evaluation of alternative strategies for Virginia oyster grounds management economic considerations in policy design

[From the U.S. Government Printing Office, www.gpo.gov]

AN EVALUATION OF ALTERNATIVE STRATEGIES FOR
VIRGINIA OYSTER GROUNDS MANAGEMENT:
ECONOMIC CONSIDERATIONS IN POLICY DESIGN

Leonard Shabman
Eric Thunberg

Department of Agricultural Economics
Virginia Tech
Blacksburg

March, 1987

Final Report in Partial Fulfillment of NOAA, Section 309 Project, "Development
of an Economic Data Base for Describing Virginia's Commercial Seafood Har-
vest Sector."

Project Officer: Janice Carter-Lovejoy
Virginia Council on the Environment

COASTAL ZONE

INFORMATION CENTER
SH
371
.S53 '
1987

Preface

The Commonwealth of Virginia's concern over the economic state of its
oyster industry has resulted in an active program to define new management
goals and strategies. Economic analysis of alternative strategies has been
conducted at Virginia Tech for a number of years. This report, a synthesis
of that reseach work, was written to provide assistance to the Virginia
Marine Resources Commission in its continuing effort to develop an oyster
fishery management plan. The analysis included in this report draws
heavily on a Masters thesis project completed by Eric Thunberg. That work
received financial support from the Virginia Sea Grant College Program,
through the Virginia Gaduate Marine Sciences Consortium. A second work
product included in Appendix B of this report was completed by Oral Capps
and Leonard Shabman as a part of that same Sea Grant supported effort.

The integration of these models, the improvement of their data base
and the addition of auxiliary models was completed with the support of
funds from the NOAA 309 program administered through the Virginia Council
on the Environment. The development of this report would not have been
possible without that support.

Finally, the development of this report has benefited from extensive
discussions with the VMRC staff. Their insights on both the technical data
in the linear programming model and the general issues facing the
Commonwealth as it develops the oyster management plan have been
invaluable.

U. S. DEPARTMENT OF COMMERCE NOAA
COASTAL SERVICES CENTER
2234 SOUTH HOBSON AVENUE
ChARLESTON, SC 29405-2413

OCZ

V- Property of L:;c Library
* CO

.ii-

Table of Contents

Preface..............................

Table of Contents .........................

List of Tables......................... v'

I ~~List of Figures .........................viii

3 ~~~Section1ag

Introduction ..1........................

Section 1: Review of Oyster Grounds Law and Policy........2

Trends in Virginia Oyster Production .........2
Virginia's Oyster Grounds Management Apprroach ....3

Section 2: The Oyster Industry Today: Policy Alternatives . ... 5

U ~~~~~The OFMP Planning Process.............. 5
OFMP Goals and Objectves............... 5

Section 3: Management Alternatives................7

Regulatory Measures................. 8
Public Grounds Repletion Measures...........9

Section 4: Overview of Oyster Fishery Economic Models . ..... 9

U ~~~~~The Linear Programming Model .............10
The Econometric Demand System.............11
The Budget Simulator .................11
The Input/Output Model ................12
Section 5: An Analytical Framework for Policy Evaluation . ...12

I ~~~~~Policy Variables ...................13
Framework Structure .....u............14
* ~~~~~The Policy Evaluation Procedure............14

Section 6: Policy Simulations: Results..............17

Conditions Without a Management Plan .........17
The Baseline Condition ................19
A Seed Transplant Policy ...............21
A Seed Transplant and Seed Dredge Polcy........23
Seed Dredging All Areas .............23
Dredge or Tong Seed ...............25

Seed Dredging in Restricted Areas Only......25
Market Oyster Dredging Policy.............27
Market Oyster Dredging-Only ...........27
Dredging Market Oysters in Restricted Areas .... 29
Combinations of All Strategies ............31
Section 7: Discussion of Results and Recommendations.......31

I ~~~~~~Competitive Production Cost..............35
Repletion Budget ...................37
Income and Access to Baylor Grounds..........37
I ~~~~~~Shall Requirements ..................39
Recommendations and Disc'uss ion ............41

References ............................45

Apoendix Rpae

I ~~A. The Linear Programming Model .................49

Part 1: Technical Structure................51

Introduction ...................51

fl ~~~~~~The Linear Programming Model ...........52

The Objective Function............52
Model Activities...............54
Primary Activities ...........54
Secondary Activities ..........56
Tax Collection and Repletion Fund
Transfer ................60
Constraints....................60

The. Computer Model ................62

3 ~~~~Part 2: Estimation of Technical Coefficients .......75

Introduction....................76

I ~~~~~~~Objective Function Coefficients..........76

Public Grounds Repletion Costs........76
I ~~~~~~~~Private Grounds Production Costs.......78
Oyster Harvesting Costs ...........81

3 ~~~~~~~Estimation of Constraint Coefficients.......83

Estimation of Matrix Elements...........85

-iv-

I B. ~~Demand for Chesapeake Bay Oysters ..............89

General Background ................90

Chesapeake Bay Oyster Demand ...........92

H ~~C. The Budget Simulator.....................99

Budget Format...................100

D. Contribution of the Oyster Industry to the Virginia
3 ~~~~Economy and Tax Base.....................107

Introduction ...................108

3 ~~~~~~Measuring the Regional Economic Impact of
the Oyster Industry ...............108
3 ~~~~~~~Virginia's Commecial Oyster Industry.......110

The Multiplier Effects of Virginia's
3 ~~~~~~Oyster Industry .................112

Estimated Impacts of the Oyster Industry. ....112

Future Research Needs ..............112

Indirect Tax Revenue from the Oyster Industry.. 113

U~~~~~~~~~~~~V

List of Tables

Table Daze

Table 1: Summary of Policies Included in Policy Scenarios ..... 18

Table 2: Pre-OFMP Conditions .................................. 20

Table 3: Scenario 2: The Baseline Condition An Increased
Repletion Budget With No Change in Repletion Strategy 22

Table 4: Scenario 3: A Seed Transplant Policy ................. 24

Table 5: Scenario 4: Seed Transplant and Seed Dredge All
Seed Beds ............................................ 26

Table 6: Scenario 5: Seed Tansplant and Seed Dredge or Tong... 28

Table 7: Scenario 6: Seed Transplant and Dredge Seed Except
In The James River ................................... 30

Table 8: Scenario 7: Seed Transplant and Dredge All
Market Oysters ....................................... 32

Table 9: Scenario 8: Seed Transplant and Harvest Market
Market Oysters by Dredge or Tong ..................... 34

Table 10: Scenario 9: Seed Transplant, Tong or Dredge Market
Oysters and Dedge All Seed .......................... 36

Table 11: Scenario 10: Seed Transplant, Dredge o Tong Market
Oysters and Dredge or Tong Seed ...................... 38

Table 12: Scenario 11: Seed Transplant, Dredge or Tong Market
Oysters and Dredge Seed Except in the James ......... 40

Table 13: Comparison of Policy Scenarios to the Baseline
Condition ........................................... 42

Table 14: Shell Requirements and Labor Requirements ........... 44

Table A. 1: Description of Model River Systems ................. 53

Table A.2: Feasible Shell-Harvest River Systems ............... 59

Table A.3: Public and Private Growing Areas ................... 61

Table A.4: Summary of Primary Activities ...................... 63

-vi-

Table A.5: Public Grounds Seed Areas .......................... 66

Table A. 6: MSX Compatability by Seed Origin and Receiving
System ............................................. 67

Table A. 7: River System Compatability by MSX Type ............. 68

Table A. 8: Tableau Key ........................................ 72

Table A.9: Annual Harvest Cost Budget ......................... 82

Table B.i: Demand Estimation: Chesapeake Bay Oyster Industry.. 96

Table B. 2: Analytically Derived Reduced Form Equations
for Chesapeake Bay Oyster Demand ................... 98

Table C.1: Budget Spreadsheet Format for a Representative
Private Grounds Planter ............................101

Table C.2: Per Acre Returns to Fixed Costs and Management for
a Representative Private Planter ...................102

Table C.3: Budget Spreadsheet Format for a Representative
Public Grounds Hand Tonger .........................103

Table C.4: Annual Returns to Fixed Costs and Management for
a Representative Public Grounds Hand Tonger ........104

Table C.5: Spreadsheet Format for a Public Grounds Tonger .....105

Table C. 6: Annual Returns to Fixed Costs and Operator Labor
for a Representative Public Grounds Dredge
Harvester ..........................................106

Table D.1: Multipliers for the Virginia Seafood Industry
from the Virginia Input-Output Model ...............114

Table D.2: Example of Calculation of Output and Value-Added
Effects ............................................ 115

Table D.3: Economic Impact of the Oyster Industry .............116

Table D.4: Virginia State Tax Revenues as a Percent of
Gross State Product ................................117

-vii-

List of Figures

Figure page

Figure 1: Policy Evaluation Flow Diagram ..................... 15

Figure A. 1: Model River Systems .............................. 55

Figure A. 2: Diagram of Model Activities ...................... 57

Figure A. 3: Cultural Technique Feasibility ................... 64

Figure A. 4: Model Tableau .................................... 69

Figure A.5: Fund Transfer Submatrix .......................... 71

Figure D.1: Product Flows in the Virginia Seafood Industry ...111

-viii-

An Evaluation of Alternative Strategies for Virginia Oyster
I ~ ~~~Grounds Management: Economic Considerations in Policy Design
3 ~~~~~~~Leonard Shabman and Eric Thunberg

Introduction

Throughout its history the oyster fishery has played an important role
inVirginia's coastal economy. Accounting for fifteen to twenty percent of
the total value of all commercial species landed in Virginia, the oyster
fishery is a source of income and employment for thousands of tidewater
I ~ ~residents (JLARC 1977). It is because of its importance to the tidewater
region that the oyster industry has been a subject of law and public policy
3 ~~since colonial times. (Santopietro, 1986).

Once the single largest producer of oysters in the United States, the
Virginia oyster industry has been in a steady decline since the early
1960's. Public and private concern over the industry's decline prompted
the State's legislature to review the condition of the oyster fishery and
the State's role in its management. In response to its findings the 1984
legislature directed the Virginia Marine Resources Commission (VMRC) to
I ~ ~develop and implement an oyster fishery management plan (OFMP). The State
set the plan's goal to maximize the biological, economic and social benefit
from the harvest and utilization of market oysters. To achieve this end
I ~ ~the VMRC can employ management alternatives that include harvest season,
harvest gear and entry regulations in combination with alternative public
investment strategies (shelling and oyster seed transplanting) to enhance
3 ~~harvestable populations. Any alternative strategy would have a different
impact on the cost of producing and harvesting oysters in the State and
upon the VMRC budget. The task before the VMRC is to select a mix of
* ~~policies that (1) satisfies the production goals of the OFMP, (2) remains
* ~~within the VMRC budget and (3) is acceptable to the State's harvesters and
processors. Balancing these three objectives is the management challenge.
3 ~~~Seeking to address this mangement challenge, this report provides a
summary of a series of research projects conducted at Virginia Tech in the
Department of Agricultural Econo mics under grants from NOAA, VHRC and Sea
I ~ ~Grant. While each project had an independent scope and purpose, the body
of empirically developed models may be used in an integrative fashion to
evaluate the VMRC's management alternatives. This integrative work was the
3 ~~product of special funding provided through the NOAA, Section 309 program.

The objective of this report is to present an analytical framework
that has been developed to evaluate the economic impacts of oyster grounds
I ~ ~policy alternatives and to draw implications for policy design based on
management strategies identified in the September 1986 draft overview of
the OFMP. This report is organized into seven sections. First an overview
3 ~~of the oyster industry and oyster management policies is presented. The
second section discusses the current state of the oyster industry and the
goals and objectives of the fishery management plan. Section three

I~~~~~~~~~~~~~~I

describes the management alternatives available to the VMRC. The fourth
section provides ~an overview of oyster fishery economic research and
describes the economic models that are used to evaluate the VMRC' s
management options. Section five presents an analytical framework for
evaluating the effectiveness of alternative management strategies. Section
six identifies the management strategies to be evaluated and presents the
results of the analysis. The final section draws conclusions and
3 ~~recommendations for oyster grounds policy design.

Section _J..0: Reviw o.f Virginia Oyster Grounds Law and Policy

I ~~~The Virginia oyster, Crassostrea virzinic-a is a sessile mollusk
attaching itself to any firm clean substrate. Oysters may be found in
intertidal zones and waters up to and sometimes exceeding 25 feet in depth.
I ~ ~A filter feeder, the oyster subsists on nutrients extracted from the water
column by passing water over its gills. The oyster's ability to select its
food and extrude unwanted materials allows it to survive in waters carrying
high silt and turbidity loads characteristic of estuarine environments.
Reproduction is triggered by temperature (Kennedy and Breisch, 1981). In
Virginia spawning begins in early July and continues into September as long
as Bay temperatures do not fall below 20 degrees celsius (Merritt, 1977).
I ~ ~Larval oysters spend their first few weeks in a free-swimming state until
they settle to the bottom and attach themselves whereupon they are termed
spat. It is at this point that the availability of a hard surface is
I ~ ~paramount. The productivity of a river system can be enhanced by placing
oyster shells, the preferred material, on the growing bottom just prior to
the spawning period.

I ~~~The oyster survives best in estuarine conditions where salinities
range between 5 and 35 parts per thousand (ppt), (Haven et. al. , 1981).
Oyster growth is most rapid in salinities exceeding 15 ppt. Unfortunately,
I ~ ~salinities of this level are favored by the oyster's principal enemies:
the disease MSX and the oyster drill (a snail-like organism). Mortalities
due to these two enemies are particularly high among spat and yearling
I ~ ~(known as seed) oysters. In waters below 15 ppt. in salinity, survival of
young oysters is greatly enhanced, however, growth rates are slowed and
oyster growth can be stunted due to overcrowding. When harvested and
transported to a different river system, however, growth resumes and the
oyster will reach the 3 inch legal size quickly. The interesting result is
that the oyster grounds in Virginia can be divided into two distinct types,
those that produce large amounts of seed and undersized oysters and those
I ~ ~that are productive of market oysters. Moreover, the location of these
grounds is well known and has changed little over time enabling oyster
biologists to identify "best" management strategies for oyster production
I ~ ~by river system (Haven et. al. , 1981).
Section 1.JA: Trends in Virzinia Oyster Production

U ~~~By today's standards market oyster production in Virginia at the turn
of this century was large, averaging over five million bushels per year

1 ~~~~~~~~~~~~~-2-

over the period 1890-1925 (Haven et. al. , 1981). Although data is sketchy
there exists sufficient evidence to suggest that oyster production prior to
this time was even greater. Due to such heavy exploitation of Virginia's
oyster resources biologists noted as early as 1881, 1910 and 1912 that the
I ~ ~natural oyster rocks were significantly depleted (Haven et. al. , 1981). By
1930 total oyster production had dipped to 1,686,914 bushels. From 1930 to
1960 harvests were relatively stable ranging between 1.5 and 3.5 million
I ~ ~bushels but never approached production levels experienced prior to 1930.
Over the last twenty-five years a dramatic decline in oyster production has
occurred following the appearance of MSX disease in 1960. M4SX, an oyster
disease active in high salinity waters, had a devastating affect on
privately leased growing bottoms. Additionally, worsening economic
conditions throughout the 1970's forced the abandonment of Virginia's most
productive private oyster beds. Public grounds harvests were also
3 ~ ~adversely affected by MSX and hurricane Agnes in 1972. The latter event
devastated brood stock in many of the lower salinity portions of the Bay
fundamentally altering the level and distribution of setting in the
I ~ ~affected river systems.
Section 1.2: Vir2inia's Oyster Grounds Management Anproach

* ~~~As a matter of policy Virginia maintains a dual approach to oyster
grounds management. Such an approach was initiated upon the completion, in
1896, of a survey of the State's natural oyster bottoms. The survey, known
as the Baylor Survey, delineated naturally productive oyster rocks on the
basis of the presence of live oysters or oyster shall. Bottoms identified
as being naturally productive were designated as Public grounds and as
I ~ ~stated in Article XI of the Constitution of Virginia:
The natural oyster beds, rocks and shoals in the waters of this
State shall not be leased, rented or sold, but shall be held in
trust for the benefit of the people of this State . ..
Bottoms not so designated were made available to private individuals for
I ~ ~the express purpose of oyster culture. The practical importance of the
Baylor Survey was to establish the limits of the natural oyster resources
* ~~and the State's stewardship over them.

