. Communities
that increase processing under rationalization will benefit
from the program. Communities that lose processing activity
would realize less benefits in a rationalized fishery. The
regionalization alternatives are intended to prevent some of
the redistribution of processing activity to protect
communities that have benefited from the distribution of
processing in the current race to fish. The extent of the
impacts of rationalization on communities depends on whether
these regionalization alternatives are adopted and whether they
succeed in achieving their goals.\3\
---------------------------------------------------------------------------
\3\ NPFMC, page 406.
Under a two-pie IFQ program, if the entire fishery is allocated
through processing shares, processor entry would require the
purchase of processing shares. Processing shares in this
program would create a regulatory barrier to entry. The extent
of the barrier would depend on the market price of processing
shares, which cannot be predicted. The relatively few number of
processors in the crab fisheries could lead to a limited market
for processing shares, which would complicate entry to the
processing sector. Crab in a fishery with fully allocated
processing privileges, however, would have a lower ex vessel
price.\4\
---------------------------------------------------------------------------
\4\ NPFMC, page 407.
At least eight communities that are home to BSAI crab
processors could be directly impacted by the regionalization
provisions under consideration. The overall impact on
communities cannot be determined until the regionalization
alternatives are selected and the communities are defined.\5\
---------------------------------------------------------------------------
\5\ NPFMC, page 428.
The small governmental jurisdictions could also be impacted by
the rationalization of the crab fisheries. As noted above,
depending on the alternatives chosen, communities might be
protected by a regionalization program. The regionalization
options would either directly protect individual communities or
would simply divide the fisheries into regions. If communities
are directly protected, the rationalization is unlikely to have
negative effects on those communities. Under a more general
regionalization program that divides the fishery into two
regions, it is likely that communities would receive limited
protection. Under this more general protection some processing
activity could move between communities in a region. This is
likely to benefit those communities that receive additional
processing activity and harm communities that processing
activity leaves. Similarly, if no regionalization program is
incorporated into the rationalization program some communities
would benefit from rationalization, while others would suffer
some losses. Fish taxes would likely be redistributed with the
redistribution of processing activity. In addition, the
provision of support services and associated sales taxes will
likely be redistributed to the extent that deliveries to
processors are redistributed in a rationalized fishery.\6\
---------------------------------------------------------------------------
\6\ NPFMC, pages 429-30.
Increased efficiency in the fisheries arising from
rationalization could also reduce the provision of support
services and sales taxes, if the fleet is able to reduce their
overall costs. These savings may occur in large communities as
well as small. Since the redistribution of activity and the
increased efficiency cannot be predicted the effects of the
rationalization cannot be fully predicted.\7\
---------------------------------------------------------------------------
\7\ NPFMC, page 430.
The analysis attempted to learn about processor quotas from other
fisheries. This was a difficult undertaking in that such a management
measure apparently does not exist in any other fishery worldwide.
``Several program aspects of the alternatives are unique. For example,
no two pie IFQ programs have been implemented in any fisheries to
date.'' \8\
---------------------------------------------------------------------------
\8\ NPFMC, page 396.
---------------------------------------------------------------------------
Analogies were drawn by many to pollock AFA co-ops. The crab
rationalization analysis presented several statements concerning those
analogies, demonstrating that pollock co-ops and processor shares are
fundamentally different:
There is no parallel for that [two-pie] system under the
current pollock co-op system.\9\
---------------------------------------------------------------------------
\9\ SIA, page 18.
Pollock co-ops are plant-specific for the shore processing
sector, which makes them effectively community specific in
terms of social impacts. Under the proposed BSAI crab co-ops,
company level rather than plant level co-ops are contemplated.
This means that shifts of landings and processing effort
between communities could occur in a way (or to a significant
degree) that they cannot under the pollock co-ops.\10\
---------------------------------------------------------------------------
\10\ SIA, page 18.
Additional information was gathered from other studies and reports.
One of the most important of such other reports is the 1999 National
Research Council's report ``Sharing the Fish''. The following excerpt
---------------------------------------------------------------------------
concerning processor shares is from the Council analysis:
The NRC report ``Sharing the Fish'' included a discussion of
the merits of allocating either a portion of the quota to
processors or alternatively creating a separate class of shares
for processors. Processors are thought to have had mixed
results under IFQ programs. When adversely affected, processors
are argued to suffer from stranded capital and lower
profitability. Processor losses could also have negative
impacts on isolated communities. Processors that are successful
in IFQ fisheries tend to obtain results through ``contractual
methods or vertical integration''. The study concludes that if
protection of processors is an ``appropriate social goal, this
could be accomplished by allocating separate harvester and
processor quota.'' The report also suggested that other
methods, such as buyouts or permitting processors to own
harvester quota might be preferred to processor quotas, if
processor protection is a concern.\11\
---------------------------------------------------------------------------
\11\ NPFMC, page 256.
The NRC report's full recommendation for shares to other entities
such as processors is more robust. While the NRC finds no compelling
reason to establish a two-pie system they did recommend consideration
of community allocations of IFQs:\12\
---------------------------------------------------------------------------
\12\ National Research Council. 1999. ``Sharing the Fish: Toward a
National Policy on Individual Fishing Quotas.'' National Academy,
Press. Washington, D.C. Pages 205-207.
---------------------------------------------------------------------------
Processor Allocations
Recommendation: On a national basis, the committee found no
compelling reason to recommend the inclusion or exclusion of
processors from eligibility to receive initial quota shares.
Nor did the committee find a compelling reason to establish a
separate, complementary processor quota system (the ``two-pie''
system). If the regional councils determine that processors may
be unacceptably disadvantaged by an IFQ program because of
changes in the policy or management structure, there are means,
such as buyouts, for mitigating these impacts without resorting
to the allocation of some different type of quota, with a
concomitant increase in the complexity of the IFQ program. For
example, coupling an IFQ program with an inshore-offshore
allocation would preserve the access of shore-based processors
to fishery resources. Whatever method is chosen, it should not
have the effect of subsidizing excess processing capacity.
Depending on regional considerations, some councils may choose
to allow processors to acquire quota share either through
transfer or through ownership of harvesting vessels that are
entitled to an initial allocation of quota.
Allocations to Communities
Recommendations: The committee recommends that councils
consider including fishing communities in the initial
allocation of IFQs, where appropriate, and that the Secretary
of Commerce interpret the language in the Magnuson-Stevens Act
pertaining to fishing communities (Sec. 303 [b][6][E] and
National Standard 8) to support this approach to limited access
management. Congress should allow, though a change in the
Magnuson-Stevens Act if necessary, councils to allocate quota
to communities or other groups, as distinct from vessel owners
or fishermen. Where an IFQ program already exists, councils
should be permitted to authorize communities to purchase, hold,
manage, and sell IFQs. These communities could use their quota
shares for community development purposes, as a resource for
preserving access for local fishermen, or for reallocation to
member fishermen by a variety of means, including loans. If the
communities chose to allocate the rights to individuals, they
could be constrained by covenants or other restrictions to be
nontransferable. Regional fishery management councils should
determine the qualifying criteria for a community that is
permitted to hold quota. A range of factors, such as proximity
to the resource, dependence on the resource, contribution of
fishing to the community's economic and social well-being, and
historic participation in the fishery, may be among the factors
that a council considers when setting criteria for establishing
which communities may hold quota. The range of criteria will
have to be carefully considered and weighted and the
implications of defining these criteria would have to be
examined fully.
One of the issues raised by the community of Kodiak and others is
the inclusion of recent participation in the analysis. This is neither
a new nor recent concern and it is one the Council is well aware of.
``In addition to the issues raised in the NRC report, NOAA GC has
emphasized that the failure of the halibut and sablefish IFQ program to
give sufficient consideration to recent participation was an important
issue in the lawsuit filed against that program.'' \13\
---------------------------------------------------------------------------
\13\ NPFMC, page 193.
---------------------------------------------------------------------------
Information Concerning Impacts to Kodiak
The Council analysis documents present a great deal of information
concerning crab fisheries and how they related to Kodiak during the
last decade of the twentieth century. This information appears
primarily in the harvesting and processing sections of Section 2.6 and
in the Kodiak section of the SIA (pages 102-129).
The crab considered for rationalization, referred to as PRP,
consist of nine species and/or fisheries.\14\ Of these, the three
historically most important to Kodiak are Tanner crab (opilio and
baridi) and Bristol Bay red king crab. There has not been a fishery for
baridi since 1996 and only a very low harvest of opilio since 1999.
Therefore, the species of most immediate interest to Kodiak is Bristol
Bay red king crab.
---------------------------------------------------------------------------
\14\ The species included in the crab rationalization plan: Bering
Sea opilio, Bristol Bay red king crab, Bering Sea bairdi, Pribilof red
king crab, Pribilof blue king crab, St. Matthew blue king crab, Eastern
Aleutian Islands (Dutch Harbor) golden king crab, Western Aleutian
Islands (Adak) golden king crab, and Western Aleutian Islands (Adak)
red king crab.
---------------------------------------------------------------------------
A specific portion of the Kodiak based fleet in particular and the
Kodiak fishing community in general are dependant on PRP crab. ``[Table
2.6-13] indicates, for example, that PRP crab accounts for 35.5 percent
of the total value harvested by the combined Kodiak fleet. As shown,
although Kodiak has a large and diversified fleet, the Kodiak community
fleet is relatively more dependent on the BSAI crab species proposed
for rationalization (by far) than any other local Alaskan Fleet.'' \15\
---------------------------------------------------------------------------
\15\ NPFMC, page 111.
---------------------------------------------------------------------------
The relative dependence of the Kodiak fleet on PRP crab, while very
high, is still understated. This is because there was no fishery for
Bristol Bay red king crab, the most valuable of the PRP species, during
two of the years used in the analysis while fisheries for salmon,
pollock, Pacific cod, and other species continued.
The analysis presents the value and poundage of rationalized crab
species processed for certain areas and communities. However, while
Kodiak is one of the two communities listed, the information for
Bristol Bay red king crab is suppressed due to confidentiality.\16\
Tables presented in Appendix 3 to the SIA provide much more information
concerning Kodiak as well as other community participation in the
fisheries. However, the PRP crab species are combined and therefore
individual species/fisheries data are not available.
---------------------------------------------------------------------------
\16\ NPFMC, Tables 2.6-20 through 2.6-26.
---------------------------------------------------------------------------
Tables 2.6-27 through 2.6-29 of the NPFMC 2002 analysis attempt to
represent processor dependency for the PRP crab species during the
period 1991-2000. The data indicate that PRP crab account for 1.6
percent of the total PRP crab processed in all areas during the period
but 4.2 percent of the value of all species processed in Kodiak. The
explanation of the data, however, leaves some doubt as to their
usefulness. ``[I]t is important to bear in mind that these data cover
the full spectrum of processing operations in a given locality, and not
only those that process BSAI crab. It thus represents community
dependence on BSAI crab at a relatively high level of abstraction and
may not reflect any specific operation in the community, and the data
may represent more in the way of collective entity dependency rather
than community dependency.'' \17\
---------------------------------------------------------------------------
\17\ NPFMC, page 133.
---------------------------------------------------------------------------
The change is relative crab landings in Kodiak is discussed in the
documents. These qualitative discussions relate both to why the level
of landings have changed and what aspects of fishery management may
affect them. The discussion also overlaps to other community impacts
such as services and taxes.
After 1996, all pots had to be removed from the fishing grounds
within 7 days of the end of the season. Many Kodiak Bering Sea
crab boats report that they are not large enough to carry both
a load of crab and all their pots, so that this change in
regulation severely limited their ability to deliver crab to
Kodiak, especially during high GHL years. Such boats were
limited even on the last (or only) delivery trip of the season.
For their last trip, such boats were essentially forced to
deliver to a Bering Sea processor, return to the grounds and
pick up their pots, and then go to Kodiak. Some harvesters also
reported that processors required them to deliver all crab to
them, by linking such deliveries to markets for other fish.
Still, by 1999 the Kodiak processors and fleet had evidently
adapted to the extent that Kodiak deliveries and processing
were at the same levels as the early 1990s. The sharp increase
in 2000 may be due to a number of factors. One would be the
great decline in the GHL and harvest, so that many vessels had
only one delivery trip, often of a partial load, that allowed
them to carry their empty pots as well. Kodiak processors may
also have offered price incentives, for various reasons.\18\
---------------------------------------------------------------------------
\18\ SIA, page 119.
Kodiak processors believe that they could compete and increase
their market share for BSAI crab, especially in a rationalized
fishery, which would reduce the incentives for quick (Bering
Sea port) delivery. They cite the 2000 crab season as support
for this contention, as they greatly increased their market
share over that of the recent past. Thus, any alternative which
limits the ability of Kodiak processors to compete for BSAI
crab could have potential adverse effects on both Kodiak
processors and crab vessels. It would be in addition to the
competitive advantage that Bering Sea processors reportedly
achieve over Gulf of Alaska processors from rationalized [AFA
pollock] fisheries in the Bering Sea in relation to the open
access fisheries in the Gulf of Alaska.\19\
---------------------------------------------------------------------------
\19\ SIA, pages 123-4.
For Kodiak, local BSAI crab processing is not taking place because
of inefficiencies in the fishery that, in turn, make the lower costs of
local processing not worthwhile in an unrationalized system.\20\
---------------------------------------------------------------------------
\20\ SIA, page 128.
---------------------------------------------------------------------------
Implications to Communities From Combined Affects of Crab
Rationalization
The analysis does not quantitatively, nor for that matter in one
location qualitatively, address the expected combined effects of the
various management measures. Certainly such analysis is not conducted
for communities. However, by drawing some of the points from the
Council analysis together it is possible to make a reasonable projected
of expected impacts.
Potential for Oligopsony
The analysis states ``[A]n oligopsony exists when a limited number
of buyers of a product exists. These buyers can influence ex vessel
prices by refraining from purchases.'' \21\
---------------------------------------------------------------------------
\21\ NPFMC, page 403.
---------------------------------------------------------------------------
The crab processing industry is dominated by several large crab
processors. The following information is for the Bristol Bay red king
crab fishery. Similar concentration occurs in the Bering Sea opilio
fishery.
Average of top four processors\22\ 15.09%
---------------------------------------------------------------------------
\22\ NPFMC, Table 3.4-13.
---------------------------------------------------------------------------
Combined concentration of top 4 processors\23\ 60%
---------------------------------------------------------------------------
\23\ NPFMC, Table 3.4-13.
---------------------------------------------------------------------------
Total processing owners\24\ 26
---------------------------------------------------------------------------
\24\ NPFMC, Table 3.4-13.
---------------------------------------------------------------------------
Median processor allocation\25\ 1.0%
---------------------------------------------------------------------------
\25\ NPFMC, Table 3.4-5.
---------------------------------------------------------------------------
Average processor allocation\26\ 3.8%
---------------------------------------------------------------------------
\26\ NPFMC, Table 3.4-5.
---------------------------------------------------------------------------
Number of processors allocated 20% \27\ <3
---------------------------------------------------------------------------
\27\ NPFMC, page 283.
---------------------------------------------------------------------------
Number of processors allocated 30% \28\ 0
---------------------------------------------------------------------------
\28\ NPFMC, page 283.
---------------------------------------------------------------------------
Number of catcher/processors \29\ 11
---------------------------------------------------------------------------
\29\ NPFMC, Table 3.4-2.
---------------------------------------------------------------------------
Total C/P allocation\30\ 8.1%
---------------------------------------------------------------------------
\30\ NPFMC, Table 3.4-2.
These data probably underestimate the level of concentration.
Processor concentrations are grandfathered at any level with verified
contracts entered into by June 10, 2002. Any such contracts are unknown
at this time and are not required to be made public until the time of
allocations.
As is the nature of manufacturing in an unrestricted industry,
participants enter and leave on a fairly regular basis. This has been
the case in crab processing. The difference between the number of
processors and how change has occurred during the 1990s is illustrated
in information from Tables 2.3-1 and 2.3-2 of the Council analysis.
Data are not presented in the analysis for the number of processors
prior to 1991. The number of processors has decreased during the period
due to a dramatic decrease in catcher processors and floaters.\31\ The
number of shorebased processors has remained relatively constant at 15
in the Bristol Bay red king crab fishery since it resumed in 1996.
---------------------------------------------------------------------------
\31\ From 9 floaters and 25 C/Ps in 1991 to 2 and 6, respectively,
in 2000.
------------------------------------------------------------------------
Year Delivery Percentage Number of Processors
------------------------------------------------------------------------
1991 71% 29 qualified
29% 27 non-qualified
2000\32\ 96% 20 qualified
4% 3 non-qualified
------------------------------------------------------------------------
Foreign ownership of processors is widespread in the crab fishery.
Several of the major crab processors including Peter Pan Seafoods,
UniSea, and Westward Seafoods are fully owned by foreign companies
while Alyeska Seafoods and Stellar Seafoods \33\ are partially (25
percent or more) foreign owned.\34\ Under the recommended
rationalization program, 47 percent of the Bristol Bay red king crab
and 36 percent of the Bering Sea opilio processing quota will be
allocated to the afore named five foreign owned processors? \35\ The
qualified crab processing facilities owned by these firms which will
receive allocations are presented in Appendix 3-3.
---------------------------------------------------------------------------
\32\ The percentage is estimated based on analysis text since the
percentage for non-qualified processors is confidential.
\33\ Figure 8.7 of the January 2000 AFA report to Congress
considers Steller Seafoods to be 100 percent owned or controlled by
Peter Pan Seafoods.
\34\ NPFMC, Table 3.14-1.
\35\ NPFMC, Table 3.14-2.
------------------------------------------------------------------------
Company Catcher/processor Shorebased Floater
------------------------------------------------------------------------
Peter Pan Seafoods King Cove Blue Wave
Port Moller
UniSea Dutch Harbor
St. Paul
Westward Seafoods Dutch Harbor
Alyeska Seafoods Dutch Harbor
Stellar Seafoods Stellar Sea
------------------------------------------------------------------------
In addition, Trident Seafoods is known to be a major processor of
Bristol Bay red king crab as is Icicle Seafoods. Their qualified crab
processing facilities listed in Appendix 3-3 include:
------------------------------------------------------------------------
Company Catcher/
processor Shorebased Floater
------------------------------------------------------------------------
Trident South Naknek Akutan Bountiful
Akutan Alaska Packer
Glacier Enterprise
St. Paul Island Independence
Northern Enterprise
Sea Alaska
Royal Enterprise
Tempest
Western Enterprise
Icicle Arctic Star
Bering Star
Coastal Star
Evening Star
Northern Victor
------------------------------------------------------------------------
Of these seven companies, only Stellar Seafoods is not an AFA
qualified pollock processor owner. Harvesters have expressed concerns
about potential loss of bargaining power if there is a ``two-pie''
rationalization system. ``Harvesters claim that under the AFA crab
sideboards imposed on processors, harvesters have faced limited markets
and been forced to be price-takers and thus received less value for
their crab than they would otherwise have expected. The belief is that
processors may acquire even more leverage in the price-bargaining
relationship under any rationalization plan other than the ``one-pie''
(harvester only IFQ) alternatives. In short, many harvesters believe
that they have already been adversely affected in this way by the crab
sideboards imposed on AFA-qualified processors by the AFA. Once the
capped processors reach their limits, there are very few alternative
markets to which a harvester can sell.'' \36\
---------------------------------------------------------------------------
\36\ SIA, page 118.
---------------------------------------------------------------------------
This brings to mind a statement from almost a century ago.
We can let Alaska become the private preserve of a few great
special interests to be developed and controlled at their
pleasure and as their profit may dictate. . . . Or we can treat
Alaska as the future home of hundreds of thousands of free
American citizens, and its resources as a trust to be developed
and conserved for their benefit and for the benefit of all the
people-who are its owners. Gifford Pinchot, Saturday Evening
Post, December 16, 1911, pp12 ff.
Market Implications of Processor Quotas
The Council commissioned a study to specifically examine the
impacts on the crab fisheries from alternate rationalization
programs.\37\ It contained a number of statements related processor
quota markets and the effects on ex-vessel values from a two-pie
system. Overall, they found that more restrictions placed on where a
harvester may deliver, the more policy benefit shifts towards
processors.
---------------------------------------------------------------------------
\37\ Walter Milon and Steve Hamilton. ``A Comparative Analysis of
Alternative Rationalization Models for the Bering Sea/Aleutian Islands
(BSAI) Crab Fisheries,'' March 2002, Discussion Paper for the North
Pacific Fishery Management Council. 66 pp.
Typically markets [for processor permits] work best when there
are a large number of buyers and sellers in the market. In
``thin'' markets for permits, a small number of firms may have
little incentive to trade if there are small differences in the
firms' productions costs and/or the firms perceived the permits
can be used for strategic advantage.\38\
---------------------------------------------------------------------------
\38\ Milton and Hamilton, Page 9. Referenced to (1) ``Small
Pollution Markets: Tradeable Permits versus Revelation Mechanisms,'' F.
Andersson, 1997. Journal of Environmental Economics and Management
32:38-50. (2) ``Raising Rival's Costs Strategies via Emission Permits
Markets. E. Sartzetakis. 1997. Review of Industrial Organization
12:751-765.
If there is a market share guarantee for processors, then there
is not a threat of losing deliveries if the ex-vessel price is
lowered. Therefore, a processor who wants to maximize profits
would lower his ex-vessel price and harvesters would still have
to deliver to him.\39\
---------------------------------------------------------------------------
\39\ Milton and Hamilton, Page 41.
In general, the less the market is segmented, the larger the
prevailing ex-vessel prices under processor price competition.
. . . [t]he more restrictions that are put in place to govern
the location to which individual harvesters must deliver, the
less the degree of ex-vessel price competition to acquire
market share.\40\
---------------------------------------------------------------------------
\40\ Milton and Hamilton, Pages 41-2.
Other studies reviewed in the Council analysis seem to agree on the
effects of thin markets. ``The authors [Matulich and Sever] caution
that transaction costs can inhibit efficiency. Specifically, they point
out that thin harvesting or processing share markets (or markets with
few participants or shareholders) could reduce efficiency gains in both
sectors. Limited numbers of participants would limit the ability of
participants to purchase the number of shares necessary to operate
efficiently. Vertical integration is argued to have the potential to
contribute to these efficiency losses.'' \41\
---------------------------------------------------------------------------
\41\ NPFMC, page 402, referencing Matulich and Sever,
``Reconsidering the Initial Allocation of ITQs,'' Land Economics
(1999).
---------------------------------------------------------------------------
Vertical integration does exist in the crab fisheries. Under the
proposed Bristol Bay red king crab rationalization plan, there are 31
vessels with documented affiliation with a total of 6 processors. This
will result in 4 processors being allocated an amount over the one
percent harvest ownership cap and at least one processor receiving over
5 Percent\42\
---------------------------------------------------------------------------
\42\ NPFMC, Table 3.4-14.
---------------------------------------------------------------------------
These 31 vessels will be allocated 12.6 percent of the total
harvest allocations.\43\ As stated previously, there may well be
greater vertical integration due to contracts entered into on or before
June 10, 2002 and not yet made public.
---------------------------------------------------------------------------
\43\ NPFMC, Table 3.4-27.
---------------------------------------------------------------------------
Processor Bargaining Tactics in a Two-Pie System
The Council analysis addressed some of the potential implications
of bargaining between harvesters and processors in a two-pie system.
Harvesters and processors in the crab fisheries are related on
several levels ranging from common ownership to simply repeated
transactions in the buying and selling of crab. Since the
relationships are often manifold, their dynamics are also quite
complicated. Understanding these relationships, however, is
critical to understanding the applicability of a
rationalization program in a fishery.\44\
---------------------------------------------------------------------------
\44\ NPFMC, page 99.
Harvesters and processors also have support relationships that
are important to both sectors. Some processors sell bait, fuel
and food to vessels (often on credit) and store gear for
vessels during the offseason. At times, vessel owners with
large debts to processors will give the lending processor a
lien on their vessels. Whether a lien is taken is dependent on
the relationship between the vessel owner and the processor.
Because of confidentiality, the number of these liens and
whether and the extent to which they are used to exert pressure
on vessel operators is not know. Vessel owners also enter
contracts to tender salmon and herring for processors outside
of the crab season. Both vessel owners and processors contend
that tendering relationships are important to their businesses.
The extent to which either side exploits the other based on
these tendering contracts is also not known.\45\
---------------------------------------------------------------------------
\45\ NPFMC, page 100.
Some harvesters also reported that processors required them to
deliver all crab to them, by linking such deliveries to markets
for other fish.\46\
---------------------------------------------------------------------------
\46\ SIA, page 119.
Pricing [also] varies regionally among processors in the crab
fisheries. Regional price differences have several sources. In
fisheries where vessels make several deliveries, the
availability of goods and services in a location can be
important to fishers. Food, bait, fuel, and a good port
facilities can make a processor more attractive to vessels
wishing to offload harvests. Processors in locations that offer
less goods and services may pay price premiums to induce
fishers to sell their harvests. Processors that are distant
from grounds may also need to pay a premium price to compensate
fishers for time away from the grounds while making deliveries.
Proximity to consumer markets can also influence ex-vessel
prices. Processors with less access to consumer markets may pay
slightly less for crab inputs than processors closer to end
markets since they must bear the cost of delivering the crab to
---------------------------------------------------------------------------
market.
Generalizations concerning the spatial distribution of ex-
vessel prices may be difficult to make. Dutch Harbor, where the
most processors are located can be used as the basis for
determining prices. The prices in Kodiak are higher
(approximately $0.20 in the recent Bristol Bay red king crab
fishery) because of the longer distance to the fishing grounds
and the proximity to consumer markets. . . . These minor price
differences between ports are thought to have little effect on
the competitiveness of vessels that deliver to the facilities a
the different ports, when the other costs are considered.\47\
---------------------------------------------------------------------------
\47\ NPFMC, pages 101-2.
To an unknown extent, price negotiations and delivery patterns
are influenced by relationships between harvesters and
processors. Some harvesters tender salmon and herring for
processors. Maintaining this contract might require the
harvester to continue to deliver crab to the processor.
Similarly, some harvesters receive financial support from
processors. Whether formalized or not, some of these harvesters
have a perceived obligation to deliver crab harvests to the
processor with whom they have the financial relationship. The
extent of the impact of these relationships and obligations on
prices and delivery patterns is not known.\48\
---------------------------------------------------------------------------
\48\ NPFMC, page 102.
---------------------------------------------------------------------------
Conclusions of Community Impacts Under a Two-Pie System
The impact to communities of allocations to processors within those
communities is not at all certain. This is due to the transferability
of processor quotas between plants owned by the same company. ``Because
allocations are to processing companies, however, and not to specific
facilities or communities, economic decisions at the corporate level to
shift production from one facility to another may have community
effects that are essentially unknowable beforehand. . . . Given the
tendency of the marketplace to reveal costs and incentives that had not
previously been well know, however, this assessment is not one with a
great deal of certainty.'' \49\
---------------------------------------------------------------------------
\49\ NPFMC, page 136.
---------------------------------------------------------------------------
The issue of processor consolidation is germane to the issue of
community impacts. If operations do consolidate they will occur in
fewer plants and, without a doubt, some communities will suffer losses.
The analysis states that ``[w]ith processor quota shares, we can not
predict if the processing sector will consolidate''.\50\
---------------------------------------------------------------------------
\50\ NPFMC, page 162.
---------------------------------------------------------------------------
There are three shorebased pollock co-ops in Dutch Harbor owned by
the companies listed above (Alyeska, Unisea, and Westward) and one in
Akutan (Trident). ``Under pollock co-ops, shoreplants have remained
more-or-less self-contained, self-sufficient enterprises in the
community. This varies from plant to plant, but operations tend to be
of an industrial enclave nature, with a relatively low volume of
purchases of goods and services from the local support sector. Crab co-
ops are not seen as likely to change this pattern.'' \51\
---------------------------------------------------------------------------
\51\ SIA, page 67.
---------------------------------------------------------------------------
Processing enclaves do not exist in Kodiak. Only two percent of the
population lived in group housing in Kodiak in 2000 \52\ compared to 51
percent in Unalaska the same year.\53\ The processing plants in Kodiak
purchase utilities, supplies and services from local venders as do the
workers who live in the community.
---------------------------------------------------------------------------
\52\ SIA, Table 2.6-4.
\53\ SIA, Table 2.2-5.
---------------------------------------------------------------------------
Therefore, to the extent that processing is reduced in Kodiak, and
shifted to processors which operate from enclaves, there will an
overall net reduction in local economic activity in Alaskan communities
related to crab fisheries. For all the reasons discussed above, this is
a likely outcome under a two-pie crab rationalization program.
______
Prepared Statement of Michael B. Laukitis, President,
North Pacific Fisheries Association
U.S. Senate Committee on Commerce, Science, and Transportation
Senator John McCain, Chairman
Dear Senator McCain:
The North Pacific Fisheries Association (NPFA) thanks you for the
opportunity to comment on the Bering Sea Crab Rationalization Plan.