Prior to the Baylor Survey all harvesters were subject to several
legal measures regulating the taking of oysters. These measures included
gear, entry (residency, licensing and season restrictions), and harvestable
size regulations, required the culling of live oysters from shall at the
time of harvest, and restricted the removal of oyster shells for road
paving and the manufacture of lime (Santopietro, 1986). I~t is important to
I ~ ~note that by 1900 all of the above harvest restrictions were implemented
and that the same regulations exist today in very much the same form. The
harvest laws apply only to watermen, as the Chesapeake Bay harvesters are
I ~ ~called, working the public grounds. The production and harvest of oysters
on leased grounds is not subject to the above regulations.

I ~~~~~~~~~~~~~-3-

An additional measure employed by the State is the Oyster Repletion
I ~ ~Program (ORP). Through the application of aquacultural techniques the
objective of the ORP is to influence the supply of seed and market oysters
available for harvest. This objective is accomplished by planting shell on
public seed beds and growing areas or transplanting seed from the James
River to areas better suited for growth. From its authorization in 1928 to
1960, however, it is doubtful that the ORP had any affect on harvestable
populations as repletion effort varied greatly and little attention was
I ~ ~paid to biologically optimal timing or placement of shell. It was not
until after the appearance of MSX that repletion effort was undertaken with
earnest, and greater care was taken to insure that shells were placed on
bottoms most likely to receive a set of larval oysters.
Since 1960, the ORP has functioned as a disaster relief program and as
an on-going oyster bed replenishment program (Baker, 1977). In any given
repletion season both functions are served as some repletion effort may be
targeted for river systems experiencing natural disaster (such as
freshwater kill due to heavy rains), while the remainder of the ORP budget
I ~ ~may be used in replenishment of oyster bed substrate depleted during normal
harvesting. The Virginia Marine Resources Commission' s success in
administering the ORP to achieve its objectives is difficult to assess due
I ~ ~to data deficiencies. Studies in 1977 and 1983, however, concluded that
the ORP's impact was positive (JLARC, 1977; JLARC, 1983). In spite of a
positive assessment of the ORP's success both studies pointed out
3 ~~deficiencies in the ORP and the State's management of public oyster
grounds.

The 1977 JLARC report documented the decline of the oyster industry
and cited the need to consider policy alternatives to enhance production.
The authors concluded that production increases would be feasible, but that
an expanded state role for management of the oyster grounds would be
I ~ ~required. The 1983 JLARC study examined several alternative management
strategies and simulated their impacts on oyster production through 1990.
The authors found izhat unless current approaches to oyster grounds
management were changed the industry would continue its current decline.
They further concluded that even if existing harvest or leasing regulations
were changed only small increases in oyster production could be achieved
unless current repletion strategies were also changed. This finding
I ~ ~reveals the interdependence of legal and repletion strategies in oyster
grounds management. The authors recommended that: 1) the state modify
existing approaches to oyster grounds management, 2) the state should
I ~ ~incorporate the use of economic modeling as a fisheries management tool to
assess the impact of alternative strategies, and 3) the state should
implement an oyster fishery management plan that employs the most effective
I ~ ~harvest and repletion strategies subject to technical, economic and
political feasibility. In response to these recommendations the 1984
General Assembly mandated that VMRC develop and implement such an oyster
3 ~~management -plan.

I ~~~~~~~~~~~~-4-

Section 2.0: The Oyster Industry Today: Policy Alternatives

The State legislature mandate to VMRC was to develop and implement an
OFMP and set as its goal to maximize the biological, economic and social
benefit from the harvest and utilization of the State's oyster resources.
The legislature -declined,' however, to make any recommendations or
provisions as to how this was to be done. In meeting the state's
requirement VMRC, in 1985, began to take steps toward development of an
OFMP.

Section .1: The OFMP Planning Process

The development of the OFMP has taken place in four phases. In the
first phase a Fisheries Management Advisory Council (FMAC) was formed with
members appointed by the commissioner of VMRC. The advisory committee is
made up of private citizens active in and knowledgeable about the seafood
industry. The Commissioner also appointed scientific advisors including
economists, biologists and management specialists to assist the committee.
While VMRC has the ultimate responsibility for the plan, the purpose of the
FMAC is to provide recommendations and citizen expertise to VMRC with
regard to OFMP provisions, goals and strategies.

The second phase in the plan's development was for the FMAC and VMRC
to meet and discuss in a broad sense what the objectives of the plan should
be. Following this, VMRC staff drafted a management plan with stated
goals, objectives and proposed management options to meet the plan's goals.
The third phase in the OFMP development process took place over the next
several months as FMAC and VMRC met, discussed and revised the proposed
management plan. Currently the OFMP is in its fourth phase, that of
evaluating the impact of alternative policies and choosing the mix of
regulatory and repletion strategies that most effectively meets the OFMP
goals and objectives. The final phase in the OFMP process will be the
finalization and actual implementation of the plan.

Section 2.2: OFMP Goals and Objectives

Selection of the "best" mix of management alternatives requires that
some standards or criteria be set against which each management strategy
may be measured or compared. In this case the appropriate standards should
be consistent with the stated goals of the OFMP and current legislative and
executive concerns.

The stated goal and objectives of the OFMP are as follows:

Goal:
To acheive and maintain a level of Virginia's Public
and Private ground oyster stock to generate the greatest
possible biological, sociological and economic benefits from
their harvest and utilization.

-5-

Obj ectives:
I ~ ~~1. By 1993, increase public grounds prouction to at least
700,000 bushels (20% above the ten year average). To the
extent possible, reduce inter-annual fluctuations in public
3 ~~~~product ion.

2. Assist private oyster producers to ensure an increase in
private grounds production to 700,000 bushels (the
approximate 20 year average) by 1995.
3. Ensure the collection of biological, sociological and
economic data, as well as, fisheries statistics to monitor
and evaluate the effectiveness of management measures.

4. Implement the Marine Resource Commission's regulatory
I ~ ~~authority to impose effective and timely managment measures.

5. Support water and habitat quality standards necessary for
natural production. Encourage reclamation, where practical,
of condemned oyster grounds through VMRC cooperation with
the State Water Control Board and other State agencies.
(Source: Overview OFMP September 1986, p. 17-18)
In addition, concerns have been expressed with the legislative and
3 ~~executive branches of State government (JLARC, 1983) that:

1. The goals and objectives of the OFMP should be acheivable
3 ~~~without undue increases in State funding requirements.

2. The repletion program should be self financing with no
unwarranted State subsidy to private planters or Baylor
I ~ ~~grounds harvesters.
The objectives and concerns listed above, while not amenable to analysis as
* ~ ~sated, form the basis of a framework within which specific performance
criteria may be set for evaluation of alternative management policies.

To cast the above objectives and concerns into a framework that is
amenable to analysis it is necessary to establish a conceptual analytical
approach. First, it must be recognized that is is not possible to maximize
two or more policy goals simultaneously. The simultaneous maximization of
biological, socoiological and economic benefits of the oyster resource,
therefore, is not an operational policy goal. Instead, it is necessary to
identify a single policy goal while setting minimum satisfactory attainment
3 ~~levels for the remaining goals.

Given that the impetus behind the OFMP mandate was declining oyster
production, it is logical that increased -oyster production be selected as
the primary policy goal. The remaining objectives and legislative concerns
then become constraints or conditions under which increased, oyster

3 ~~~~~~~~~~~~~-6-

production is to be acheived. Within this context the policy goal of the
I ~ ~OFMP can be restated in the following general way:
The State should seek to implement management
I ~ ~~strategies to increase oyster harvests above their currently
depressed levels, subject to the following conditions.
i. Total production should be divided between public
I ~ ~~~and private grounds.

ii. The production goal should be attained in the least
costly manner possible, recognizing the increasingly
competitive national oyster market.

iii. Repletion program expenditures should stay within
a predetermined annual repletion budget.
iv. The repletion program should be self sufficient,
with no unwarranted state subsidy to private planters or
Baylor grounds harvesters.

I ~~~~v. Regulatory measures should assure continued access
to the Baylor grounds for those watermen who wish to earn
income from oystering.
vi. Regulatory measures should assure watermen an
income sufficient to permit their continued participation in
* ~~~the oyster fishery.
Stated in this manner a single policy goal is identified and the
conditions under which every alternative management strategy are to meet
I ~ ~that goal are clearly defined. Furthermore, these policy constraints
derive from the goals and objectives of the OFMP and from the legislative
concerns listed previous ly*.
Section 3.0Q: Management Alternatives
The preceding discussion pointed out the specific goal and policy
I ~ ~constraints that are implied by the broadly stated OFMP goal. Of the many
management options available to VMIRC, it is unlikely that any single policy
will be capable of satisfying all the management conditions set for the
3 ~ ~OFMP. Rather, it is likely that a mix of regulatory and management
strategies will be required to attain OFMP goals at levels considered
I~~~~~~~~~----------
*Note, however, that the OFMP objectives 3, 4 and 5 are excluded.
Objectives 3, 4 and 5 are administrative in nature and are assumed to be
carried out in addition to the specific management strategies that are
implemented to meet the oyster production goal.

I ~~~~~~~~~~~~~-7-

satisfactory by VMRC managers and industry representatives alike.

Section 341: RegulatoryZ Measures

3 ~~~The policy alternatives from which VMRC may choose are either
regulatory in nature or involve direct public investment in the oyster
fishery. VMRC's regulatory options fall into three broad categories,
market oyster-harvest regulations, bottoms designation, and seed oyster
harvest regulations. Market oyster harvest regulations include: minimum
size, open season, gear, quotas and restricted entry. These proposed
regulations are not in themselves a departure from traditional approaches
I ~~to regulating the harvest of oysters in Virginia. The proposed means of
administering these regulations, however, represents a significant change
in oyster grounds management policy. In order to allow for increased
I ~ ~management flexibility, the 1985 Legislature granted VMRC the regulatory
authority to implement all harvest related restrictions as they deemed
appropriate. Prior to this action all harvest regulations were
legislatively set limiting VMRC's ability to respond quickly to changes in
environmental considerations. Today, rather than representing limits or
legal confines within which VMRC must work, the setting of harvest
regulations is an important management tool to be molded and used as
management circumstances dictate.
The purpose of bottoms designation policies is to increase the
quantity of harvestable acreage for public hand tongers or to increase the
number of acres of growing bottom that are available for lease by private
individuals. QFMP provisions propose to increase harvestable acreage by
opening formerly condemned public oyster beds. In the event of shellfish
contamination the Bureau of Shellfish Sanitation will close the affected
bottoms to all harvesting. VMRC proposals include reviews of condemned
oyster grounds to assess the possibility of reopening condemned oyster
I ~~grounds that are no longer contaminated. Policies aimed at increasing
leasable acreage include the opening of nonproductive public grounds for
lease and requiring proof of active cultivation on oyster grounds currently
under lease. The former policy would represent a radical departure from
N ~~traditional approaches to oyster grounds management and would face
considerable opposition (Santopietro, 1986). Proposed proof of use
policies would prohibit the nonproductive holding of leases, thereby,
either forcing the choice between production or abandonment. The latter
choice, it is believed, will increase the number of oyster grounds
3 ~~available to individuals who are willing to plant seed oysters.

Although, in principle and application, seed oyster harvest
regulations are similar to market oyster harvest regulations they serve at
I ~~least one distinctly different purpose. The purpose of proposed seed
harvest regulations is to reduce the cost of harvesting seed. The
important consequence being that ORP costs would be reduced and lowered
seed costs might attract increased investment in private oyster grounds
production. The most important seed oyster harvest policy considered for
the OFMP is permitting the use of a dredge on specifically designated seed
I~~~~~~~~~~~~~~8

beds. Currently, the only harvest gear permitted on public grounds seed
beds is the hand tong. Permitting the more labor-efficient dredge would
substantially reduce seed harvest costs and seed prices.

Section 3.2: Public Grounds Reuletion Measures

In addition to regulatory policy measures, VMRC is also authorized to
oversee direct public investment in the oyster fishery, i.e. the oyster
repletion program. Although, management policy toward administration and
the basic aquacultural techniques used in the program are unchanged, OFMP
provisions represent significant changes in VMRC's approach to public
U ~ ~grounds repletion. The important change in oyster grounds repletion policy
called for in the OFMP is the evaluation of alternative repletion
techniques to identify the most cost-effective mix of repletion activities.
I ~ ~To make such an evaluation possible increased use of economic and
biological analysis is called upon, and to support these efforts OFMP
provisions emphasize the need for improved and expanded data collection.
It is in these three provisions that the proposed management plan marks a
change in ORP policy. For the first time the utility of a formal analysis
of the repletion program has been recognized.

I ~~Section 4.0O: Overview of Oyster Fishery Economic Models

Beginning in January 1980 and continuing through December 1985, the
I ~ ~Virginia Sea Grant consortium supported a broadly scoped fishery economic
studies program, focused upon the Chesapeake Bay fisheries. The first
project lasted three years and was entitled "Alternative Future Directions
for Viginia' s In-Shore Fishery." One product of this effort was the
econometric model results used to develop the oyster management planning
recommendations of the 1983 JLARC study. Following from this study a
second project, more clearly focused upon the oyster fishery was funded by
I ~ ~Sea Grant. The study, "Evaluation of Public Policy Options for Increasing
Virginia Commercial Oyster Harvest" was the basis for the development of
the models described in Sections 4.1 - 4.4. By 1985, when the OFMP
I ~ ~planning process was initiated these various economic models were, in large
part, developed. Given the OFMP mandate to consider economic factors in
the plan evaluation process the potential for use of these models in the
OFMP process seemed to be significant. However, a realistic appraisal of
their usefulness identified three potential problems.
First, some aspects of the modeling had not yet been reviewed by
I ~ ~oyster management specialists. Second, some of the model data bases could
be made more current. Third, the separate models had not been intergrated
into a complete analytical system. Therefore, during late-1985 and 1986
I ~~actions were taken to address each of these three concerns. Financial
support for this effort was provided by VMRC and by the Virginia Council on
the Environment through the NOAA Section 309 program. The structure and
functioning of the analytical system developed under this funding is
described in Section 5. The separate models, (a linear programming model,
an econometric demand system, a budget simulator and an input/output

I ~~~~~~~~~~~~~-9-

model), *that comprise the system are breifly described below. A detailed
treatment of each of the models can be found in the technical appendices
included in this report.

Section 4. 1: The Linear Programminiz Model

The first model used in policy evaluation is a Linear Programming (LP)
model. LP models, a subset of all mathematical programming models,
I ~ ~describe an economic system in terms of linear -equations. The economic,
biological and political relationships contained within the model's
technical structure are assumed to be linear. LP is a - mathematical
3 ~~technique that optimizes an objective function subject to limits of
resource availability and technical feasibility. Specifically, the LP
model described here minimizes the public plus private cost of producing
and harvesting a prespecified level of market oysters (Thunberg, 1985).
Framed in this way the policy goal of the OFMP is captured through the
preset harvest level and the policy constraint that the goal be attained in
the least costly manner is also incorporated. The model captures the
I ~ ~remaining policy considerations by placing restrictions on the model's
ability to acheive the production goal. In a LP context these restrictions
take the form of constraints on the model's ability to select specific
U ~ ~management strategies or by placing limits on the availability of resources
that are needed in order to acheive the management goal. For example, the
requirement that the repletion program stay within a specified annual
budget is modeled by placing a limit on the availability of repletion funds
for each repletion season. For a complete description of the model's
activities, and constraints see Appendix A, Part 1.

I ~~~The technical information incorporated into the LP model is reflective
of specified legal and repletion policies existing as of 1983. The
technical information was initially developed under a Sea Grant funded
I ~ ~project. Revisions to the model were made through consultations with VMRC
management personnel under the Section 309 funding. A detailed discussion
of the technical information included in the model can be found in Appendix
A, Part 2. Proposed management strategies are evaluated by altering the
models technical information to reflect the changed management environment.
For example, a regulatory change permitting the use of a new harvest gear
would alter the cost of harvesting oysters. To examine the economic
1 ~ ~impacts of such a management policy the analyst must identify the magnitude
of the change in harvest cost and then adjust the model to reflect the new
information. Solving the model yields an estimate of the cost and
I ~ ~repletion funding savings or increases associated with the new gear
regulation. In addition to repletion and cost information the model's
output determines tax collections and production levels by river system on
public and private grounds. The LP model, therefore, provides quantitative
estimates of the mix of market oyster production between public and private
grounds, the total cost of acheiving the harvest goal, the total amount of
repletion funding required to acheive the goal and the amount of tax that
is collected to offset the repletion progam costs.

* -~~~~~~~~~~~~10-

I ~~Section 4. 2: The Econometric Demand Sy~stem

Using the production levels determined by the LP model associated with
I ~ ~a policy change, oyster ex-vessel, wholesale and retail prices can be
predicted using a system of demand equations. The demand system expresses
oyster prices at each level of the marketing chain as a function of
Chesapeake Day and Gulf State harvest, U. S. consumption of oysters,
population, income, time trends and an index of intermediated goods and
services. Price changes, therefore, can not only be forecast for changes
in Virginia production but for changes in other factors affecting price as
well.
The demand model is estimated using simultaneous equations econometric
techniques. The resulting individual parameter estimates are interpreted
as being the change in the dependent variable associated with a one unit
change in any one independent variable holding all other independent
variables constant. Since changes in Virginia management policies have an
impact only on Virginia production, prices can be forecast by multiplying
the Virginia production regression coefficient by the new level of oyster
harvest while maintaining all other independent variables at their
I ~~specified levels. In this manner oyster ex-vessel, wholesale and retail
prices are estimated. For a detailed description of this demand model see
Appendix B. The new price estimates can then be applied to the production
I ~ ~levels determined in the LP solution to compute harvester, processor and
retailer total revenues. While most policy changes will likely have some
effect on oyster prices and, therefore, on industry total revenues, in
order to determine whether or not industry participants have been made
better off by a policy change it is necessary to examine net revenue

I ~~Section 4.3: The Budget Simulator

The third model used in the analytical framework, a budget simulator,
I ~ ~is used to evaluate changes in net revenue due to a policy change. The
budget simulator estimates changes in net revenues for a typical public
grounds harvester and individual harvesters and private planter given
specified relationships for each budget item. Changes in net revenues are
assessed by determining the impact of a proposed policy on particular
budget items. For example, private planter net revenue changes associated
with policies resulting in lowered seed prices would be attained by
I ~ ~adjusting the seed price coefficient in the budget simulator. Lowered seed
prices would likely result in increased production and lowered oyster
prices. Adjusting the private planter's budget to account for the input
and output price changes yields an estimate of net revenues for the private
planter attributable to the proposed policy. Comparing the new level of
net revenue to net revenues that would have existed without the policy
change provides the opportunity to determine whether or not industry
participants are made worse off or better off because of the policy change.
A desciption of the budget model can be found in Appendix C.