NPFA was founded the year Alaska became a state and is one of the
oldest fisheries organizations in the State of Alaska. Our members
include Bering Sea crabbers, halibut and sablefish IFQ longliners, Gulf
of Alaska fixed-gear groundfish harvesters, fishermen who custom
process their own catch for export markets, and salmon fishermen. Much
like family farmers we are independent fishermen and fishing families
who go out on our own boats and work to receive the highest value for
our catch.
NPFA supports the conservation and safety aspects of the Bering Sea
Crab Rationalization Plan which are much needed for the benefit of
future sustainable crab stocks and for the benefit of fishermen who
risk their lives to harvest American seafood. We also support
individual fishing quotas (IFQ's) for harvesters. Many of our members
are involved in the halibut and sablefish IFQ fisheries, and we have
seen that this carefully designed program is benefiting everyone
involved in the industry. The halibut industry went from chaos prior to
1995 to a stable and safe fishery where the industry is putting very
high quality seafood products on American's kitchen tables. We believe
the majority of the crab plan could go forward as written.
There is one very important aspect of the Crab Rationalization Plan
that NPFA does not support, and that is Processor Quotas, or PQ's. NPFA
and many people in Alaskan coastal communities see corporate
consolidation in our fisheries and the drive by multi-national food
companies to control Alaska's rich fishery resources, our management
agencies, our coastal communities, and eventually our fishermen as the
number one threat facing us today. Our fishermen and communities want
the full protection of Federal anti-trust laws, and we want to be able
to develop new markets in our communities and to be able to sell our
catch in open markets. We believe healthy competition brings
innovation, and although we are in an old profession many of us pride
ourselves as small business entrepreneurs.
Processor quotas by definition are anti-competitive. They give a
market share guarantee that eliminates any incentive to maintain ex-
vessel prices above the monopoly level. We do not believe the binding
arbitration program contemplated by the North Pacific Fisheries
Management Council is adequate to protect fishermen's interests. The
Council's expertise is in fishery management and conservation.
Obviously its expertise is not in planned economies and anti-trust law.
This program shifts benefits to the processing sector. Also, after the
American Fisheries Act we have seen that the processing sector uses the
market power derived from processor shares in the Pollock fishery to
further shift benefits away from harvesters in other non-Pollock
fisheries. We fear that this trend towards processor quotas only begins
here and will eventually affect all of Alaska's fisheries to the
detriment of independent fishermen.
Awarding cartel status to the snapshot of companies who happen to
process crab in Alaska at this time would also harm other Alaskan
processors who might not be active in that region of the state
currently. (The short history of the American Fisheries Act is a case
in point. Several prominent Alaskan processing companies who argued for
free markets for Pollock leading up to the Act but were ``left out'',
were then obviously disadvantaged versus their competitors who had a
guarantee share and equity in the Pollock fishery, so they are now
leading the charge for processing quotas in the crab fishery.)
Furthermore this crab cartel would harm competing processors in other
coastal regions of the United States who would then have to compete
with already powerful vertically integrated multi-national food
companies who would be granted a guaranteed share of the resource as
well as pricing power from the boat all the way to the consumer.
Although it may seem to be good for Alaska to have the companies who
operate here given such dominance in the marketplace, it is telling
that the overwhelming majority of communities and fishermen in Alaska
do not support processor shares. It needs to be asked whether granting
processor shares to companies who happen to process in Alaska's Pollock
and crab fisheries eventually triggers uncompetitive consequences for
all seafood companies in the United States.
In conclusion NPFA sees the benefits of rationalizing the Bering
Sea Crab fishery for conservation and safety reasons. We do not feel
rationalization should take place at any cost. The economic analysis
used by the North Pacific Fisheries Council to justify processor shares
called the Matulich study is seriously flawed as was reported by the
GAO earlier this year. Alaskan processors and communities gain
protection from other aspects of the Crab Plan (i.e., regionalization
requirements for delivery of crab and that catcher boats must deliver
their catch in Alaska).. The processor share aspects of the program are
unacceptable. They upset the balance between processor and independent
fishermen and potentially trigger consequences for all fisheries
throughout the United States.
Again, thank you for the opportunity to provide written testimony
to this hearing.
Sincerely,
Michael B. Laukitis,
President,
North Pacific Fisheries Association.
______
Prepared Statement of Phillip Lestenkof, President,
Central Bering Sea Fishermen's Association
My name is Phillip Lestenkof. I am the President of the Central
Bering Sea Fishermen's Association (CBSFA), the management organization
for St. Paul Island under the Western Alaska Community Development
Quota (CDQ) Program. By way of background, the CDQ program was
established by the U.S. Congress in 1992 as a mechanism for ensuring
that remote and impoverished Western Alaska communities would benefit
from rich Bering Sea fisheries. Initially, 7.5 percent of the fisheries
were set aside for the 6 CDQ groups that were created, including CBSFA.
I want to thank the Committee Chair and members for the opportunity
to provide testimony on the proposed Bering Sea Aleutian Island (BSAI)
Crab Rationalization Program (the program) developed by the North
Pacific Fishery Management Council (NPMFC) which CBSFA has steadfastly
supported over the last four years together with the City of St. Paul,
AK.
I. Impacts of Crab Rationalization Program on St. Paul
I will refer to the testimony of Simeon Swetzof, the Mayor of the
City of St. Paul for a description of the impacts that the collapse of
the opilio and other Bering Sea crab stocks has had on St. Paul
Island's economy.
Suffice it to say that, as the local fishermen's association, CBSFA
has been stressed by the direct and indirect impacts that the collapsed
crab stocks have had on economic activity on St. Paul. Crab processing,
for example, enables local processors to cover overhead costs and
allows them to process other species valuable to our community and
local fishermen, in particular IFQ and CDQ halibut. The low volumes of
opilio crab being processed, therefore, have threatened other economic
activities including those upon which our local fishermen make their
livelihoods for most of the year.
The NPFMC's crab rationalization program will stabilize the
situation by ensuring, through regionalized shares, that crab
processing will continue to be a viable business activity on St. Paul
even under a scenario of reduced crab stocks. More importantly, with a
rationalized and regionalized crab fishery, CBSFA's private sector
partners will be able to plan and predict outcomes to business
decisions more easily. Numerous projects that intend to take advantage
of St. Paul's unique location amidst the Bering Sea fisheries to
diversifY into multispecies processing, are on hold pending enactment
of legislation by the U.S. Congress that will implement the program.
In addition, CBSFA is working on a largely CDQ-owned, vertically
integrated crab processing and harvesting venture that depends on
implementation of the program. This new venture has been approved by
the State of Alaska and NMFS as the core project in CBSFA's 2003-2005
Community Development Plan. As part of the plan, CBSFA is negotiating
for equity ownership in various crab industry assets. These include:
1. Five percent of the total Opilio crab harvesting quota.
2. Five percent of the total Bristol Bay Red King Crab harvesting
quota.
3. More than five percent of the Adak Brown harvesting quota.
4. Five percent of the Pribilof Reds, Pribilof Blues, and St.
Matthews harvesting quotas.
5. CBSFA is also negotiating with processing companies to acquire
equity stakes in approximately I 0 percent of the entire
Processor Quota Share pool created by the program.
If successful, CBSFA's plan will represent a significant step in
increasing the interests of native-owned Alaskan companies crab
industry assets, many of which are currently owned by Japanese
interests.
II. The Need for Regionalization
Like the City of St. Paul, CBSFA views the regionalization scheme
in the program as the most effective tool for fulfilling the guidelines
of Standard 8 of the Magnuson-Stevens Act and accommodating the diverse
interests within the BSAI crab fishery. Standard 8 directs the fishery
councils to take into account the need for sustained participation of
fishing communities in the fisheries. The need for regionalization into
northern and southern shares depending on historic activity arose in
order to protect St. Paul's (and other BSAI communities') livelihood
and the considerable infrastructure investments made to develop a port
which has served the BSAI crab industry very effectively and
profitably.
During the 1990s, our community benefited from a derby-style, race-
for-fish scenario where vessels delivered crab to the nearest available
harbor, in order to immediately depart for further loads of nearby
abundant crab stocks. In a rationalized fishery however, the race-for-
fish, would be eliminated and furthermore, with the Guideline Harvest
Levels (GHL) dramatically reduced as a result of the biomass collapse,
harvesters would be limited to single trips and would therefore be more
likely to deliver their crab to other ports such as Dutch Harbor/
Unalaska or to their homeports including Kodiak, where a number of
harvesters are based.
Just as harvester and processor investments have been recognized
and protected in prior rationalization schemes, both CBSFA and the City
of St. Paul took the novel approach of insisting that community
considerations needed to be legitimately factored into this process.
The stakes were considerable. St. Paul Island, unlike economically
diversified communities in the Gulf of Alaska, is almost entirely
dependent on crab processing. While St. Paul's strategic location and
its existing infrastructure make it a competitive port, most of the
processing operations in the Pribilof Islands are·''satellite
plants'' for diversified, parent seafood companies. Therefore, with the
incentive for decapitalization and consolidation in a rationalized
fishery, St. Paul was at risk of seeing local processing operations
retreating south to larger facilities in Dutch Harbor, Akutan, and
Kodiak.
Regionalization will stabilize the fishery and ensure that existing
processing operations will remain in the northern region in a
rationalized scenario. One of the added benefits of regionalization is
that the longer runs involved between the fishing grounds and ports
such as Dutch Harbor or Kodiak in a non-regionalized scenario would
have implied additional fuel costs to harvesters and greater deadloss,
to the detriment of the resource.
III. Benefits to the Resource and the Fleet
Environmental benefits will accrue as well from the mechanisms to
reduce excess harvesting and processing capacities. With slower seasons
in which to prosecute the fishery, fewer crab pots will be lost at sea
and deadloss and bycatch will be reduced. A healthier, better managed
resource is critical to St. Paul as well, since my community's long-
term livelihood depends on the sustainable use of the surrounding
Bering Sea fisheries Finally, experience demonstrates that to the
extent that a rationalized management system results in a sustainable,
economically viable fishery, then improvements in safety should follow.
In a scenario where the ``race-for-fish'' is eliminated, fatalities
resulting from having to fish in poor weather conditions should be
reduced.
IV. The ``Third Pie'': Community Protections and Benefits
Under the proposed program, allocations of harvest shares would be
made to harvesters, the CDQ program, and captains. Moreover, processors
would be allocated processing shares and harvesters would be permitted
to consolidate their efficiencies by forming cooperatives. From a
community perspective, establishment of processor shares protects the
considerable investments made by processing companies in remote coastal
Alaskan communities such as St. Paul. These investments typically
include the processing facility itself, labor housing, docks, cold
storage, tender vessels, and other infrastructure.
Communities for their part have spent significant resources on
harbor improvements, docks, water/sewer systems, roads, emergency
support services, and community administration to attract and
accommodate processors.
In effect, processor shares are necessary for regionalization to
work for communities. As a largely processing town, St. Paul's future
is linked to the processors that have invested or will invest there.
St. Paul (as is St. George) is heavily dependent on crab processing
business in its harbor and on the floating processors within the state
boundaries around the island. The community receives benefits from
taxes, fuel sales, village corporation land leases, store sales,
transportation and employment opportunities provided by the processors
operating in the community.
Under the program, processor allocations and the corresponding 90
percent of the harvester allocations will be designated into the
earlier mentioned northern and southern regions, based on historic
participation. Provisions allowing for the non-regionally designated 10
percent of harvester allocations to be delivered to any processor at
any port ensure that a substantial portion of the crab resource will be
subject to competition among ports and processors seeking to attract
deliveries.
However, as specified in the NPFMC's trailing amendments, a two-
year ``cooling off period'' has been established during which
processing shares must remain in communities where processing was
historically conducted, so as to avoid a ``rush'' to consolidation in a
few ports immediately after implementation of the program. In addition,
communities and CDQ groups such as CBSFA have been granted a right of
first refusal on the sale of processing shares to protect crab-
processing dependent communities.
To accommodate periods of abundance in the crab stocks and provide
opportunities for new processors and communities to compete, caps of
175 million pounds for opilio and 20 million pounds for Bristol Bay red
king crab have been established on the amount of IPQs granted for the
two largest crab fisheries. However, if stocks continue to remain low
(and below those caps) the IPQs will provide stability to the
processing sector and crab dependent communities.
Other important community friendly provisions from CBSFA's
perspective include an increase in CDQ crab allocations from 7.5
percent to 10 percent of the Total Allowable Catch (TAC). CDQ groups
will be required to deliver at least 25 percent of the allocation to
shore based processors. Moreover, CBSFA and other CDQ groups have the
potential to benefit from provisions extending community purchase
rights of harvesting and processing shares to community and CDQ groups.
These purchases would be conditioned on the use of these shares for the
benefit of community residents. Finally, CBSFA and other CDQ groups
have been granted higher ownership caps than individuals purchasing
shares in order to give such groups the wherewithal to consolidate
their economic interest in the crab fisheries to the benefit of Western
Alaska residents.
VI. Conclusion
The NPFMC's BSAI crab rationalization program is a carefully
balanced plan that requires the unique structure that has been
developed over a four year period to protect all of the participants in
this fishery. It is clear that the program increases the economic
stakes of communities and CDQ groups in the proper management of
fisheries on which they depend for survival. Unlike harvesters and
processors, communities cannot ``sail away'' in pursuit of better
fishing grounds once a fishery has collapsed.
We urge Congress to support the NPFMC's plan by enacting
legislation as expeditiously as possible. After four years of economic
crisis, remote communities in the Aleutian Chain and particularly in
the Pribilof Islands are highly dependent on this program for their
economic well-being and survival.
Mr. Chairman, and distinguished members of the Subcommittee, thank
you for this opportunity to provide written testimony on behalf of the
CBSFA. We look forward to discussing these issues with you and your
staffs.
______
Prepared Statement of Rudy A. Petersen, Fisherman, CEO, and Owner,
Fishermen's Finest, Inc., F/V American No. 1/F/V U.S. Intrepid
Bering Sea Crab Rationalization
With reference to my letters of August 26, 2002, September 6, 2002,
December 13, 2002, and March 6, 2003 (copies enclosed) opposing the
North Pacific Fishery Management Council's (NPFMC) proposed ``two pie''
Bering Sea Crab Rationalization Plan, a plan which endorses ``processor
shares'' for the Bering Sea crab fishery, I hereby submit my written
testimony to the U.S. Senate Committee on Commerce, Science, and
Transportation; The Honorable John McCain, Chairman.
There should be no doubt on the part of the Committee Members of my
opposition to this plan as well as the opposition of the Independent
crab fishermen, concerned citizens and Alaskan communities. My letters
and petition with over 700 signatures of Americans opposing this plan
have been faxed and mailed as set forth above to the following:
House Resources Committee Members
Senate Resources Committee Members
House Appropriations Committee Members
Senate Appropriations Committee Members
All members of Washington State Delegation (Governor,
Senators, Congressmen, NPFMC members, etc.)
All members of Alaska State Delegation (Governor, Senators,
Congressmen, etc.)
I am submitting this written testimony as I will be unable to
attend the hearing in Washington, D.C. In person. I urge you to review
my materials once again and to oppose any Bering Sea crab
rationalization plan that includes ``processor shares'' or a ``two
pie'' formula.
Processor quota shares are much bigger than a mere ``fish grab.''
It strikes to the very heart of our fishing industry and if allowed to
invade our free market driven industry, the processor quota two pie
allowances will negatively Impact our fishing Industry forevermore.
The Idea of a few fishing companies having control over our crab
and other fisheries is not in the best interest of the fishermen nor in
the best interest of the public. The resources are after all a public
resource; protection of this and all other public resources should be
maintained by allowing our competitive free market forces to work, not
by granting the processing sector a virtual monopoly as proposed by the
North Pacific Fishery Management Council.
______
Fishermen's Finest, Inc.
Seattle, WA, August 26, 2002
Hon. John McCain,
Washington, DC.
Re: Bering Sea Crab Rationalization
I am writing to you as CEO of Fishermen's Finest, Inc., an
Independent, wholly U.S. owned fishing company located in Seattle,
Washington. I operate four small catcher/processor fishing vessels
which harvest and process groundfish in North Pacific waters and
support some three hundred fishing families from across the country. I
am also writing on behalf of the many who share my concerns and oppose
the North Pacific Fishery Management Council's proposed Bering Sea Crab
Rationalization Plan that endorses processor shares.
Recently, the North Pacific Fishery Management Council voted in
support of the Bering Sea Crab Rationalization Plan that would award
the processing sector with an anti-trust exemption by allowing
processing shares (referred to as a ``two pie'' system). We oppose the
proposed plan which will create a virtual monopoly for a few
processors. The Honorable Leon Panetta, head of the Pew Ocean
Commission, said in a recent Interview for the Alaska Fisheries Report,
``I'm just very concerned any time that competition is restricted. What
happens when you concentrate power in any one segment of our economy,
is that the little guy is bound to be hurt . . . I was very surprised
that they (the House Resources Committee) provided an exception for
Alaska (and nowhere else in the country). If it's not good for the rest
of the country, it's not good for Alaska . . .''
It is wrong to push through this ``two pie'' system, which has been
rejected in all other fisheries, when one sector of the industry--the
processors--has the upper hand financially and politically and the
small independent financially depressed harvesters have no organized
defense.
It is implicitly unfair to the harvesters that originally built the
crab fishery and who are currently experiencing great economic hardship
due to depressed resources and market conditions. Support of the ``two
pie'' system will give even more control of this fishery to the
processors and ultimately will hurt the harvesters even more. It will
be the end of the independent fishermen. The small independent business
people of this country need your support and therefore I am looking for
your help to stop this proposed ``two pie'' system that would unfairly
grant the processing companies, most of which are foreign owned, the
upper hand at the expense of the independent fishermen.
Monopolies are bad for America, especially when they involve our
natural resources. It is wrong to grant exclusive access to a select
few at the expense of the small independent fishermen/businesses of
this country. Please do not allow Congress to include wording in the
Magnuson-Stevens Act Reauthorization, or in any other Bill,
appropriations or otherwise, that allocates crab rights to processors.
Concerned U.S. citizens have signed a petition opposing processor
shares and we are continuing to gather more signatures. I would welcome
the opportunity to talk with you and discuss this important issue in
more detail should you need any additional information. Thank you for
your consideration.
Sincerely,
Rudy A. Petersen
Enclosures: Various Media Articles opposing processor shares (via 1st
Class Mail)
Organizational Charts Illustrating processor foreign ownership (via 1st
Class Mail)
Signed Petitions against processor shares (Via 1st Class Mail)
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
______
Fishermen's Finest, Inc.
Seattle, WA, September 6, 2002
Via 1st Class Mail
Re; Bering Sea Crab Rationalization
With reference to my previous letter faxed and mailed to you late
last month, I am writing to follow up with additional signed petitions
opposing the proposed ``two pie'' crab rationalization plan proposed by
the North Pacific Fishery Management Council that would grant
``processor shares'' in the Bering Sea crab fishery.
If by chance you have not seen this previous letter, I will provide
a brief overview of my opposition to this proposed plan. I am writing
to you as CEO of Fishermen's Finest, Inc., an independent, wholly U.S.
owned fishing company located in Seattle, Washington. I am also writing
on behalf of the many who share my concerns and oppose the North
Pacific Fishery Management Council's proposed Bering Sea Crab
Rationalization Plan; a plan that endorses ``processor shares.''
We have gathered in excess of 500 signatures of U.S. Citizens who
share our concern for fairness and justice and oppose the proposed
``two pie'' Bering Sea Crab Rationalization Plan. We will continue to
bring this issue to light with the public, as awarding the processing
sector with an antitrust exemption through ``processor shares'' is
wrong. Monopolies are bad for America, especially when they involve our
natural resources. This plan, if enacted as proposed would grant
exclusive access to a select few processors at the expense of the small
independent fishermen/businesses of this country; this is wrong.
I offer the following reasons to oppose the ``two pie'' Bering Sea
Crab Rationalization Plan:
This type of rationalization of our public natural resources
has been rejected by all other Fishery Management Councils
across the country.
The processing sector already has the upper hand financially
and politically and the small independent financially depressed
harvesters have no organized defense.
Support of the ``two pie'' system wilt give even more
control of this fishery to the processors and ultimately will
hurt the harvesters even more.
Processor shares equate to a virtual monopoly for the
processing sector.
A majority of the crab processing companies are foreign
owned/controlled while the Independent harvesters must meet
Maritime Administration required 75 percent U.S. ownership.
Please do not allow Congress to include wording in the Magnuson-
Stevens Act Reauthorization, or in any other Bill, appropriations or
otherwise, that allocates crab rights to processors. Concerned U.S.
citizens have signed the enclosed petitions opposing ``processor
shares'' and we are continuing to gather more signatures. I would
welcome the opportunity to talk with you and discuss this important
issue in more detail should you need any additional information. Thank
you for your consideration.
Sincerely,
Rudy A. Petersen
Enclosures: Over 500 Signed Petitions against ``processor shares/two
pie''
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
______
The Seattle Times--Editorials & Opinion: Saturday, November 02, 2002
Crab industry bakes a monopoly pie
The Alaska crab industry is asking Congress for a law that would
grant Individual fishing and processing quotas. The fishing quotas, if
paid for, may be wise. Processing quotas are monopoly privileges, and
should be rejected.
Individual quotas began with halibut. It is a valuable fish, and in
the open-access fishery, owners kept piling in boats until the season
shrank to a few days. Fishermen stayed awake with pills, they put hooks
through their palms, and if there was a storm, sometimes their boats
roiled over. Halibut. available in theory for most of the year, entered
the market in a flood and were sold tothe consumer frozen.
The system was crazy. And so came the idea of setting a quota per
boat. That way, a skipper would no longer be In a rush, and would treat
nature, his catchand his personal safety with respect. The new system
was tried, and ail these public benefits were achieved. But there was
also a private benefit: the quota shares, provided free to every owner
of a halibut boat, were a new asset that could be bought and sold. That
meant that owners of unneeded boats could sell their shares and be paid
to leave.
Nowhere else In American Industry was capital paid to leave. The
companies that bought the halibut also had excess investment, and they
had to write it offas a loss.
Not any more. The proposed crab plan has quotas for processors,
What Is being done here is the creation and division of a pie: The
pie does not consist of the crab, which was always there, but the
exclusive rights to the crab. It is something like a stock offering,
except that the original owner of the shares--the public--is paid
nothing. The transaction Is political and the public is not
represented. The dickering is between the boat owners, the processors,
the Alaskan coastal communities and the skippers. All get a share.
The buying and sailing of a right to catch fish makes a certain
amount of sense. Buying and selling a right to buy fish is selling a
monopoly privilege. It will be remarkable if it is even legal.
Anchorage Daily News--Monday, May 19, 2003
Fish Processing Shares Are Bad For Alaska
To Governor Frank Murkowski, Senator Ted Stevens, Senator Lisa Murkowski, and Congressman Don Young:
We are Alaskans urging you to reject processing quota because of the controlling effect on markets, fishing
families, coastal communities, and the public process.
WE OPPOSE A RETURN TO PRE-STATEHOOD DAYS WHERE A FEW FISH PROCESSORS CONTROL ALASKA'S FISHERY RESOURCES.
GOVERNMENT SHOULD NOT ALLOCATE MARKET SHARES TO PROCESSOR OR FORCE ALASKA FISHERMEN TO SELL THEIR CATCH TO A
SELECT GROUP OF PROCESSORS.
Danny Powell Sergius Robert Kara Lagasse Danielle Nathan Dean Clark Mack P.
Krukoff Linville Santillana Jankowiak Kinney
Vanessa Norman Mullan Heather Malina Kenneth Bill Webber Alexander Charles
Layamana Johnson Nicklas Simpson Ivanov Polak
Peter Togiak Alfonso Brett Robert Earnest D. Richard Stephanie Nikolai
Suralta Zirlott Beedle Linton Thummel Pitzman Basargin
James Russell Ryan Johnson James Curran Katrina Carl Ernest Wahl Anton Tom Wood
Higbee Nostrand Sanarov
Tim Quirk Susan Jeffrey John Nevin Greg Moyer Kenneth Lon A. White Robert Le Roy
Mears Williams Martin
David Cline Timothy Stage Howard Brian John Margaret Richard Linda Kozak
Henning Dierich Whittier Bolger Sheard
Ray Bellamy Jeremy Shiok Dan Bodey Lee Don Dumm Mary Taylor Madeleine Lucio
Veerhusen Winningham Grant Gonzalez
Michael Van Hans Christopher Jimmy Weaver John Spute Jennie Julie Lisa Ruoff
Note Tschersich Lopez Jacobsen Millier
Tracy Lohman Atiana Konev Frederico Rex Willhite Marie Lowe Laurie Rilee Higbee Ken Woodman
Sabado Bermudez
Shelley Philip Wiley J. Porter Bruce Colleen Betsy Wilson Nancy Vera
Rainwater O'Hara Schactler Rankin Behnken Basargin
Dorothy Holm Avram Kalugin Jack Cushing Brian Elizabeth Daniel Richard Peter
Blankenship Sencac Harbour Tyler Galaktionof
f
Marie Murray Pepper Nelson Chuck Kyle Crow Berry Spivey Gregory Dave Kaplan Leon J.
Thompson Spalinger LeClaire
III
Pete Janstory Nathan Benton Robert Harry Robert James Kenneth Geoffrey
Johnson Elliott Livingston Brennan Hoisington Hershberger
Michael Margaret Palmer Ron Knop Paul Seaton Burt Faustina Jules
Oleksa Short Thomassen Lekanoff Tejada Tileston
Lisa Frederic Tracy Craig Kathleen Stephen Hall Edmund E. Sandra Vicente Phillip
Ballenger Pasana Willhite Naanos Favadil
Kathryn Dave Magone Lavro Ivanov Mimi Hogan Ron Huffnage Frances Jon Glazner Ivan Reutov
Swiderski Ashford
Feodor Matt Strametz Richard William Foma Efimov Kenneth N. Casey Eads James
Martushev Davis Tener Holland Mahaffey
Tim Shen Delonio Dale Banks Mari Reeves Brad Shocky Kenneth Bill Schauff
Agonoy Faulkner Greenberg Quinn
Rick DeGroot Eric J. Mark Hibbard Ronald Eads Duncan Francine Mark R. David Reutov
Weathers Fields Bennis Lyman
Frank Lloyd Kent Helligso Monica Eladio Malcolm Twyla Hunter Duane B. Kathryn
Zacharof Francisco Milne Perk Hansen
Vivian Candace Ozols Johanna David Susan Vasily Wendy Beck Rion Vanek
Menaker Tellman Woodruff Luzadden Reotov
Jeff LaPlant Vladimir Ruth Laura Mark Edens Carl Byers Geoffry Louis Barr
Polushkin Longhenry Sheldon Peterson
Sergei Reutov Nikit Reutov Bob Zenaida Susan Payne Benjamin Jeremy C. Charles
Marquardt Latonia Golodoff Hansen Kramer
Karen DiBari Walter Sonen Ben John Hamik John Robert Mead Roberta Rick Ladd
Ballenger Luzadden Highland
Lavender Locke Finley Leonard Katya Kirsch Sandra Leo Linda Richard
Beattie Kelley Hudson Robichaud Sargent Powell
Suzi Golodoff Tuan Tran Trifilyi Konstantin William Phillip Sandra Beth
Reutov Basargin Barnes Wyman Stepanoff Pattinson
James Sandin Sherri Mike Hough Bruce Paul Dan Clare Larry Dooley Hal Long
Pristash Hendrickson lattimore
Sanford Crane Kevin Lowell Cecil Ranney Abi Eleanor Kathie Gregory Jeff Yates
Woodbridge Werbe Veltre Hoffman
John Bahe Peter Kojin Amanda Bill Choate Ralph Ash Mark A. Hida Kudrin Klaus
Appell Lanchuk Czerweny
Gary Jackson Barbara Alexus Grant Miller Nikolai Thomas Ely Thor Scot R.
Thompson Kwachka Kuzmin Skulstad Cummings
Mark Galdu Eric Brese Jeff Cline David Hulien Pete KJ Herman Marga Raskin Nikita
Hendrickson Kuzmin
Henry Cooper Shawnee Daren M. Nina Faust Bob Johnson Alex Jeanne See Kurt
Kinney Traxinger Martushoff Marquardt
U.R. Johnson Pete David S. Dolph Robin Engel Charles Mary Lentfer James Wild
Cummiskey Goldstein Rehder
Janis Harsila Mike Annette Daniel Betsy Goll Robert Allen Greg LoForte Peter Jack
Martushoff Stepetin Gleason Wagner
Sam Powell Kim Menster Mike Clark Ian Pitzman Henry M. Noel Johnson Michelle D. Floyd
Wieve Frank Tomkins
Scott Nora M. Dorothy Brenda Nicholas Dan Oliver Ray Tremblay Jimmy
McCracken DeSylva Childers Greenbank Dirks Charliaga
Ashley Eads Bruce Short James Sung Kim Johnny Lind Douglas Reutov Mike Consolasion
Skonberg Deplazes Ramirez
Gary Phil Gordon Albert Sergey Lisa Janelle Ruperto D. Patrick
Marincovich Fredericks Serebrekoff Martushoff Everett Gonzales Dineen
Brian Scully Henry Timm Janel Jason Tony M. Joseph Kelso Rosanne Craig Brault
Keplinger Suralta Stouers Curran
Stephen Rebecca Edward R. Joseph Dora Suetzed Constance Kevin Paul
Howell Bartee Hansen Emerson Jensen Stalker Castellani
Laura Lucas Dan Kondak Zahary Paul Jim Stewart Ivan Chuck Vendt Vasily
Martushev Saunders Usoltsoff Reutov
Vikki George Reutov Marvin Yagie Robert Chris R. Brad Sapp Artemon Nancy Jones
Garroutte- Schell Branch Basargin
Simpson
Bruce David Coleman Bert Barry Nicole George Rika Mouw Ivan
Flanigan Stammerjoha Fredericks Caputo Krukoff Martushev
n
Ray Blanco George Eric David Mazen Danny Donald Penny Ripple Robert King
Owletuck Bergloud Clarion Hanson
Richard J. Tom Dooley Oliver Owen Bill Rotecki Theresa Kevin Kiel Freddie Foma Reutov
Sharpe Peterson Cnossen
Dave Beebe Robert Jack Lentfer Ellen Mahle Rena Stuart Mach Hazel Jones Bobbie Jo
Pletnikoff Peterson Skibo
Nicholas A. Michelle Dennis Young William Patricia Joni Ken Zafren William H.
Troxell Church ``Webber, Olsen Whitmore Barker
Sr.''
Akaky Yaroslav Anita R. Danny Arseny Pelalgia Bill Prout Kenneth
Martushev Eglevsky Carpenter Carpenter Kuzmin Reutov Duckett
Marina Alice Schell Theresa M. C. Cleary John LeClair Feoktista Isa Wirz Andrea
Cabangbang Gurske Kuzmin Hernandez
Peter Stanley Nestor Anissa Berry- Ellen Flint Eileen Kent Joseph
Lekanoff Jhemas Ivanov Frick Mullen Herschleb O'Gorman
David Donald Lazar Reutov Jim Smith Mary Kay Maureen Cora Holmes Kandi Powell
Sperbeck Lawhead Cichoski Knutsen
Kevin Thomet Michael Felicia Alan Kelly Peter Konev Deborah Scott Kerr Edward
Grunert McAuley Williams Shaishnikof
f
Kondra Kuzmin Dean Harker Linnea Penni M. Douglas Annmarie Jessie Leo Soucy
Osborne Hees Anderson Squires Wilson
Robert Tim Marsh Grant Trask June Sargent Chris Burns Steve Webb Sandy Gary Squires
Russell Barclay
Teri Poulos Philip Jun Blanco Susan Mary Monroe Al Weathers Teresa C. Karen
Freitag Byersclorte Bishop Williams
r
Mark Stu Pechek Julia Forte Natividad S. Greg Johnson Jason Geck John DiBari Charlie
Tarabochia Canete Parsons
John Anthony Denny Kay Richard Winslow Dan Farrer Martin Teresita Nikolai
Weathers Schultz Hoffman Robards Layamana Reutov
Bill Sullivan James Canitz Thomas Finn David Otte Eric Stacy Girard William A.
Weathers Arbelovsky Lucore Bunter
Bobby Storrs John Chapple Lynn Collier Bob Wood Tim Longrich Sarah Blue Angela Stefan
McBride Horwath
Sandra Nelson Eva Holm Rob Dierich Annie Justin Fager Charles Stuart Kiril
Poirier Edwards Schaaf Basargin
Mary Mark Kozak Gordon Blue Mako Jeff Coleman Pete Hannah Brad Stevens Emil
Rosenthal Haggerty Berikoff
Sr.
James Johnson Matt W. Clark Sue Libenson Wendy Close Eldor Lee Logan Hull Sherry Hoyt Alexey
Eskew Besargin
Thomas Elino Culler Lee Martin Raymond Joshua H. Steve Marty Remund Jerry Bongen
Solenberger Hinkl Conn Schoonmaker
Kathy DJ Vinberg Thea Thomas Doug Hogen Ginna Willy Dunne Dan Gritman Rebecca
Kurtenbach Purrington Poulson
Michael Robert T. Peg Tileston Claire Spence Lewis W. Bob Bacolas Rodolfo Baru
Bentley Gustafson Laukitis Ellsworth Myers
Jennifer Chris Myrick Margaret Eric Eppie Kenneth Steve Haney Garland
Castellani Runser Almocera Seczesny Simpson Blanchard
Jacob Slaven Iosif Mike McCune William Steven Horn David Harris Shila Hough Tom Lohman
Martishev Paulo
Sandra Marker William Eros Kuzmin Ralph Charles Song Perk Michael Morgan
Dushkin Guthrie McCallum Yourkowski McBride
Oliver Felton Christine Charles Borghy R. Sam Eads Jody Spivey Erica C. Greg Forte
Hoen Harvey Holm Thompson
Layne Wilde Ben Enticknap Linda James Ralph Lohse Lori Lee Vern Culp Robert
McCullough McCullough Dentler
Bert Wheeler Kenneth Dewey John Munns Peter Steven Ilia Kuzmin Caroll Hamik Sarah
Thompson Campbell Gregson
Valerie Ellis Fran Latham Lacey Berns Karoly Geneva Gene Bill Rome R. Martens
Gaspar Macinko Anderson
Lisa Reutov Nancy Berland David Brad Sapp Richard Paul Bowen Tracey Michael
Whitmire Berns Wooden McConnell
Sonny Nguyen Suzanne Ian Russell Marietta Martin Kari Johnson Joe Macinko Walter
Abraham Coons Budnick Tellman
David Little Geredine Taan Alex Ladutko Chris Lynch Meri Holden Michael Clinton Tonya M.
Poole Mullins Weth
Wade Buscher Terry Johnson James Barton Wendy Stanley Nicholas Jennifer A. Leonard Efta
Johnson Houghton Dowie Creelman
Joe Newhous Wesley J. Steve Fish Jeff Jensen Glen Carroll Shirley Bob Nehus Harvey
Humbyed Snyder Goodell
Jolene Holl Jim Tolson Sharon Brown Paul Clay Frick Carl Burton Lenon J. David
Robancho II Norman Stanton
John Faris Edgar Callen Ron Hepler Royce Curry Leigh Thomet Nick Mazin Gerald Sharon
Brookman Shears
Bill Gurske Greg Brown John Lee III Iris Kessler Annie Wilson Paul Shadura Toras Fisk Karl Ohls
II
Susan John Ludvick Christopher Amy Lewis Andrew Jeffrey Don Lane Gordon
Houlihan Richards Jacobsen Yeoman Wilson
Jr.
Laura Schmidt Joni Lenz Jody Adams James Brown John Robin Faulk Annette Oliver Holm
Reisdorf Blankenship
Severean Jennifer Shirley Liss Ernesto Joshua Shawn Terry Stacy
Reutov Butler Casano Williams Knudeson Cummings Studebaker
John A'Piak Nick Dull Barbara Diana William Greg Wilson Betsy Trina
Marsh DeFazio Terry Pitzman Smallwood
Beaver Nelson James Sprague Brenda Christina Milton Frank Kavik Dan Falvey
Tellman Henslee Holmes Gwartney Anderson
Ken Jones Mike Brooks Luckida Ken G. Linda Bill Murray Summer Mack Philip
Boatwright Graham Behnken Lindley
Fadey Kuzmin Carl Holman Chris Awythe Steve Lawrence William Chris Stiehm Cliff Ward
Kalugin Trani Fiorentino
Roy Smith Alexander Andrew Harry Curran David Reutov Eric Meyer Jim Webber Nathan S.
Basargin Edgerly Dall
Dana MacAuley Cid Blase Roberta Per Nolan Ray Kranich Donald C.O. Edgar Smith
Copeland Roberts GigDecker
Dave Florence Paulino Wayne Clare Sergei Marjorie Kris
Schuckman Collins Jackson Stockert Litvin Hermans Anderson
Sister Diane Richard Ken Roosa Wendy Beck Leonard Don Thomas Jacob
Bardol Musgrove Jones Hernandez Wischer Stepetin
Derek Rusan Salvador S. Patricia Nick Budnick Herta Philip Oman Dean Cramer Lilia Howard
Taboy Phillips Tschersich
Sherry Ibarra L. Rodrigo Lee Neel Walter Mim John Ketelle Lori Sword
Hermeling Francisco Cadillo Sargent McConnell
Harold Martin Tarri Thurman Jeff Carr Vincent Ernest Norman Blank Rachel Simian
Shavender Shaishnikof Alinsunurin Basargin
f
William Erik Fellows Robert John Lyle Richard James Constance Chris
Ourada Arnold Nelson Niemela Griffith Whittington-
Evans
Mike Litzow Doug Sackett Debra Stanley Ivan Konev David Graciano Kathy Kuletz
Corbett Wolrich Hermeling Suralta
Boris Howard Nikola William Jason Ginter Susan McLane James Mary Timm
Galaktionoff Peterson Basargin Auger Wheeler
Alan Parks Astafey Kurt Michelle Cesario Evtropii Charlie Linda
Afonin Goetzinger Powell Suralta Matveev Schollenes Ellsworth
Julia Walker Shellie Clay Durham Phyllis James Clauss Taylor Jones Richard Elizabeth
Mathews Perry Kaminsky Johnson
Mimi Tolva Stacey Clark George Michelle Keegan Eric Myers Coral Gerald
Overpeck Ridgway Keplinger Chernolt Sanger
Nancy Lord Blair Martens Larry Steve Brown John Mouw Thia Falcone Matthew Jenny Norris
Shaishnikof Wiley
f
Pasha Isadore Russell Bienvenido Alan Ryden Scott Seaton Marie Finn David Pray Leroy
Betterton Ricafrente Cossette
David Le Cathy Gleason Joyce Levine Joseph Bob Hooyman Marc Meindl Tom Walters Susana
Reutov Macadaan
Kent Barkhau David K. Art Schultz Bill Lindow Thomas Crane Colleen George Alfred
Edens Newman Smallwood Peeler
Allan Hayes Lena Lawhead Edesy Victoria David P. Ivan Reutov Elizabeth Kate Finn
Fefelov Ellis Aers West
Frederick Nikita Shelley Darlene Port Kristin Neil Sargent Kathy Nugent Lauralee
Deveau Basargin Suydam Smith Matlock
Andy Endresen Vasily Kuzmin Lawrence Scott Seaton Roberto Paula Terrel Randy Baker Claire
Prokopeufh Mendoza LeClair
Jason Remedios A. Becky Horton Ruben Coral Christine Wendy Willis Marie
Gsertsien Small Topacio Ricketts- Holland Minkoff
Fotter
Juliana Michael Charles Chris Skelly Laura Viac Ben Marsh Stephen Rosa Valarie
Martin Bender Wilber Donaldson
James Nestic Colleen Jason Mack James M. Angie Perry Deborah Troy Alvin
Helligso Bailie Goodell Zimmerman Bereskin
Mike O'Meara Greg J. Bateman Michael John Renner Tom Miller Peggy J. Michelle
Streveler Boskofsky Smith Squires
Dave Beam Yakov Reutov K.E. Jr. Kip Thomet Jeremy Bob Childers Robin Kevin
Berggren Millen Stevens McDougall
James Gregg Jones Robert Mee Joshua Lowes Ronald John Gregory Cyn Muhs Terese Jones
Forsberg Tennison
Todd Hoppe Jill Lynn Johnson Mike Mahoney Frederick Bob Charles Frank Coma
Wittenbrader Reynolds Shavelson Mclinn
______
Fishermen's Finest, Inc.
Seattle, WA, December 13, 2002
Hon. John McCain
Washington, DC.
Re: Bering Sea Crab Rationalization
Dear Senator McCain:
With reference to my previous letters faxed and mailed to you
earlier this year (copies enclosed for your reference), I am writing to
follow up with additional signed petitions opposing the ``two pie''
crab rationalization plan proposed by the North Pacific Fishery
Management Council (NPFM Council) that would grant ``processor shares''
in the Bering Sea crab fishery.
To date, we have collected in excess of 700 signatures opposing
this plan. Many prominent fishermen as well other concerned Americans
have signed my petition, for example:
John Boggs of Lynnwood, WA Richard Hastings of Langley, WA
Stanley Hovik of Edmonds, WA Karl Hansen of Seattle, WA
David Little of Sammamish, WA Einar Longesater of Shoreline, WA
Frank O'Hara, Sr. of Thomas Parks of Edmonds, WA
Rockland, ME
Alf Sorvik of Edmonds, WA Konrad Uri of Seattle, WA
I understand from response to my previous letters that many feel I
oppose rationalization of the crab fishery altogether. This is not the
case. The fishery can be adequately and fairly rationalized at the
independent fishermen's level; it is not necessary to offer processor
shares as doing this will equate to a virtual monopoly for the
processing sector. In fact, the very antitrust laws of our Country that
were put in place to protect our free economy would need to be re-
written to allow this type of rationalization of the Bering Sea crab
fishery.
In addition to my opposition and that of many other concerned
Americans, fishermen and otherwise, I understand the following Alaskan
Communities oppose this plan as well: Akutan, Anchor Point, Chignik,
Chignik Lagoon, Clam Gulch, Cold Bay, Cooper Landing, False Pass,
Homer, Hope, Kachemak Bay, Kasilof, Kenai, King Cove, Kodiak, Moose
Pass, Nanwalek, Nelson Lagoon, Nikiski, Ninilchik, Port Graham,
Portage, Seldovia, Seward, Soldotna, Sterling, Tyonek, and Whittier.
Please do not allow the NPFM Council to set a precedent for other
fisheries in our Country by allowing their ``two pie'' crab
rationalization plan, which would grant ``processor shares'' in the
Bering Sea crab fishery, to move forward either as an amendment to an
Appropriations Bill or as part of the Magnuson-Stevens Act
Reauthorization. This type of rationalization of our public natural
resources has been rejected by all other Fishery Management Councils
across the country. If it Is wrong for the rest of America, it is wrong
for Alaska as well--we, the Independent fishermen, concerned citizens,
and Alaskan Communities need your help to stop this plan.
I would welcome the opportunity to talk with you and discuss this
important issue in more detail should you need any additional
information. Thank you for your consideration.
Sincerely,
Rudy A. Petersen
Enclosures: My previous letters dated 8/26/2002 and 9/6/2002
Signed Petitions against ``Processor Shares/Two pie''
Seattle Times Editorial ``Crab Industry Bakes a Monopoly Pie''
______
Fishermen's Finest, Inc.
Seattle, WA, March 6, 2002
Hon. John McCain, Chairman,
U.S. Senate Committee on Commerce, Science, Transportation,
Washington, DC.
Re: Bering Sea Crab Rationalization
Dear Senator McCain:
With reference to my previous letters faxed and mailed to you last
year (copies enclosed via First Class Mail, for your reference), I am
writing to urge you to oppose the North Pacific Fishery Management
Council's (``NPFMC'') proposed ``two pie'' crab rationalization plan
that would grant ``processor shares'' in the Bering Sea crab fishery.
Please do not allow Congress to allow wording in the Magnuson-
Stevens Act Reauthorization, or in any other Bill, appropriations or
otherwise, that allocates crab rights to processors. The fishery should
be rationalized at the harvester level, as all other fisheries have
been rationalized in our Country.
As this is a very important issue that will set a dangerous
precedent for the entire Country, should the NPFMC plan be adopted, I
will continue to oppose the ``two pie'' crab rationalization plan for
the following reasons:
Violates Anti-trust.
Grants the Processing Sector an unfair advantage.
Contrary to all other Fishery Management Councils' actions.
Places U.S. natural resources in foreign owned companies
control.
Many Alaskan Communities oppose plan.
Over 700 concerned citizens have signed my petition opposing
plan.
Sets a dangerous precedent that may be applied in other
fisheries.
The Bering Sea crab fishery should be rationalized at the harvester
level only because:
Consistent with all other Fishery Management Councils'
actions.
Provides some protection for the economically depressed crab
fishermen.
Fishing vessels must meet U.S. ownership requirements under
MARAD.
Please contact me should you need any additional information
regarding this very important issue. Thank you for your consideration.
Sincerely,
Rudy A. Petersen
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
______
Prepared Statement of Ron Philemonoff, Chairman and CEO,
Tanadgusix (TDX) Corporation of St. Paul Island Alaska
My name is Ron P. Philemonoff. I am Chairman and Chief Executive
Officer of Tanadgusix Corporation, the ANSCA Village Corporation for
the Aleut Community of St. Paul Island. St. Paul Island is home to the
largest Aleut community in the world, and a focal area for both crab
catching and crab processing. We speak in opposition to the plan before
your Senate Committee, as it is written and presented.
Our corporation owns 90 percent of the land on St. Paul Island as a
result of ANCSA (Alaska Native Claims Settlement Act) entitlements. As
the former hunters and processors of the Northern Fur Seal, the Aleut
people have inhabited the Pribilof Islands since being forcibly placed
on these remote islands prior to the purchase of Alaska by the United
States from the Russian Emperor. The Aleuts of the Pribilof Islands
harvested and processed fur seals thru the Russian occupation, and
under United States oversight until 1984. Our people were wards (some
say slaves) of the United States government until fur seal harvesting
and the Fur Seal Treaty of 1911 was abolished in 1984 due to national
and international environmental politics.
At the time of the cessation of the harvest, St. Paul Aleuts were
specifically promised that we would enjoy the fruits of freedom and
free enterprise, through the enjoyment of an economy based on the
fisheries resources of the Bering Sea. By the providence of nature,
more than 75 percent of the fisheries resources of the Bering Sea are
located within 200 miles of the Pribilof Islands. With such abundance
within close proximity, we were optimistic that an equitable share of
those resources would be made available to us to meet the economic
needs of the Aleut people of the Pribilof Islands.
In the mid-1980s, TDX began writing to NPFMC and to the government,
notifying them that the realization of Aleut dreams and the government
promise of economic development would be severely compromised by the
heavy capitalization of the fisheries in the Bering Sea, which was
proceeding rapidly, supported by government loans and NPFMC policies.
We duly registered our serious concerns that our people would be left
out of participation, by not having the historical participation so
commonly used under fishery management regimes, to qualify corporations
and large vessel owners for entitlement to resource rights. Our
concerns were largely ignored. By 1989, overcapitalization had created
a crisis for government loan portfolios, Seattle corporations and
foreign corporations, such that threats to the entire resource were
perceived. The inshore/offshore resource split was invoked in 1992, as
a means to ``rationalize'' the groundfish fisheries, because the larger
companies were fighting over respective share and threatening the
entire resource. In less than a decade, the ``rationalization'', called
inshore/offshore, was converted to a full scale property entitlement to
the pollock fishery resource, now enjoyed by the same beneficiaries who
were first ``rationalized''.
Crab Processing North of 56 Degrees. In 1987, while all the furor
over groundfish was starting to occur, there were no crab facilities or
processing on the Pribilof Islands, nothing North of 56 degrees with
exception of floating processors that followed the crab fleet. Harbors
promised to us by the government in 1984 had become bogged down in
engineering and the politics of contracting and government. Our
corporation perceived back then that Aleuts would not get a chance in
any fishery if we did not accelerate our participation as a community.
TDX thereupon embarked on a program to convert port proximate
properties on St. Paul Island to participate in the crab fisheries. The
heyday of Bering Sea King crab fishing had occurred in the 1970s and
early 1980s, and our people had watched helplessly from the shore as
hundreds of vessels, foreign and domestic, lighted our islands from
offshore, using our island as a lee in the Bering Sea wind, working and
exploiting the crab resource off our shores. We were determined by the
late 1980s to mount participation in the crab fisheries, for the
economic good of the Aleuts.
With our own entitlement and settlement funds, and with the help of
a willing Japanese entrepreneur, TDX cleared and refitted an abandoned
fur seal processing facility, converting it to crab processing
facilities. While the government sponsored breakwater and harbor
facilities on St. Paul Island were barely functional, TDX Corporation
spent millions of dollars to build a 300 foot ocean dock and to dredge
immediate proximate tidelands areas so that we could invite into St.
Paul, the Bering Sea crab fleet and process their product. By 1992,
about the time the inshore/offshore debate had reached its height, we
had succeeded in cornering about 25 percent of the Bering Sea crab
resource, purchased and processed right on St. Paul Island. We planned
and developed a new shoreside processing facility that is now the
largest crab processing facility in the world. Use of that facility was
ultimately purchased from our partner in 1993 by Trident Seafoods, a
member of the pollock and salmon clubs, and the major processor of crab
in the Bering Sea.
While TDX and Aleuts may have been inexperienced in fisheries and
the politics of NPFMC in the 1980s, we suffer from no current illusions
regarding what ``rationalization'' is all about, and what it means:
resource entitlement and property rights. Our opposition to this plan
is based on the fact that our corporation and our shareholders, as
pioneer investors and venture capitalists of processing North of 56
degrees, to the tune of over $15.0 million, have been left out of this
rationalization plan, which has, as a key component and foundation of
its rationale for processor quota, a rewarding of investments made to
support the crab industry, with quota property rights. If investments
in the industry are going to be considered as a rationale for
distributing crab processing rights into private ownership, all in the
name of ``rationalization'', then we respectfully request that our
investment also be rationally considered as a basis for Processor
property right PQS for TDX. We own the dock, the plant, the land and
the tidelands leases upon which the processing infrastructure for both
Trident and Icicle sits. We view such a settlement that does not
include TDX, as both unfair and inequitable, and a severe injustice to
the Aleut people.
We assert that any assessment of NPFMC and the State of Alaska that
Aleut needs are served by Community Development Quota participation in
this plan, is a flawed and technically deficient analysis, that
disenfranchises both TDX Corporation and its shareholders, as major
equity investors in crab processing North of 56 degrees, from rightful
participation in full and equitable sharing of this proposed resource
distribution. We cannot support, and must vigorously object to this
plan as a matter of simple fairness and justice.
We should clarify these statements. TDX Corporation does not blame
participant processors for seeking a share of Bering Sea resources. We
do not claim that we are solely deserving of resource entitlement, any
more or less than any deserving vessel owner who annually risked life
and limb to participate in these dangerous crab fisheries. Trident
Corporation and Icicle Seafoods, both current St. Paul Island tenants
of TDX Corporation, are responsible and reliable tenants, as well as
owners of catcher vessels. But their rights in this plan to crab
processor quota rights, north of 56 degrees, derives from use of TDX
property, TDX dredged tidelands, and TDX facilities on St. Paul Island,
which were developed with corporate resources of Tanadgusix Aleut
shareholders, unacknowledged, unrewarded and inequitably excluded from
participation in this plan of distribution.
There are so-called ``community protection clauses'' in this plan
which will be lauded as preserving and meeting Aleut interests on St.
Paul Island. In fact, they are last minute adaptations to the growing
awareness that community and regional protections in the plan, are
filled with loopholes: loopholes that exempt bairdi crab from the
program, that allow PQS to be leased to floaters, that allow PQS to be
used within the same company outside of so-called regions, sold to
other companies, and sold for custom processing. These are ambiguous
and smoky illusions which will assure that community protections will
be fish planned out of existence, in the name of ``fisheries
efficiency''. We repeat our contention that, if allocations are going
to be made based on investments in our St. Paul Island community, those
allocations should go to those who made the investments, not in the
form of a nebulous first right of refusal to a CDQ group that has not
invested the first dime in crab processing on St. Paul Island, and that
is consumed with political posturing and bureaucratic dithering. Aleut
survival is much too important to base our community participation on
such a wobbly foundation.
Our request for equity PQS recognizing the TDX investment has been
made at the last two NPFMC meetings, and has been totally ignored.
Specifically, TDX requested before the NPFMC that its original
investment in the Anderson Plant be recognized by an 8 percent
allocation of the Bering Sea crab PQS, all species, recognizing the
investments and levels of processing achieved by TDX St. Paul prior to
there ever being either Trident or Icicle or Unisea in the community,
or processing in St. George for that matter. TDX went on to say that if
there is to be no allocation for its investment, then the PQS assigned
to and derived from processing at our plants and facilities on our
lands and leased tidelands, needs to be tied permanently and
inseparably to those facilities.
Honestly speaking, the crab ``rationalization'' plan is a self-
serving resource entitlement plan concocted under a democratic veil at
NPFMC, by an already well vested group of influential large fishery
resource owning companies, whose own interests already include a
substantial portion of the crab catching industry, and the Bering Sea
pollock resource. These companies held the independent crab fishermen
and crab vessel owners hostage to the threat, that any rationalization
or IFQ program in crab was dead on arrival, due to their substantial
political influence, unless processor quota share was included. The
losers here are small businessmen, small processors, individual vessel
owners, smaller vessel participants, Alaskans, and Bering Sea Aleuts.
This program should not be confused with or compared to the Halibut/
Sablefish IFQ program, as a rational panacea to ``too many vessels
chasing too few fish''. We repeat again our request that NPFMC
reexamine the history of crab processing development and participation
north of 56 degrees, and implement more equitable distributions if such
distributions are to be invoked in the name of rationality.
We speak in frustration. This plan is an unfair vesting of
resources to which Aleuts residents of the Bering Sea, if anyone,
should be given priority access. Unless and until direct and
unambiguous protections are provided for communities north of 56, free
of loopholes, and unless and until, the investments of Alaskans into
the industry are considered on an equal par with the investments of
large non-resident corporations, we must strongly request that your
withdraw any support of the crab rationalization in its current form.
We cannot support processor quotas being distributed when the chance
exists, multiple avenues, that those quotas will leave the island, for
the sake of processor convenience or fishing efficiency.
TDX has earned the rights to participation with our investments. We
request that Congress look into these questions, and request the
National Marine Fisheries Service or the NPFMC to document and support
our historical participation and investment in crab fisheries north of
56 degrees. If our equity cannot be recognized, then we ask you to send
the NPFMC back to the drawing board for development of a crab
distribution plan that doesn't disenfranchise legitimate ground
investments, and market freedom for long time fishermen, by
distributing property rights to the few processors still standing in
the game. Rationalizing the fishery has never been explained as taking
all risk out of the fishing business, in the name of supporting a few
remaining processors.
Tanadgusix Corporation urges the respected Senators of the Commerce
Committee to dump this badly flawed plan that disenfranchises and gives
no credence to, the true and verifiable investment and equity of the
Aleut people of St. Paul Island in the crab industry. Or to make sure
that it does so, before authorizing another distribution of public
resource to large industry, without considering the needs of those who
live in the middle of the resource.
Respectfully Submitted,
Ron P. Philemonoff,
Chairman and CEO,
Tanadgusix Corporation.
______
Prepared Statement of Byron L. Pierce
Dear Senators,
My name is Byron L. Pierce. I am a 53 year old Ex-Marine and
Vietnam Vet. Both my older brothers were Marines. My middle brother
lost his life in Vietnam. After my tour of duty I returned to my home
in Oregon and attended college for a few years. I felt my life wasn't
quite right and felt I needed more so I headed north and ended up on
the most beautiful Island, Kodiak Island in Alaska.
It was shortly after arriving here I had a chance to go King Crab
fishing in Kodiak. I immediately fell in love with crab fishing and
decided then and there that this is what I wanted to do with my life.
Every aspect of the crab fishing business intrigued me. It is man
against the sea and his ability to overcome tremendous odds while
working exceptionally hard and hoping to come home with a good payday,
and then doing it again and again. That was 30 years ago.
In my career, I have owned three crab boats. I started with a small
boat and over the years have upgraded to larger boats. I have lost many
close friends over the years and in 1993 my family and I lost 3 crewmen
when our crab boat Massacre Bay sank. We currently own 1 boat now, the
Nuka Island, and are a family owned business. I have literally given my
heart and soul to the crab industry in Alaska for 30 years.
I have watched the crab industry change drastically over the years
and now we are down to Rationalization, which I believe is needed!
Senators, the big issue is the whole concept of Processor Quotas
and the changing of the Anti-Trust laws to allow them. The concept of
Processor Quotas tells me I have to deliver to just certain processors,
and just take what price they choose to pay me for my crab. To me this
is crazy!! It's like telling a chicken farmer he can only sell to
certain markets. This is totally Un American to me!!
The only people Processor Quotas are good for are Processors. To
change the Anti-Trust laws on order to get this accomplished is just
plain wrong!
Senator, please don't tell me that this is what I went to war for
or that this is what my brother lost his life for!
In the early 90s, an Individual Quota Program (IFQs) was put in
place for Alaskan halibut and sable fish fisheries that has proven to
work well. I feel strongly that this same program would work perfectly
for our crab fishery to the men that earned them. The fishermen, Not
the processors.
Again, please vote this down and vote Not to change the Anti-Trust
Laws.