~~~~~~~~~~~~-II

I ~~Section 4.4: The Inrout/OutDut Model

The preceding effects, production, cost, repletion funding, price,
total and net revenue effects relate specifically to the oyster industry
I ~~alone. Recognizing that linkages between the oyster industry and other
sectors of the Virginia economy exist, it is important to consider how
management changes within the oyster producing sector affect the Virginia
economy as a whole. These state-wide effects can be evaluated through the
use of multipliers generated with an Input/Output (I/O) model.

I ~~~An I/O model describes the interdependencies that exist between
different sectors of an economy in terms of the purchases that are made by
each industry from all other industries. Oyster harvesters purchase fuel
I ~ ~and materials from several other economic sectors. These purchases in turn
become income to the sellers of the materials which may be spent on
purchases of seafood products or numerous other products produced in other
sectors. The total amount of economic activity in an economy induced by an
initial transaction in a single sector is termed a multiplier effect and
can be computed for each economic sector. Multipliers for the seafood
industry are available from a Virginia Input/Output model (Regional Science
I ~ ~Research Institue, 1982). Using these multipliers it is possible to
estimate changes in value-added, or addition to net State income. A second
adjustment allows computation of indirect tax revenues. Changes in value-
I ~ ~added and indirect tax revenues indicate state-wide net (gross sales less
production costs) changes in economic activity due to a management change
in the oyster industry alone. Indirect tax revenues are useful in
assessing whether or not the repletion program can be made self-financing.
For example, a change in management strategy may result in increased direct
oyster taxes but may still not recover all costs of the repletion program.
The collection of indirect tax revenues (tax revenues generated in sectors
I ~ ~other than the seafood sector) may be sufficient to recover the remaining
repletion budget deficit. F~r a discussion of this modeling work see
* ~~Appendix D.

Section 5.Q: An Analytical Framework for Policy Evaluation

The policy goal and constraints for the OFMP were stated in Section 2
I ~ ~and the management alternatives available to VM4RC to acheive that goal were
discussed in Section 3. In the absence of any policy constraints any given
management strategy or mix of strategies would likely be capable of
I ~ ~satisfying the OFMP policy goal. However, several policy constraints do
exist and selection of any given management alternative would have
different implications for the degree to which these constraints are met.
I ~ ~The "best" choice of specific policies and combinations of policies
requires, therefore, that VMRC be able to quantify the degree to which
proposed management strategies can satisfy the goal and policy constraints
set for the OFMP. In addition, informed selection of policy alternatives
requires that potential policy conflicts be identified. In this section,
an integrative analytical framework based on the economic modeling

I ~~~~~~~~~~~~-12-

described previously will be presented that is capable of accomplishing
I ~~this task. First, however, what is meant by an integrative approach?

The analytical framework is integrative in two senses. First,
I ~ ~economic modeling often incorporates biological and political information
in addition to economic considerations. The economic models described in
Section 4 are integrative because they treat the oyster industry as a
I ~~system in which limits to biological feasibility and political
acceptability are built into the models' structure. Secondly, the
analytical framework is itself integrative in that individual models
interface with each other as output from one model is used as input to
I ~ ~other models. The analytical framework is integrative, therefore, with
respect to the technical information incorporated into the system and with
3 ~~respect to the economic models themselves.

The analytical framework employs the four economic models described in
Section 4 to estimate the economic impact of a given management policy or
I ~ ~mix of policies. The aim of the framework is to obtain quantitative
estimates of changes in selected policy variables due to a change in
management strategy. Given these, the ability of each policy or policy mix
to meet the goal and various constraints of the OFMP can be assessed and
I ~~policy conflicts may be identified. Relative performance for different
management strategies can be evaluated and compared, enabling VMRC to
identify the most effective mix of management strategies for the OFMP. The
policy variables of interest are identified below.
Section 5..j: Policy Variables

The goal of the OFMP is to increase the production of market oysters.
The first policy variable of interest is, therefore, the total state
harvest resulting from any given management strategy. The policy
constraints of the OFMP! require estimates of several more variables.
First, the distribution of harvest between public and private grounds must
be estimated. Second, in order to satisfy the condition that the harvest
I ~ ~goal be attained in the least costly manner possible, the total cost of
acheiving the harvest goal must be estimated. Third, to determine whether,
under any given management strategy, the repletion program can be self
fnancing, the ORP funding requirements and the level of direct and
indirect tax collections must be estimated. Fourth, the income effects of
increasing harvest and changing management policies require that oyster
prices at the ex-vessel, wholesale and retail level be estimated. These
prices can then be used to estimate changes in industry gross revenues at
the harvest, wholesale and retail level and to evaluate changes in net
revenues for individual Baylor grounds tongers and private planters. The
I ~ ~last policy variable of interest is the total impact on the Virginia
economy associated with increased market oyster production.

I ~~~~~~~~~~~~-13-

Section 5.2: Framework Structure

The structure and flow of the analytical framework is depicted in
Figure 1. The policy variables that are being evaluated are shown within
the circles of Figure 1. The models used to estimate the impact on the
policy variables due to a policy change are represented by triangles. The
policy analysis is initiated by specifying values for the choice variables
I ~~listed in the rectangles of Figure 1.

The choice variables represent either policy choices or exogenous
factors. The allowable limit on repletion spending or the level of the
I ~ ~repletion tax, for example, set the conditions under which each management
option is to be evaluated. The availability of productive oyster beds or
the level of U. S. consumption of oysters represent limits of physical
resource availability or factors that are unaffected by the particular
management strategy under scrutiny.
Both aspects of the integrative nature of the analytical framework can
I ~ ~readily be seen in Figure 1. Political considerations and biological
realities are incorporated as choice variables that place' explicit
restrictions on the linear programming model that are then -carried
I ~ ~implictly through to the demand system and the budget simulator. For
example, political considerations may require that minimum historical
repletion patterns be maintained. These requirements place restrictions on
the LP model that affect the model's results and are implicitly carried
through the entir'e analytical system. The relationships, as indicated by
the direction of the arrows, between the models of the system can be seen
as output from the linear programming model is imposed on the demand
I ~ ~system. Likewise, the demand system output is imposed on the Input/Output
model and the Budget model. A final dimension included in the analytical
framework is the qualitative evaluation of specific components of the
I ~ ~output to determine the feasibility of the production levels determined in
the policy analysis. For example, prices may fall below that required for
private planters or public harvesters to operate profitably. Similarly,
production levels may exceed available processing capacity. In either
I ~ ~case, the analytical framework suggests the need to go beyond the
quantitative results of the models to offer a qualitative interpretation of
3 ~~the results before drawing management implications.

Section 5.2: The Policy Evaluation Procedure

The policy analysis begins by setting the conditions under which the
policy is to be evaluated. Specifically, the policy goals and restrictions
under which the management policy is to be evaluated must be set. The
policy goal set for the OFMP is to increase total State harvest to an
annual level of 1,400,000 bushels for a period of ten years. This total
harvest is to be allocated among public and private grounds in the
following manner: public grounds harvest must be at least 700,000 bushels
annually and private grounds production may not be greater than 700,000
bushels each year. Setting the distribution of production in this manner

3 ~~~~~~~~~~~~-14-

Choice Variables Models Outputs Evaluation of Output

Policy Choices

VMRC Repletion Budget Production by
Public Rocks Repletion Tax / ProgrammingN River System
Public OFHP Policy Options Public/

Technical Coefficients Will Private Planters Plant;
and Constraints Pricest Will Public Tongers Harvest?
Ex-Vessel /
Availability of Bottoms Demand Wholesale 1. Econometric Analysis
Productivity of Bottoms Systemail 2. Budget Analysis
Availability of Inputs
Political Constraints

Revenu
Exogenous Variables arvest Processing Feasibility
Processor
Regional Income Retail Budget Analysis
Consumption
Maryland Harvest
Gulf States Harvest

State Impacts
Value Added Will Consumers Purchase?
Production Costs Indirect Taxes
Pt Econometric Demand Analysis
Average Product Budget
Public Costs Simulat
Private Planter Costs

uNet Revenuel
Public Tonger
Private Plant

Figure It Policy Evaluation Flow Diagram

insures that any increased State management efforts will not unduly benefit
I ~ ~private planters at the expense of the public grounds harvester.
In addition to the restrictions placed an oyster grounds production
two conditions are placed on the oyster repletion program. First river
systems that have historically been shelled are required to continue to
receive a minimum annual level of repletion effort. This requirement
acknowledges political pressures that may exist to maintain repletion
I ~ ~activities in river systems that have historically received attention and
insures that at least some repletion effort be expended throughout the
Virginia portion of the Chesapeake Bay. The second condition set for
I ~ ~repletion program is that repletion funding is set at no more than
$12,500,000 for the entire ten year planning period to be allocated on an
annual basis as repletion conditions require. This condition permits the
* ~~flexibility to allocate the repletion budget in the most cost-effective way
while remaining within a prespecified spending limit.