Sincerely,
Byron L. Pierce,
Crab Fisherman,
Kodiak, Alaska.
______
Prepared Statement of Richard Powell
Mr. Chairman, my name is Richard Powell and I have been a resident
of Kodiak, Alaska for 39 years and have been involved in the Alaskan
king crab fisheries for that entire time.
I would like to thank you for the opportunity to submit written
comments before the Senate Commerce Committee regarding the issue of
rationalization of the Bering Sea/Aleutian Islands crab fisheries. My
specific comments will be in relation to the proposed element of
individual processing quota (processing shares).
I would like to provide a bit of history on how I perceive this
issue has evolved over the last several years. The crab industry has
been through a difficult time in the last several years, with the
severe downturn of the red king crab and opilio crab stocks of the
Bering Sea. The fishing industry has been working for years to attempt
to develop an industry-funded buyback program. This, coupled with a
license limitation program designed to only allow recent participants
to remain in the fishery, can provide some short-term relief to the
fleet.
However, it is apparent that this fishery needs a quota based
program, to allow individual harvesters the ability to determine when
their fishery will occur. As you are aware, the job of fishing crab in
the Bering Sea is the most dangerous occupation in the United States.
In recognizing this, crab harvesters have worked for over four years in
attempting to create a quota based program, which will lead to a much
safer fishery, as shown by the results of the halibut IFQ program in
Alaska.
The processing sector in Alaska is a combination of U.S. and
foreign owned companies. These companies have a strong lobby and work
closely together to achieve their goals. When the issue of crab
rationalization began to be discussed amongst the harvesters, the
processors formed a coalition and began working as a unit to achieve
their goal. My view of this goal is that the processors want to make
sure to control the fleet as much as possible, in regard to time and
location for delivery, as well as the pricing structure. This benefits
companies who are diversified in their processing products, and want to
create efficiencies for themselves.
In order to achieve this goal, they determined that the only
solution to their concerns would be processing shares. They needed to
begin with a limited entry system of their own to create a pool of
eligible processors. But this wouldn't be enough. They determined that
they didn't even want to compete between themselves for the product.
The answer was, of course, guaranteed deliveries based on their
previous processing history. The processors discussed this with each
other and determined that they would only agree to a minimum of 90
percent guaranteed deliveries. They further agreed that this was the
only solution and that no other would be acceptable. This created a
very uncomfortable situation for the negotiators of the crab
rationalizaton committee, which worked for over a year to try and
determine elements and options for the Council's analysis.
Other options were presented and repeatedly rejected. In fact, the
processing sector representatives went so far as to say that if they
didn't get their way, they would block the buyback program (which is
still in OMB limbo), as well as stop any rationalization program from
being implemented in the crab fishery. The processing lobby is powerful
and well-funded. The results are that we are in this dilemma.
The North Pacific Council, just one year ago, voted 11-0 to send
their required report to Congress, along with the preferred
alternative. The preferred alternative included the 90 percent
guaranteed delivery requirement to eligible processors, as they
intended. While the Council debated and voted many times over a period
of days on the harvester elements of the program, there was no single
vote on the policy of granting processor shares. It was simply included
as part of a many-page motion. It was not amended or discussed, but
simply approved in the final package. The Council made it clear that
the cornerstones to this program were the binding arbitration and
community protection portions of the plan. These elements were to be
followed as ``trailing amendments'', with their own committees and
analytical documents. This past month, the Council saw the final
documents and reports on the ``cornerstones'' of the crab plan. They
were barely able to get a majority to approve them. It is clear that
with closer scrutiny, the Council's preferred alternative in June 2002,
has serious problems.
The entire issue of processing shares was discussed briefly by the
National Academy of Sciences and the concept was rejected as
unnecessary. The Committee was not convinced that processors should be
granted processing shares, or even if they should be granted harvesting
shares for eligible vessels they might own.
I am frankly surprised that this issue has continued to receive
serious consideration. It proves to me that the processor lobby is as
strong as it was before Alaska became a state and they controlled the
seafood industry through the salmon traps. Their successful lobby
resulted in Alaska remaining a territory for years after it should have
become a state. The fishermen cannot hope to compete with the power and
funding of this group. We can only hope that our elected
representatives will continue to use good judgment and reject this
proposal. This is not fishery management, it is fishermen management.
As a crab vessel owner who has a catcher vessel and a catcher/
processing vessel, I have in years past delivered some of my crab from
the catcher vessel to processors and some of my crab, I have processed
myself with my catcher/processor. Under the processing share proposal,
I will not have the option in the future to process my own crab from my
catcher vessel. Now, the last time I checked, this was still America
and I had some freedoms to buy or sell as I choose. I shouldn't be
forced to sell my crab to a specific, eligible processor, just like
they shouldn't be forced to sell the processed product to a specific
store or wholesaler. But, this is what is being proposed. It just
doesn't make sense.
I have every confidence that Congress will carefully evaluate this
issue on the merits, not the political push and pull that a well-funded
lobby can accomplish. There is a reason that we have anti-trust laws,
let's not overturn them on a whim.
It is vital that the Bering Sea/Aleutian Islands crab fisheries be
rationalized and a quota based program be implemented. It is not right
that a few powerful companies can stop this from happening unless they
get their own guaranteed lock on the product.
Thank you for considering my comments.
______
Prepared Statement of Robert E. Pries
``Crab Rationalization'' is nothing less than a legalized cartel.
The fishermen have no choice but to sell to the cartel regardless of
how low the price or they do not sell at all. Even if a portion of the
catch is reserved for free market the results will be the same: the
free buyer could not compete with the cartel if the cost of his fish
was more than the cartel. If 20 percent of the catch was reserved for
the free market then only 80 percent of the fishery will be dominated
by small group of monopolies. What a deal.
I am not a crab boat owner in this particular fishery but I do
participate in crab and numerous other fisheries. We are shaking in our
boots because we see this plague engulfing the entire fishing business.
Processors everywhere are salivating at the prospect.
The fishermen are not getting a fair shake as it is and this would
only make it permanent and legally condemn us to the lowest rung of the
economic ladder. That is not fair for men who work some pretty mean
hours, live in some pretty Spartan conditions in the most dangerous
occupation in the country.
Ever since Exxon Valdez Oil Spill it has been bad news for nearly
all of the fishermen in Alaska, even those who were not directly
affected by the spill. That was a 9.0 disaster on the Richter Scale for
a lot of fishermen who still have not been compensated not to mention
those who never will which may include nearly all of us if it grinds
along for another 15 years.
We have more Pink Salmon than our processors seem to want but do
you think Alaska will allow foreign processor vessels into our
fisheries? Think again. It is all about protecting the soft money
contributors.
Senator McCain, I was a dedicated supporter of your bid for the
Presidency mostly because of your fight for campaign reform. I believe
that that this legalized bribery system of ours is the root of
corruption in this country. It sets a bad example for the leaders of
the country who have not already been corrupted and may be the best
example of the trickle down theory yet.
I cringe when I hear our President talk about spreading democracy.
Make no mistake we are governed by a Plutocracy. The ratio of CEO to
worker compensation increasing from 400-1 to 4,000-1 during the past 20
years should convince anyone that this country is more and more being
run by the wealthy for the benefit of the wealthy. Those CEO's make
those fantastic salaries even when their Companies are failing.
I served my country just like you Senator McCain. I was a Navy
fighter pilot too and I am not blind nor am I deaf. I hear Perot's
sucking sound. I see that the law is for the highest bidder. I see
antitrust as a bad joke. I see outright white collar criminals all
over. The politicians revolving door . . . and greed heads everywhere.
I think that we may be in bad trouble. I don't think it was always like
this. We are in dire need of truth, honesty, integrity all over this
country.
The only hope for the boat owners and crews is for a free market.
WE only ask for a fair price for our catch something we will never see
under this scheme which was contrived by a processor and his favorite
politician. Best I can tell this processor has become very wealthy
while his boat owners barely got by if they did and the crews were
asked to live on a pitiful wage.
Please look past the attempt to cloak this greed under a
respectable name like ``Rationalization''. The Processors want a shoe
in deal and my 25 years experience with them tells me that they will be
just as greedy as Enron. Each one of these processors want their own
monopoly and nothing less. Free market used to be the mantra of our
economy playing field this ``Rationalization'' couldn't be further from
a level playing field in a free market.
The proper name is CRAB ABOMINATION.
______
Prepared Statement of David Soma,
Deep Sea Fishermen's Union of the Pacific
The Deep Sea Fishermen's Union of the Pacific (DSFU), a labor
organization that has represented working fishermen for 90 years, is
grateful for the opportunity to provide a statement for the record of
this hearing on the Bering Sea and Aleutian Islands (BSAI) Crab
Rationalization Plan (Plan). The long history of DSFU has been marked
by a clear vision for the longline halibut and sablefish fishery, a
progressive outlook for the industry, an emphasis on safety and the
immense professionalism of its members. DSFU is justifiably proud of
its unbroken service to working fishermen. The Union has amassed an
unparalleled string of successes in safety, fishing management,
conservation and equitable treatment for our members throughout our
history.
The almost 300 current members of the Union represent a growing
number of the longline schooner fleet crewmen and an even faster
growing number of crab fishermen from Washington and Alaska.
The Union was formed, when the crews were confronted by extremely
low prices for halibut, poor resource conditions, and fishing in very
bad weather and unsafe conditions that led to several vessels sinking
and the loss of several crewmen. The time was right to take some
action. They created the Union with goals that were straight-forward
and practical; fair wages, better living conditions, improved benefits,
and a closed shop.
As working partners in fishing operations, the members of the DSFU
took responsibility for the management and well being of their
livelihoods. The Union's principles continue to be: The proper long-
term management of resources for conservation and efficiency; the fair
and straightforward treatment of vessel, skipper and crew; professional
work standards; and the integrity of fishermen to stand together.
After a turbulent beginning with the Fishing Vessel Owner's
Association (FVOA), initial differences were settled and the two
organizations established a simple, elegant and precise contract, now
known as the Set Line Agreement. It became the basis for payment and
working conditions for all longline fishermen, Union or otherwise,
working from the mouth of the Columbia to the Bering Sea.
Relations between these two organizations have been candid and
collegial ever since. There were many good reasons for the Union and
FVOA to cooperate and collaborate. Many captains, boat owners and
crewmen realized that their goals were inextricably linked. It was
rapidly apparent that there were a sizeable number of key issues that
would need the full cooperation and lobbying power of both the Union
and management.
In 1962, the North Pacific Fisheries Commission ruled that the
Bering Sea halibut stocks were underutilized. They then became fair
game for the Japanese. American fishermen strongly objected, but to no
avail, and the decision negatively impacted every American fishermen
working in the North Pacific. From 1966-1974, the halibut quota dropped
precipitously. Longlining reached its lowest point in 1974 with just
over 21 million pounds caught for the entire year. Trips took 20-25
days--fish were old when delivered and ``hole'' trips (ones that made
no money) were common.
In the Senate, Senator Warren Magnuson from Washington State and
Senator Ted Stevens stepped up to the plate. Under their leadership,
the foundation was laid for the Americanization of the fisheries within
200 miles of our coasts.
However, a race ensued among American fishermen, as the domestic
fleets grew in size and efficiency. For halibut, openings grew shorter.
In 1980, area three was open for eight days and by 1983 it was five.
Crew fitness and endurance became increasingly important. The Union
grew with young men who were drawn by the reputation of the schooners
and the Union's professionalism. An old sideline, black cod,
increasingly became a mainstay of the longline fleet.
FVOA and DSFU worked together to stop the inequities involved in
fishing for black cod. All hook and line fishermen in the North Pacific
again benefited from the effort.
By 1991, fishing time for both black cod and halibut had shrunk to
a scant number of days. The Union entered into a dialogue with FVOA
about a plan centered on Individual Fishing Quotas (IFQs). After
lengthy discussion and lots of effort over a period of seven years,
IFQs were adopted for the halibut and sablefish (blackcod) fisheries.
The decision was not an easy one. The prime-mover for this process was
the North Pacific Fishery Management Council, which began considering
options in the late 1970s. Dangerous ``derby'' fisheries that caused
safety, care of gear and quality of product issues were resolved by
IFQ's.
The Set-Line Agreement was renegotiated between the Union and FVOA
in 2002, again demonstrating its utility and the ability to stand the
test of time.
DSFU continues to set the standard for longlining in the North
Pacific. The Union has unhesitatingly stepped up as a key part of the
most powerful lobbying body for hook and line operations in the North
Pacific. The recent addition of a growing number of crab crewmen and
skippers to its ranks is a testimonial to the worth, stability and
advocacy for working fishermen DSFU has always demonstrated. From
dories to schooners, from lay-up to IFQ--the Union has really been
about people--its incredible membership.
Background and Need
Owners of the crab boats operating in Alaska, most of which are
based in Seattle, have been struggling for a decade to stay solvent
despite plunging populations of most crab species. In September 2002
the National Marine Fisheries Service (NMFS) and the Alaska Department
of Fish and Game made the situation even more critical when they
announced that the allowable catch for snow crab, the most populous
species, was lowered for the 2003 season. This was a 15 percent
decrease from the 2002 catch. Without enabling legislation from
Congress, the fishery will continue to operate under the Olympic quota
system. The Olympic system forces boats to race for the fish, usually
resulting in uneconomical marathon fishing sessions that are damaging
to the crab stocks and dangerous to crews. This system is the main
reason crabbing is considered the most dangerous fishery in the United
States.
The proposed alternative, a quota allocation system, would give
catchers, processors, communities and skippers specific quotas of the
available catch. Under the system, quota owners will be able to catch
crab according to market and weather conditions and their own best
operating efficiencies as opposed to the dangerous and innefficient
``derby'' system.
Versions of the quota allocation system have been implemented by
the halibut, pollack and whiting fisheries, all with positive results.
The Pacific halibut fishery, once poorly controlled and managed, is now
considered well-managed, environmentally responsible and sustainable
for the forseeable future. The fishery currently fishes 20 percent of
the available biomass while providing good wages for the crewmen and a
profitable business for the owners. In addition, the fishery has seen a
marked decrease in accidents and deaths since rationalization. In a
recent Seattle Times article, both Pacific halibut and sablefish were
noted as well-managed and received Monterey Bay Aquarium's green, or
``Best-choice,'' rating. The DSFU believes these remarkable results are
directly related to the implementation of a quota allocation system.
It is interesting that, with the decline of the crab fishery and
the onset of a rationalization program initiative, crab skippers and
crewmen strongly petitioned DSFU for representation in 2002. DSFU has
stepped up, in adherence to its by-laws, to represent working
fishermen.
Position Statement
The Union, and its members feel strongly that crab rationalization
is the correct mechanism to revitalize the industry and encourage a
safe, healthy and economically strong crab fishery. However, the
members also have some very serious reservations about the plan as
currently designed. The inclusion of processor shares and more
particularly the 90:10 split is worrisome to our members. The fact that
the Council passed an amendment that significantly weakens the
arbitration safeguards for the harvesters is also of concern to our
members. In our opinion, the processor share portion of the plan comes
close to running afoul of the anti-trust laws and particularly the
restraint of trade provisions. The membership is also concerned about
the precedent it sets for the next fishery that rationalizes. Once
processor shares are legitimized, we see them becoming a part of all
rationalizing plans. When this is coupled with an almost negligible 3
percent share for skippers, it becomes even more difficult for the
Union to not be concerned about the long-term consequences of the plan.
Finally, the absence of crew shares has to be of extreme concern to
a Union representing working fishermen. A plan that, once again, gives
no recognition to those who perform the overwhelming majority of the
work resulting in the successful prosecution of any fishery can only be
considered short-sighted and inequitable by the Union representing
these working fishermen. We do recognize and appreciate the inclusion
of a loan program intended to give crewmwmbers the opportunity to
purchase quota. However, as can be seen in other fisheries, this is not
as easy as envisioned and we feel initial crew shares would be a more
equitable approach.
For these reasons, the DSFU believes it is in our members' best
interests to not support or oppose the current plan at this time. The
Union is heartened by the Council's letter of 6 May stating its
intentions to closely monitor the arbitration and community protection
sections of the plan and its intention to ask congress for assistance
if the plan does not work as anticipated. However, the DSFU believes we
should reserve our judgement until legislation is enacted and the plan
is allowed to work for a period of time. We believe that the
anticipated positive effects or feared negative results will emerge
over time and if the results are negative, we would respectfully like
to reserve our right to petition the Council and ultimately the
Legislature for relief and modifications as necessary.
______
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
______
Prepared Statement of Alf O. Sorvik, Owner, F/V Rainier,
F/V Paragon, F/V Isafjord
As a former member of the Alaska Crab Coalition (``ACC''), I am
submitting this written testimony to refute the position of the ACC, as
presented today by it's executive director, Arne Thompson. I, along
with many other independent crab fishermen, left the ACC when it became
apparent the group no longer represented the interests of the
independent fishermen and was simply a voice for the processor
controlled and/or CDQ controlled crab fishing vessels.
I am a fisherman and I have been active in the crab fishery in the
Bering Sea and Aleutian Islands since the 1960s. My vessels have
harvested Red, Brown, and Blue King Crab; Opilio and Bairdi Tanner Crab
for over 30 years. While I have fished for many of the processors that
stand to gain an unfair advantage by the North Pacific Fishery
Management Council's (``NPFMC's'') proposed rationalization plan which
endorses ``processor shares,'' I cannot support this plan.
If the Commerce Committee allows this plan to move forward, you
will be awarding the processing sector a monopoly. Further, this
proposed plan goes against the wishes of almost all the Alaskan
communities, violates our antitrust laws, sets a dangerous precedent
for other fisheries, and illustrates just how precarious our Council's
position in the management of our fisheries has become.
I ask you to send a clear message to all the Councils in our
Country that the Commerce Committee will not stand for proposed
rationalization plans that violate the law. The NPFMC needs to do what
the crab sector asked and agreed upon; provide a rationalization plan
in conjunction with a buyback program and let the crab sector solve
it's current economic problems without further diminishing the sector's
position by granting the processing sector an unfair advantage through
a ``two pie/processor share'' scheme.
In closing, this is bigger than just a crab rationalization
program, it sets the course for further rationalization plans across
the Country. Set the process on course and do not allow the NPFMC's
proposed Bering Sea Rationalization plan to move forward with
``processor shares.''
______
Prepared Statement of Capt. Gary Stewart, F/V Polar Lady
My name is Gary Stewart. I have fished crab in the Bering Sea since
1969. I have been a boat owner since 1978. I am president of the Alaska
Marketing Association, the collective bargaining association of Bering
Sea crab fishermen and the largest single group representing
harvesters. The AMA negotiates crab prices for the entire Bering Sea
crab fleet with the exception of the catcher-processor vessels.
The crab rationalization program seems to be slowly working its way
through the bureaucratic process. It is looking more to me like the
deck is being stacked against the crab fishermen of the Bering Sea. I
have watched and participated in this process. I think this once
promising program will reduce our fishing fleet to the status of serfs
and sharecroppers. I believe that getting our IFQs may not be worth the
price we will be paying. Do any of you think we will be better off with
processor quota shares? I know that processors will be. I know that CDQ
groups will be. Some fishing dependent communities may be while others,
like Kodiak, will be harmed. I have strong doubts the fishing fleet
will benefit.
The North Pacific Fisheries Management Council has sided with the
processors on every phase of the rationalization process. The two-pie
that the Alaska Crab Coalition sold me on was a program that gave
processors a quota share for 80 percent of the crab. I didn't like it
but thought it might be workable if a mechanism were in place to allow
us to negotiate a fair price. Without public discussion, the NPFMC
voted unanimously to give the processors 90 percent, with a binding
arbitration process written by a processor attorney. A committee was
formed to discuss arbitration models, and they had many meetings and
spent a lot of time trying to hammer out a program we could live with--
an ultimately is was all in vain as the NPFMC rubber-stamped the
processors plan, ignoring the AMA and all other harvester groups.
For the first time that I can remember all the crab fishing groups
agreed on something-the council's preferred binding arbitration model
did not adequately protect harvesters. Jake Jacobsen, the AMA manager,
went to the April council meetings and testified in favor of changes to
the plan in order to better protect harvesters. The council voted
against us 6-5.
I had hoped the processor quota share might get tossed out when the
Alaska Crab Coalition (a harvester political group of 10-15 boat
owners) pulled its support. That didn't last long.
I believe we are stuck with a binding arbitration program that will
not be fair for harvesters. It was written by processors for
processors.
The Unalaska City Council recently voted to endorse processor quota
shares. Privately, some. council members have said that they endorsed
the plan only because they did not want to upset the all-powerful
processors.
Questions I would like to have answers to:
1. Why did the NPFMC disregard public testimony on two-pie.
2. Why did the NPFMC go against the findings of their own analyst in
regard to binding arbitration?
3. If processors need the protection of two-pie then why are they
going to be able to buy 5 percent of the quota, each?
Fishermen only get to own 1 percent of the fishing quota
each!
4. Why was the AMA not invited to testify at the Senate committee
hearing?
By my own unofficial survey eight out of ten boat owners are
against processor quota shares, but many are afraid of retaliation from
processors so they think they must keep quiet. They understand where
the power will be in the future.
Processors have been telling us all along that if we want
rationalization we better get in line and not object to anything they
want. They must be telling the NPFMC the same thing.
If rationalization were to go away tomorrow who would Jose the
most? Would it be us fishermen? We would be back where we started. As a
father who raised my family in the crab fishery and whose son is now
fishing in the Bering Sea crab fisheries, I pray for a rationalized
fishery. Processor quota shares are a bad idea. Sure, they make a big
deal about the 10 percent harvesters will be able to deliver on the
open market, but in truth the 10 percent is really much less
(processors own many fishing boats)-and with processors holding boat
loans and tendering contracts over the heads of harvesters, there is
very little that will not be delivered to them.
Processors just about have the golden goose wrapped up. But would
they really ``pull the plug'' on rationalization if fishermen ask for a
system that is not so one-sided? With all that processors stand to
lose, I think not.
Gary Stewart,
F/V Polar Lady.
______
Prepared Statement of Bob Storrs, Vice President,
Unalaska Native Fisherman Association (UNFA)
Chairman McCain, Senator Stevens, and members of the Committee:
First, we would like to thank you for holding this hearing and
receiving testimony regarding the Bering Sea crab rationalization plan
put forward by the North Pacific Fishery Management Council (NPFMC),
and in particular--the portion of that plan that deals with the gifting
of exclusive processor quota shares to specific, largely foreign owned
corporations.
There are some who would tell you that processor quota is an item
that can rightfully be handled by the NPFMC. Nothing could be further
from the truth. This is a question of national policy that reaches far
beyond the domain of the Council.
The decision that you, as our elected representatives, make will
reach far beyond price negotiations for seafood--far beyond economics
and anti-trust issues, it is a matter of who will control not only our
resources, but also the very social and political structure of our
communities forever.
With that in mind, we only wish that you could have talked with
some of the many people who have privately apologized to us for not
taking a public stand on this issue because they fear for their markets
or their jobs. This is a sad fact of life in the company town scenario.
Hopefully it is not the road that you would send us down.
The Unalaska Native Fisherman Association, along with other
fishermen's groups, city and village councils, independent processors,
and ordinary citizens from across Alaska stands absolutely opposed to
processor quota shares, and we feel that it is unfortunate that some
with the State of Alaska and the NPFMC have accepted the arguments of
several large companies as justified through the theories of one
economist.
Dr. Matulich's views are based on data selectively released by the
processors, and neither these findings nor his methodology find broad
approval among other economists or by the independent review of the
General Accounting Office.
Some people have asked us how processor quota would directly affect
the members of UNFA. We offer a case in point:
Seven years ago, fishermen from Unalaska and some of the outlying
villages met to discuss the formation of a co-op. The University of
Alaska sent down two economists who helped us draw up by-laws and
formulate a business plan. A lease arrangement and was made for the
last small family-owned processing plant in the Aleutians, and
enthusiasm for the project ran high.
One week before we were to take possession of the facility, it
burned to the ground in our town's most spectacular blaze since the
Japanese attack of World War II.
Recently, people in our town have again begun to express an
interest in re-activating the co-op, perhaps by using a small floating
processor. One problem though--if processor quotas become a reality,
that will be impossible.
We have repeatedly been told that the 10 percent of the quota left
open to the free market would accommodate any future plans as well as
lend some capitalist flavor to this attempt at a planned economy. That
is ridiculous. As an economist who specializes in co-ops pointed out to
us--you cannot have a viable co-op when your members are required by
Federal law to deliver 90 percent of their product to the competition.
That is not the American way of doing business.
The fact of the matter is, if you allow processor quota to become a
reality, you will have made it functionally illegal for us to have a
viable co-operative. You will have completed the job that the fire
began.
After 100 years of American anti-trust legislation, is that really
where you want us to be?
It is important to remember that all of the goals of fleet
reduction, increased safety, community stability, as well as
conservation concerns could be achieved without granting cartel status
to a handful of companies. Conservation and safety issues are dealt
with on deck and in the wheelhouse--not in a far away corporate
boardroom.
In fact, all that has prevented the communities and the fishermen
from achieving a safer, more rational fishery has been the fact that
these corporations have held the public process hostage to their
demands for control of the resource. They called it their ``poison
pill'' and from the beginning, they made it clear that it had to be
their way or no way. They realize that they have tremendous political
power and that they are about to acquire much more.
This is particularly galling because huge questions continue to go
unanswered: How great is this so-called corporate distress? Who owns
how much of whom? Where are the profits--or the losses--being realized?
How great are the losses to the U.S. Treasury due to abusive transfer
pricing? What lessons, if any, are to be gleaned from the Bristol Bay
price fixing case?
The Unalaska Native Fisherman Association has, from the beginning,
insisted that for there to be any meaningful discussion of relief to
these companies, the books--from Seattle to Tokyo--must be opened.
Citing proprietary concerns, the companies have failed to do this.
They have released only select data through their own select economist.
We can see where that got us with Emon and Worldcom.
We find it ironic that any Alaskan fishermen who applies for a loan
must completely bare his or her books--including all IRS records for
years past, yet we stand on the verge of granting virtual ownership of
a once publicly-owned American resource to a handful of trans-national
companies--all purely on faith.
Even as we look beyond the forest of ledger sheets and corporate
relationships, it is very difficult for many of us who make our living
at sea to equate the financial distress of a crab fishing family to the
pains of Nippon Suissan Kaisha--who just over year ago purchased
Gorton's Seafoods and a similar Canadian company for 175 million
dollars, and in so doing virtually completed the vertical integration
of their operations in North America.
Make no mistake about it--we want these companies to prosper.
Perhaps there are things we can do to assist them through what they
claim are tough times; maybe with open books and open minds we can work
together to find some new ideas that will benefit all of us. Let one
thing be clear--granting the companies such unprecedented control over
our resources and our lives should not even be an option.
The very idea of facilitating the passage of the processor quota
abomination by initially applying it solely to our area is--for those
of us UNFA members who have served our country in combat--particularly
offensive.
Please remember, Alaskans deserve the same economic freedoms and
protections enjoyed by all other Americans. We are citizens, not
pariahs. That is why we urge you to reject the idea of processor quotas
and allow us all to move foreword in a public process unhindered by
further threats of corporate veto.
Again, we thank you for your attention to this matter.
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
______
Prepared Statement of Simeon Swetzof, Jr., Mayor,
City of Saint Paul, Alaska
Mr. Chairman and Members of the Subcommittee, thank you for the
opportunity to provide testimony on the proposed Bering Sea Aleutian
Island (BSAI) Crab Rationalization Program. When implemented, this will
be an effective tool for the proper management of the BSAI crab
resources and their sustainable commercial use by harvesters,
processors, and communities.
My name is Simeon Swetzof. I was born on St. George Island, but was
raised and have spent most of my life on St. Paul Island. My wife,
Phyllis, and I have raised our four children on St. Paul.
I have served my community as a member of the City Council since
1994 and I have been Mayor of the City of St. Paul since 1999. In
addition to my duties as Mayor, I am a commercial fisherman and
subsistence hunter. Finally, I have represented St. Paul at the North
Pacific Fishery Management Council (NPFMC) since 1998 and have been
intimately involved in the development of the crab rationalization
program being considered today and the effort to protect community
interests in this process.
I. Introduction
Following the federally-mandated transition from fur sealing in
1983 and the completion of the St. Paul Harbor in 1990, St. Paul
successfully developed a fisheries-based economy. From 1993 to 1999 St.
Paul was the primary crab processing location in the Bering Sea and the
second fishing port in Alaska, after Dutch Harbor/Unalaska, in terms of
state tax revenues. In 1998 and 1999, crab deliveries to St. Paul
exceeded 40 percent of the total harvest and the Harbor generated $6
million per year in national economic benefits from opilio Tanner crab
(also known as ``snow'' crab) alone, by providing an alternative to at-
sea processing or delivery to more distant locations. The strategic
value of the harbor, located in the middle of various Bering Sea
fisheries, speaks for itself.