Procedurally, any particular management strategy can be evaluated
I ~ ~subject to the above conditions in the following manner. The values for
the choice variables are selected and incorporated into the technical
information of each affected economic model. For example, a seed dredging
I ~ ~policy would alter the price paid for seed. The demand model and the I/O
model would not be affected by this policy change but the coefficients of
the LP model and the budget simulator would have to be adjusted to reflect
I ~ ~the new policy. The policy analysis begins by using the linear programming
model to evaluate the least cost mix of production activities that are
required to meet the goal, subject to the, specified constraints. The
results of the linear program determine total production between public and
private grounds, the total cost of acheiving the harvest goal, total
repletion program costs and total repletion tax collections. As indicated
in Figure 1, the total production figure is then incorporated into the
demand system to estimate ex-vessel, wholesale and retail prices. ,Given
these, gross revenues at each level of the industry can be computed.
Applying the multipliers from the I/O model to the gross revenue figures
results in an estimate of the net value of increased State-wide economic
market oyster harvest. Imposing the estimated prices and any changes in
production costs associated with a specific management policy on the budget
simulator results in estimated annual net revenues for a representative
public grounds harvester and net returns per acre for a representative
* ~~private planter.
The final phase in the policy analysis is a qualitative analysis
phase. In this step the results of the policy simulation are examined to
I ~ ~determine whether or not they seem reasonable in terms of the likely demand
for increased oyster production, and the likelihood that sufficient labor
will exist to harvest the increase in oyster production. The following
section provides the results of policy simulations based on this analytical
framework for the management alternatives discussed in Section 3.

* ~~~~~~~~~~~~-16-

Section 6.0: Policy Simulations: Results

In this section the results of eleven possible policy scenarios are
presented based upon the consideration of the policies and combinations of
policies that were discussed in Section 3. Table 1 provides a tabular
summary of these policies. The column headings of Table 1 indicate the
various policies that are available for consideration. The row labels
indicate the different policy scenarios that were evaluated. The cell
entries in Table 1 indicate the scenario and policy alternatives that are
evaluated in that scenario. If an "X" appears in any row then the
corresponding management policy, as indicated by the column heading, is
incorporated into that paticular analysis. For example, Scenario 6
evaluates a management strategy that: i) limits repletion spending to a
maximum of $12,500,000, ii) uses a repletion strategy that includes
shelling oyster grounds and transplanting seed, iii) permits the taking of
market oysters with hand tongs only and iv) permits the harvest of seed on
all seed beds and the harvest of seed with a dredge on all seed beds except
those in the James River. For each scenario the basic conditions as
discussed in Section 5.3 were set. In addition to these, it is assumed
that the mangement area concept as described in the September Draft Summary
of the OFMP is adopted for the entire Virginia portion of the Bay and its
tributaries (Draft OFMP Summary, 1986) Given these initial conditions the
economic impacts of each of the policy scenarios were evaluated, the
descriptions and results of which, are presented below.

Section 6.1: Conditions Without a Management Plan

In this scenario the production, price and cost conditions that
existed before the OFMP mandate were simulated. The average 1983, 1984 and
1985 production, repletion budget and tax collections were used to
establish the pre-OFMP conditions. The results of this scenario are
presented in Table 2.

These results do not replicate the pre-OFMP conditions in any year but
represent a composite for the 1983-1984-1985 time period. The results can
provide useful information for comparison with the alternative management
strategies incorporated in the OFMP. Under the pre-OFMP conditions total
harvest was 641,972 bushels with 341,870 and 300,103 bushels harvested from
public and private grounds respectively. The predicted prices associated
with this harvest were $0.937, $1.307 and $2.151 per pound of meat at ex-
vessel, wholesale and retail respectively.* The repletion program spent

*A conversion rate of 4.6684 pounds of meats per bushel was used to convert
bushels of oysters harvested to meat yields. This conversion rate
represents a weighted average based on reported conversion rates in the
NMFS Fisheries Statitics of the United States and harvest levels reported
in VMRC Annual Reports for the years 1960 to 1977 (NMFS, 1977; VMRC Annual
Reports, 1982).

-17-

Table 1
SLummary of Policies Included in Policy Scenarios

Repletion Budget Repletion Strategy Market Oyster Seed Oyster Harvest Policies
Spending Limit Plant Transplant Harvest Tong Dredge Dredge All Areas
Scenarios $12,500,000 Shell Seed Tong Dredge All Areas All Areas Except James River

Scenario I NA X X X
----------------------------------------

Scenario 2 NA X X X
----------------------------------------_____________________________________________________________________________________

Scenario 3 X X X X X
----------------------------------------_____________________________________________________________________________________

Scenario 4 X X X X X
----------------------------------------_____________________________________________________________________________________

Scenario 5 X X X X X X

Scenario 6 X X X X X X
----------------------------------------_____________________________________________________________________________________

Scenario 7 X X X X X

Scenario 8 X X X X X X
.............................................................................................................................

Scenario 9 X X X X X X

Scenario 10 X X X X X X X

Scenario 11 X X X X X X X
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

$828,257 and tax revenues from the public grounds oyster repletion tax were
$418,125 leaving a budget deficit of $410,132. The dollar value of the
State-wide addition to net income (value added) associated with the oyster
fishery under pre-OFMP conditions was estimated to be $2,359,671.70.
Indirect tax revenues were estimated to be 11.75% of value added or
�$277,261.34. Under pre-OFMP conditions, therefore, all but $75,875 of
repletion program expenditures were recovered either through direct or
indirect sources.

Total revenues at the harvester, processor, and retailer levels were
estmiated to be $2,808,531, $3,971,565 and $6,446,088 respectively. Net
returns to fixed costs and the operator for watermen working the public
grounds were estimated to be $1,252.69 per year. Private planters holding
high productivity beds (1,500 bushels per acre) were estimated to net
$1,502.87 per acre per year while planters holding moderate productivity
grounds (1,125 bushels per acre) earned $489.93 per acre per year and
planters on low yielding grounds (750 bushels per acre) were estimated to
lose $523.004 per acre per year.

The scenario described above typifies what the condition of the oyster
industry might be in the absence of a fishery mangement plan or any of the
policy changes such a plan might entail. The remaining scenarios assume
the existence of an oyster fishery management plan but in each case a
different set of management provisions incorporated in the plan is
evaluated.

Section k.2: The Baseline Condition

In this scenario the effect of increasing the repletion program budget
without any other changes in management strategy is evaluated. Therefore,
the repletion program uses a shelling-only strategy. This scenario will
be used as a baseline condition because it considers no management changes
other than an increase in repletion program expenditures. Indeed, the
decision to increase repletion spending has been made. The management
options considerd in the remainder of this section can be compared with
this baseline.

Initially a total repletion budget of $12,500,000 was given to the
model to be allocated across the ten year planning period considered in the
analysis. The annual harvest goal of 1,400,000 bushels, however, could not
be attained at this repletion program funding level. The repletion budget
was then increased and the models were resolved. The results of the
analysis are presented in Table 3.

Under a repletion strategy that relies on a shelling program alone the
annual harvest goal of 1,400,000 bushels is feasible only if repletion
program expenditures are $13,836,105 over the ten year planning horizon.
Total harvest is 13,748,040 bushels with 8,143,138 and 5,604,402 coming
from public and private grounds respectively. The total cost of acheiving
this level of production is $47,268,580. The total cost is split between

-19-

Table 2
Pre-OFMP Conditions

Total Harvest (Bushels) 641,972.0
Public Harvest 341,870.0
Private Harvest 300,103.0

Repletion Budget ($) 828,257.0
Tax Collections ($) 418,125.0
Budget Deficit ($) 410,132.0

Prices ($/Pound)
Ex-Vessel 0.93712
Wholesale 1.30717
Retail 2.15086

Total Revenues ($)
Harvester 2,808,531.85
Processor 3,917,565.07
Retailer 6,446,088.88

Net Revenues ($/Year)
Private Planter
750 (bushels/acre/year) -523.00
1125 (bushels/acre/year) 489.93
1500 (bushels/acre/year) 1502.87

Hand-Tonger 1252.69
Dredger NA

Value Added 2,359,671.70
Indirect Tax Revenues 277,261.34

-20-

public and private grounds in the amounts of $28,756,978 and $18,511,602
respectively. The predicted prices associated with the annual 1,400,000
bushel harvest goal are $0.85889, $1.23894 and $2. 10822 per pound at ex-
vessel, wholesale and retail respectively. These prices are below those
which exist in the pre-OFMP situation of Scenario 1. Total revenues over
the ten year period for harvesters, processors and retailers are estimated
to be respectively, $55,124,719.60, $79,516,841.60 and $135,308,405.50, an
increase over pre-OFMP conditions. Annual returns to fixed costs and
mangement for a representative public grounds hand tonger are estimated to
be $3,736.84, over two times the pre-OFMP figure. This figure is based on
the assumption that the increased harvest goal will, on average, double the
productivity of the oyster grounds and, therefore, double the productivity
of the average hand tonger. The change in productivity more than offsets
the reduction in prices resulting in a net increase in returns to public
harvesters. Net returns for private planters are lower under this
scenario, versus the pre-OFMP condition, because increased harvests result
in lower prices received and there are no offsetting cost savings realized
by private planters. Planters on high productivity grounds are estimated
to receive net revenues of $964.03 per acre per year while moderate and low
productivity grounds are estimated to yield net returns of $85.80 and
-$792.42 per acre per year. Over the ten year planning period the
increased harvest goal is estimated to result in an increase in State-wide
net income of $47,729,529 and indirect tax collections of $5,608,219.70.
These indirect tax collections repesent nearly 60% of the repletion budget
deficit of 9.5 million dollars.

Section 6.3: A Seed Transplant Policy

The previous scenario assumed that the only repletion activity was the
shelling of public rocks to produce market oysters. This scenario permits
the transplanting of seed subject to the requirement that minimum
historical shelling levels are maintained. Once these minimum repletion
requirements are met the choice of transplanting seed or planting shell is
based on the relative cost-effectiveness of each activity. The repletion
budget is set at a maximum of $12,500,000 to be allocated over the ten year
planning period. The results of this scenario are presented in Table 4.

The total harvest of market oysters is the same as that under a
shelling only repletion strategy. With the option of transplanting seed,
however, the distribution of production between public and private grounds
changes because with a seed-transplant strategy the public grounds become
more competitive with private grounds on a cost-effectiveness basis. Under
this strategy public grounds production increases to 9,248,037 bushels
while private grounds production decreases to 4,500,003 bushels. The seed-
transplant option also allows the harvest goal to be attained whithin the
limits of the repletion budget on expenditures of $12,050,558. With a
budget deficit of $7,140,125 the repletion program defecit amounts to
$1,531,906 when indirect tax collections are considered. Additionally the
total public plus private cost of producing the harvest goal declines to
$42,929,639. The total private grounds cost falls to $14,132,969 while

-21-

Table 3
Scenario 2: The Baseline Condition An
Increased Repletion Budget With No
Change in Repletion Strategy

Total Harvest (Bushels) 13,748,040.0
Public Harvest 8,143,138.0
Private Harvest 5,604,402.0

Total Cost ($)
Public 28,756,978.0
Private 18,511,602.0
Total 47,268,580.0

Average Cost ($/bushel)
Total 3.4382
Public 3.5314
Private 3.3030

Repletion Budget ($) 13,836,105.0
Tax Collections ($) 4,370,228.0
Budget Deficit ($) 9,465,877.0

Prices ($/Pound)
Ex-Vessel 0.85889
Wholesale 1.23894
Retail 2.10822

Total Revenues ($)
Harvester 55,124,719.6
Processor 79,516,841.6
Retailer 135,308,405.5

Net Revenues ($/Year)
Private Planter
750 (bushels/acre) -792.42
1125 (bushels/acre) 85.80
1500 (bushels/acre) 964.03

Hand-Tonger 3736.84
Dredger NA

Value Added 47,729,529.00
Indirect Tax Revenues 5,608,219.70

-22-

public grounds costs rise slightly to $28,796,670 as compared to the
baseline condition. The drop in private grounds costs and rise in private
grounds costs are attributable to the change in the distribution of
production on public and private grounds that takes place under this
scenario,
The production goal is the same as that in the baseline condition.
Since this is the case the predicted prices, total revenue figures, hand
tonger net revenue, value added and indirect tax revenues remain the same
for this scenario and will be the same for all the other scenarios
considered below. For this scenario net revenues for private planters are
* ~~also the same as the baseline condition because a seed transplant strategy
does not affect the cost or price conditions faced by private planters.

Section 6.4.0: A Seed Trans-plant and Seed Dredge Policy

A policy that permits the dredging of seed is expected to result in
3 ~~lowered seed prices resulting in increased private planting and possibly
lowering replIetion program costs as well. In the recent past all seed
dredging policies had been vigorously opposed by watermnen and viewed with
skepticism by VMRC managers as well. Recent survey work demonstrates,
however, that opposition to seed dredging may be weakening among watermen
(Santopietro, 1986). Additionally, in public hearings held in the fall of
1986 watermen expressed little opposition to the principle of seed dredging
I ~~under certain conditions. Recognizing the biological and political
concerns expressed by both management and watermen groups three alternative
seed dredging policies are evaluated.

Section 6. 4. 1: Seed Dredgin~ All Areas

In this scenario the only method of harvesting seed considered is
dredging. Under such a policy it is assumed that a seed bad, once
harvested, would have to be reshelled before it could be expected to be
productive again. To evaluate the effect of a seed dredge-only policy this
3 ~~assumption was incorporated into the technical structure of the linear
programming model. In addition to this assumption, it was also assumed
that a bushel of dredge seed would be less productive than a bushel of
tonged seed. Specifically, it was assumed that it would require two
bushels of dredged seed to match the productivity of a single bushel of
tonged seed. The harvest cost per bushel for dredged seed was set at $1.00
per bushel (twice that paid by the Potomac River Fisheries Commission in
U ~~1984). Allowing seed-transplant activities and setting the repletion
budget at $12,500,000 the results of the policy simulation are presented in
* ~~Table 5.

Under a seed dredge only policy the total cost of acheiving the
harvest goal is $43,990,092. The total private cost increases slightly
over that in Scenario 3 to $14,894,132 but so too does total private
grounds production at 4,777,712 bushels. The increase in private grounds
production compared to Scenario 3 occurs because the reduced cost of seed

I ~~~~~~~~~~~~-23-

Table 4
Scenario 3: A Seed Transplant Policy

Total Harvest (Bushels) 13,748,040.0
Public Harvest 9,248,037.0
Private Harvest 4,500,003.0

Total Cost ($)
Public 28,796,670.0
Private 14,132,969.0
Total 42,929,639.0

Average Cost ($/bushel)
Total 3.1226
Public 3.1138
Private 3.1406

Repletion Budget ($) 12,050,558.0
Tax Collections ($) 4,910,432.0
Budget Deficit ($) 7,140,125.0

Prices ($/Pound)
Ex-Vessel 0.85889
Wholesale 1.23894
Retail 2.10822

Total Revenues ($)
Harvester 55,124,719.64
Processor 79,516,841.68
Retailer 135,308,405.56

Net Revenues ($/Year)
Private Planter
750 (bushels/acre) -792.42
1125 (bushels/acre) 85.80
1500 (bushels/acre) 964.03

Hand-Tonger 3736.84
Dredger NA

Value Added 47,729,529.00
Indirect Tax Revenues 5,608,219.70

-24-

makes some private grounds relatively more cost-effective to the next best
public grounds production alternative. Note, however, the public grounds
cost is greater in this scenario than under any previous one because of the
assumed need to completely reshell the public grounds seed beds after each
harvest. This is also reflected in repletion program expenditures of
$12,500,000 required to meet the harvest goal.

Private planters are made better off with a seed dredge policy because
I ~ ~their production costs are reduced. The lower production costs are
reflected in increased annual per acre returns of $1,249.03, $663.55 and
$-507.42 for high, moderate and low productivity beds respectively. All
other economic impacts, prices, total revenues, tonger net revenue, value
added and indirect tax revenues remain unchanged.

Section 6.4.2: Scenario 5: Dredge or Tong Seed

Recognizing that repeated shelling for seed may be more costly than
necessary this scenario permits the choice between dredging or tonging seed
on a cost-effectiveness basis. The linear programming model was 'adjusted
to allow either seed dredging (under the conditions described above) or
tonging. Areas that should be dredged and areas that should be tonged are
I ~~determined from the results of the linear program. In principle, the
linear program makes this determination by comparing the cost per bushel of
seed produced for all the public grounds seed beds for dredged seed and
I ~ ~tonged seed. The model then selects the lowest cost combination of harvest
gear and river system in which to produce seed subject to limits of
technical feasibility and resource availability. All other policy
conditions remain the same as that for Scenario 4. The results of the
analysis are presented in Table 6.
Permitting the choice between dredging or tonging seed reduces the
I ~ ~repletion cost of attaining the harvest goal. The repletion budget deficit
is likewise reduced and when indirect tax revenues are considered the
repletion program is nearly self-financing with a modest budget deficit of
I ~ ~$271,035. The total public costs are reduced as compared to Scenario 4
even though the distribution of public and private grounds production
favors increased public grounds production. All other factors prices,
revenues, etc. are the same as they were in Scenario 4 and the baseline
condition.
Section 6.4.3: Seed Drdedgiun in Restricted Areas Only

The opinions expressed at the Fall public hearings reflected the view
that the establishment, by VMRC, of seed dredging areas in the Piankatank
I ~ ~and Great Wicomico would be acceptable to the watermen whereas similar
areas in the James would be strongly opposed. This view has its origin in
the fact that prior to VMRC's shelling efforts in the Piankatank and Great
Wicomico neither of these river systems had been productive of seed. Since
the VMRC was the agency to initiate the shelling watermen view the VMRC as
having property rights to the seed resources that were developed there. To

I -~~~~~~~~~~~~25-

Table 5
Scenario 4: Seed Transplant and Seed Dredge
All Seed Beds

Total Harvest (Bushels) 13,748,040.0
Public Harvest 8,970,328.0
Private Harvest 4,777,712.0

Total Cost ($)
Public 29,095,960.0
Private 14,894,132.0
Total 43,990,092.0

Average Cost ($/bushel)
Total 3.1997
Public 3.2435
Private 3.1174

Repletion Budget ($) 12,500,000.0
Tax Collections ($) 5,049,821.0
Budget Deficit ($) 7,450,179.0

Prices ($/Pound)
Ex-Vessel 0.85889
Wholesale 1.23894
Retail 2.10822

Total Revenues ($)
Harvester 55,124,719.64
Processor 79,516,841.68
Retailer 135,308,405.56

Net Revenues ($/Year)
Private Planter
750 (bushels/acre) -507.42
1125 (bushels/acre) 663.55
1500 (bushels/acre) 1249.03

Hand-Tonger 3736.84
Dredger NA

Value Added 47,729,529.00
Indirect Tax Revenues 5,608,219.70

-26-

reflect this policy environment this scenario permits the dredging of seed
only in the Piankatank and Great Wicomico rivers while allowing seed to be
hand tonged in all river systems. Seed transplanting is still considered
and the repletion budget remains at $12,500,000. The results of this
scenario are presented in Table 7.
Under a restricted area seed dredging policy the distribution of
production between public and private areas is the same as it was under
Scenario 5. This result indicates that limiting the areas in which seed
oysters may be dredged does not alter the relative cost-effectiveness of
public as opposed to private grounds production. Limiting seed dredging
I ~ ~does, however, increase the cost of attaining the harvest goal slightly to
$41,831,773 as compared to Scenario 5. The repletion cost associated with
restricting seed dredgeing is also increased over the unrestricted case.
I ~ ~$11,526,678. Under this scenario, however, public rocks oyster repletion
tax collections increase and when combined with indirect tax collections a
budget deficit of $756,718 results. Like the previous two scenarios
prices, revenues, net returns, value added and indirect tax revenues are
unaffected by the particular form of the seed dredging policy adopted.
3 ~~Section 6.,2.0: Market Oyster Dredging Policy

The management area concept includes the possibility that market
oysters areas may be identified and dredging of market oysters could be
I ~ ~permitted in specific portions of river systems on a permanent basis or in
response to changing resource abundance. Existing policy does permit the
dredging of market oysters on a limited basis in carefully designated
regions of the Day. The following analysis examines two different
management approaches to a market oyster dredging policy. In each case the
dredging of seed is not permitted but seed transplanting is allowed and the
repletion budget is set at $12,500,000 for the ten year planning period.
I ~ ~It is assumed in each case that if a market oyster bed is harvested by
dredging it will not become productive again until it is reshalled or
reseeded. The cost per bushel of harvesting market oysters by dredging 'Was
I ~ ~estimated using the budget analysis to be $1.60. For each scenario the
linear programming model was adjusted to reflect these changes.

Section 6. 5. 1; Market Oyster Dred~in2-Onlv

The first market oyster dredging policy to be examined is one that
would permit the dredging of market oysters on all oyster rocks in the
State. The results of this scenario are presented in Table 8.
Under this scenario it is assumed that all market oysters harvested on
public grounds will be harvested with a dredge. Permitting the taking of
oysters with a dredge results in a greater share of total production going
to the public grounds than under the baseline condition. The allocation of
production is, however, identical to that determined in Scenario 3, the
seed-transplant policy. The total public grounds cost of acheiving the
harvest goal decreases to $26,852,969 largely due to the decreased cost of

I ~~~~~~~~~~~~-27-

Table 6
Scenario 5: Seed Transplant and
Seed Dredge or Tong

Total Harvest (Bushels) 13,748,040.0
Public Harvest 9,248,036.0
Private Harvest 4,500,004.0

Total Cost ($)
Public 28,216,013.0
Private 13,473,271.0
Total 41,689,284.0

Average Cost ($/bushel)
Total 3.0323
Public 3.0510
Private 2.9940

Repletion Budget ($) 11,057,677.0
Tax Collections ($) 5,178,423.0
Budget Deficit ($) 5,879,254.0

Prices ($/Pound)
Ex-Vessel 0.85889
Wholesale 1.23894
Retail 2.10822

Total Revenues ($)
Harvester 55,124,719.64
Processor 79,516,841.68
Retailer 135,308,405.56

Net Revenues ($/Year)
Private Planter
750 (bushels/acre) -507.42
1125 (bushels/acre) 663.55
1500 (bushels/acre) 1249.03

Hand-Tonger 3736.84
Dredger NA

Value Added 47,729,529.00
Indirect Tax Revenues 5,608,219.70

-28-

harvesting the market oysters. The repletion budget, however, is nearly
exhausted because of the increased repletion effort required to maintain
the productivity of the public oyster rocks. Prices, total revenues, net
returns to private planting, value added and indirect tax revenues are the
U ~ ~same as they were for the baseline condition. Since it is assumed that all
harvest will be done with a dredge net returns to hand tongers is excluded
from consideration. Instead, using the budget simulator net returns to
fixed costs and the operator of a dredge boat are estimated to be
$21,309.52 per year. Private planter net revenues are the same as they
were in the baseline condition because the dredging of seed is not
* ~~permitted in this scenario.

Section 6.5.2: Dredgin~ Market Oysters ini Restricted Areas

It is doubtful that pursuit of an unrestricted dredging policy to
harvest market oysters would be politically acceptable or biologically
responsible. A more sound approach may be to permit the dredging of
oysters in limited areas of stock abundance or in areas where the
sustainability of natural production is questionable. Scenario 8 evaluates
the potential impacts of implementing a limited area market oyster dredging
I .~policy. The location and acreages of market oyster dredging areas and
tong areas were determined by using the results of the linear programming
model. As described earlier, the linear program selected the dredge and
tong areas by selecting the least costly combination of river system and
I ~ ~harvest gear to produce market oysters. Table 9 presents the results of
the analysis for this scenario.

Establishing market oyster dredge areas results in a higher portion of
I ~ ~total harvest going to the public as compared to the baseline condition or
any other scenarios examined previously. At $39,990,807 the total cost of
attaining the market oyster production goal is lower than that of any
I ~ ~scenario evaluated previously. The reason for this result is that the
majority of the public grounds production is attributable to the
transplanting of seed. It was assumed, upon the recommendation of VMRC
I ~ ~repletion officers, that only those areas that did not receive a natural
set would be candidates to receive transplanted seed. This condition is
precisely that which would make a particular market oyster bad a suitable
area for dredging. Since the majority of the public grounds harvest comes
from these dredged areas the total cost of acheiving the harvest goal is
lower than it would be otherwise.

I ~~~Permitting the dredging of market oysters in limited areas makes those
areas cost-effective relative to comparable private grounds. This result
c an be seen as the public grounds production in this scenario is greater
I ~ ~than that in the baseline condition or any other scenario previously
examined. In this scenario public grounds prodution in which oysters were
harvested with a drege were selected over alternative private grounds
because the harvest cost reduction afforded by the dredge policy made these
grounds more cost-effective than would be the case if they were tonged.
The repletion budget required to meet this public grounds production level

I ~~~~~~~~~~~~-29-

Table 7
Scenario 6: Seed Transplant and Dredge Seed
Except In The James

Total Harvest (Bushels) 13,748,040.0
Public Harvest 9,248,045.0
Private Harvest 4,499,995.0

Total Cost ($)
Public 28,317,594.0
Private 13,514,179.0
Total 41,831,773.0

Average Cost ($/bushel)
Total 3.0427
Public 3.0620
Private 3.0031

Repletion Budget ($) 11,526,678.0
Tax Collections ($) 5,161,741.0
Budget Deficit ($) 6,364,937.0

Prices ($/Pound)
Ex-Vessel 0.85889
Wholesale 1.23894
Retail 2.10822

Total Revenues ($)
Harvester 55,124,719.64
Processor 79,516,841.68
Retailer 135,308,405.56

Net Revenues ($/Year)
Private Planter
750 (bushels/acre) -507.42
1125 (bushels/acre) 663.55
1500 (bushels/acre) 1249.03

Hand-Tonger 3736.84
Dredger NA

Value Added 47,729,529.00
Indirect Tax Revenues 5,608,219.70

-30-

falls within the prespecified spending limit of $12,500,000 but exceeds the
I ~ ~amount at which the repletion program could be made self financing through
direct and indirect revenue sources. Since total public plus private
grounds production is equivalent to that in the baseline condition prices,
I ~ ~revenues, private planter. net revenues, value added and indirect taxes are
the same as that estimated for the baseline condition. Net returns to
fixed costs and management for a representative dredger are the same as
U ~ ~they were in Scenario 7. The cost and price conditions faced by public
grounds hand tongers have not changed and, therefore, estimated net returns
for an average hand tonger are equivalent to that in the baseline
* ~~condition.

Section 6.6: Combinations Of All Stratezies

I ~~~~The remaining policy simulations represent three diffferent
combinations of all the individual policies examined above. The first
combination permits the transplanting of seed, the dredging of market
3 ~~oysters in limited areas and allows the dredjing of seed oysters in all
seed areas. The second policy combination is equivalent to the first with
the exception that seed dredging is permitted in limited areas only as
determined by the results of the linear programming model. The last policy
combination is the same as the second except that seed dredging is not
permitted in the James river. The results of these respective policy
combinations are presented in Tables 10, 11 and 12.

Under all three policy combinations total State harvest is the same
and, therefore, prices, revenues, value added, indirect tax revenues and
net returns for public grounds harvesters are the same for each scenario.
Private planter net returns are the same for all three scenarios as well,
but net revenues are higher than in the baseline condition because of the
availability of dredged seed. In scenario 9 where all seed oysters are
I ~ ~harvested with a dredge total costs and rep ltion costs are higher than when
only limited dredging is allowed. Scenarios 10 and 11 attain very similar
results with respect to the allocation of production between public and
I ~ ~private grounds, the total public and private costs of attaining their
respective harvest levels and the repletion funding required to meet the
harvest goal. Direct tax revenues are slightly higher, however, in
I ~ ~Scenario 10 making it, of the three policy combinations, the one that comes
closest to resulting in a self financing oyster repletion program.

I .SeSction 2.0: Discussion of Results and Recommendations

This section will compare the results of the alternative model
simulations. The comparisons will be to the pre-OFMP condition (Scenario
I ~ ~1) and the baseline condition (Scenario 2). Recall that the baseline
condition begins with a recognition that the state is committed to an
expanded repletion program and to increasing oyster harvests as a result.
I ~ ~The challenge to VNRC is to utilize this increased budget authority to meet
the goals of the stated management plan. In Section 2. 2 the goal statement
was summarized as follows:

Table 8
Scenario 7: Seed Transplant and Dredge All
Market Oysters

Total Harvest (Bushels) 13,748,040.0
Public Harvest 9,248,043.0
Private Harvest 4,499,997.0

Total Cost ($)
Public 26,852,969.0
Private 14,077,380.0
Total 40,930,349.0

Average Cost ($/bushel)
Total 2.9771
Public 2.9036
Private 3.1283

Repletion Budget ($) 12,467,420.0
Tax Collections ($) 5,029,292.0
Budget Deficit ($) 7,398,128.0

Prices ($/Pound)
Ex-Vessel 0.85889
Wholesale 1.23894
Retail 2.10822

Total Revenues ($)
Harvester 55,124,719.64
Processor 79,516,841.68
Retailer 135,308,405.56

Net Revenues ($/Year)
Private Planter
750 (bushels/acre) -792.42
1125 (bushels/acre) 85.80
1500 (bushels/acre) 964.03

Hand-Tonger NA
Dredger 21,309.52

Value Added 47,729,529.00
Indirect Tax Revenues 5,608,219.70
Value Added

-32-

I ~~~~~The State should seek to implement management
strategies to increase oyster harvests above their currently
depressed levels, subject to the following conditions.

i. Total production should be divided between public
and private grounds.

ii. The production goal should be attained in the least
costly manner possible, recognizing the increasingly
* ~~~competitive national oyster market.