The considerable investments made by the community, the State of
Alaska, and the Federal government to build a harbor and the
accompanying infrastructure necessary to service the Bering Sea fishing
fleet contributed to the generation of considerable wealth for the
community as well as for harvesters and shore-based processors.
Unfortunately, as a late entrant into the fishing industry, St. Paul
was in the process of diversifying its crab dependent economy when the
opilio stocks collapsed in 1999, taking with them 85 percent of St.
Paul's economy and affecting related service industries such as air
freight, refueling, supplies, and crab pot storage. The economic crisis
has been compounded by a severe decline starting in 2002 of the halibut
CDQ fishery, the second most significant activity on the island.
II. Collapse of the Bering Sea Crab Stocks and Impacts on St. Paul
a. Status of the Fishery:
The Bering Sea snow crab fishery, which was the subject of the year
2002, 2001 and 2000 commercial fishery failure determinations by the
Secretary of Commerce, has continued in a state of severe decline. The
National Marine Fisheries Service (``NMFS'') trawl surveys for several
years have determined that the stocks had not recovered to any
significant extent. The 2003 harvest amounted to about 26 million
pounds. Last year (2002), the State of Alaska, which manages the crab
fishery, set the Guideline Harvest Level (``GHL'') at 31 million
pounds. Barely 23 million pounds were harvested in 2001, an 88 percent
decrease from the 1999 harvest level of 192 million pounds.\1\
---------------------------------------------------------------------------
\1\ The year 2001 commercial fishery suffered further from
extremely harsh weather conditions in which storms with hurricane force
winds, together with some of the largest tides of the year, produced
extremely dangerous sea conditions. The Alaska Department of Fish and
Game reported to the NPFMC in April 2001 that ``a number of vessels had
wheelhouse windows blown out and other structural damage caused by
large waves.'' As a result, many vessels were forced to head for safe
port a full 12 hours before closure of the season, and the fishery was
unable to harvest the full GHL.
In addition to the 86 percent reduction in the GHL from 1999 to
2001, and less than full harvest recovery due to severe weather
conditions, the year 2001 commercial snow crab fishery faced a
reduction in prices. At an ex vessel price of $1.55 per pound, 30 cents
per pound less than the previous year, the estimated 2001 snow crab
fishery value was $35.9 million. This compares to an overall fishery
value in excess of $55 million in 2000, and $162 million in 1999.
(ADF&G Report, April 2001).
---------------------------------------------------------------------------
The cause of the collapse of the Bering Sea crab stocks remains
undetermined. NMFS' best available scientific information continues to
suggest that the decline in the stocks is due to natural and
environmental factors and not to fisheries management policies.\2\
---------------------------------------------------------------------------
\2\ See 66 Fed. Reg. 742, 742-46 (Jan. 4, 2001) (discussing NMFS'
approval of the Snow Crab Rebuilding Plan, Amendment 14 to the
Fisheries Management Plan for Bering Sea/Aleutian Islands King and
Tanner Crab).
---------------------------------------------------------------------------
b. St. Paul--the ``most affected'' community:
The NPFMC's Snow (opilio) Crab Rebuilding Plan recognizes St. Paul
as the coastal community ``most affected by the low stock sizes of snow
crab.'' \3\ Crab landings and processing accounted for 85 percent of
the cash entering the community in 1999. The City receives a 3 percent
sales tax on crab delivered to and processed by floating processors
within three nautical miles of the Island and a 3 percent sales tax on
crab delivered inside the Harbor for processing. St. Paul also receives
half of the fisheries revenues that the State receives as a 3 percent-5
percent tax on vessels fishing outside and inside of the harbor. In
addition, the City receives sales tax on fuel and supplies that are
sold in the Harbor, and derives revenue and jobs from the crab fishery
in-harbor processors and service support to the crab vessels calling
St. Paul.\4\
---------------------------------------------------------------------------
\3\ The finding is made as part of the NPFMC's examination of the
importance of the fishery resource to fishing communities, mandated
under National Standard 8 of the Magnuson-Stevens Act.
\4\ The development of St. Paul's crab dependent economy and a
forecast for the future of the economy, in light of the collapse of the
crab stocks, is discussed in the NRC Report, Attachment 4 hereto.
---------------------------------------------------------------------------
The community has suffered a loss of 82 percent to 90 percent in
revenues from offshore and onshore processors, and harbor services in
2000 and 2001 as compared to 1999 due to the crab collapse. The total
loss in revenues based on the fisheries collapse in those two years, as
compared to 1999, was 85 percent and 83 percent, respectively. Those
losses can be summarized as follows (in thousands, rounded to the
nearest one-tenth of thousand dollars):\5\
---------------------------------------------------------------------------
\5\ These figures do not include other revenue sources which are
not dependent on state of the Opilio crab harvests.
City of St. Paul Revenue Impacts--Opilio Crab Fishery Collapse 1999-2002
[in thousands of $]
----------------------------------------------------------------------------------------------------------------
revenue revenue revenue
Revenue Source 1999 2000 2001 2002 decline 1999 decline 1999 decline 1999
v. 2000 v. 2001 v. 2002
----------------------------------------------------------------------------------------------------------------
Onshore processors 1,706.1 254.2 320.1 214.4 -86% -82% -88%
----------------------------------------------------------------------------------------------------------------
Offshore processors 1,011 156.8 84.1 331.1 -85% -92% -68%
----------------------------------------------------------------------------------------------------------------
Fuel distributors 85 11.4 30.8 32.7 -87% -64% -62%
----------------------------------------------------------------------------------------------------------------
Harbor services 759.5 77.5 137.3 199.9 -90% -82% -74%
----------------------------------------------------------------------------------------------------------------
Local businesses 75.4 28.4 63.1 63 -63% -17% -17%
----------------------------------------------------------------------------------------------------------------
Grand totals 3,637.1 528.3 635.4 841.1 -85% -83% -77%
----------------------------------------------------------------------------------------------------------------
The revenue losses were directly felt by the Island's 650 native
Aleut inhabitants by loss of jobs, loss of community health and safety
services, loss of the community day care facilities, and curtailment in
air passenger, cargo and bypass services to the mainland. As reported
in the Wall Street Journal,\6\
---------------------------------------------------------------------------
\6\ On This Alaska Island, Survival Is More Than Just a TV Game--A
Cutback in Crabbing Quotas Leaves St. Paul Scrambling To Save its Way
of Life, THE WALL STREET JOURNAL, Jan. 15, 2001. The irony that St.
Paul would be looking to subsistence hunting of the fur seal as a
source of sustenance underscores the severe impact of the collapse of
the crab stocks on St. Paul's crab dependent economy.
St. Paul has been thrown into crisis. Layoffs are mounting, a
food bank has opened and an exodus from the island has begun
that could cripple one of the last intact native communities in
the U.S. . . . Particularly hard hit has been the mainstay of
the community, the town government of St. Paul, which has axed
about half of its 90 workers. A day-care center for city
workers was closed, killing five jobs, as was a town-run dental
clinic, wiping out two more jobs and islanders' only access to
dental care. [With protein supplies on the island dwindling,
the native Aleuts of St. Paul are turning to subsistence
hunting of seal,] ``the island's last source of sustenance if
all else fails.''
III. Actions Taken by St. Paul in Response:
a. Securing a Commercial Fishery Failure Determination:
St. Paul's first step was to seek a determination by the Secretary
of Commerce that a commercial fisheries disaster had taken place under
Section 312(a) of the Magnuson-Stevens Act and that St. Paul be
recognized as an affected fishing community.\7\ The Secretary made
affirmative determinations for 2000, 2001, and 2002. As a result,
disaster relief funding was set aside, providing affected communities
in the Pribilofs and the Aleutian Chain with some degree of relief.
---------------------------------------------------------------------------
\7\ Under Section 312(a) of the Magnuson-Stevens Act the Secretary
is authorized to exercise discretion in determining whether there is a
commercial fishery failure due to a fishery resource disaster resulting
from: (a) natural causes; (b) man-made causes beyond the control of
fishery managers to mitigate through conservation and management
measures; or (c) undetermined causes. Such a determination authorizes
the Secretary to provide funds appropriated for the purpose of
``assessing the economic and social effects of the commercial fishery
failure, or any activity that the Secretary determines is appropriate
to restore the fishery or prevent a similar failure in the future and
to assist a fishing community affected by such failure.''
---------------------------------------------------------------------------
The City of St. Paul used its initial share of these funds to
increase water storage capacity for fish processing and fire
protection, both necessary steps to develop the infrastructure required
for diversification into other fisheries in order to reduce the
island's overwhelming dependence on crab. Together with the Corps of
Engineers, the City has also pursued the Harbor Improvements Project in
part in order to attract a multispecies processing, shore-based,
facility to the island. Negotiations with processing companies are
ongoing and their success rests in part on the implementation of the
crab rationalization program and the economic stability that will
accompany it.
b. Vessel Buy-Back Program and BSAI Crab Rationalization:
As part of its multi-pronged response to the crab collapse, St.
Paul also played an active role in promoting initiatives supported by
most of the crab industry to reduce the capitalization of the fishery
by: (1) promoting a fishing capacity reduction program whose goal is to
implement a license and vessel buyback program, and (2) rationalizing
the BSAI crab fisheries. Efforts with regards to the latter resulted in
Congress directing the NPFMC, as part of the Consolidated
Appropriations Act of 2001 (P.C. 106-554), to:
``. . . examine fisheries under its jurisdiction, particularly
the Gulf of Alaska groundfish and Bering Sea crab fisheries to
determine whether rationalization is needed. In particular the
NPFMC shall analyze individual fishing quotas, processor
quotas, cooperatives, and quotas held by communities (emphasis
added). The analysis should include an economic analysis of the
impact of all options on communities (emphasis added) and
processors as well as the fishing fleets. The NPFMC shall
present its analysis to the appropriations and authorizing
committees of the Senate and House of Representatives in a
timely manner.''
After undertaking the congressionally mandated analysis and an
exhaustive four year process of public input, and meetings by industry
sectors and NPFMC members, the Council concluded that rationalization
of the crab fishery would improve economic conditions substantially,
for all sectors of the crab industry. At its June 2002 meeting in
Unalaska, the NPFMC by a unanimous 11 to 0 vote, identified a specific
rationalization program as its preferred alternative for
rationalization of the BSAI crab fisheries. The vote on its preferred
alternative known as a ``Three-Pie Voluntary Cooperative Program''
reflects the fact that this is a comprehensive program that balances
the interests of communities, harvesters, processors, captains, and the
resource, and incorporates short and long-term safeguards for all of
them.
IV. St. Paul: A Leading Proponent of Regionalization
From the outset, St. Paul has been a leading advocate for the
inclusion of crab-dependent community concerns in any BSAI crab
rationalization scheme. One of the ground-breaking concepts that the
community I represent has helped to promote is the concept of
regionalization of BSAI crab landings consistent with recent industry
behavior.
St. Paul views regionalization as the most effective tool for
fulfilling the guidelines of Standard 8 of the Magnuson-Stevens Act and
accommodating the diverse interests within the BSAI crab fishery.
Standard 8 directs the fishery councils to take into account the need
for sustained participation of fishing communities in the fisheries.
The need for regionalization arose in order to protect St. Paul's (and
other BSAI communities') livelihood and the considerable infrastructure
investments made to develop a port which has served the BSAI crab
industry very effectively and profitably. During the 1990s, St. Paul
benefited from a derby-style, race-for-fish scenario where vessels
delivered crab to the nearest available harbor, in order to immediately
depart for further loads of nearby abundant crab stocks. In a
rationalized fishery however, the race-for-fish, would be eliminated
and furthermore, with the Guideline Harvest Levels (GHL) dramatically
reduced as a result of the biomass collapse, harvesters would be
limited to single trips and would therefore be more likely to deliver
their crab to other ports such as Dutch Harbor/Unalaska or to their
homeports including Kodiak, where a number of harvesters are based.
Just as harvester and processor investments have been recognized
and protected in prior rationalization schemes, St. Paul took the novel
approach of insisting that community considerations needed to be
legitimately factored into this process. The stakes were considerable
for St. Paul, which, unlike economically diversified communities in the
Gulf of Alaska, is almost entirely dependent on crab processing. While
St. Paul's strategic location and its existing infrastructure make it a
competitive port, most of the processing operations in the Pribilof
Islands are ``satellite plants'' for diversified, parent seafood
companies. Therefore, with the incentive for decapitalization and
consolidation in a rationalized fishery, St. Paul was at risk of seeing
local processing operations retreating south to larger facilities in
Dutch Harbor, Akutan, and Kodiak.
Regionalization will prevent this migration from taking place,
allowing St. Paul and other northern communities to survive. In
addition, the longer runs involved between the fishing grounds and
ports such as Dutch Harbor or Kodiak in a non-regionalized scenario
would have implied additional fuel costs to harvesters and greater
deadloss, to the detriment of the resource.
More importantly, as is highlighted by the crab disaster, St.
Paul's long-term survival is dependent on its diversification into
other commercially valuable fish species. Even moderate volumes of BSAI
crab valued at $1.50/pound or better go a long way to supporting a
processor's overhead costs, allowing it to engage in lesser-value cod,
salmon, and groundfish processing. Therefore, consolidation of crab
processing away from the Pribilofs would severely impair St. Paul's
capacity to diversify since this activity provides a foundation for
cod, IFQ halibut, and CDQ halibut processing.
V. Benefits to the Resource and the Fleet
Prior experiences in other rationalized fisheries indicate that the
crab resource will benefit from reductions in bycatch and associated
mortality rates. Environmental benefits will accrue as well from the
mechanisms to reduce excess harvesting and processing capacities. With
slower seasons in which to prosecute the fishery, fewer crab pots will
be lost at sea and deadloss and bycatch will be reduced. A healthier,
better managed resource is critical to St. Paul as well, since my
community's long-term livelihood depends on the sustainable use of the
Bering Sea fisheries that surround us.
Moreover, within the high-risk Alaska commercial fishing industry,
the BSAI crab fisheries had a disproportionately high level of injuries
and deaths. From 1990-2001 a total of 25 vessels were lost resulting in
forty deaths. Twenty-one more lives were lost as a result of being
swept overboard, crushed by crab pots or entangled in lines. Experience
demonstrates that to the extent that a rationalized management system
results in a sustainable, economically viable fishery, then
improvements in safety should follow. In a scenario where the ``race-
for-fish'' is eliminated, fatalities resulting from having to fish in
poor weather conditions should be reduced.
VI. The Rationalization Program from the Community Perspective
Under the proposed program, allocations of harvest shares would be
made to harvesters, communities, and captains. Moreover, processors
would be allocated processing shares and harvesters would be permitted
to consolidate their efficiencies by forming cooperatives. From a
community perspective, establishment of processor shares protects the
considerable investments made by processing companies in remote coastal
Alaskan communities such as St. Paul. These investments typically
include the processing facility itself, labor housing, docks, cold
storage, tender vessels, and other infrastructure.
Communities for their part have spent significant resources on
harbor improvements, docks, water/sewer systems, roads, emergency
support services, and community administration to attract and
accommodate processors. As a result of these types of investments, St.
Paul is today the most indebted community in Alaska on a per capita
basis. Furthermore, the processing of crab on-shore leaves behind
numerous benefits to the host community and to the State through taxes,
job, and infrastructure development. In effect, processor shares are
necessary for regionalization to work for communities. As a largely
processing town, St. Paul's future is linked to the processors that
have invested or will invest there.
Processor allocations and the corresponding 90 percent of the
harvester allocations will be designated under the program into
northern and southern regions, based on historic participation.
Provisions allowing for the non-regionally designated 10 percent of
harvester allocations to be delivered to any processor at any port
ensure that a substantial portion of the crab resource will be subject
to competition among ports and processors seeking to attract
deliveries. The BSAI crab rationalization plan provides the harvesting
sector with more flexibility, for example, than is allowed under the
highly successful American Fisheries Act. Movement is also allowed
within the regions to provide for some flexibility and competition for
private enterprise.
However, as specified in the trailing amendments, a two-year
``cooling off period'' has been established during which processing
shares must remain in communities where processing was historically
conducted, so as to avoid a ``rush'' to consolidation in a few ports
immediately after implementation of the program. In addition,
communities and CDQ groups have been granted a right of first refusal
on the sale of processing shares to protect crab-processing dependent
communities.
To accommodate periods of abundance in the crab stocks and provide
opportunities for new processors and communities to compete, caps of
175 million pounds for opilio and 20 million pounds for Bristol Bay red
king crab have been established on the amount of IPQs granted for the
two largest crab fisheries. However, if stocks continue to remain low
(and below those caps) the IPQs will provide stability to the
processing sector and crab dependent communities.
Other important community friendly provisions from St. Paul's
perspective include an increase in CDQ crab allocations from 7.5
percent to 10 percent of the Total Allowable Catch (TAC). CDQ groups
are required to deliver at least 25 percent of the allocation to shore
based processors. On St. Paul, this will encourage further economic
development by our local CDQ organization, the Central Bering Sea
Fishermen's Association (CBSFA). As there are few significant economic
activities that can be engaged-in in geographically isolated
communities in the Bering Sea, these increases in allocation have the
potential to be significant if and when the various crab stocks
rebound.
Moreover, either St. Paul or CBSFA will benefit from provisions
extending community purchase rights of harvesting and processing shares
to community and CDQ groups. These purchases would be conditioned on
the use of these shares for the benefit of community residents.
Finally, CDQ groups have been granted higher ownership caps than
individuals purchasing shares in order to give such groups the
wherewithal to consolidate their economic interest in the crab
fisheries to the benefit of Western Alaska residents.
VII. Conclusion
As stated earlier, the crab rationalization program is a carefully
balanced plan that requires the unique structure approved by the NPFMC
to protect all of the participants in this fishery. In this regard, we
applaud the work patiently undertaken by the Council members and
staffpersons for over four years.
Concerning BSAI crab dependent communities, reducing or eliminating
IPQs would prevent this framework from working and would leave
communities unable to safeguard the processing activities that are
critical to the economic health of remote Bering Sea communities. For
this reason, we urge Congress to support the NPFMC's plan by enacting
legislation as expeditiously as possible. After four years of economic
crisis, this rationalization program is needed in order to revitalize
severely depressed Bering Sea communities. Unlike other parts of
Alaska, remote communities in the Aleutian Chain and particularly in
the Pribilof Islands are highly dependent on this program for their
economic well-being and survival.
Mr. Chairman, and distinguished members of the Subcommittee, thank
you for this opportunity to provide written testimony on behalf of the
City, and the community, of St. Paul. We look forward to discussing
these issues with you and your staffs.
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
______
Prepared Statement of Stephen R. Taufen, Founder,
Groundswell Fisheries Movement
Dear Commerce Committee Senators & Chairman McCain:
The Groundswell Fisheries Movement is primarily concerned about
networks of foreign owned seafood companies and their global
affiliates, and ``abusive transfer pricing'' practices that shift
profits offshore to avoid U.S. taxes. Crab ``rationalization'' is just
one more scheme that will harm the U.S. Balance of Payments and allow
more hidden tax evasion that deprives the U.S. of the ``maximum net
national benefit'' from its fisheries resources.
We believe that the Senate should not allow illicit drains on the
U.S. Treasury, derived by unfettered tax evasion (hidden behind
taxpayer nondisclosure protections that wrongly protect foreign evaders
doing business in the USA) and by granting exemptions from antitrust
scrutiny.
Instrumental rationalization depends upon fair economics and a
balance of market powers. Just as for the NASDAQ, mathematicians and
economists know that for investment markets to fairly benefit everyone,
resource commoditization economics must remain a ``win win'' game for
all investors.
Furthermore, Congress should not institute 'binding arbitration'
processes that protect the commercial secrecy of tax evading foreign
multinationals in ways that guarantee a losing negotiation game for
highly invested U.S. harvesters, whose costs are fully known to these
buying adversaries.
Groundswell is highly concerned about further ``regulation-
negotiation style'' political intrusions promoting concentration on the
buying side of Alaska's seafood markets while the economic problems are
generally on the selling side. Properly addressing problems on the
selling side (harvester grounds prices), the Department of Commerce is
currently reviewing undertaking a $100 million vessel buyback program
to lessen pressure on crab fisheries.
It is a matter of too many boats currently chasing too few crabs.
It is not rational to resolve this by forcefully collectivizing fleets
and forever misbalancing market powers.
The Department of Justice is currently looking into Crab
Rationalization and has been communicating with the North Pacific
Fisheries Management Council on this matter. The Senate should consider
waiting until that DOJ report is available, and remember that the DOJ
wrongly dismissed a 1997-era set of criminal indictments recommended by
its San Francisco branch against Japanese fisheries multinationals
operating in the U.S. on the idea that ``one species does not an
industry make.''
Conveniently, since that time, these foreign firms and others have
been marching one-species-at-a-time to take over our North Pacific
resource rights through their U.S. affiliates and subsidiaries. Those
firms feel that they were granted immunity from prosecution under
pollock schemes, which goes a long way to explain why they believe the
Congress should grant further antitrust exemptions in Crab
Rationalization schemes, too.
Accordingly, Congress should also consider asking the GAO to
further examine the overall structural economic problems of Alaska's
fisheries--especially the current role of foreign financing and effects
of practices of Abusive Transfer Pricing.
Likewise, there is much to be learned about the economic players
involved from a major ``restraint of trade'' trial, L. Alakayak et al
v. BC Packers et al (Alaska superior court case #3AN-95-04676 Civil),
which began February 3 and lands this week in the jury's hands. It
involves dozens of U.S. seafood processors, Japanese importers and
their American subsidiaries as defendants. Congress must wait for its
verdict.
Many defendants are the same firms who ask for exclusivitv in
controlling crab-buying markets in Alaska. Before giving away national
crab rights to them, Congress should know if they are law-abiding, and
about their entire global resource strategy.
There are also serious conflicts-of-interest that have plagued the
North Pacific Fishery Management Council, which has been for many of
the defendants' a control point for Crab Rationalization and other
policy-making. Groundswell has repeatedly testified on the public
record regarding the matters of abusive pricing practices, to no avail.
This Council is not listening and incapacitated in it policy making
(unable to protect the United States' interests) because of its
conflicts-of-interest.
The salmon antitrust case evidence has already revealed that the
free market (an interplay of ``supply and demand'' and arm's length
transactions) for much of our seafood is a conspiratorial fraud, and
that in the case of salmon, the oligopsony of buying companies runs
parallel offshore markets in farmed fish to lower Alaska wild fish
prices.
The abundant evidence of ``price verification'' phone calls among
buyers removes all doubt about the interrelated workings of the
oligopsony that salmon harvesters have faced. Economists have estimated
the market losses to be over $380 million-before applying treble
damages in this $1.4 billion case. We must not forger that an earlier
crab antitrust case showed the price-fixing behavior exists across
species in Alaska fisheries.
Many of these firms were under a Federal ``consent decree'' yet
continued to communicate on crab issues, and on pollock and salmon,
acting unfettered in forwarding their restraints of trade.
One need look no further for ``the grounds price problem'' than the
pollock fisheries, where similar, exclusive rights were granted under
the American (sic) Fisheries Act. On the Japan side of the North
Pacific, catcher fleets receive nearly three times the price for
deliveries as do Alaskan fleets in U.S. ports, for similar product.
That caused an estimated shortfall of $1.9 billion over the past seven
vears to the U.S. pollock fleet compared to what they logically should
have received.
Granting similar oligopsony power for crab will worsen the negative
economic shock on U.S. fleets and our economy, only benefiting the
Japanese financiers and processors. They repatriate about a billion
dollars annually through abusive transfer pricing (ATP) techniques to
foreign shores, harming the U.S. economy directly and through
multiplier effects of the lost dollars.
This is an alternative essence of management from that promised by
the original Magnuson Fisheries Act. where harvesters and processors
would together ``Americanize'' our Alaskan fisheries and pay taxes.
Otherwise, one would assume that a national fish price would be the
logical means to protect U.S. interests in our fisheries.
Congress should not grant exclusive rights to a few corporations on
the crab buying side because it also affects their price-making powers
over other species. too. They are already promoting a non-competitive
``salmon rationalization'' scheme that promises to take more rights
away from small businesses and forever eliminate large competitors from
new entry into the market. And they are talking of similar powers over
other species. There is no end to their greed.
It is not rational to stifle competition and innovation. especially
as our naturally healthy seafood supplies become increasingly valuable.
To ensure that these few buyers are granted exclusive national resource
buying rights, in perpetuity, from captive fleets-especially in light
of their alleged conduct in other seafood markets--is unimaginable.
Thank you for your considerations regarding today's Commerce
committee meeting.
Stephen R. Taufen,
Founder,
Groundswell Fisheries Movement.
Cc: Senator Maria Cantwell
Senator Olympia Snowe
______
Prepared Statement of Mimi Tolva
Dear Senate Commerce Committee,
The Crab Rationalization program was originally started to allow
Bering Sea crab fishermen to harvest the resource in a safer,
economically sound manner while drastically reducing mortality
ofbycatch and harvested product. By allocating a certain allowable
catch to each harvester, (an Individual Fishing Quota, or IFQ), the
derby-style fishery each winter would be eliminated. Fishermen could
choose to fish in better weather, reducing the accidents and injuries
that have plagued the fishery, currently ranked as the most dangerous
occupation in the United States. With more time and better weather, the
quality of the product would be enhanced, mortality would be reduced,
and bycatch levels would drop dramatically.
The North Pacific Fishery Management Council (NPFMC) was appointed
to develop a plan to implement these ideas for the long-term health of
the fishery. Much time and effort and many years have gone by, and a
pretty good plan has emerged.
However, one very controversial issue threatens to undermine the
entire plan. A Processor Quota Shares (PQS) proposal has been added to
offer protection to existing processors already in the crab market.
Under the PQS proposal, harvester shares would be designated ``A-
shares'' and ``B-shares''. Harvesters would be forced to sell all of
their ``A-shares'' to a specified processor. The remaining ``B-shares''
are intended to give harvesters negotiating power, provide a market
price indicator at the beginning of each season, and allow harvesters
to sell that portion to any buyer they wished.
In June of 2002, the NPFMC voted to accept this proposal at a ratio
of 90 percent A shares and 10 percent B-shares. (You read that right)
With a 90 percent A-share ratio, processors will not need to
competitively bid on product. Fishermen will not be able to negotiate a
fair price. Almost total control by a few canneries will squelch any
new developing crab markets and negatively affect price negotiations
for the entire crab industry. This system would be anti-competitive and
would severely curtail or eliminate opportunities for small processing
businesses and independent fishermen. It goes completely against the
free market system and would require a change in the current anti-trust
laws to even be legal.
Crab Rationalization is not complete. The Processor Quota Share
component is not necessary to implement the rest of the plan. The
processor quota component could be pulled from the overall plan if
enough people weigh in on the issue. It could also be modified to a
more fair and equitable split of 50 percent each of ``A'' and ``B''
shares.
Please do not allow this very biased proposal to go through as it
stands now, with the 90:10 ratio.
Mimi Tolva
______
Prepared Statement of Peter Tyack, Senior Scientist and Walter A. and
Hope Noyes Smith Chair, Biology Department, Woods Hole Oceanographic
Institution
Madame Chair and distinguished members of the Committee, my name is
Peter L. Tyack. I am a Senior Scientist and Walter A. and Hope Noyes
Smith Chair in the Biology Department of the Woods Hole Oceanographic
Institution in Woods Hole Massachusetts. Thank you for the opportunity
to provide my views on the Marine Mammal Protection Act (MMPA)
reauthorization as it relates to scientific research.
I have been fascinated since I was a child in the social behavior
of marine mammals and how they use sound to communicate and explore
their environment. I have spent much of the last 25 years following
these animals at sea, listening to their sounds and watching their
behavior. As I started my career in basic research it never occurred to
me that chasing my personal interests would ever become central to such
an important policy issue. In my testimony I address issues concerning
regulation of harassment takes under the MMPA, especially those for
scientific research and incidental takes resulting from exposure to
manmade noise.
Introduction
Three committees of the National Research Council (NRC) of the
National Academy of Sciences have reviewed issues concerning low
frequency sound and marine mammals. Each of these NRC committees has
published a report:
National Research Council (NRC). 1994. Low-Frequency Sound and
Marine Mammals: Current Knowledge and Research Needs.National
Academy Press, Washington, D.C.
National Research Council (NRC). 2000. Marine Mammals and Low-
Frequency Sound: Progress Since 1994. National Academy Press,
Washington, D.C.
National Research Council (NRC). 2003. Ocean Noise and Marine
Mammals. National Academy Press, Washington, D.C.
I was a member of the first two committees and reviewed for the NRC
the report produced by the third committee. I would like to take this
opportunity not only to give my personal views, but to reiterate some
of the repeated suggestions of the NRC committees for changes to the
MMPA.
Regulations to protect marine mammals need to be drawn to focus scarce
regulatory resources on situations where ``takes'' are most
likely to risk adverse impacts to marine mammals.
One of the most important suggestions of the NRC reports on marine
mammals and ocean noise is to regulate harassment in the same way for
all activities, allocating regulatory effort where harassment takes are
most likely to risk adverse impacts to marine mammals. Currently we are
far from this goal. For commercial fisheries, the MMPA allows
incidental taking of endangered marine mammals as long as there is
negligible impact from incidental mortality and serious injury. NMFS
interprets this as an exemption for commercial fisheries from the
prohibition of harassment. Harassment takes are ignored for effects of
propulsion noise from vessels, and a NMFS enforcement officer reported
to the Marine Mammal Commission last year that his region will not
prosecute cases of level B harassment for companies that take tourists
to swim with wild dolphins. This growing industry based upon
intentional harassment thus can count on freedom from prosecution of
its violations of the MMPA. On the other hand, marine mammal biologists
are required to wait half a year or more for permits covering the
slightest possibility that their research may disrupt the behavior of
marine mammals. Once they receive a permit, the permitting process
itself may trigger litigation that can block urgently needed
conservation research. Many other users of sound in the sea, from the
Navy to geophysical contractors to academic oceanographers, find
themselves in a no man's land, where the appropriate regulatory process
for harassment takes is obscure. So far the solutions of the regulatory
agencies have fared poorly in court.
Congress speaks through the MMPA to give commercial fisheries a
special exemption with much more scope to harm marine mammals than
other activities such as conservation research, naval exercises, or oil
exploration. This is in effect a statement of national priorities,
ranking activities for which the United States is most willing to risk
the well being of marine mammals. I would ask all members of this
Committee to stop and think whether commercial fishing should
automatically rank as a higher national priority than scientific
research, the search for domestic sources of petroleum, or the ability
to protect ourselves from enemy submarines.
Problems with permitting scientific research on marine mammals.
As a biologist personally concerned with protecting marine life, I
believe that the double standards in the MMPA have led to a
counterproductive situation for permitting scientific research designed
to protect marine mammals. The permitting process was created to allow
an exemption for scientific research from the MMPA prohibition on
taking marine mammals. It is ironic that, far from exempting research
from an effective prohibition, the permitting process restricts for
researchers, activities that are unregulated for other users. As early
as 1985, NMFS stated in its Annual Report on the MMPA that ``one of the
most extensive administrative programs in NMFS is the permit system
that authorizes the taking of marine mammals for scientific research
and public display.'' From my perspective, this is backwards. Scarce
regulatory resources should only be devoted to minor harassment takes
for research after the much more significant takes of activities that
do not benefit marine mammals are controlled by regulations that are
effectively enforced. As I pointed out in a 1989 article, a scientist
playing back the sounds of a tanker to monitor responses of whales
requires a permit to cover any ``takes'' for animals whose behavior has
changed, while the thousands of tankers entering U.S. ports are
unregulated. This is particularly ironic since the first warning about
effects of noise on marine mammals concerned the risk that increased
shipping noise might significantly reduce the range over which whales
could communicate, a warning issued in 1972, the year the MMPA was
enacted. Not only can they ignore the likely disruption of behavior
caused by noise, but even the lethal ``impacts'' caused when a vessel
collides with a whale are completely unregulated. Nothing we have
learned in the following decades has reduced scientific concern, yet in
spite of three decades of warnings, NMFS has not taken the first step
to protect whales from the risks posed by vessel traffic.
It has been recognized for over a decade that the regulatory focus
on research activities is interfering with research needed to obtain
critical information to evaluate risk factors for noise exposure in the
sea. As the 1994 National Academy report on Low-frequency Sound and
Marine Mammals put it:
Scientists who propose to conduct research directed toward
marine mammals are aware of the permitting requirements of the
MMPA and of the Endangered Species Act (ESA) and the associated
regulations. Most of their research can be conducted under the
scientific permitting process. They routinely apply for and
obtain such scientific research permits. However. the lengthy
and unpredictable duration of this process can create serious
difficulties for research. . . . In addition to permit delays,
certain types of research that are considered ``invasive'' or
``controversial'' either are not allowed under the current
permitting process or may require an Environmental Assessment
or even an Environmental Impact Statement under the National
Environmental Protection Act (NEPA). Such a regulatory burden
actively discourages researchers from pursuing those lines of
study. (p 29)
The committee strongly agrees with the objective of marine
mammal conservation, but it believes that the present emphasis
on regulation of research is unnecessarily restrictive. Not
only is research hampered, but the process of training and
employing scientists with suitable skills is impeded when
research projects cannot go forward. Experienced researchers
are the ultimate source for expanding our knowledge of marine
mammals. A policy that interferes with the development of this
resource appears to be self-defeating. (p 30)
Things were bad in 1994, but they have recently become much worse.
The delays for permitting have become much longer, over 21 months in
some cases. In addition, the judge in a recent court case regarding the
permitting process ruled that all acoustic research on marine mammals
is controversial. This led him to rule that a permit for acoustic
research requires an accompanying Environmental Assessment or
Environmental Impact Statement. This decision means that all of the
research that can help resolve the marine mammal issues raised by the
National Academy reports is subject to much more regulatory burden than
before. Unless Congress changes the regulatory process or provides new
funds to the NMFS Office of Protected Resources to conduct the analyses
required under NEPA, the permitting process will not only discourage
research, but may make it almost impossible to conduct some research
that is urgently needed for conservation biology and that has
negligible effects.
Let me illustrate with an example from the research of Scott Kraus,
a biologist at the New England Aquarium who has studied North Atlantic
right whales for decades under a series of research permits from NMFS.
In August of 2001, he applied for a new permit, as his old one was set
to expire 31 December 2001. In November 2001, after the end of the
public comment period, the Permit Division received a letter from a
self styled ``environmental warrior'' claiming, incorrectly in my
belief, that the research would harm right whales. In early December
2001, operating under his old permit, Kraus started aerial surveys to
keep ships from hitting whales, and he was told the biological opinion
for the new permit was almost done. Kraus never received his permit by
the time his old one expired, and on 24 January 2002, NMFS informed him
that they would defer decisions on a permit until an Environmental
Assessment was conducted following NEPA rules. This was a complete
surprise for Kraus, who had to cancel a research program designed to
develop whale-safe lines for fishing gear. During 2002, eight right
whales were known to get entangled in fishing gear, and six were
thought to have died. It is now May 2003. Kraus had to cancel another
attempt to repeat the whale safe fishing line project in 2003, and he
still has no prediction from the NMFS Permit Division as to when his
permit will be issued. There may be a new determination of a need under
NEPA for an Environmental Impact Statement for his permit, not just an
Environmental Assessment.
Let me recap. The survival of right whales in the North Atlantic is
threatened because so many are killed from entanglement in fishing gear
and from vessel collision. Unlike any airline, as a scientist, Kraus
needs a permit to fly over right whales, in case the whales might hear
the plane and somehow be disturbed. Delays in permitting endanger his
ability to fly surveys designed to warn ships of the presence of
whales. The ships that regularly kill whales are subject to no
regulation, and travel wherever they please at any speed through
critical habitats of the most endangered whale in U.S. waters. In spite
of some fisheries regulations, whales are dying in fishing gear at
alarming rates. Fishermen can continue to place lethal fishing gear
where it can kill whales, but Kraus cannot test new ideas for whale
safe fishing gear, because the environmental paperwork for his research
is not sufficient, even after 21 months of delay. Is there something
wrong with this picture?
I have also personally had experience with the mad world in which
Federal actions block the research needed to protect marine mammals
from poorly regulated impacts of human activities. We cannot protect
marine life from intense underwater noises until we get better at
detecting when a marine mammal or sea turtle is in the danger zone.
Recently, there have been promising developments for whalefinding
sonars. These are high frequency sonars that work like fish finders to
detect echoes from animals close enough to be harmed by unintentional
exposure to intense sounds. When these whalefinding sonars reached the
point in their design process where they were ready to be tested at
sea, I submitted an application to amend my research permit to test how
well a whalefinding sonar could detect migrating gray whales. We know
how migrating gray whales respond to noise, and we expected little if
any behavioral response to the whalefinding sonar. The study was
designed with very sensitive methods to detect whether whales avoided
the sound source by a hundred meters or so, and we requested permission
to ``take'' the whales by harassment.
The Permit Division of NMFS issued the amendment to my permit in a
timely fashion, but only after deciding that the environmental
assessment it had conducted for my original permit sufficed, i.e., that
the amendment did not require a new environmental assessment. The
wording allowing ``takes'' of gray whales alarmed an animal rights
advocate in Australia, who gathered a few small fringe groups to
request an injunction against the research the day before the study was
to begin. The study was delayed by a temporary restraining order and
the entire field team and one of the research vessels in our national
oceanographic fleet were tied up for most of the month planned for the
research. In the end, the judge ruled that the amendment to my permit
was invalid because the NMFS Permit Division had not prepared a new
Environmental Assessment under NEPA not just for my original permit,
but for each amendment to the permit. In the end, hundreds of thousands
of taxpayer dollars were wasted and we are a year behind in developing
more effective methods for monitoring marine mammals.
The NMFS Permit Division of the Office of Protected Resources has
just nine personnel and is increasingly inundated. In 2001 they advised
scientists applying for a permit to expect processing times of at least
90 days for most marine mammal permits with an additional 135 days for
permits affecting endangered species. However, some permits have been
subject to greater delays. NMFS currently advises scientists to allow
at least 6 months for processing, longer for research involving
endangered species. In the cases of my and Kraus' permits, it appears
that last minute complaints by a fringe extremist could trigger a
``public controversy'' condition requiring exhaustive environmental
assessments. Given these precedents, I consider that only permits
backed by environmental analyses acceptable under NEPA are solid enough
to protect research from nuisance lawsuits. My understanding is that it
typically takes several months and $50,000-$100,000 to produce an
Environmental Assessment, and $500,000-$1,000,000 and 1-2 years to
produce an Environmental Impact Statement. Due to the increasing number
of scientific research permits, and the renewed emphasis on NEPA
analysis, some permit applications may be delayed much beyond 6 months,
with dramatic increases in the burden on the Permit Division and on the
applicants.
Congress has in the past few years taken strong steps to fund
research to help resolve urgent conservation problems such as declining
populations of Steller sea lions, or the threat of extinction for the
North Atlantic right whale, and I applaud these actions. Yet both of
these research efforts were delayed by more than a year because of
delays in the permitting process for scientific research. Recent
litigation has highlighted the importance of adequate NEPA analysis in
order to issue legally defensible permits. If Congress wants to support
critically needed conservation research, it is not enough to fund the
science. Congress will also have to mandate significant increases in
funding to the Permit Division.
The time required to obtain a research permit has swelled from 3
months to 6 months to 21 months and counting. A very important change
suggested by the NRC would be for Congress to specify a fixed maximum
time for NMFS to process permits and authorizations. The 1994 NRC
report suggested 10 days for initial processing, 30 days for the public
comment period, and 10 days to issue or deny the permit. The Permit
Division used to use a more liberal 30 days for initial review, 30 days
for the public comment period and a concurrent 45 days for review by
the Marine Mammal Commission, and 30 days to issue or deny the permit.
This totals to 115 days. Additional limits would need to be set for
preparation of environmental analyses under NEPA and conducting a
public hearing. Yet delays of more than 3-4 months pose a serious
burden to research. The only way for the permitting process to proceed
in a timely fashion will be for the Permit Division to conduct
programmatic environmental analyses for most typical research
activities well before applicants request a permit. This additional
burden must be achieved while the ongoing flow of permit applications
is expedited. If NMFS is to issue timely and legally defensible
permits, the permit division and other supporting divisions in the
Office of Protected Resources will need additional program staff, with
specialists in many areas such as environmental law, NEPA, marine
mammal population biology, acoustics, animal health and welfare.
Congress will also have to mandate significant increases in funding for
the Office of Protected Resources to hire contract personnel or to
outsource the analyses required under NEPA and the ESA.
Ironically, it appears that the more serious the conservation
problem addressed by a research project, the more likely the project is
to be attacked by extremists and delayed or cancelled. One side effect
of the permit process is that it personalizes a project in the name of
a scientist. When a ship hits and kills a whale, when dolphins die in
fishing nets, when a sea turtle is killed in an underwater explosion,
the impact is no-fault and impersonal. But when a scientist applies
personally for a permit to help solve these problems, he or she is
front and center in a very public process. This makes the scientist an
all too easy target for uninformed emotional attacks against the bigger
problem. The ``Tyack permit'' is the subject of misinformation in
websites from Australia to the UK.
Some animal rights groups have specialized in attacking research;
it has become all too easy for less scrupulous groups to move from
attacking suffering and pain induced by experiments in captive animals,
to raise funds by misrepresenting research directed at helping to
protect wild animals from serious threats. Activists have actually
tried to sabotage some conservation biology projects. It may reduce the
attractiveness of these cynical ad hominem attacks if research
institutions or consortia were to apply for general authorizations for
different kinds of research, much as other activities that may ``take''
marine mammals are authorized.\1\
---------------------------------------------------------------------------
\1\ A problem with the language of the MMPA involves the use of the
word ``take'' to cover the potential for an activity to cause slight
and temporary changes in behavior. In this age of the internet, it is
quite easy for people all over the world to hear of a permit allowing
thousands of ``takes'' of marine mammals. It is difficult for people
from many countries to find it credible that the U.S. would regulate
the potential for any change in behavior, so it can easily appear that
this permit allows ``taking'' in the normal English sense, which sounds
quite drastic. I urge the language of the permitting process be changed
to use ``take'' for lethal take, ``injury'' for level A harassment, and
``disrupt'' or ``disruption'' for level B harassment.
---------------------------------------------------------------------------
The failure of NMFS to prevail in recent challenges to their
attempts to exempt the permitting process from further environmental
review under NEPA suggests the need for Environmental Assessments or
Environmental Impact Statements for each activity that may be permitted
or authorized. I cannot imagine that even a newly invigorated Permit
Office could perform these analyses for every project, although there
is considerable overlap between the permitting process under MMPA and
the environmental analyses under NEPA. Given how similar the two
processes are, perhaps Congress could specify the categorical exclusion
of these permits under the MMPA. Otherwise, the MMPA or regulations
might specify programmatic environmental analyses of specific research
procedures, such as aerial or vessel survey, tagging, biopsy sampling,
sound playback, etc. I must emphasize that many of the most serious
problems with marine mammal research permits have not been MMPA
problems as much as NEPA problems. As I mentioned above, the Office of
Protected Resources will require a considerable injection of funds and
highly skilled personnel to be able to oversee the production of the
required NEPA documentation.
Suggested unified procedure for authorizing any takes under the MMPA
As my testimony has gotten deeper into the problems with the
current regulatory situation, the discussion has gotten further and
further from the basic goal of equal treatment of all seagoing
activities under the MMPA. I believe that it would be much better if
Congress could correct the deficiencies in the MMPA so that one or two
simple regulatory processes for authorizing takes could be applied
evenly to all seafaring activities. This should be designed to target
situations of potential adverse impacts while minimizing the regulatory
burden for activities with negligible effect. If a streamlined and more
inclusive authorization process were accompanied by better monitoring
and reporting requirements, then we would be in a much better position
to identify and devote scarce regulatory resources to situations where
marine mammals are most at risk from human activities.
Please allow me to sketch an outline of such an approach based upon
suggestions from the 1994 and 2000 NRC reports on Marine Mammals and
Low-frequency Sound. The 2000 report reviews a suggestion made in 1994
regarding amendments to the MMPA that were adopted in 1994:
The NRC (1994) suggested that the regulations governing the
taking of marine mammals by fishing activities should be
broadened to include other user groups that might take marine
mammals. This concept was incorporated into the 1994 MMPA
amendments. The MMPA now requires calculation for each species
of a conservative number of animals that might be taken by
humans from marine mammal stocks, while ``allowing that stock
to reach or maintain optimum sustainable population,'' called
the potential biological removal (PBR) level (MMPA, sec
1362[t]; see Appendix C). NMFS is required to tally all human-
induced mortality for its stock assessments (MMPA Sec. 1386[a])
and uses this number to estimate PBR. (p. 66)
Here is the concept for regulating takes advocated by the 1994
National Academy Report on Low-frequency sound and marine mammals:
The proposed regime is designed to redirect regulation to focus
on human activities with the largest impact on marine mammal
populations, scaling the extent of regulation to the risk the
activity poses to populations. The proposed regime was
initially developed primarily for commercial fishing, but it
was designed to allow the inclusion of other ``user groups''
for PBR. If such a mechanism is adopted in the revised
legislation, this committee recommends that Congress and NMFS
consider it for regulating most ``takes'' of marine mammals by
research as well. Since the objective of the law is to protect
marine mammals, it is difficult to understand applying
different, and less stringent, rules to activities that kill
marine mammals than to activities that are known to benefit
them or to have negligible effects on them. For any population
in which harassment is considered to be a serious risk to
populations, taking by harassment may be included in these
regulations. Where taking by fishing is considered to have
negligible impact compared to other activities, regulatory
attention should focus on these more significant risks. (p. 35)
Now we have almost a decade of experience with this PBR regime. The
incidental take provisions of the MMPA for commercial fisheries require
determination of whether the incidental mortality and serious injury
from commercial fisheries will or will not have a negligible impact on
marine mammal stocks. Fisheries are categorized as to whether they have
frequent, occasional, or remote likelihood of causing mortality or
serious injury, and each fishery receives an authorization for
incidental takes subject to conditions. As long as a fisher registers
with this authorization process, complies with the conditions, and
reports any takes, s/he is exempt from the prohibition against taking.
This regime for regulating fisheries takes that may kill animals
has been quite successful in highlighting situations where populations
are threatened by fishing. Fishers in low impact fisheries have a
simple and streamlined regulatory process, and regulation ramps up
corresponding to the threat, up to closing down fisheries that threaten
the survival of marine mammal populations. However, this regime ignores
effects of harassment, and is not systematically organized to include
takes such as vessel collision, explosions, etc.
The MMPA has a complex tangle of different authorizations for
taking marine mammals. Each kind of take authorization has had
successes and failures. The basic goals of the Act clearly have not
been well served by such different standards for regulating takes for
different activities. As the NRC said in 1994, ``it is difficult to
understand applying different, and less stringent, rules to activities
that kill marine mammals than to activities that are known to benefit
them or to have negligible effects on them.'' The only case that I
think justifies a lower level of regulation involves takes for
scientific research that enhances the survival or recovery of species
or stocks. After three decades of experience, I believe that the Act
would be much more likely to meet its goals if it condensed all
incidental take authorizations to one process that incorporates the
best features of PBR and the best features of the other mechanisms.
I suggest that different user groups that may take marine mammals
could either voluntarily form together or be designated by NMFS. The
list of user groups must include all activities that may take marine
mammals. Such a list would have to include all vessels large enough to
be registered in the U.S. that operate in waters inhabited by marine
mammals. Either the user groups or NMFS should be required to prepare a
Programmatic Environmental Impact Statement, an Environmental
Assessment, or some simpler form of analysis depending upon NEPA
criteria, including whether takes, including harassment takes, were
anticipated to be frequent, occasional, or occur with a remote
likelihood. After this stage, some activities might be judged so low
risk that they would just need to specify when the user registers the
vessel, an obligation to report all takes. For activities judged at
higher risk for takes, each vessel or member of the user group could
apply for incidental take authorization similar to that used now by
commercial fisheries, but including takes by harassment. Before
embarking to sea, each user in a high impact activity could be required
to notify NMFS where they are going and what they plan to do. All users
should be required to report any takes, including level A or B
harassment takes, with strict requirements for prompt and complete
reporting. For activities that might cause harassment takes beyond the
range of detection of the vessel, a monitoring program could be
established to study animals at different ranges from the activity in
order to better estimate the number of harassment takes.
My understanding of the PBR process is that it only counts lethal
takes or injury so serious that it poses a risk of death. For the
purposes of incorporating harassment takes into this regime, NMFS could
start by just requiring complete and accurate reporting of all
potential takes, including any disruption of behavior. The inclusion of
any disruption of behavior should not be interpreted to signify that
all of these constitute ``takes'' under the MMPA. Rather, accurate
reporting of behavioral disruption could be used to help identify what
exposures pose a risk of adverse impact. Ultimately, the significance
to the population of any take is the effect on the demography of the
population, the ability of the population to grow or remain a healthy
size. I strongly encourage Congress to adopt wording requesting that
NMFS attempt to account for harassment takes conservatively in terms of
just the same demographic effects as lethal takes. This is currently a
challenging scientific problem, but the correct wording would stimulate
the appropriate science, while focusing attention on the critical issue
of keeping marine mammal populations healthy. As I discuss more fully
in the last section of my testimony, the best way to do this is to
define harassment in terms of biological significance of the take.
Ultimately a demographic accounting of harassment takes would
require population modeling that relates the dosage of exposure to
harassment to population parameters. There has been great progress in
this kind of population modeling in the past decade, and the PBR
process has forced NMFS to sharpen its stock assessments for marine
mammals, including summarizing all known lethal takes. However, right
now the critical analyses could not be performed for harassment takes
because we know so little about exposures of marine mammals to
harassment. A critical aspect of the PBR regime is that it exempts
registered fishers from the prohibition on taking as long as they
accurately and fully report any takes. A similar clause for all vessels
that may be involved in harassment would ultimately give scientists the
data needed to regulate harassment in terms of biological significance
of impacts to populations. As in the terms of permits for scientific
research, the user should report any observed disruption of behavior,
but the regulations should be clear that not all of these will
ultimately be considered ``takes'' by harassment.
A timely reporting requirement may also make it easier to prosecute
cases of intentional harassment, as failure to report would violate the
terms of the exemption.
This kind of program would allow NMFS to identify situations where
A stock was at risk from a particularly high number of
takes.
An area or activity caused a high number of takes for a
variety of species.
There were particular hot spots of takes.
The cumulative takes pose a risk to the population
Where the sum of takes, lethal, injury, or harassment, pose a risk
to a population, this regime should require a take reduction team to
reduce the problem. This kind of regulatory regime would reduce the
burden on activities that pose little risk, while focusing attention on
species, areas, or activities that pose the greatest risk to the most
endangered populations.
Some may be concerned that the regulatory process I sketch out
would lead to reduced protection. It would certainly streamline the
regulatory process and make it more predictable for most activities,
but I agree with the National Academy (2000) report on Marine Mammals
and Low-frequency Noise that such a change would, if done correctly,
increase protection from the status quo:
The Committee also suggests that activities that are currently
unregulated, but which are major sources of sound in the ocean
(e.g., commercial shipping) be brought into the regulatory
framework of the MMPA. Such a change should increase protection
of marine mammals by providing a comprehensive regulatory
regime for acoustic impacts on marine mammals, eliminating what
amounts to an exemption on regulation of commercial sound
producers and the current and historic focus on marine mammal
science, oceanography and Navy activities. (p. 72)
This change would be all the more effective if it was not limited
to acoustic impacts, but included all sources of takes including
harassment into an integrated workable regulatory structure. The
current MMPA has unbalanced criteria for authorization, allowing some
fisheries to kill animals with no requirement beyond reporting, while
having no procedure available to other activities to authorize more
than a small number of insignificant harassment takes. This does not
meet the conservation goals of the Act.
During the past several years, there have been efforts to address
the very real problems with the MMPA by developing new exemptions for
specific activities such as military readiness. I do not think that
complicating the Act by creating yet another special exemption is the
best answer. I strongly urge Congress to respond to the problems
highlighted by DOD by trying to fix the underlying flaws in the
regulatory procedures of the MMPA before granting a special exemption
that does nothing for marine mammal conservation and leaves many other
producers of sound in the sea with no way to meet the regulatory
requirements. If done correctly, the regulations might be able to
include all activities in a streamlined regulatory approach that
focuses attention on those situations that pose the most risk to marine
mammal populations. Ideally this should include a common mechanism for
dealing with harassment takes in fisheries as well as other activities.
Suggested rewording of incidental take authorization for effects of
noise.
While I believe there is an opportunity to improve the MMPA by
reducing the maze of take authorizations, this may be politically too
difficult to achieve this year. If Congress cannot achieve a common
mechanism for authorizing takes, I would advocate simple changes to the
existing incidental take authorizations in sections 101.a.5.A and
101.a.5.D that I believe would make them appropriate for regulating
acoustic impacts. When the MMPA was first written, it emphasized takes
in commercial fisheries. Certainly no one at that time was thinking
about whether the regulatory process would work for issues such as
incidental harassment takes resulting from unintentional exposure to
noise. Nor was there much experience with issues under NEPA of whether
the impacts of entire activities needed to be evaluated together, or
whether it was better to authorize each time a ``take'' was possible.
Since the MMPA was passed, many studies have demonstrated that
marine mammals respond to ships, dredging, icebreaking and
construction, and sound sources such as pingers, air guns, and sonars.
Most of these sound sources are currently unregulated simply because
NMFS chooses not to enforce the prohibition against taking marine
mammals by harassment. I doubt that many of these activities could find
a regulatory procedure under the current wording of the Marine Mammal
Protection Act that would allow activities with negligible impact while
controlling those that might have an adverse impact. As has been
pointed out by each of the three National Academy reports on this
topic, the dominant source of manmade noise in the ocean is the
propulsion sounds from ships. Yet this has not been regulated by NMFS.
As the National Academy 2000 report Marine Mammals and Low-frequency
Sound put it:
If the current interpretation of the law for level B harassment
(detectable changes in behavior) were applied to shipping as
strenuously as it is applied to scientific and naval
activities. the result would be crippling regulation of nearly
every motorized vessel operating in U.S. waters. (p. 69)
One response to this conundrum is for each activity to seek special
exemptions if their activities become targets of regulation. However,
the National Academy 1994 report Low-Frequency Sound and Marine Mammals
discouraged that approach:
``However, it seems unreasonable that an exemption from the
``take'' prohibitions of the MMPA should be available for some
human activities, including some that kill marine mammals,
without being available for other human activities whose goal
may include the acquisition of information of potential value
for the conservation of marine mammals.'' (p 38)
The first two reports of the National Academy of Sciences on marine
mammals and low frequency sound specifically suggest a broader solution
to this problem: removing the requirements for small numbers of takes,
while retaining a criterion of negligible impact:
Reword the incidental take authorization to delete references
to ``small'' numbers of marine mammals, provided the effects
are negligible. (p. 39)
Low frequency Sound and Marine Mammals (1994)
In addition to making the suggested change m the level B
harassment definition, it would be desirable to remove the
phrase ``of small number'' from MMPA section 137l(a)(5)(D)(i).
If such a change is not made, it is conceivable under the
current MMPA language there would be two tests for determining
takes by harassment, small numbers first, and if that test were
met, negligible impact from that take of small numbers. The
suggested change would prevent the denial of research permits
that might insignificantly harass large numbers of animals and
would leave the ``negligible impact'' test intact. (p. 71)
Marine Mammals and Low-frequency Sound (2000)
My understanding of the judge's ruling in the legal challenge to
operation of the SURTASS LFA sonar, NRDC vs Evans, is that the judge
ruled against the interpretation followed by NMFS that ``small'' can be
interpreted in terms of population size, and exactly following the
fears of the National Academy panel, ruled that the current MMPA
language does require both negligible impact and small numbers, where
the meaning of the word small could not be interpreted in terms of size
and status of populations.
The restriction in the MMPA authorizations for incidental takes to
``a specified geographical region'' may also rule out this
authorization process for most impacts of noise. If ``specified
geographical region'' is taken to mean areas small enough to involve
the same assemblage of species and oceanographic conditions, the
requirements of the incidental take authorizations would be
incompatible with the NEPA requirement to consider all possible uses of
a system. Many sound sources are on a large number of vessels, each of
which may cross the ocean in weeks. Many marine mammals also migrate
thousands of miles through very different habitats. This makes it
difficult to specify a geographical region for a whale that may be in
the Caribbean one day, and off New England a few weeks later. I do not
think that the wording specifying a geographical region is easily
reconciled with the potential numbers and movements of both the animals
and the noise sources.
The propulsion sounds of ships elevate the ambient noise over the
world's oceans, and this global impact is likely to reduce the ability
of whales to detect calls at a distance. I see no process by which such
takes could be authorized under the current wording of the MMPA. Depth
sounders and fish finders have sounds that do not carry as far, but
they are used by tens of thousands of vessels. These sounds have the
potential to disturb marine mammals, and therefore may take animals by
harassment, but did Congress intend to require authorization for each
user? How far could a vessel go before its takes move out of the
``specified geographical region?'' Oceanographic research, much of
which uses motorized vessels and uses sound as a tool to explore the
ocean, also has a global scope, and may be difficult if not impossible
to authorize under the current regulatory procedures.