iii. Repletion program expenditures should stay within
3 ~~~a predetermined annual repletion budget.

iv. The repletion program should be self sufficient,
with no unwarranted state subsidy to private planters or
Baylor grounds harvesters.
v. Regulatory measures should assure continued access
to the Baylor grounds for those watermen who wish to earn
income from oystering.
vi. Regulatory measures should assure watermen an
income sufficient to permit their continued participation in
the oyster fishery.

The results of the model analysis give insights into the management
approaches needed to assure attainment of these goals. The discusion which
follows will draw upon the results reported in Section 6 and Table 13,
which is a summary of certain results from Section 6.
The modeling results suggest that a 1.4 million production target is
I ~ ~acheivable. However, this target can not be acheived within the expected
$12.5 million ten year repletion budget if the only repletion strategy is
shelling for market oysters. The management strategies which include at
I ~ ~least some seed transplanting will permit the production goal to be
attained within allowed repletion program spending limits.

Compared with pre-OFMP conditions, any management strategy will result
in reduced oyster prices but increased gross revenue, state economic
activity of over $47 million for the ten year period, and indirect tax
revenues on this economic activity of $5.6 million for the period occur
I ~ ~because of increased harvests. For the individual hand tonger the
scenarios all promise increased income from oystering compared with the
pre-OFMP condition. This occurs despite the price depressing effect of
I ~ ~increased production. For the private planter the same scenarios result in
OFMP implementation making them worse off financially, compared with pre-
OFMP conditions. This occurs in those cases where the OFMP induces a price

I ~~~~~~~~~~~~-33-

Table 9
Scenario 8: Seed Transplant and Harvest Market Oysters
By Dredge or Tong

Total Harvest (Bushels) 13,748,040.0
Public Harvest 9,646,834.0
Private Harvest 4,101,206.0

Total Cost ($)
Public 27,146,840.0
Private 12,843,967.0
Total 39,990,807.0

Average Cost ($/bushel)
Total 2.9088
Public 2.8140
Private 3. 1317

Repletion Budget ($) 12,299,954.0
Tax Collections ($) 5,248,199.0
Budget Deficit ($) 7,051,755.0

Prices ($/Pound)
Ex-Vessel 0.85889
Wholesale 1.23894
Retail 2.10822

Total Revenues ($)
Harvester 55,124,719.64
Processor 79,516,841.68
Retailer 135,308,405.56

Net Revenues ($/Year)
Private Planter
750 (bushels/acre) -792.42
1125 (bushels/acre) 85.80
1500 (bushels/acre) 964.03

Hand-Tonger 3,736.84
Dredger 21,309.52

Value Added 47,729,529.00
Indirect Tax Revenues 5,608,219.70

-34-

depressing production increase, but does not offer offsetting reductions in
private planting costs through seed price reductions. Thus, although the
model suggests that the production goal might be acheived for any scenario
and that a mix of private and public production is possible, the economic
feasibility of private oyster planting is reduced from pre-OFMP conditions
in some cases. Specifically, Table 13 shows $85 per acre per year net
returns for the baseline condition (Scenario 2) and Scenarios 3, 7 and 8.
Pre-OFMP returns for a planter harvesting 1125 bushels per acre were
estimated at $490 per year. With the economic conditions existing in the
oyster industy at this time it seems likely that reductions in returns to
private oyster culture will discourage industry entrants and lead to the
I ~ ~exit of existing growers. Therefore, without further analysis it appears
that these management approaches would not acheive the OFMP production
goal, because of the adverse effect on the economics of private planting.
U ~ ~The common characteristic of all these scenarios is that they rely oaily
upon tong harvest of seed. Only by dredge harvest, recognizing that dredge
seed may be lower productivity per bushel and will require more frequent
seed bed shelling, does the production goal, with respect to its private
grounds production component, appear acheivable. Note that dredge harvest
need not be in the James River (Table 1). Further discussion of these
points is included in sections below. However, the first management
implication is clear.
Implication J1: The OFMP must adopt policies which lower
I ~ ~~seed prices to private planters by permitting dredge harvest
of seed. The OFMP will directly affect the economic
feasibility of private planting by lowering market oyster
prices and this can only be offset by policies to lower seed
prices.
Section 7.1: Competitive Production Cost

The national market for oyster production is becoming increasingly
competitive. Production from the Pacific Coast is now being added to South
I ~ ~Atlantic production as a competitor to Chesapeake Bay oysters. One
important aspect of a competitive advantage will be oyster prices which
will in turn depend on cost of production. Therefore, the need to attain
production targets at lowest cost is understandable. Table 13 presents the
result that the least cost strategy is scenario 11. Scenarios 7 and 8 run
a close third and fourth behind Scenarios 11 and 10. Scenarios 7 and 8,
however, result in substantially lowered returns to private planting (as
I ~ ~compared to pre-OFMP conditions) and would, as noted earlier, be likely to
discourage private oyster planting. Thus, scenarios 7 and 8 are acceptable
only if the public grounds are increasingly the focus of the OFMP and the
I ~ ~private planters are allowed to exit from the industry. This is not
consistent with the policy statement. Scenarios 4, 5, 6, 9, 10 and 11 all
are cost minimizing strategies which raise private planter?'s returns over
the pre-QEMP condition. These scenarios all have the common characteristic
of allowing seed dredging. Thus a second management implication is

* ~~~~~~~~~~~-35-

Table 10
Scenario 9: Seed Transplant, Tong or Dredge
Market Oysters And Dredge All Seed

Total Harvest (Bushels) 13,748,040.0
Public Harvest 8,926,984.0
Private Harvest 4,821,056.0

Total Cost ($)
Public 28,983,169.0
Private 15,035,181.0
Total 44,018,350.0

Average Cost ($/bushel)
Total 3.2017
Public 3.2466
Private 3.1186

Repletion Budget ($) 12,500,000.0
Tax Collections ($) 5,345,782.0
Budget Deficit ($) 7,154,218.0

Prices ($/Pound)
Ex-Vessel 0.85889
Wholesale 1.23894
Retail 2.10822

Total Revenues ($)
Harvester 55,124,719.64
Processor 79,516,841.68
Retailer 135,308,405.56

Net Revenues ($/Year)
Private Planter
750 (bushels/acre) -507.42
1125 (bushels/acre) 663.55
1500 (bushels/acre) 1249.03

Hand-Tonger 3,736.84
Dredger 21,309.52

Value Added 47,729,529.00
Indirect Tax Revenues 5,608,219.70

-36-

Implication 2: The national competitiveness of the Virginia
oyster industry will be enhanced if seed dredging is allowed
and if seed transplanting is the basis for the repletion
program for the public grounds.

Section 7.2: Repletion Budget

There is no scenario that results in a zero deficit in the repletion
budget without consideration of indirect tax revenues. However, Scenarios
5, 6, 10 and 11 result in modest repletion budget deficits if indirect tax
revenues are considered. The characteristic that all these scenarios have
in common is that they all include seed dredging. Given the imprecision of
the indirect tax revenue estimates it appears that Scenarios 5, 6, 10 and
11 all permit near full recovery of relatively low repletion spending from
direct and indirect tax. However, a shelling only repletion strategy (the
baseline condition) will result in substantial deficits.

Implication 3: Self sufficiency of the repletion budget is
possible, if indirect tax revenues are considered and dredge
seed harvest and seed transplant are part of the OFMP
repletion strategy.

Section 2.3: Income and Access to Baylor Grounds

It was noted in Section 7.0 that hand tong labor benefited by
increased net income. Table 13 shows this to be $3,736 per year to
operator labor and the boat. Also, shown in Table 13 is the case where
dredge harvest of oysters on public grounds offers a return to the
harvester of $21,309 to labor and the boat. Of course the boat costs would
be higher for the dredge operator so the difference in returns between hand
tong and dredge effort is less than suggested by these two figures.

These net returns figures must be evaluated in order to determine
whether limiting the access to Baylor grounds to hand tongers will permit
the plan to succeed. To assess this, consider that for the pre-OFMP
condition public grounds harvest was 341,870 bushels. Assuming that an
average hand tonger fishes 75 days each year and harvests 10 bushels per
day (as included in the cost budgets), it would take 331.42 full time
tongers to harvest the 341,870 bushels. The first two columns of Table 14
show the number of full time hand tongers and dredgers that would be
required to harvest market and seed oysters for each scenario. These
figures were computed assuming that due to the OFMP production targets the
average hand tonger would harvest 20 bushels per day while still fishing
for 75 days and the average dredger would fish for 75 days per year and
harvest 160 bushels of market oysters per day. Note that only in Scenario
7 (where all market oysters are harvested by dredge), does the annual hand
tong labor required for the production goal fall below that of the pre-OFMP
condition. All other scenarios, including Scenario 11 where the dredging
of seed and market oysters is permitted, result in greater hand tong labor

I ~~~~~~~~~~~~-37-

Table 11
Scenario 10: Seed Transplant, Dredge or Tong
Market Oysters and Dredge or Tong Seed

Total Harvest (Bushels) 13,748,040.0
Public Harvest 9,248,044.0
Private Harvest 4,499,996.0

Total Cost ($)
Public 28,216,012.0
Private 13,473,271.0
Total 41,689,283.0

Average Cost ($/bushel)
Total 3.0323
Public 3.0510
Private 2.9940

Repletion Budget ($) 11,410,809.0
Tax Collections ($) 5,350,003.0
Budget Deficit ($) 6,060,806.0

Prices ($/Pound)
Ex-Vessel 0.85889
Wholesale 1.23894
Retail 2.10822

Total Revenues ($)
Harvester 55,124,719.64
Processor 79,516,841.68
Retailer 135,308,405.56

Net Revenues ($/Year)
Private Planter
750 (bushels/acre) -507.42
1125 (bushels/acre) 663.55
1500 (bushels/acre) 1249.03

Hand-Tonger 3,736.84
Dredger 21,309.52

Value Added 47,729,529.00
Indirect Tax Revenues 5,608,219.70

-38-

demand than the pre-OFMP condition. However, opportunity costs of labor
and changing seashore community structure suggests that few new persons
will enter the hand tong fishery (Santopietro, 1986). It seems unlikely
that the additional numbers needed to harvest 1,400,000 bushels will be
found with the modest increase in returns for hand tong market oyster
harvest.

Alternatively, if some dredge harvest were permitted the increase in
hand tong labor required to harvest the production goal would be reduced.
For example, in the baseline condition it would require a near doubling
over pre-OFMP levels of the number of hand tongers needed to harvest the
public grounds production that would occur under the management plan. It
is only when managed dredging of market oysters is considered that the
numbers of hand tongers required to harvest the public grounds production
seem feasible given the lack of entry into the tong labor force, and the
modest increase in returns to hand tongers. Note that the scenarios that
include managed dreging of the public grounds (7, 8, 9, 10 and 11), with
the exception of Scenario 7, compare favorably with the number of hand
tongers in the pre-OFMP scenario (Scenario 1) suggesting that few persons
would be replaced with a dredge policy. As oyster production grows
increased use of dredge harvest could be permitted. Indeed, it seems that
without dredge harvest of market oysters the production goals cannot be
met.

Implication 4: It is possible to meet production goals,
maintain access to Baylors and assure adequate income to
harvesters only by permitting a mixed strategy of hand tong
and managed dredge harvest of market oysters on the Baylor
grounds.

Section 7.4: Shell Reauirements

Given the relative difficulty VMRC has had with obtaining sufficient
shell supplies for its current repletion activities it is important to
assess the shell requirements for each scenario. Table 14 lists the amount
of shell that would be required in order to meet the harvest goal. In the
pre-OFMP condition (Scenario 1) 1,661,194 bushels of oyster shell were
planted. On a ten year basis this level of shelling would amount to
16,611,940 bushels. If pre-OFMP repletion strategies are maintained
(Scenario 2) a shelling-only strategy would require 36,338,287 bushels of
oyster shell over the ten year period, more than double the pre-OFMP
amount. It seems unlikely that such a shelling level can be sustained
given limited access to house shell and uncertain quantities of reef shell.
Only Scenarios 5, 6, 7, 8, 10 and 11 call for shell quantities that are
reasonably close to that used in the pre-OFMP condition. At a minimum each
of these scenarios calls for a repletion program heavily oriented toward
the transplanting of seed.

-39-

Table 12
Scenario 11: Seed Transplant, Dredge or Tong
Market Oysters and Dredge Seed Except in The James

Total Harvest (Bushels) 13,748,040.0
Public Harvest 9,248,036.0
Private Harvest 4,500,004.0

Total Cost ($)
Public 28,347,624.0
Private 13,502,133.0
Total 41,849,757.0

Average Cost ($/bushel)
Total 3.0440
Public 3.0652
Private 3.0004

Repletion Budget ($) 11,563,183.0
Tax Collections ($) 5,282,057.0
Budget Deficit ($) 6,281,126.0

Prices ($/Pound)
Ex-Vessel 0.85889
Wholesale 1.23894
Retail 2.10822

Total Revenues ($)
Harvester 55,124,719.64
Processor 79,516,841.68
Retailer 135,308,405.56

Net Revenues ($/Year)
Private Planter
750 (bushels/acre) -507.42
1125 (bushels/acre) 663.55
1500 (bushels/acre) 1249.03

Hand-Tonger 3,736.84
Dredger 21,309.52

Value Added 47,729,529.00
Indirect Tax Revenues 5,608,219.70

-40-

Implication 5: Given the limited availability of house
shell and uncertain supplies of reef shell, only a repletion
strategy that, at a minimum, emphasizes transplanting seed
will be capable of meeting and sustaining the production
goal set for the OFMP.
Section 7.5: Recommendations and Discussion

I ~~~Only Scenarios 10 and 11 are capable of meeting the multiplicity of
goals set for the OFMP. The management characteristics associated with
these scenarios are that all repletion strategies and all harvest
I ~ ~strategies are possible. Because of the political sensitivity of the issue
of seed dredging in the James, Scenario 11 which does not permit dredging
the James River seed beds would be favored.

The general outline of a management plan following Scenario 11 and
drawing upon the results of the model can be described. First, in 11 the
repletion program concentrates its activities on shelling for a set in good
seed areas and then harvesting seed by dredge in all areas except the James
for grow out in other areas. Seed areas that might be considered are the
Piankatank and Great Wicomico Rivers. Once transplanted market oyster
harvest would be by dredge. Tong harvest would be reserved for those areas
that are shelled to receive a natural set for market oyster production.
The number of dredge operators for both seed and market oysters would be
I ~ ~limited. One means of limiting numbers could include giving dredging
rights to holders of hand tong licenses who would then dredge or sell their
right to others who wish to dredge. Other possibilities could be explored.
Associated with this strategy is the fact that seed prices would be
lowered, encouraging more private planting.
This management strategy is realistic in that it results in a
technically attainable goal and is consistent with the economic realities
that can keep the Virginia harvest sector and private planter competitive
with other oyster producing regions. In this regard, it is important to
note that the final model structure has been formed to favor tong harvest
of seed and market oysters. As a result the fact that dredge activity is
still suggested is that much more significant. The primary examples of the
* ~~intentional bias toward tong harvest are the following:

1. Seed bed shelling requirements: For tong harvest
shelling is assumed to be required only once every four
I ~ ~~years. Setting success only declines 10% per year between
shelling. For dredge harvest shelling is assumed to be
required every year. Setting success goes to zero if not
reshelled.
2. Productivity of Seed: Tong seed is assumed to be twice
as productive 'of market oysters as is dredge seed. In other
words, spat count per bushel is one half that of tong seed.

I ~~~~~~~~~~~~-41-

Table 13
Comparison of Policy Scenarios to the Baseline Condition

Repletion Budget Deficit Public Grounds Private Planter
Average Total Repletion N/O Indirect WV Indirect Net Revenues Net Revenues
Scenarios Cost per Bushel Budget Taxes Taxes Tonger Dredge (1125 Bushels/Acre)

Baseline 3.4382 13,836,105 9,465,229 3,857,080 3,736.84 NA 85.80

Scenario 3 3.1226 12,050,558 7,140,125 1,531,906 3,736.84 NA 85.80

Scenario 4 3.1997 12,500,000 7,450,179 1,841,960 3,736.84 NA 663.55

Scenario 5 3.0323 11,057,677 5,879,254 271,035 3,736.84 NA 663.55
'3
Scenario 6 3.0427 11,526,678 6,364,937 756,718 3,736.84 NA 663.55

Scenario 7 2.9771 12,467,420 7,398,127 1,798,908 NA 21,309 85.80

Scenario 8 2.9088 12,299,954 7,051,755 1,443,536 3,736.84 21,309 85.80

Scenario 9 3.0880 12,500,000 7,154,218 1,545,999 3,736.84 21,309 663.55

Scenario 10 2.9055 11,410,809 6,060,806 452,587 3,736.84 21,309 663.55

Scenario 11 2.8653 11,563,183 6,281,126 672,907 3,736.84 21,309 663.55

Logic suggests that these are extreme assumptions. To illustrate, if
it is the case that shelling for dredging occurred every four years and if
it was assumed that productivity declined 50% (as distinct from the extreme
model assumption of the need to reshell yearly) the model would always
choose dredge over tong harvest. Thus, even this radical decline in
setting rates (50%) on dredged beds relative to the tonged beds (10%) still
favors dredging.

I ~~~The productivity of seed also was biased toward tonging. To recognize
this, assume a spat count of 600-800 per bushel for hand tong seed. The
model assumes a spat count of 300-400 for dredge seed. However, the
I ~ ~Potomac River Fisheries Commission has permitted the dredge harvest of seed
for a number of years. Their spat counts per bushel were 477, 514 and 503
3 ~~for 1982, 1983 and 1984 respectively.

Taken together modest, and probably defensible, adjustments to seed
bed shelling requirements and seed productivity for tong vs. dredge harvest
would tilt the model solution entirely toward dredge harvest of seed. The
effects would be to (1) reduce average costs below those shown in Table 13,
(2) reduce the repletion budget and repletion budget deficits shown in
Table 13, (3) enhance the access to and income potential from the Baylor
grounds and (4) enhance the returns to private planting shown in Table 13.
The results of the models developed for this work can only be a guide
to change. Any change which is instituted will come slowly and must be
carefully monitored. However, change is needed if there is to be increased
oyster production, a labor force that is willing to harvest the oysters and
a private sector that is willing to invest in oyster planting. In summary,
the needed changes are:
(1) A repletion program emphasis on seed transplant..

(2) A program for managed dredging of seed and market
3 ~~~~oysters.

The alternative is the continued decline of the industry as labor and
capital resources leave the sector and as other areas capture Virginia's
3 ~~oyster market.

I ~~~~~~~~~~~~-43-

Table 14
Shall Requirements and Labor Requirements

Annual Annual Tonged Dredged Tonged Public Private
Tong Dedge Seed Seed Market Market Market Shells
Scenario Harvesters Harvesters Oysters Oysters Oysters Oysters Oyster Planted
Ifull time) (full time) (bushels) Ibushels) (bushels) (bushels) (bushels) (bushels)

1 331.42 341,87 0.0 300,103 1,661,194

2 642.43 25.01 2,986,596 0.0 8,143,13 0.0 5,604,902 36,338,287

3 712.01 46.71 2,864,140 0.0 9,248,03 0.0 4,500,003 21,057,255

4 616.54 37.50 0.0 5,646,569 8,970,32 0.0 4,777,712 22,399,938

5 684.35 63.34 2,304,281 3,509,766 9,248,03 0.0 4,500,004 16,951,666

6 706.18 52.12 2,689,511 2,687,674 9,248,04 0.0 4,499,995 15,426,108

7 148.42 48.70 4,452t706 0.0 0.0 9,248,043 4,499,997 13,770,270

8 492.11 114.57 4,247,813 0.0 5,257,734 4,389,099 4,101,206 13,379,659

9 350.62 70.75 0.0 8,641,326 5,259,293 3,703,999 4,784,742 22,353,055

10 414.04 87.10 1,848,154 4,131,721 5,286,534 4,217,498 4,244,010 17,475,537

11 439.41 82.10 2,702,734 2,688,914 5,239,784 4,246,099 4,262,157 15,430,812

References

Bailey, R. S. and F. C. Biggs, Let's Be Oyster Farmers. Virginia Institute
of Marine Sciences. Gloucester Point, Virginia. 1968.

Baker, Samuel H., Frederick H. Harris and Carlisle E. Moody. A Benefit-
Cost Analaysis of the Virginia Oyster Subsidies: An Historical
Appraisal and Proposals for the Future. Department of Economics,
College of William and Mary. Williamsburg, Virginia. December 23,
1977.

Belongia, M. "Application of Ridge Regression with Verification of New
Procedures." Aricultural Economics Research. 31:36-39. 1979.

Belsley, D. A., E. Kuh and R. W. Welsch. Regression Diagnostics:
Identifivin2 Influential Data and Sources of Collinearitv. John Wiley
and Sons. New York, New York. 1980.

Brown, W. G. "Effect of Omitting Relevant Variables Versus Use of Ridge
Regression in Economic Research." Oregon Akricultural Experiment
Station Special Report 394. Corvallis, Oregon. 1973.

Brown, W. G. and B. R. Beattie. "Improving Estimates of Economic Parameters
by Use of Ridge Regression with Poduction Function Application."
American Journal of Azricultural Economics 57:21-32. 1975.

Campbell, D. W., and R. H. Forste. The Biological and Economic Aspects of
Oyster Propogation in Maryland. Mayland Dep. Nat. Res. (no. FA-
MRR-78-2). Annapolis, Maryland. 1978.

Capps, O., Jr. "Treatment of Multicollinearity in Simultaneous Systems of
Linea Equations." Staff Paper SP-82-1, Department of Agricultural
Economics, Virginia Polytechnic Institute and State University,
Blacksburg, Virginia. 1982.

Cox, Robert W. "Virginia Annual Gross State Product, 1958-1985." Tayloe
Murphy Institute, University of Virginia, Charlottesville, Virginia.
December, 1986.

Durbin, J. "Testing for Serial Correlation in Systems of Simultaneous
Regression Equations." Biometrica 370-377. 1957.

Fishery Statistics of the United States. U.S. Department of Commerce,
National Marine Fisheries Service, Washington D. C. 1950-1976.

Fox, K. The Analysis of Demand for Farm Products. USDA Technical Bulletin
No. 1081. 1953.

-45-

Gardner, B. L. "The Farm-Retail Price Spread in a Competetive Food
Industry." American Journal of Azricultural Economics 57:399-409.
1975.

Haven, Dexter S., W. J. Hargis and P. C. Kendall. The Oyster Industry of
Virginia: It's Status; Problems and Promise, Virginia. Virginia
Institute of Marine Sciences Special Papers on Marine Science No. 4.
Gloucester Point, Virginia. April 1981.

Haven, Dexter S., J. P. Whitcomb and P. C. Kendall. The Present and
Potential Productivity of the Baylor Grounds in Virginia. Virginia
Inst. Mar. Sci. Mar. Sci. and Ocean Eng. (no. 243). Vol. I&II.
Glocester Point, Virginia. May 1981.

Heien, D. M. "Markup Pricing in a Dynamic Model of the Food Industry."
American Journal of Agricultural Economics. 62:10-18. 1980.

Hoerl, A. E. and R. W. Kennard. "Ridge Regression: Applications to
Nonorthogonal Problems." Technometrics 12:69-82. 1970

Joint Legislative Audit and Review Commission. Program Evaluation Marine
Resource Management Programs in Virginia. Richmond, Virginia. June
1977.

. The Economic Potential and Management of Virginia's Seafood
Industry. House Document No. 2. Commonwealth of Virginia. Richmond.
1984.

Kennedy, Victor S. and L. L. Breisch. Maryland's Oysters: Research and
Management. Maryland Sea Grant Pub. (no. UM-SG-TS-81-04). College
Park, Maryland. 1981.

Maasoumi, E. "A Ridge-Like Method for Simultaneous Estimation of
Simultaneous Equations." Journal gof Econometrics 12:161-176. 1980.

Malinvaud, E. Statistical Methods of Econometrics. Rand McNally and
Company, Chicago, Illinois. 1970.

Merritt, Donald W.. Oyster Spat Set on Natural Cultch in the Maryland
Portion of the Chesapeake Bay. UMCEES Special Report No. 7. Horn
Point Environmental Laboratories. Cambridge, Maryland. Feb. 1977.

Moscardi, E. and A. de Janvry. "Attitudes Toward Risk Among Peasants: An
Econometric Approach." American Journal of Aericultural Economics
59:710-716. 1977.

Regional Science Research Institute. Basic Regional Input-Ouput for
Transportation Ipact Analysis: Handbook .1. Regional Science Research
Institute, Amherst, Massachusetts. July, 1982.

-46-

Santopietro, George D.. The Evolution of Property Rights to a Natural
Resources: the Oyster Grounds of the Chesapeake Bay. Department of
Agricultural Economics, Virginia Tech, Blacksburg, Virginia. May 1986.
(Unpublished Dissertation).

Shabman, Leonard A. and Oral J. Capps. Econometric Demand and Supply
Models of Selected Chesapekae Bay Seafood Products: Report 1: Oysters.
Department of Agricultural Economics, Virginia Tech, Blacksburg,
Virginia. 1986. (Unpublished Working Paper).

Shabman, Leonard A. and Thomas Johnson. Contribution of the Commercial
Seafood Industry to the Virginia Economy: A Preliminary Assessment.
Department of Agricultural Economics Virginia Tech, Blacksburg,
Virginia. October, 1986. (Unpublished Staff Paper #SP-82-8).

Shellfish Market Review. U.S. Department of Commerce, National Marine
Fisheries Service, Washington, D. C. 1981.

Thunberg, Eric. A Decision Model to Evaluate the Cost-Effectiveness of
Alternative Virginia Oyster Grounds Management Strategies. Department
of Agricultural Economics Virginia Tech, Blacksburg, Virginia. 1985.
(Unpublished Thesis).

Vinod, H. D. "Application of New Ridge Regression Methods to a Study of
Bell Systems Scale Economics." Journal of the American Statistical
Association 71:835-841. 1976.

Vinod, H. D. and A. Ullah. Recent Advances in Regression Methods. Marcel
Dekker, Inc., New York, New York. 1982.

Virginia Marine Resources Commission. Laws of Virginia Relating to the
Marine Resources of the Commonwealth. Newport News, Virrginia. 1980.

Virginia Marine Resources Commission. Annual Reports for 1981 and 1982.
Newport News, Virginia. 1982.

Viginia Marine Resources Commission. An Overview of the Virginia Oyster
Fishery Management Plan. Fisheries Plans and Statistics Section,
Fisheries Management Division, Virginia Marine Resources Commission.
Newport News, Virginia. 1986. (Unpublished Draft)

-47-

Watson, D. E. and K. J. White. "Forecasting the Demand for Money Under
Changing Term Structure of Interest Rates: An application of Ridge
Regression. if Southern Economic Journal 43:1096-1105. 1976.

Waugh, F. V. Demand and Price Analysis. Technical Bulletin No. 1316, U.S.
Department of Agriculture, Economic Research Service, ESAD. 1964.

.-48-

l
I
I
I