The House of Representatives has proposed changes in wording of the
sections of the MMPA for taking and importing marine mammals in HR
1835. My understanding is that the primary effect of these changes is
to remove the conditions of small numbers and specified geographical
region for the incidental take provisions. I believe that as long as a
sharp focus is maintained on the issue of negligible impact, these
changes would make the process work for effects of noise on marine
mammals, while still protecting marine mammal populations from adverse
impacts. Since millions of sources such as depth sounders and the
propulsion noises of every motorized vessel could cause harassment
takes under the current definition, I believe that it will be essential
for the process to authorize general activities, rather than individual
vessels or sound sources. This is incompatible with restricting the
authorization to ``small numbers,'' if this is taken literally to mean
just a few individuals, or ``specified geographical region,'' if this
is taken to mean small areas.
Definition of harassment
Until the last paragraph, I have kept my testimony quite general,
but I have some specific comments on the definition of harassment in
the current MMPA and the changes suggested by the Administration and
House of Representatives in HR 1835.
The current definition of level B harassment in the MMPA is:
``has the potential to disturb a marine mammal or marine mammal
stock in the wild by causing disruption of behavioral patterns,
including, but not limited to, migration, breathing, nursing,
breeding, feeding, or sheltering.''
The 1994 NRC report on Low Frequency Sound and Marine Mammals
succinctly reviewed the problem of how harassment has been interpreted
under the MMPA:
Logically, the term harassment would refer to a human action
that causes an adverse effect on the well being of an
individual animal or (potentially) a population of animals.
However, ``the term `harass' has been interpreted through
practice to include any action that results in an observable
change in the behavior of a marine mammal. . . .'' (Swartz and
Hofman, 1991). (p. 27)
The 1994 NRC report goes on to note that many minor and short-term
behavioral responses of marine mammals to manmade stimuli are simply
part of their normal behavioral repertoire. There is clearly a need for
some standard of negligible effect, below which a change in behavior is
not considered harassment.
The change in the definition of level B harassment proposed in HR
1835 is:
``disturbs or is likely to disturb a marine mammal or marine
mammal stock in the wild by causing disruption of natural
behavior patterns, including, but not limited to, migration,
surfacing, nursing, breeding, feeding, or sheltering, to a
point where such behavioral patterns are abandoned or
significantly altered.''
As a biologist who has studied the behavior of marine mammals for
more than 25 years, I find this wording confusing, and I do not see how
it addresses the problem identified by the NRC. The last phrase added
to the definition does add a criterion of significant alteration.
However the point of the NRC reports was biological significance, a
disruption that could have an adverse impact. My dictionary defines
significant as ``likely to have influence or effect.'' The addition of
the word ``significant'' in the new definition therefore does not give
the same standard as suggested by the NRC. As our techniques to study
marine mammals have grown in sophistication and sensitivity, it is now
possible to demonstrate statistically significant alerting or orienting
responses that in my opinion fall well below the negligible impact
standard.
I find the addition of the word ``abandoned'' particularly
confusing in the new definition. It certainly makes sense to add a
criterion for abandonment of critical habitat, but what does this
wording mean for behavior patterns? A sperm whale or elephant seal can
dive for an hour or more, but any marine mammal that abandons surfacing
behavior cannot breathe. If it abandons surfacing for more than a few
hours, it is certainly dead. If a sperm whale group is sheltering a
young calf from a killer whale attack, even a momentary abandonment of
the behavior could be lethal. Calves may be able to survive for days or
weeks if their mother abandons nursing, and many whales could survive
for years without feeding, but what is the time period implied by
``abandon.'' My understanding of ``abandon'' is that it means a
permanent change. By this definition, the ``abandonment'' wording turns
level B harassment into a lethal take. Far from distinguishing
negligible from potentially significant effects, it muddies the waters
further.
Another problem with the use of the term ``abandon'' is that I take
it to mean ``giving up''--a 100 percent cessation of an activity. Yet
since the definition of harassment also applies to stocks, this
definition is not conservative enough for actions that may affect a
large portion of a stock. For example, suppose an activity caused a 50
percent reduction in foraging rates in a majority of the population, or
caused animals to be 50 percent as effective in finding a mate for
breeding. Such reductions would not ``alter'' the form of the behavior,
nor would they meet an abandonment criterion, but few populations could
sustain such changes on a long term basis.
I would like to take this opportunity to reiterate the suggestion
of the National Academy of Sciences second report (2000) on Marine
Mammals and Low Frequency Sound on the definition of level B
harassment:
``NMFS should promulgate uniform regulations based on their
potential for a biologically significant impact on marine
mammals. Thus, level B harassment should be redefined as
follows:
Level B--has the potential to disturb a marine mammal
or marine mammal stock in the wild by causing
meaningful disruption of biologically significant
activities, including. but not limited to, migration,
breeding, care of young, predator avoidance or defense,
and feeding.
The Committee suggests limiting the definition to functional
categories of acttv1ty likely to influence survival or
reproduction. Thus, the term ``sheltering'' that is included in
the existing definition is both too vague and unmeasurable to
be considered with these other functional categories.'' (p 69)
This definition was written by scientists and may require some
rewording to fit legal and legislative requirements. But if the
definition of harassment is to be changed, it should be done so in a
way that makes biological sense and that corrects the need for a
negligible impact standard. I do not think that the changes proposed in
HR 1835 for the definition of harassment succeed in this task. I urge
the Senate to consider using the definition of harassment suggested by
the National Research Council in any amendments to the MMPA.
Conclusion
Madam Chair, I sincerely appreciate your attention to this
difficult and complex issue. There are real problems with current
implementation of the MMPA in our changing environment. However, I am
convinced that Congress and the responsible Federal agencies can make
real progress to create permitting and authorization processes that are
more predictable and efficient, while improving the protection for
marine mammals from adverse impacts of human activities.
Thank you, and I look forward to your questions.
______
Prepared Statement of Konrad S. Uri
I am providing this written testimony to follow up on my letter
this past December to each member of the Commerce Committee as well as
each member of the Appropriations Committee opposing the ``Two Pie''
rationalization plan as proposed by the North Pacific Fishery
Management Council (NPFMC).
I am a lifelong fisherman who has participated in the bottomfish
(trawl, longline & pots) fisheries off the coast of Washington and
Alaska for over 50 years. I also have been a crab fishermen for over 35
years, harvesting and processing Red, Golden, & Blue King Crab as well
as Opilio and Bairdi tanner crab; currently, I fish pacific codfish by
longline in the Bering Sea/Aleutian Islands area about 6 months out of
the year.
I have seen the development of the Bering Sea/Aleutian Island
fisheries from the 1960s, dominated by foreign vessels, to present with
the U.S. public resource harvested by American vessels. I was active in
supporting and implementing the original Magnuson Act, as well as many
other fishery management, allocation, political issues over the years.
With the advent of the American Fisheries Act, and now the proposed
Bering Sea Aleutian Island Crab Rationalization plan, I am discouraged
with the direction of the government to rationalize the various
fisheries of the North Pacific. If you allow the NPFMC's proposed crab
rationalization plan to move forward, you will award the processing
sector with a monopoly.
Providing the processing sector with ``Processor Shares'' violates
the antitrust laws of our land, grants the processing sector an unfair
advantage over the independent fishermen, is adverse to the position of
all the other Fishery Management Councils across the Country, provides
quasi ownership interest in the public's resources under the control of
foreign owned companies, goes against all but one Alaskan communities'
resolutions, and sets a very dangerous precedent which I fear will be
applied in other fisheries in the future.
I therefore ask the Commerce Committee to reject the proposed plan
and send it back to the NPFMC for revamping. Beyond a buyback program,
the fishery should be rationalized at the harvester level, with no
``two pie'' and/or ``processor share'' provision.
______
Aleutian Pribilof Island Community Development Association
Juneau, AK, May 19, 2003
Hon. Ted Stevens
United States Senator
Washington, DC.
Re: Crab Rationalization
Dear Senator Stevens:
There is little dispute that Alaska's Bering Sea crab fisheries are
in crises. The combination of low guideline harvest levels, generally
weak ex-vessel and wholesale prices, and overcapitalization in both the
harvesting and processing sectors have brought the crab industry and
many of the communities that depend upon it to their knees. Many of the
participants face bankruptcy if nothing is done and, for some, it may
be too late already.
Crab rationalization is the only remedy available. Status quo is
not an option. The design and implementation of any fisheries
rationalization scheme is very controversial. That is because there are
winners and losers. There is no way around that. In this case, most of
the controversy swirls around the creation of crab processor quota
shares.
Like Kodiak, we do not particularly like processor quota shares.
From our perspective, however, they are a necessary ingredient to make
crab rationalization work. Without processor quota shares, the
political will within the industry as a whole that is necessary to
bring about crab rationalization does not exist. Therefore, without
processor quota shares, there is no crab rationalization and the
industry may be doomed. The impact upon coastal communities in that
scenario is unacceptable.
The logic behind the Council's crab rationalization program is
reasonable. Effort in both the harvesting and processing escorts need
to be reduced so that the inefficient and wasteful race for crab is
eliminated. This presumably, will bring about some semblance of
stability to the industry. The industry consists of three parties:
harvesters, processors and communities--the program most to be designed
so that each component experiences stability, otherwise chaos results.
Crab harvesters argue they need quota shares to protect their
capital investment by allowing them to fish crab wisely and
efficiently. In that manner they are able to survive financially.
Crab processors argue that if harvesters receive quota shares,
processors must as well. Otherwise, the processors' ability to compete
and survive in an industry they have participated and invested in for
decades disappears. They believe they need quota shares to protect
their capital investment.
Coastal communities dependent upon crab offer the same rationale.
Without adequate community protection in a rationalization program, the
capital they have invested in infrastructure is wasted and the economic
foundation of their community is destroyed.
All three legs of the stool have the same legitimate need for
stability. And like any three legged stool, this one will tilt at best,
and collapse at worst, if one of the three legs is not treated equally.
APICDA's focus during the North Pacific Fishery Management
Council's crab rationalization process has been to protect the
interests of our member communities. Initially, we proposed that all of
our communities, including those that had never processed a crab, be
allowed to construct crab shore plants and purchase crab without
requiring crab processor quota shares. The non-APICDA communities
argued against that approach, reasoning that every pound of crab that
would go into our non-historic crab processing communities would be
stolen from a community dependent upon crab processing. We found that
was a reasonable argument, particularly during times when crab quotas
are low, and agreed to drop our proposal. Forthwith, our focus was to
ensure that the crab processing history earned in our communities (St.
George in particular) would stay in that community. '
Further, we argued that there should be a limit on the amount of
processor quota shares, with the limit tied to the size of the crab
quota; crab harvested above those amounts could be delivered to any
community and any processor regardless of whether or not they have
processor quota shares. Our reasoning was simple: when crab quotas are
above a certain level, there is plenty of crab to address both the
historic crab processing communities' needs and to allow for other
communities and processors to participate. That approach has been
adopted by the Council.
We, as well as other communities argued for a variety of other
community protection provisions. These provisions included: that
communities in which processor quota shares are allocated have the
right of first refusal to purchase processor quota shares from a
processor who is selling; a prohibition on the transfer of processor
quota shares away from a community during the first two years of the
program; and, a host of other community protection mechanisms. All have
been adopted by the PM.
We understand the concerns voiced by the City of Kodiak. One of our
member communities is St. George. St. George is a crab dependent
community. Unlike Kodiak, there is no other commercial fishing in St.
George except a small amount of halibut. There has been no crab
processing in St. George since 2000 due to the collapse of the opilio
stocks. The City of St. George has lost 98 percent of its revenues
since 2000, and is near default on its bond obligations.
It is disturbing to us to contemplate the future of St. George in
the absence of the NPFMC's crab rationalization program. St. George
will flounder. Unlike many other communities, St. George does not have
a shore based seafood processor, nor does it have access to salmon or
groundfish. St. George is essentially 100 percent dependent upon crab.
Without processor quota shares and without crab rationalization, there
is no fishery-related future of consequence for the community. There is
nothing else of economic significance to fall back upon.
We understand very well how controversial and difficult crab
rationalization is. We understand the risks and the pitfalls. But we
are standing on the abyss with status quo, and cannot survive the fall
if rationalization does not happen soon.
Sincerely,
Larry Cotter,
CEO.
Cc: APICDA Board of Directors
______
Prepared Statement of Michele Longo Eder, F/V Michele Ann
Hon. John McCain, Chairman
Committee on Commerce, Science, and Transportation
Dear Mr. Chairman, Senators Wyden and Smith, and Members of the
Committee:
I am a fisherman's wife from Newport, Oregon, a town of about
10,000 people on the central Oregon Coast. My husband is a commercial
fisherman, and has been fishing for 30 years. He owns his own boat, and
has a crew of 3 or 4 men. Usually, he fishes for Dungeness crab and
sablefish. We also have permits to fish for pink shrimp, halibut,
salmon, and albacore tuna.
Let me state from the outset: We are opposed to fish processors
being allocated quota or shares, or in any way be guaranteed a certain
amount of, or access to fish that fishermen like my husband and his men
go to sea to catch.
Now let me tell you why. The only way our crewmen can earn a
living, and the only way our business can make a profit and continue to
survive in this industry, is to get the very best price for our fish
that we can possibly fmd.
I can't quote statistics per se, but I can tell you that in the
last 15 years, on the West Coast of the United States, there has been
an absolute collapse in the number of independent fish processors that
a fisherman can sell their fish to. In towns where there had once been
booming bayfronts, now there is but one or two processing plants, at
most. For example, in Newport, OR, as recently as 15 years ago, there
were 5 different processors that competed for product from fishermen.
Each plant had to provide good service, in terms of efficient unloading
and quick turnarounds, fresh bait, quality ice, and most of all
COMPETITIVE PRICES or they wouldn't get the fish.
Today, for a variety of reasons, there is one fish processor that
processes the majority of West Coast bottomfish and crab, and that is
the Pacific Seafood Group.
Although I can congratulate them on their success for being an
efficient business, buying up money-losing plants and shutting them
down and generally cornering the market, what it has done to the family
fishing business has been devastating. Other buyers are at risk of
losing their shirt if they offer fishermen a better price for their
product, for the largest processor can, and does, immediately sell at a
lower price in order to recapture any market share they may have lost.
This happens season after season, fishery after fishery, and year after
year.
Fishermen have it hard enough as it is, trying to find new and
better markets for their fish on a boat by boat basis, without
GOVERNMENT giving to FISH PROCESSORS another tool that would allow them
to further control and depress prices-and the tool that I am referring
to is ``processor quotas'' or ``processor shares.''
In my opinion, processor quotas have absolutely nothing to do with
either fisheries management, safety, or conservation-the purposes for
which we think government resources and laws should be directed.
Instead, quota shares for processors are an anti-competitive measure
that will make the cost prohibitive for new entrants into the
processing business, and further, will force fishermen to sell only to
those that receive an initial allocation or have the capital resources
to buy processor quota.
The work that my husband and our men do is brutal. I can't begin to
tell you what our life is like. I don't pretend to know what the rest
of the Nation needs, but the last thing our West Coast fisheries need
is the Federal government entwined with fish processors to give them an
even stronger control over the prices paid for our fish. Please--don't
make our proud fishing communities nothing but ``company towns.''
Thank you for the opportunity to offer this testimony.
Sincerely,
Michele Longo Eder.
______
Prepared Statement of Dennis Petersen
Senator John McCain, Chairman
U.S. Senate Committee on Commerce, Science, and Transportation
Dear Senators:
I've been a participant in the Bering Sea King & Tanner crab
fisheries from 1963 until l982. In those years I served in all
capacities aboard my own & other vessels. I also was a board member of
the Alaska Marketing Association representing vessel owners & crab
fishermen in price negotiations with all processors who bought crab
from all Bering Sea boats. During those years, there were few conflicts
of interests between being a fisherman & being a processor. Today that
line has become inexplicably blurred.
Processors own a very high percentage of the crab catcher boats &
in any price negotiating process will ultimately dominate that process
with independent vessel owners finding their voices being stifled by
those processors who will ultimately buy all the crab. The vessel owner
who wouldn't go along with his processor would stand an excellent
chance of being cast adrift with no market: truly an economic disaster
for the small independent fisherman who risks all.
Further, in the only group representing Seattle vessel owners &
fishermen, ACC, there is a makeup, I understand, within that
association whereby a small group who have ownership in multiple
vessels, including a very large crab processor, which dominates that
association on whatever position might come up for a vote.
Consequently, that group is highly suspect as to its loyalties within
the Seattle crab fleet & will probably suffer a real loss of membership
for their support of the two (2) tier crab rationalization plan. Yet,
politically, they are given credence for being the ultimate voice of
the Seattle fleet (which represents roughly 70 percent of the Bering
Sea crab fleet). Not so . . . their testimony should be taken for what
it is . . . completely tainted & not representative of a majority of
the Seattle fleet.
Obviously, I completely oppose the two tier crab rationalization
plan even though I am completely retired from the industry. It is a
product of big money influencing decisions where the men who work in
the world's most hazardous profession will loose. Do not be swayed by
the big moneyed processor advocates who have dominated the North
Pacific Fisheries Council's decision making process as regards this
super important issue. Please reject this abomination!
Respectfully submitted,
Dennis Petersen.
Cc Senator Maria Cantwell
______
Testimony of Victor Smith
May 20, 2003
To:
U.S. Senate Committee on Commerce, Science and Transportation
Senator John McCain, Chairman
Re: Crab Rationalization Hearings
Dear Senator McCain:
Congress should not grant exclusive rights to a few corporations on
the crab buying side because it also affects their price-making powers
over other species, too. As a salmon fisher, I am concerned about
negotiating against networks of offshore companies and global
affiliates. Their abusive ``transfer pricing'' practices that shift
profits offshore to lower fish prices and avoid U.S. taxes should be
more closely examined, for all species.
There was talk in Petersburg, AK., after Senator Stevens met with
Icicle Seafoods fishermen there last June, that Processor Quota Shares
were assured for Pacific Cod too. And I am most concerned because some
are already promoting a non-competitive ``salmon rationalization''
scheme, as well.
Fishers are also worried about systemic conflicts-of-interest
plaguing the State of Alaska, the North Pacific Fishery Management
Council, and the fishers own associations. These comprise the power
base for Crab Rationalization and other policy-making. I have
information on these conflicts of interest that I believe would
convince you that at best there is only the appearance of a power base,
and that that appearance was illegally acquired and has been used to
defraud the public.
This thinly veiled appearance of fishermen's support for
legislation that will be harmful to fishermen is the result of a long
campaign of intense opinion shaping and silencing led by conflicted
leaders of the Alaskan fishing community.
Opposition has been silenced by leaders outright misrepresentations
of memberships opinion, as well as leaders preempting association votes
by publicly announcing association positions without polling their
members. They have also been involved in passage of legislation and
policy making detrimental to fishermen without fishermen's approval. As
a result, millions of dollars have been taken out of the harvesting
side of Alaska's fisheries.
Members of these associations (United Fishermen of Alaska, South
East Alaska Seiners Association, Puget Sound Vessel Owners Association,
and possibly others) appear to have been engaged with Icicle Seafoods,
Trident Seafoods, Norquest Seafoods, and possibly other processors, as
well as some Joint Legislative Salmon Industry Task Force members, and
others, in a pattern of mutual favoritism.
Conflicted members of these associations appear to have:
(1) Assisted processors in attempts to acquire permanent Processor
Quota Shares and market exclusivity.
(2) Assisted processors in transferring more production costs to
fishermen, lowering prices, and reducing public benefits
derived from fisheries resources.
(3) Helped processors elect and support politicians seemingly
willing too exchange support for favors.
(4) Proposed and led a legislative agenda favoring processors that
ignores real solutions for fishermen and fishing communities.
(5) Refused to protect Alaska's fishermen and wild salmon from
pressure from fish farmers and other interests and in some
cases even supported others interests.
(6) Engaged in actions to undermine fishermen and public support in
a fishermen's class action court case looking into charges of
price fixing in Bristol Bay.
(7) Violated association bylaws, misrepresented associations'
positions, and created the appearance of fishermen's consensus
where none existed.
(8) Discriminated against non-resident Alaskan fishermen.
(9) Used libel, character assassination, and intimidation to
silence opposition.
(10) Engaged in lobbying efforts that undermine public support for
fishermen.
(11) Used public money to fund some of their efforts.
I believe these abuses and more have been going on for at least the
last several years. I urge you to pay extra close attention to the
voices of opposition coming from the few fishermen that still have the
will to speak out. If I may, I would like to forward you this
information. We (fishers) have been left behind on our small business
rights.
It is a matter of too many boats currently chasing too few crabs,
and fishermen not receiving a fair percentage of the retail price of
their product. The $100 million vessel buyback program should lessen
pressure on crab fisheries, without also forcefully collectivizing
fleets and forever misbalancing market powers.
To institute the crab ``rationalization'' plan will merely stifle
competition and innovation, as well as lead to even lower prices for
harvesters. Concentrated major grocers do not yet control assigned
collectives of American farmers, who are then forced to supply products
only to those particular stores. So, should crab harvesters not also
share the same antitrust protections as other fishing fleets and
businesses?
We hope that you do not grant processor Quota shares and even think
about asking the GAO to study Alaska's economic structure more
thoroughly, so that we all understand more about the overwhelming power
foreigners and unchecked corporate greed now have in the U.S. seafood
industry.
I urge you to ground-truth the testimony you hear by listening
carefully to all sides. Thank you for considering protecting my
independent, small business and our community concerns.
Sincerely,
Victor Smith.
______
Testimony of Victor Smith
May 25, 2003
To:
U.S. Senate Committee on Commerce, Science and Transportation
Senator John McCain, Chairman
Re: Crab Rationalization Hearings
Dear Senator McCain:
I have read some of the presentations made to your Committee at the
Hearing on the 20th. I was not surprised to see that Senator Murrary
was involved. I had predicted her involvement last year when I first
learned of her and Senator Stevens working together on the Boeing
tanker deal. There was an article on the tankers in the Saturday
Seattle Times. (Right along side an article on the tax cuts.) It was
also interesting to hear Rush Limbaugh belittling you on Friday on his
program. I think I know what's involved there and I believe you're
being charitable referring to the Boeing Deal as a handout.
I don't think the timing of these events is a coincidence and I
believe there's linkage with the Crab hearings. I believe Patty Murray
is Senator Steven's new consensus building WA. Democrat. I doubt she
knows much more about crab than she's been told. Not so with Kevin
Duffy.
Mr. Duffy was the fishermen's associations' choice for Commissioner
of Alaska Department of Fish and Game. Despite the fact that he'd
previously voted for Processor Quota Shares on the Council. Note in the
following e-mail how David Bedford, who was then General Manager of
South East Alaska Seiners Association and is now Deputy Director of
ADF&G, denies endorsing PQs and denies that SEAS support of Mr. Duffy
was support of PQs.
Now look at where Mr. Duffy is. He is no longer just one of eleven
unanimous voters on the Council; he is now the expert presenter of the
plan. And Gov. Knowles is no longer pulling his strings, so who is
manipulating him this time?
It was wrong how Mr. Duffy acquired fishermen's support to become
the processors embedded Commissioner, and his actions now prove this.
These people and the action they are proposing is a fraud. Mr. Bedfords
record of misdeeds goes back at least a year, and I have similar
material on several other key proponents of this plan.
Have you received any testimony from Oliver Holm, of Kodiak? If you
have not, I would like to forward you letters from him that outline how
United Fishermen of Alaska misrepresented fishermen's support to
Senator Stevens two years ago. I have also sent his material to the
Dept. of Justice. Have they forwarded any of it to you?
This is an incredibly flawed process, and an incredibly involved
one. The Bay Case was no test of this collusion; in fact, the high
level tampering in the press highlighted it. Do you have any advise
that might help me decide what to do with this information? I've
received a number of threats that have me wondering now about what I
should be doing.
Victor Smith
Original Message
From: The Smiths
To: Gordon Blue
Sent: Friday, May 23, 2003 5:21 AM
Subject: Re: CR Hearings (2)
Gordon-
Remember that Sen. Stevens voiced almost the exact same concerns
last year when he announced the Senate wouldn't be taking it up and
he's still at it. We predicted Sen. Murray would be involved in this
last fall and here she is. I don't think they are any less sure about
what they want. Stevens bas got a huge investment in Murray. I agree
about the importance of the Bay Case but am pessimistic about the
outcome. Fingers are crossed though. In the euphoria of a win the
processors might be given just about anything. Kevin Duffy? Read the
following e-mail:
This e-mail (following) was received March 7, '03, from David
Bedford, then SEAS General Manager.
Dave-
If, as you say, SEAS has never supported processor quotas, why
are we supporting for commissioner Kevin Duffy, who voted for
them?
You confuse support for the individual with support tor one
specific action. SEAS believes that Kevin will be responsive to the
concerns of commercial fishermen, particularly seiners, as he has been
as Alaska's Commissioner to the Pacific Salmon Commission and Deputy
Commissioner for Fish and Game. Therefore SEAS supports him. Note also
that Kevin supported processor shares in the crab fishery, a fishery
with little direct relevance to the mission of SEAS, because the
Governor, his boss, instructed him to do so. Further you should
recognize that the Governor arrived at this stance because his door was
closed to commercial fishermen. The crucial issue now is whether we are
in a position to work with and influence the current administration. We
are.
Why are we marching in lock step with the UFA, who's president
last spring, mischaracterized the UFA boards position as being
consistent with processor quotas, just as he relayed this same
message to Senator Stevens?
I don't know what Bob told anyone but I do know that UFA's position
on the AFA with its processors quotas is that the organization backs
the North Pacific Council and opposes having resource management or
allocation done by the Congress. As I recall, the NPFMC voted
unanimously for processor quotas, including the commercial fish
representatives, one of whom is an AMCC member from Kodiak.
Weren't you one of the UFA exec committee who voted to take no
action on the UFA boards vote to oppose processor quotas?
Victor, the truth of the matter is that some crab fishermen
supported processor quota and some opposed them. A UFA board member
sneaked a proposal on processor shares in without giving any background
materials to the board and without alerting the affected crab
fishermen. My position was neither in favor nor opposed but rather that
there should be a full and fair and informed discussion before UFA took
a position against one group of fishermen or another. I wanted to talk
to the fishermen before I voted either for or against a resolution that
affected them.
Wasn't SEAS board advised not to interfere in the crab
rationalization plan?
Not only no but Hell No.
The Gilchrest Amendment, which we weren't opposing, was about
considerably more that the crab fishery as everyone involved
knows.
You are wrong that everyone knows what the Gilcrest amendment is
about. I don't. I take it from the context that it is the legislation
that would permit processor quotas in the Bering Sea crab fishery. Is
it an amendment to Magnuson-Stevens? There are a thousand things a day
that I don't have time to oppose, or even find out about. If every
seiner paid his dues SEAS could afford another one of me and there
might be time to get to North Pacific Council and Federal
waterfisheries issues that have no immediate relevance to the seine
fishery.
Why did we support a Governor who wouldn't take e position
against PQ's?
There are no PQ proposals for the seine fishery or any other salmon
fishery. If it happens, SEAS will be in the thick of it. The governor
does support the economic prosperity of the fishery and the coastal
communities. Bread and butter issues seem to resonate with the fleet
right now.
Why wasn't SEAS calling its members to action last spring when
this issue was before congress?
What issue? The crab fishery? Gilcrest? Processor quotas? How much
free time do you think I have? Do you have any idea how much unpaid
time I put in tor SEAS and for you as a fisherman in the seine fishery?
How much more do you want? If you want my limited time spent on the
Bering Sea crab fishery instead of the Board of Fish or the permit
reduction program or the MMPA law suit or state fishery legislation or
the Federal subsistence program or any of the dozen other things I keep
tabs on, get the board to tell me to shift my focus from the seine
fishery to the North Pacific Council.
Actions speak louder than words. Please explain why one should
interpret our record on PQs as one of non-support.
Because we never supported them.
Insteed of asking me to prove a negative why don't you prove the
assertion you make. Show me one printed word, other than spurious e-
mails from Kodiak, where SEAS supported processor quotas. Never
happened. All there is innuendo from one disaffected UFA board member
who tried to sneak a resolution past the UFA board without full
discussion--a resolution that was opposed by some fishermen in the
relevant fishery, the proposal was advanced by the representative of an
aquaculture association--a group with no connection whatsoever with the
crab fisheries. It sure looked like a personal agenda to me so I felt
the need to discuss the matter with some crab fishermen.
How would you like it if UFA took a position on the Southeast seine
fishery without letting any seiner know? Is that the kind of conduct
SEAS should support in UFA, policy making by ambush?
How do you claim SEAS opposes fish farms when we have sided
with the UFA and their approval or a Senator who has sponsored
a fish farming bill. SEAS did nothing to alert our members to
the code of conduct NMFS was developing for EEZ fish farming in
WA. and AK, whose comment period ended in Oct. last year.
Let's double the dues and get more members, then we can hire more
help and deal with more things. Apparently you have different
priorities that the SEAS Board. Instead of stating dissatisfaction over
SEAS lack of action on a specific issue, send me the information to
persuade the board to work on mariculture instead of the current
priorities.