~~~~~~~I ~~Appendix A
I*~~~~ ~~The Linear Programming Model

I
I
I
I
I
I
I
I
I
I
I
I -49-
I

Appendix A

The Linear Programming Model

Appendix A is split up into two parts. Part one provides a technical
description of a linear programming model that was developed under Sea
Grant funding. Part two of Appendix A presents a discussion of the
technical infomation incorporated into the models described in Part 1. For
more information, a complete description of all aspects of the model a
Master's Thesis eiititled, "A Decision Model to Evaluate the Cost-
Effectiveness of Alternative Oyster Grounds Management Strategies,"
(Thunberg, 1985) should be consulted.

-50-

Appendix A, Part I

I ~~~~~The Linear Programming Model: Technical Structure

I ~~Section .1.0: Introduction

The Virginia Marine Resources Commission (VMRC) is responsible for the
I ~~management of all oyster bottoms within the State's domain. VMRC is
charged with the administration of the Oyster Repletion Program (ORP) and
the formulation and enforcement of policies and regulations governing the
production and harvest of oysters on public and private oyster beds. In
attempting to achieve repletion and total State market oyster production
goals VMRC must allocate limited resources over time among alternative
aquacultural practices and river systems. In addition, VMRC must evaluate
and recommend regulations on harvest of seed and market oysters.
These resource allocation and policy decisions should provide for the
attainment of oyster production goals at minimum cost to public and private
interests over time. VMRC's repletion activities and policies affect both
public and private costs of producing market oysters. VMRC, therefore,
should concern itself not only with the costs of administering and
conducting its activities, but also how its actions affect the cost to
private individuals harvesting oysters on public grounds or producing
market oysters on leased bottoms. The decision problem faced by VMRC is
I ~~one of minimizing costs over time subject to, limits of resource
availability, market oyster production potential, and the range of policy
options available. Linear Programming (LP) is a mathematical optimization
I ~~technique that can be used to determine an optimal solution to this
decision problem.

A multi-period LP model permits the simultaneous consideration of many
I ~ ~different repletion and policy alternatives while recognizing that VMRC' s
decisions take place over time. In addition to being an optimization
technique, LP provides useful information pertaining to shadow prices for
included constraints, and the sensitivity of the model's solution with
respect to changes in objective function or constraint coefficients.

Within a linear programming framework, VMRC's management problem may
be stated in the following way. The VMRC has limited funds for repletion
purposes, ie. planting shell or seed. VMRC may also choose to substitute
repletion activities with policy changes designed to increase market oyster
harvest such as, permitting dredging of seed or leasing of unproductive
Baylor grounds. By allocating its funds and implementing alternative
policies, VMRC can provide many combinations of aquacultural practices and
policies to achieve a given market oyster harvest goal. These combinations
form the activities of an LP model. Knowing how much each activity
contributes to the harvest goal, and recognizing its limited resources, the
Commission must determine the unique mix of activities required to meet a
harvest goal at minimum cost over its planning horizon.

* ~~~~~~~~~~~~-51-

I ~~Section 2. 0: The Linear Prozramminz Model

The linear programming model described below is the model which has
I ~ ~been formulated to analyze VMRC's decision problem. The model considers
the production of market oysters from public and private grounds over a
period of ten years. All model activities are assumed to be initiated at
the beginning of the year in which they occur.

The ten year planning horizon may be thought of as consisting of two
distinct parts. The model is formulated as if no repletion program existed
I ~ ~prior to the first year of the model's planning period. The first part of
the planning horizon, therefore, is an interim period over which no market
oyster harvest goal is set. This interim period is necessary because
I ~ ~oysters take, on average, 2-5 years to reach the three inch minimum legal
harvestable size (Haven et. al. 1981). A lag, therefore, exists between
the time production activities are initiated and market oysters become
available for harvest. The length of the lag depends on whether seed (2-3
year lag) or shall (3-5 year lag) is planted and the growth potential of
the receiving system. The second part of the planning horizon is the
period of years over which a specified market oyster harvest goal may be
set.
In the model all public and private oyster bottoms in Virginia have
been disaggregated into the 13 river systems listed in Table A.I1.
The indicated number for each river system corresponds to the system
location depicted in Figure A. 1*. The Potomac river is managed jointly by
Maryland and Virginia and is not considered in the model. Separate river
systems were identified using two criteria; i) geographic location and ii)
similarities in environmental characteristics important for oyster growth.

I ~~Section 2..j: The Objective Function

The objective of the linear programming model is to minimize the
I ~ ~present value of public plus private costs of producing a prespecified
level of market oyster harvest over the ten year planning horizon. The
public costs included in the model are expenditures for contracted services
incurred by VMRC as it carries out its repletion program. Private costs
considered in the model are all costs incurred by private planters in the
production and harvesting of market oysters and the cost of harvesting
market oysters to watermen working the public grounds. It is often assumed
that the public sector has a longer-term investment horizon relative to the

*Originally only ten river systems were identified for modeling. Upon the
recommendation of VIMS oyster biologists, the James and York rivers were
split into two regions each, the Rappahannock was split into upper, mid-
and lower sections and the Corrotoman river was incorporated into the Lower
Rappahannock forming the 13 river systems listed in Table A. 1.

I ~~~~~~~~~~~~-52-

Table A. 1

Description of Model River Systems

1 Upper James above James River Bridge

2 Lower James below James River Bridge

3 Upper York

4 Lower York

5 Mobjack Bay including North, East, Ware
and Severn Rivers
6 Piankatank

7 Upper Rappahannock

8 Mid-Rappahannock

9 Lower Rappahannock including the Corrotoman to
Windmill Point
10 Great Wicomico

11 Upper Management Area Windmill Point to Smith Point

12 Pocomoke and Tangier including Baylor bottoms on
the Bayside of Eastern Shore
13 Eastern Shore Seaside

-53-

private sector. Given this assumption, an interest rate of 4.51 percent
was used to discount public costs to the end of the first period. This
public rate reflects the 1983 quarterly average interest rate offered on
long-term state and local government securities (Federal Reserve Bulletin
1984) adjusted by a 5 percent inflation rate. Private costs were
discounted by an interest rate of 6.0 percent. The private rate is
equivalent to the 1983 prime rate on short term business loans adjusted by
a 5 percent inflation rate. The short term private interest rate was used
because it is assumed that such loans would be used to cover operating
expenses incurred between harvest periods.

Section Z.2: Model Activities

There are two general types of activities considered in the model.
Activities that result in market oyster production, hereafter referred to
as primary activities, are represented in Figure A.2 by the rectangular
boxes. The primary activities of Figure A. 2 are the final products of the
production process depicted by a flow chart for each activity. Activities
which do not directly result in the production of market oysters are
represented by the circles in Figure A.2.

Section 2.2.1 Primary Activities

The four primary activities are based on current harvest regulations
and aquacultural practices used in the Chesaspeaske Bay by public and
private culturists. Each of these activities is uniquely defined by
aquacultural technique, harvest technology, bottom ownership and production
period.

The first primary activity is a shell-to-harvest activity where fresh
or dredged oyster shells are planted on Baylor bottoms by VMRC to receive a
set. The resulting young oysters are left to grow to a harvestable size.
This aquacultural technique has for many years been the most important part
of VMRC's repletion strategy. For this reason and because setting rates on
private grounds are usually insufficient to make shell-to-harvest
economically feasible, the model includes only the Baylor bottoms as
candidates for oyster production using this technique. Setting rates vary
from river system to river system. A shell-harvest activity is included in
the model for only those river systems (listed in Table A.2), determined by
a literature survey and consultation with oyster management professionals,
with sufficient setting rates to make shell-to-harvest a feasible oyster
production activity.

The shell-to-harvest activity is carried out on Baylor grounds.
Harvesting, therefore, is done by hand tongers. A four year growing period
is assumed to be required from the time the shelling is initiated and the
time the first harvestable oyster is taken. Upon consultation with VMRC
repletion personnel it was further assumed that the shelled oyster bed will
remain productive, but at a declining rate, for four years after the first
oyster is harvested. A annual decline in productivity of ten percent of
the previous years standing crop was assumed.

-54-

I~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

V~~~~~~~~~~~~~~~ .-
F11 M 13
I~~~~~~~~~~~

m
I p~~~~~~~~~~~~~~~
P~~~~~~~~~

I~~ 3

I~S_

I ~~~~~~~~Figure A. 1 Model River Systems

* ~~~~~~~~~~-55-

I ~~~The second and third primary activity is a seed transplant-to-harvest
activity. This activity involves harvesting seed from seed areas and
transplanting the seed to grow-out areas on Baylor grounds or to growing
areas on leased bottoms. Where setting rates are considered sufficient to
support self-sustaining oyster populations on public grounds, and where
losses to diseases and predators to non-indigeneous seed, seed transplant-
to-harvest activities were eliminated from model consideration. Public
grow-out areas and leased growing areas are listed in Table A. 3.
The cultural technique for transpant-to-harvest activities are the
I ~~same whether seed is planted on public or private grounds. The distinction
between primary activity number two (seed transplanted onto Baylor grounds
by VMRC) and primary activity number three (seed transplanted by private
individuals on leased bottoms) is one of harvest technology. All
harvesting on Baylor grounds is assumed to be done with hand tongs, while
all harvesting on leased bottoms is assumed to be done with an oyster
3 ~ ~dredge. The production period for both public and private seed-transplant
activities is three years, where seed oysters are planted at the beginning
of the first year of the production period and are harvested at the end of
the third year of the production period. The only exception to this is
I ~~seed transplanted on the Eastern Shore Seaside. Seed transplanted to these
leased bottoms grow very rapidly and losses to disease and predation are
heavy if the oysters are left on the bottom for more than eighteen months.
A two year production period is assumed for leased bottoms in this region.
According to VMRC repletion officers, seed would only be transplanted to
areas that do not receive a natural set. It is assumed, therefore, that
3 ~~the oyster bed will remain productive for only one season.