Why are we sitting by as BC fish farmers assault Alaska with a
lobby effort for their farm program? Why haven't we raised the
issue of farmed salmon traffickers being on the ASMI board?
Ditto the last comment.
And what of our habitat move position, you didn't even mention
that? All for now.
Victor Smith
______
City of Saint Paul
Saint Paul Island, AK, May 29, 2003
Linda Freed,
Kodiak City Manager,
Kodiak, AK.
Dear Ms. Freed:
I am writing to you concerning representations that you made
regarding the community of St. Paul during the May 20, 2003 hearings
before the U.S. Senate Commerce Committee regarding the BSAI Crab
Rationalization Program developed by the North Pacific Fishery
Management Council (NPFMC).
Although you have recently become involved in this process at the
NPFMC level, you may be aware that St. Paul has for over four years
played a leading role in developing a program that rationalizes the
Bering Sea crab fisheries and seeks to find balance among the competing
interests held in these fisheries by harvesters, processors, and
communities.
My community's considerable investment of time and effort in this
process is understandable given that 85 percent of our economy is
dependent on the crab fisheries (primarily opilio processing) and that
since the 1990s St. Paul has been the harbor for more than 40 percent
of the average annual landings of opilio crab (as opposed to under 1
percent for Kodiak). The point is that whereas for Kodiak's diversified
economy, Bering Sea crab processing is a minor economic activity, for
St. Paul and other Bering Sea communities it is a matter of survival.
It is unfortunate that, contrary to the facts, during your
presentation before the U.S. Senate Commerce Committee you chose to
mention that St. Paul's backing of the BSAI crab rationalization
program was in question. To support your unfounded assertion you
highlighted the opposition of a local private corporation. While TDX,
as a local private enterprise seeking private gains out of this
process, is entitled to its opinion, it does not reflect the voice of
the community of St. Paul. My community's numerous local interests are
represented at the City Council. Two years ago the St. Paul City
Council passed a resolution supporting the NPFMC's work on the crab
rationalization program. This support remains solid.
As the Mayor/representative of a fellow Alaska city government I
would have been pleased to meet or discuss with you my community's
position prior to the hearings. We have enjoyed an excellent
relationship with members of your community for years and have enabled
many of them to make their livelihoods as harvesters in the crab
fishery to your community's benefit. In my view, your action last week
would be akin to me pointing to the numerous voices on Kodiak in
support of the BSAI crab rationalization program to draw conclusions
about Kodiak's ''true'' position. This is all the more surprising,
since as a member of the NPFMC's Community Protection Committee you
negotiated and participated in the unanimous vote to adopt the NPFMC's
community protection provisions. Then last week, you appeared to
reverse yourself when you indicated during testimony that the program
did not have community protections.
I am attaching for your information, a copy of my testimony as well
as the St. Paul City Council resolution which affirms St. Paul's
position in support of the aforementioned program. Should you have any
questions or concerns please do not hesitate to contact me.
Sincerely,
Simeon Swetzof, Jr.,
Mayor,
City of St. Paul.
Cc: Matt Paxton--Office of Senator Ted Stevens
Bill Woolf--Office of Senator Murkowski
Anna Knudson--Office of Senator Murray
Dan Sakura--Office of Senator Cantwell
Dave Whaley--Committee on Resources, U.S. House of Representatives
David Benton--Chairman North Pacific Fishery Management Council
Kevin Duffy--Commissioner Alaska Department of Fish and Game
______
Prepared Statement of Jeffrey R. Stephan, United Fishermen's Marketing
Association, Inc.
Senator John McCain, Chairman,
U.S. Senate Committee on Commerce, Science, and Transportation,
United States Senate,
Washington, DC.
Re: May 20, 2003, hearing on the Bering Sea/Aleutian Islands Crab
Rationalization Plan: U.S. Senate Committee on Commerce,
Science, and Transportation
Dear Chairman McCain,
On behalf of the United Fishermen's Marketing Association, Inc.
(UFMA), I respectfully submit this testimony to the Record of the May
20, 2003, hearing on the Bering Sea/Aleutian Islands (BSAI) Crab
Rationalization Plan in the U.S. Senate Committee on Commerce, Science,
and Transportation.
UFMA does not support the provision of permits or other
authorization that allocates exclusive rights to U.S. fishery resources
to, or other exclusive use of such resources by, the U.S. Processing
Sector. UFMA does not support the authorization of exclusive processing
rights or exclusive processing access for the U.S. Processing Sector
with respect to BSAI crab. including Processing Shares, Individual
Processing Quotas (IPQs), ``Two Pie'', etc.
Moreover, UFMA does not support the provision of authority to the
Fishery Management Councils, or to the U.S. Secretary of Commerce
(Secretary), to extinguish market access and market freedom for the
U.S. Harvesting Sector by requiring U.S. fishing vessels to sell and
deliver their legally harvested and legally owned fisheries products to
a specific port, thereby, prohibiting such vessels from selling and
delivering their products to buyers or ports of their choice
(``Regionalization'', or mandatory ports of landing).
We respectfully request that Congress should not judge the perilous
and unsound notions of Processor Shares and Regionalization only as
regional contrivances for Alaska or for BSAI Crab Rationalization, nor
without first having comprehensively examined these concepts as
important issues of National Policy. Free and open markets, vigorous
competition and the enduring principles that underlie our antitrust
laws should be as relevant today, and in Alaska, as when the Sherman
Antitrust Act was passed.
We respectfully request that Senators and Representatives from
coastal states with commercial fishing economies should not wreak
Processor Shares or Regionalization on Alaska if they are not willing
to do so on their own states, and should understand that these schemes
will ultimately contaminate their own fishing industries.
Processor Shares and Regionalization are not fisheries management
devices. They represent market allocation, market regulation and
economic protectionism. The North Pacific Fishery Management Council
earns great respect for their dedication and noteworthy
accomplishments. The FTC, FCC, SEC and other entities, not the
Councils, address ``market regulation'', ``market allocation'' or
``preservation of competition''; these terms do not appear in the
Findings, Purposes or Policy of the Magnuson-Stevens Act (MSA), and
distinguished Council members nationwide are not chosen because of
their knowledge, experience and understanding of these concepts.
UFMA is the longest established association of BSAI crab
harvesters. The membership of UFMA includes vessel owners and operators
who harvest crab, sablefish, halibut, salmon, herring, p. cod and other
groundfish in Federal and state waters of the Gulf of Alaska (GOA) and
the BSAI. Moreover, UFMA is actively involved and impacted by
rationalization initiatives that are currently underway with respect to
several Fishery Management Plans (FMPs) that govern U.S. fishery
resources within the jurisdiction of the North Pacific Fishery
Management Council (NPFMC), including the FMPs for BSAI King and Tanner
Crabs, BSAI Groundfish, and GOA Groundfish.
I. National Study of Processor Shares and Regjona!ization
We respectfully request, prior to any further Congressional
consideration of authorizing Processing Shares and Regionalization for
application in the U.S. Fishing Industry, including the BSAI crab
fishery, that Congress direct a full and comprehensive examination and
report of Processor Shares and Regionalization. Such an examination and
report should address foreign ownership, divestiture, anticompetitive
concentrations and combinations, why already dominant processing
entities need additional help and protection, etc.
In the Sustainable Fisheries Act of 1996, Congress directed the
National Academy of Sciences (i.e., National Research Council, or
``NRC'') to examine and report on the issue of Harvesting Sector
``Individual Fishing Quotas'' (IFQs). It makes sense for Congress to
direct a similar examination and report of Processor Shares and
Regionalization, since Congress did not direct in 1996, nor did the NRC
provide in 1999, a comprehensive focus on or evaluation of the impacts,
effects and mechanisms of Processor Shares or Regionalization,
especially with respect to the significant anticompetitive and economic
power issues that are undeniably present when Processor Shares are
contemplated for use in fisheries management.
Such an examination and report on Processor Shares and
Regionalization should be considered in the context of antitrust and
trade statutes and policy, instead of in the context of MSA. We
respectfully submit that oversight of any further consideration of
Processor Shares and Regionalization, and of the above-suggested
Congressionally mandated examination and report, should have Judiciary
Committee oversight, in consultation with the Commerce Committee,
Federal Trade Commission and the Secretary of Commerce.
If after the completion and Congressional consideration of a
comprehensive examination and report of Processor Shares and
Regionalization, Congress decides that these concepts are reasonable
devices for use in the management of U.S. fishery resources, then we
respectfully request that Congress should hold hearings to guide the
development of National Standards that would apply to the use and
application of Processor Shares and Regionalization in the management
of U.S. fishery resources.
As previously referenced. in 1999 the NRC published ``Sharing the
Fish; Toward a National Policy on IFQs'' (``Sharing the Fish''). While
Sharing the Fish focused primarily on the impacts, effects and
mechanisms of Harvesting Sector IFQs, in those instances where the NRC
did address Processor Shares, it was clear that they found no valid
rationale for Processor Shares: ``. . . Nor did the committee find a
compelling reason to establish a separate, complementary processor
quota system (the 'two pie' system).'' (Sharing the Fish, page 205);
``Processors also complained . . . The committee was not convinced,
however, that the solution to the perceived problem lies in the
allocation of either harvesting or processing quota to processors.''
(Sharing the Fish, page 155).
We would be pleased to provide a suggested outline for a
Congressional mandate of a comprehensive study and report of Processor
Shares and Regionalization.
II. Lack of Adequate, Quality and Relevant Data, and Associated
Analysis and Consideration
We ask Congress to closely examine the lack, quality,
comprehensiveness, inclusion and availability of data and other
information that is important and essential to understanding and
analyzing the distributional and cumulative impacts of Processor
Shares, and to making informed decisions that are associated with the
potential application of Processor Shares. This is especially relevant
when consideration is given to the combination of Processor Shares and
Regionalization, and in the context of the BSAI Crab Rationalization
Plan.
There is an incomplete understanding and analysis of the actual
factual ownership of, and other vehicles that provide for the exercise
of economic power and control over, BSAI Crab Harvesting Sector IFQs by
the BSAI Crab Processing Sector. There is no clear enumeration of the
distribution of Processor Shares by individual processing entity,
entity location, or port. There is an incomplete understanding of the
ownership structure of Harvester Sector vessels, and of the cumulative
impacts that will result from the distribution of Harvester Sector IFQs
to Processing Sector ``affiliated vessels'' (i.e., vessels that are
associated with Processing Sector entities through a variety of
ownership interests, notes, loans, etc.), combined with the
distribution of Processor Shares to those BSAI crab processors who are
so affiliated with such vessels.
There is a significant lack of understanding and analysis of the
substantial opportunities that exist in the BSAI Crab Rationalization
Plan for the exercise of economic power and control that can and will
be used by the Processing Sector over the Harvesting Sector generally,
over the use of Harvesting Sector IFQs, and over competitive and fair
price formation. It is not now possible to reasonably or accurately
understand the substantial economic power and control that will be
vested in the BSAI Crab Processing Sector as a result of Processor
Shares. This is especially worrisome when Regionalization is combined
with Processor Shares.
We respectfully request that Congress direct the Councils and the
Secretary to request essential and relevant data that resides with the
U.S. Maritime Administration (MARAD); such data that will generally
permit the Councils and the Secretary to comprehensively evaluate and
analyze the cumulative impacts of Processor Shares, Regionalization and
Harvester Sector IFQs, and specifically with respect to the BSAI crab
fishery. Possession of this information by the Councils and Secretary
would significantly assist the Secretary and the Councils in their
understanding of the social, policy and economic implications and
impacts of their regulatory action with respect to Processor Shares.
Further, we request that Congress direct MARAD to release and make
available the important and relevant information that they collect and
compile with respect to the underlying ownership structure of U.S.
Harvesting Sector vessels. Vessel ownership information that is
collected and held by MARAD could be of significant benefit and use to
the Secretary and to the Councils in support of making informed
decisions with respect to the underlying ownership structure of
entities that may receive IFQs or Processor Shares, including
information that is necessary for evaluating participation in U.S.
fisheries, the important distributive impacts and effects of IFQ or
Processor Share ownership, the anticompetitive impacts of vertical
integration, and the extent of foreign ownership of U.S. fishing
resources.
III. Fair and Competitive Price Formation and Arbitration
Processing Shares and Regionalization invest the BSAI Crab
Processing Sector with a significant increase of political and economic
power, control and influence that will greatly obstruct the free flow
of inputs to an otherwise but already variably competitive marketplace,
including inputs upon which competitive prices for the Harvesting
Sector depends (price formation). This is especially true for
processors who benefit from the largesse of the American Fisheries Act
Binding Arbitration as associated with price formation is characterized
as a mitigation device. The BSAI Crab Processing Sector fashioned the
pretension that Arbitration was meant to address ``failed price
negotiations'', thereby masking the true and underlying reasons why
Arbitration was investigated. That is, successful price formation
relies on the complex conditions and facts that exist in a competitive
marketplace; however, price formation and Arbitration, as a result of
Processor Shares and Regionalization, must rely on information from an
artificial and significantly misaligned marketplace. Arbitration,
albeit a politically expedient concept, provides a false sense of
security and mitigation.
The Council was offered several meaningful and realistic
opportunities to attempt to mitigate the obvious and expected negative
impacts of Processor Shares and Regionalization on fair prices, and on
competitive and open markets, including, for example, provisions that
allowed: (1) all BSAI crab vessels to freely sell all their legally
harvested and legally owned crab to Kodiak (the ``Kodiak Open Port''
concept''; (2) all BSAI crab vessels to freely sell their last load of
legally harvested and legally owned crab to Kodiak; (3) Kodiak-based
BSAI crab vessels to freely sell all their legally harvested and
legally owned crab to Kodiak; (4) Kodiak-based BSAI crab vessels to
freely sell their last load of legally harvested and legally owned crab
to Kodiak. The Council discarded all these suggestions.
An ``Open Port'' designation for Kodiak would provide some
realistic mitigation, and at least one competitive market and port, to
counter the anticompetitive impacts on prices and markets that are
created by the implementation of Processor Shares and Regionalization.
IV. Foreign Ownership
We respectfully request Congress to evaluate the impacts and
implications of Processor Shares with respect to competition of, and
methods to prevent, limit and control, foreign ownership interest in,
and economic control of (1) harvesting vessels; (2) permits or licenses
that permit the Harvesting Sector to operate a vessel in a fishery
where there exists a limitation on the number of vessels that are
permitted to harvest a U.S. fishery resource (e.g., license limitation
programs); (3) permits or licenses that permit the Harvesting Sector to
have exclusive rights to harvest a quantity of fish, expressed by a
unit or units representing a percentage of the total allowable catch of
a fishery that may be received or held for exclusive use by a person
(e.g., IFQs); (4) fishing history of the Harvesting Sector (i.e.,
landings, participation and other criteria that will determine the
qualification of Harvesting Sector entities to receive IFQs, licenses,
permits etc); etc.
V. Divestiture of IFQs and BSAI Crab Vessels in the BSAI Crab
Processing Sector
We respectfully suggest that Congress evaluate the use of
divestiture as a means to mitigate the anticompetitive impacts that
result from the application of Processor Shares in U.S. fisheries.
Processing Sector entities that receive Processor should be required to
divest themselves of harvesting vessels and Harvesting Sector IFQs as a
reasonable precondition for receiving Processing Shares.
VI. Conclusion
Processor Shares will negatively impact the success and economic
underpinnings of other fisheries. For example, we believe, as do others
in the industry, that Processor Shares in the BSAI crab fishery,
coupled with the benefits of the American Fisheries Act that accrued to
several large dominant Alaskan processing entities, will negatively
impact the ability of the Alaskan Salmon industry to recover from
current market and structural related challenges.
The impacts that result from the application of Regionalization and
Processor Shares either separately, or in combination, are difficult to
analyze and understand. However, it is imperative that a thorough
examination and understanding of these anticompetitive and other social
and economic impacts are thoroughly investigated, explored and
understood prior to a serious consideration of applying Regionalization
or Processor Shares in fisheries management, including in the BSAI crab
fishery.
UFMA has consistently supported rationalization of the BSAI Crab
fishery. We believe that it is time to deliver the safety, conservation
and management benefits of BSAI Crab Rationalization, and we hope that
the BSAI Crab Processing Sector will loosen their grip on these
benefits. We hope that Congress will let the concepts of Processor
Shares and Regionalization fade away.
Thank you for your consideration of our comments.
Sincerely,
Jeffrey R. Stephan
______
Testimony of Victor Smith
June 6, 2003
To:
U.S. Senate Committee on Commerce, Science, and Transportation
Senator John McCain, Chairman
Re: Crab Rationalization Hearings
Dear Senator McCain and Committee Members:
Regarding the testimony of Kevin Duffy, it is preposterous that
Mr. Duffy was the fishermen's associations' choice for
Commissioner of Alaska Department of Fish and Game despite the
fact that he'd previously voted for Processor Quota Shares on
the Council. The support he received from fishermen's
association executives to become Commissioner was unethical and
possibly illegally granted.
Note in the following email how David Bedford, who was then
General Manager of South East Alaska Seiners Association and
has now moved up to become Deputy Director of ADF&G appointed
by Mr. Duffy, denies endorsing PQs and denies that SEAS support
of Mr. Duffy was support of PQs.
This email (following) was received March 7, '03, from David
Bedford, then SEAS General Manager. It is Mr. Bedford's
response to questions about SEAS endorsement of Mr. Duffy for
Commissioner.
Dave-
If, as you say, SEAS has never supported processor quotas, why
are we supporting for commissioner Kevin Duffy, who voted for
them? (V.S.)
You confuse support for the individual with support for one
specific action. SEAS believes that Kevin will be responsive to the
concerns of commercial fisherman, particularly seiners, as he has been
as Alaska's Commissioner to the Pacific Salmon Commission and Deputy
Commissioner for Fish and Game. Therefore SEAS supports him. Nota also
that Kevin supported processor shares in the crab fishery, a fishery
with little direct relevance to the mission of SEAS, because the
Governor, his boss, instructed him to do so. Further you should
recognize that the Governor arrived at this stance because his door was
closed to commercial fishermen. The crucial issue now is whether we are
in a position to work with and influence the currant administration. We
are. (D.B.)
Now look at where Mr. Duffy is. He is no longer just one of eleven
unanimous voters on the Council; he is now the expert presenter of the
processors PQ plans. And Gov. Knowles is no longer pulling his strings
as Mr. Bedford claimed, so who is manipulating him this time? The
Council process is flawed, and its' goals don't match those of the rest
of the country.
It was wrong how Mr. Duffy acquired fishermen's support to become
the processors embedded Commissioner, and his actions now prove this.
Executives of the seiners associations and the United Fishermen of
Alaska have played a big role in hiding from fishermen that there have
been plans afoot in the industry to expand PQ's to all Federal
fisheries and even State salmon fisheries. These actions now
demonstrate that there has been a concerted effort by numerous people
to advance this plan.
Fishermen's association Executives violations of association's by-
laws and running their associations to achieve their own agendas by
silencing fishermen and misrepresenting fishermen's interests are
theft. For policy makers and processors to have knowingly relied upon
this illegally crafted appearance of fishermen's support as a
foundation for advancement of acquisition of resources from the
American people is unethical at best. It should also highlight what has
been going on in the fishing industry on numerous other issues. It is
not possible that what we're seeing is the result of independent
actions.
Sincerely,
Victor Smith.
______
Foremost, we oppose any type of rationalization plan that rewards
over capitalization.
We strongly oppose processor shares. They are not warranted, This
is so because of the infrastructure already in place by the processors
in question. Most of these capital investments (infrastructure) have
long been paid for. This provides them with an advantage against any
future competition.
We believe all coastal communities dependant on commercial fishing
will suffer negative economic impact. This is so because it, (BSAI Crab
Rationalization), would be president setting legislation mapping out
future rationalization plans. This plan guarantees control of our
fisheries resources to a select group of processors both FOREIGN and
domestic. This plan, even if limited only to BSAI Crab, will give these
select processors a huge competitive and economic advantage. The
majority of these processors own plants in multiple coastal
communities: thus, giving them advantages over competing plants in
communities outside of the BSAI Crab region.
We believe any legislation that contradicts our current Anti-Trust
Laws should be reviewed in depth with representatives of all groups
whom are affected both directly and indirectly.
In closing, we would like to submit that economics itself would be
enough to rationalize the BSAI Crab fishery. The pace and safety of
this fishery is a concern. These concerns can be addressed through trip
limits and gear restrictions (i.e. Pot Limits). Any further over
capitalization (i.e. Capital Construction Funds) should not be
encouraged by government. We cannot support any further legislation
that forever gives control of a public resource to a select few.
We are a commercial fishing family residing in Kodiak, Alaska. Our
above comments are brief. Familarity with the BSAI Crab Rationalization
proposal would be needed to understand our letter. We are not directly
involved in this fishery, but understand the effect the BSAI Crab
Rationalization will have on future rationalization of the Gulf of
Alaska and Bering Sea Ground Fisheries which we are economically
dependant on.
Sincerely,
Ron and Julie Kavanaugh
Ron Jr--age 22,
Miranda--age 14,
Sylvia--age 11,
Garrett--age 4
______
Prepared Statement of Hon. Ron Wyden, U.S. Senator from Oregon
Thank you, Mr. Chairman, for holding this hearing today.
There is a long history of fishing vessels from Oregon traveling up
to Alaskan waters to share in the crab fishery that has flourished
there. These boats are crewed by both Oregonians and Alaskans and
deliver their catch to the coastal communities of Alaska to be
processed. If done fairly, the assigning of quotas within a fishery can
lead to both a safer fishing season and a more sustainable fishery.
However, I share the concerns of Oregon fishers about the assigning of
processor quota shares particularly if they are done in a manner that
will benefit a few processors at the expense of many fishers and other
processors. Assigning quota share in this manner raises issues of anti-
trust and could lead to the consolidation of processing capacity.
I also am wondering why the North Pacific Fishery Management
Council has devised a plan that they do not have the statutory
authority to implement without the approval of Congress.
I thought the purpose of the management councils was to make
decisions at the regional level under the guidelines provided by
Congress. I am not sure Congress wants or should get into the habit of
debating individual management plans and making what should be regional
decisions at the national level. Nor should management councils get in
the habit of choosing to ignore Congressional instruction.
I would like to submit the following questions for the record:
Response to Written Questions Submitted by Hon. Ron Wyden to
Kevin Duffy
Question 1. As a representative of the Alaska Department of Fish
and Game and the interests of all Alaskans, was your testimony reviewed
by Mr. Alan Austerman, the governor's fish policy advisor? And if not,
would you please explain why?
Answer. Witness did not respond.
Question 2. As a member of the Council, do you have any insight as
to why there were numerous motions on elements of harvesting shares but
no motions made on elements of the processing share aspects of the
program such as the ratio of A/B shares or the qualifying years used to
award processing history?
Answer. Witness did not respond.
Question 3. The analysis of Dr. Charles R. Plott, the respected
economic modeler from the California Institute of Technology,
demonstrated that the ``last best offer'' arbitration process supported
by the processors would be less apt to result in a competitive price
for ``A'' shares, yet you voted for this ``last best offer'' approach
over the alternative ``fleet-wide'' approach. Could you explain why?
Answer. Witness did not respond.
Question 4. A National Academy of Sciences panel recommended that
``if the regional councils determine that processors may be
unacceptably disadvantaged by an IFQ program because of changes in the
policy or management structure, there are means, such as buyouts, for
mitigating these impacts without resorting to the allocation of some
different type of quota. For example, coupling an IFQ program with an
inshore-offshore allocation would preserve the access of shore-based
processors to fishery resources.'' What was your rationale for voting
against the recommendations of the National Academy of Sciences?
Answer. Witness did not respond.
Question 5. You and other proponents of processor shares have
frequently cited the 11-0 Council vote as an argument for why this plan
should be accepted. However, the 11-0 Council vote preceded the
decision on the binding arbitration trailing amendment which was a
contentious 6-5 decision favoring the plan supported by the processors.
It seems to me that once all of the parts of the proposed plan were
revealed that the plan was endorsed in the Council by the thinnest of
margins. Am I wrong?
Answer. Witness did not respond.
Question 6. According to the proposed processor share allocation
formula, which looks at a narrow two-year window, the top 12 processors
with one exception would receive more quota allocation than they
historically processed, 99.4 percent to 75.66 percent respectively. Why
did the Council decide on such a narrow window for determining
historical levels of processing?
Answer. Witness did not respond.
Question 7. The General Accounting Office (GAO) has been unable to
find any credible evidence that processors in any fishery have been
disadvantaged by the allocation of harvester shares. Why should
Congress endorse a plan that not only is contentious, but appears to be
unnecessary?
Answer. Witness did not respond.
Question 8. We know what fishing quotas achieve for resource
conservation and safety; what do processor quotas achieve for resource
conservation and safety?
Answer. Witness did not respond.
______
Response to Written Questions Submitted by Hon. Ron Wyden to
Linda Freed
Question 1. Chairman Stevens has reminded us that the original
intent of the Magnuson Fishery Conservation and Management Act was to
protect the fisheries of the United States from foreign interests. Are
there crab processors who are owned directly or indirectly by foreign
interests? Will any processors that are directly or indirectly owned by
foreign interests benefit from this proposed plan at the expense of
American owned processing companies?
Answer. Witness did not respond.
Question 2. We know what fishing quotas achieve for resource
conservation and safety; what do processor quotas achieve for resource
conservation and safety?
Answer. Witness did not respond.
______
Response to Written Questions Submitted by Hon. Ron Wyden to
Dave Fraser
Question 1. Chairman Stevens has reminded us that the fishery
management councils were set up to manage fish populations. Is there
any reason processor shares would be necessary to manage fish
populations?
Answer. Witness did not respond.
Question 2. Has a management plan that allocates processor shares
in a manner similar to this proposed plan ever been tested in a crab
fishery? If so, what is the current state of that fishery?
Answer. Witness did not respond.
Question 3. The National Academy of Sciences recognized that
processors may be disadvantaged by fishing quotas. What steps other
than processor quotas could be used to protect processors?
Answer. Witness did not respond.
Question 4. We know what fishing quotas achieve for resource
conservation and safety; what do processor quotas achieve for resource
conservation and safety?
Answer. Witness did not respond.
______
Response to Written Questions Submitted by Hon. Ron Wyden to
Frank Kelty
Question 1. The proposed plan would take a public resource, Alaskan
crab, and allocate this resource among harvesters. The proposed plan
then allocates the harvested crab to processors even though the
harvested crab has become private property in the hands of harvesters.
As a supporter of this proposed plan, are you telling me that the
United States government and not market forces should decide what
fishers may do with their private property? If the issue is fairness
and protecting processors from the perceived advantage that harvesters
will obtain with the allocation of harvester shares, what about
consumers? Should the United States Government also be setting prices
for the products the processors are selling to protect consumers from
the advantage that processors obtain with the allocation of processor
shares? If consumers don't need protection from processor shares, why
do processors need protection from harvester shares?
Answer. Witness did not respond.
Question 2. The General Accounting Office (GAO) has been unable to
find any credible evidence that processors in any fishery have been
disadvantaged by the allocation of harvester shares. Why should
Congress endorse a plan that not only is contentious, but appears to be
unnecessary?
Answer. Witness did not respond.
Question 3. We know what fishing quotas achieve for resource
conservation and safety; what do processor quotas achieve for resource
conservation and safety?
Answer. Witness did not respond.
______
Response to Written Questions Submitted by Hon. Ron Wyden to
Arni Thompson
Question 1. Chairman Stevens has reminded us that the fishery
management councils were set up to ensure that fisheries decisions
would be made at the regional level and not at the national level or in
the Federal courts. Of the witnesses that support this proposed plan,
you seem to qualify your support for the plan more than Mr. Kelty and
Mr. Duffy who seem to support the proposed plan unconditionally. In
your opinion which specific parts of the plan are the most untested?
Answer. Witness did not respond.
Question 2. Once the proposed plan is accepted, you noted in your
testimony that you expect continual oversight by the Council and
Congress to correct or modify the processor quota portion of the plan.
Based on your experience with the North Pacific Fishery Management
Council and the United States Congress, do you expect that any
necessary changes can be implemented in a span of less than two or
three years?
Answer. Witness did not respond.
Question 3. Why should Congress support a plan that is supported by
only one of the four major associations of vessel owners, which in
turn, represents fewer than 15 percent of issued crab licenses?
Answer. Witness did not respond.
Question 4. Of the vessels that you list as members of the Alaska
Crab Coalition, how many are current members, and of these, how many
qualify as ``non-processor affiliated''?
Answer. Witness did not respond.
Question 5. The General Accounting Office (GAO) has been unable to
find any credible evidence that processors in any fishery have been
disadvantaged by the allocation of harvester shares. Why should
Congress endorse a plan that not only is contentious, but appears to be
unnecessary?
Answer. Witness did not respond.
Question 6. We know what fishing quotas achieve for resource
conservation and safety; what do processor quotas achieve for resource
conservation and safety?
Answer. Witness did not respond.