Not all market oyster harvest results from man-induced production. A
fourth primary activity is included in the model which permits the harvest
U ~~of market oysters from naturally occurring stocks. Natural production is
simply defined as any production of market oysters that is not directly the
intentional result of Man's repletion or cultural activities. The Baylor
grounds were established on the basis of natural productivity. It is
assumed, therefore, that all market oyster harvest of natural production
takes place on the public oyster beds. No attempt is made to determine the
natural productivity of each river system. For the model's purpose,
therefore, -a given level of annual Bay-wide natural market oyster
production is assumed. This yearly production is harvested by the end of
each year of the planning horizon. Table A.4 provides a summary of all the
primary activities by cultural technique, harvest gear, bottom ownership
and production period.

Sectio _Z. 2 : Secondary Activities

Secondary activities do not directly result in market oyster
production. The secondary activities are, instead, support activities
providing the means through which i) seed is brought into the market oyster
production process and ii) funds are made available to carry out VMRC's

I ~~~~~~~~~~~~-56-

Harvest natural Shell-harvest Seed-transplant Seed-transplant
market oyster on Baylor on Baylor on leased
production on Grounds Grounds bottoms
Baylor Grounds

I~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

I T .
Hares with Hretwith Harvest with rnp Harvest with
Hand-tongs Hand-tongs Hand-tong oystr dredge

Shelling by Seed planting HarvesPrivate planter
VMRC by VMRC Seed seed planting

FreA Da epleted ba
seed seed

Figure A.2; Diagram of Model Activities

repletion program. Included in the model are four seed related secondary
I ~ ~activities and two repletion fund related activities.
The model permits the harvest of seed from the Baylor grounds on
I ~ ~unrepleted and repleted bars. Although some seed is produced by private
planters, the majority of seed comes from public seed beds. For this
reason private seed production is not considered in the model. Seed from
unrepleted bars represents a stock which may be exploited. A given level
I ~ ~of annual natural recruitment is assumed for modeling purposes. If in any
year this annual stock is exhausted and more seed oysters are demanded,
then the model permits seed to be produced by including a set of seed
I ~ ~production activities. Unlike seed from unrepleted bars which can enter
the market oyster production process as soon as it is required, seed from
repleted bars must first be produced by shelling Baylor bottoms to catch a
strike. River systems identified as seed areas are listed in Table A.5.
Once set, the immature oysters grow rapidly enough to be harvested by fall
of the year they were spawned. All activities are assumed to take place at
the beginning of any given period. A one year growing cycle for seed
production is assumed even though it may well be the case that the seed
oysters will be harvestable before the year is up. Like the market oyster
shelling activities, a seed bed is assumed to remain productive at a ten
I ~~percent declining rate for four years after the first seed oyster is
harvested. Also included in the model is a seed harvest activity from
unrepleted bars for each of the seed transplant-to-harvest activities.
I ~ ~Figure A.3 provides a pictorial summary of Tables A.2, A.3 and A.5.
Cultural techniques determined to be feasible are indicated for each river
system.

I ~~~The harvest of repleted-bar seed and the transport of seed from its
origin to the planting site remain to be discussed. Both of these
activities will only enter the model solution if seed production enters
I ~ ~first. Seed that is produced must be harvested. Current regulations
permit the harvest of seed only with hand tongs. A separate seed harvest
activity is included for each seed production activity to maintain
continuity in the seed to market oyster production process. Separate
activities are not required, but their absence could mask differing harvest
technologies or regulations associated with the harvest of seed in one area
3 ~~versus another.
A set of seed transport activities is incorporated into the model for
two reasons: i) to. define the unique cost of transporting seed from one
I ~ ~region to another, and ii) to prohibit the transport of seed from a given
seed area to an MSX incompatible receiving system (as shown in Table A.6).
A seed transport activity is included to move seed from each seed producing
I ~ ~area (from Table A.5) to each grow-out area (Table A.3) conditional on MSX
compatibility.
Seed resistance to MSX is assumed to improve the higher the disease
incidence in the system of seed origin. Seed that originiates in a system
of relatively low MSX incidence, however, has very little resistance to the

I ~~~~~~~~~~~~-58-

Table A.2

Feasible Shell-Harvest River Systems

Great Wicomico

Mid-Rappahannock

Lower Rappahannock

Mobjack Bay

Seaside Eastern Shore

Piankatank

Upper Management Area

Pocomoke and Tangier

-59-

disease and will likely perish before reaching a harvestable size. The MSX
categories defined in 1968 and reported in Haven et al (1981), are used to
determine MSX compatability. It is assumed that seed originating in a Type
I MSX area can be transplanted to any river system of Type I or higher.
Similaly, seed originating in a Type II river system may only be
transplanted to Type II or higher river systems and so on as depicted in
Table A.7.

Table A.6 is a matrix of seed and grow-out area compatibilities by
river system of origin and receiving system. MSX compatibility is assumed
to be the same on public and private grounds. Over 70% of the seed
I ~ ~transplanted to leased bottoms in 1981 came from the James river seed beds
(VMRC Annual Reports 1982). For this reason, only James river seed is
3 ~~allowed to go to all leased bottoms.

Section 2.2.3: Tax Collection and Repletion Fund Transfer

The final secondary activities are: i) oyster repletion tax
I ~ ~collection, and ii) repletion fund transfer from one year to another. One
source of repletion funds is the Public Rocks Oyster Repletion Tax. The
r epletion tax is levied on all seed and market oysters harvested on the
I ~ ~aylor grounds. Any harvest of oysters on public grounds, therefore,
generates revenue which goes into the State's repletion program budget.
The model is designed so that whenever a market oyster harvest or
I ~ ~production activity (all production is assumed harvested) or a seed harvest
activity is initiated a tax collection activity is initiated in the year
the oysters are harvested.

U~ ~ ~~I t is possible that OR? budgetary requirements will differ from year
to year depending on the mix of repletion activities undertaken by VKIRC.
One would expect, therefore, years of funding shortages and years of
funding surpluses. The model permits the transfer of funds from years of
budgetary surplus to years of budgetary shortage.

3 ~~Section 3.0: Constraints

In the linear programming model, maximum constraints are placed on the
availability of private and public oyster bottoms, the OR? budget, and the
natural production of seed and market oysters on the Baylor grounds.
In addition to these resource availability constraints, there is a set
I ~ ~of transfer rows which force an activity or sequence of activities into the
solution whenever a primary activity requires an auxiliary secondary
activity. If, for example, a seed-transplant activity enters the basis and
I ~ ~natural seed availability is exhausted, the transfer rows force the
following sequence of secondary activities into the basis; seed production,
seed harvest and seed transport. Another set of transfer rows initiates a
tax collection activity any time seed or market oysters'are harvested on
public grounds. Also included in the model is a set of transfers which-
effect the transfer of OR? funds from one year to another.

I ~~~~~~~~~~~~-60-

Table A.3

Public and Private Growing Areas

Baylor Grounds Private Grounds

Great Wicomico Upper Management Area

Upper Rappahannock Great Wicomico

Mid-Rappahannock Upper Rappahannock

Upper York Mid-Rappahannock

Lower York Lower Rappahannock

Upper Management Area Mobjack Bay

Piankatank Upper York

Lower York

Piankatank

Upper James

Lower James

Pocomoke and Tangier

Seaside Eastern Shore

-61-

I ~~~~The final set of constraints is the setting of the harvest goal. The
harvest goal is constrained with equality for each of the ten years of the
planning horizon.

Section 4. 0: The Computer Model

I ~~~The linear programming model just described is now presented in
tableau form in Figure A.4. The columns of the tableau represent the
alternative activities which may be chosen to achieve the prespecified
level of market oyster harvest.
The'objective function for the model is described in the "ICOST"i row.
Included in this row are the public plus private costs of initiatinga
public grounds repletion and harvest activity or a private bottoms market
oyster production and harvest activity. The coefficients in the objective
function row represent the total cost of undertaking a single unit of a
given activity discounted to the first year of the ten year planning
period.

I ~~~The bottom acreage, shell, seed, and budget requirements for each
activity are included in the rows section. Also included in this section
are tax revenues generated by each activity, the quantity of seed resulting
I ~~from a given seed production activity, and the bushels of harvested market
oysters generated by each primary activity.

Maximum levels of private and public acreage are listed under the
I ~ ~column labeled "RHS.t' Limitations on repletion funds, and natural seed and
market oyster harvest are also listed in this column. The budget for any
given year can also be increased by transferring unused funds from one year
to be used in another. This particular activity is represented in Figure

In this submatrix the budget for the year receiving the fund transfer
is increased while the budget year from which the funds are taken is
decreased by the amount of the transfer. This is indicated by a negative
one in each "BT" row of any year (j) receiving a transfer and a positive
one in the "BT" row of the year (i) making the transfer. It is only
possible to transfer funds forward in the planning period. A fund transfer
activity, therefore, is included for all years i and j such that j > i.
I ~ ~Also included in the RHS section is the prespecified level of market
oysters, as indicated by the row labeled "GOAL."

I ~~~~~~~~~~~~-62-

Table A.4

Summary of Primary Activities

Bottom Production
Cultural Technique Harvest Gear Ownership Period*

Shell-Harvest Plant shell on Baylor Bottoms Hand-tongs Public 4 Years
to receive a set and grow-out
to a harvestable size.

Public Seed- Transplant seed on Baylor Hand-tongs Public 3 Years
Transplant- bottoms and allow grow-out
Harvest to a harvestable size

Private Seed- Transplant seed on leased Oyster Dredge Private 3 Years**
Transplant- grounds and allow grow-out
Harvest to a harvestable size
ON
Natural Hand-tongs Public
Production

*The production period includes harvesting the oysters.

** Except for Eastern Shore Seaside where growth is rapid
and losses to predation are heavy if seed is left on the
bottom for more than 18 months.

none~~~~~~~~~~~'
u r'"I~~

I - ?..,-w.I,,- . ..-'I-..~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~.,.;.

I ~ ~ ~ ~ ~ ~ ~ ~ ~~~~~~�::~; c- -.., ..

000006~~~~~~~~~~~~~~)ij

o~~~~~~~~~~~wn
I~~~~~~~~~~~~

Figure A. 3 Cultural Technique Feasibility

I ~~~~~~~~~~~~-64-
I,~~~~~~~~~~~~~~~'

The technical coefficients for the model define the relationship
between a given activity and a given constraint. The coefficients denoted
by the letter "F" in Figure A.4, for example, define the quantity of fresh
shell required in order to undertake a particular shell-to-harvest
activity. The estimation of these technical relationships will be
discussed in detail in Part 2 of this appendix.

The input data analyzed by the linear programming model is represented
by generalized notation in the matrix tableau. Table A.8 explains the
notation used.

-65-

Table A.5

Public Grounds Seed Areas

Upper James

Lower James

Mobjack Bay

Lower Rappahannock

Piankatank

Great Wicomico

-66-

Table A.6

MSX Compatability by Seed Origin and Receiving System

Upper Lower Mobjack Pianka- Upper Lower Mid- Great Upper Pocomoke tipper Lower
York York Bay tank Rapp. Rapp. Rapp. Wicom. Mgt. Area &Tangier James James

Upper James C C C C C C C

Lower James C C C C C C C C C

Lower Rapp. C C C C C C C C

Piankatank C C C C C C C C C

Great Wicom. C C C C C C C C

Hobjack Bay C C C C C C C C C C C C

Source: Haven at. al. (1981)
A "C" indicates MSX compatability.

Table A.7

River System Compatability by MSX Type

MSX Type of

Receiving System

I I II III IV
ml~~~~~~~~~~I

I X x x x

II X X X
MSX Type of

Seed Origin
III x x

IV IX

This table shows the MSX Type relationship between system of
seed origin and receiving seed origin and receiving system.
An "X" indicates compatible MSX types.

-68-

SHFF r SHFD)r SHSF rt SISD rt TSF rt TSSF rt TS Srt TSL rt PSFF rt PSFD r

RBF I
RSt

RL t1

FRS F F F F

DRS 0 D 0 0

NS PC

BT B 8 B B B B B
t
GOAL G G C G C C G

TR -T -T -T -T -T -T -T

TS T SR SR s ISR

TSTL rtSRL
THrt

THT rt-SP' -SP'

COST c SH1FF CSHFO CSHSF CSHSD TSFr cTSsr IS cTSL cPSFF cPSFD
r t r t rt r t rt rt TSrt rt r~t r t

Figure A.4: Model Tableau

PSSF rt PSSDr0 HAR Sr SMr t STORL t HNSP t HNMOP TAXC t TFUND t RH;

RB F t<RBF,~

RBS t R6 S'
RLt < RL

FRS tF >0

DRS~ 0 1)0

NSPC 1

BiTt B B - I < RB*

GOAL C G

T R t- T T -T 10

TST -10

TSTL 1- =0

TSH -11 =0

THT rt-SP -SP I

rt rt t tt

Figure A.4: Model Tableauu (continued)

TFUND12 TFUND13 TFUND14 TFUND15 .....TFUNDij RHS
I
BT1 1 1 1 1 < RB*

BT21 -1 < RB*

BT3 -1 < RB*

BT4 -1 < RB*

BT5 -1 < RB*

BTiI 1 < RB*

I
BTjI -1 < RB*

* Note that the submatrix presented here includes only
the first four years of the planning period. The sub-
matrix continues in a similar manner for the remaining
years of the planning period.

Figure A.5
Fund Transfer Submatrix

-71-

Table A. 8
Tableau Key

Subscripts and Suoerscrints

I ~~t = time period; t =1-10

r = river system; r =the river systems listed in Table A. 1

i = the discount rate. An interest rate of 4.51 percent was
utilized to discount the future cost of public expendit-
I ~~~ures for oyster bed repletion. Private grounds product-
ion expenditures are discounted by an interest rate of
*~~~6pret

n =the number of years over which discounting occurs.

* ~~Coefficients

C =the cost in current dollars of undertaking any secondary
* ~~~or primary activity.

F = the number of bushels of house shell required by an activity.

D =the number of bushels of reef shell required by an activity.

B = the amount of public expenditures required to initiate
one unit of any public repletion program activity.

G = the contribution toward satisfying the harvest goal made
by a unit of any given primary activity.

T = the tax revenue generated by a unit of seed or market
oyster production on Baylor grounds.

SR =the number of bushels of seed required by a public
seed-transplant activity.

I ~~SRL =the number of bushels of seed planted on private-grounds.

SP the bushels of seed produced on public grounds from a
I ~~~unit of a seed production activity.

RBF* = the maximum acreage of firm public bottoms in a given river.

RBS* = the maximum acreage of soft public bottoms in a given river.

I ~~RL* =the maximum leased acreage in a given river.

I ~~~~~~~~~~~~-72-

RB* =the maximum amount of repletion funds available in any
I ~ ~~~given year.
G= the market oyster harvest goal.
U ~~Columns
SHFF = a shell-harvest activity using fresh shell on firm
Baylor grounds. Rivers included in this activity
and the next three activities are listed in Table A.2

I ~~SHFD = a shell-harvest activity using dredge shell on firm
Baylor bottoms.
SHSF = a shell-harvest activity using fresh shell on soft
Baylor bottoms.
SHSD = a shell-harvest activity using dredge shell on soft
I ~ ~~~Baylor bottoms.

TSF =transplanting seed to grow-out areas on firm Baylor
grounds. River systems included in this and the
next two activities are listed in Table A. 3.

TSSF =transplanting seed to grow-out areas on soft Baylor
bottoms where shelling with fresh shell used.
TSSD =transplanting seed to grow-out areas an soft Baylor
I ~ ~~bottoms where shelling with dredge shell is used.
TSL =transplanting seed on private grounds, river systems
I ~ ~~included in this activity are listed in Table A. 3.
PSFF =shelling firm Baylor grounds with fresh shell to
I ~ ~~produce seed, river systems included in this and the
next three activities are listed in Table A.5.

PSFD =shelling firm Baylor grounds wi-th dredge shell to
I ~ ~~~produce seed.

PSSF =shelling soft Baylor grounds with fresh shell to
I ~ ~~~produce seed.
PSSD =shelling soft Baylor grounds with dredge shell to
I ~ ~~~produce seed.
HARS = harvest seed produced by the previous activities.
I ~~STOR = transport the harvested seed (from the HARS activity)
to a receiving system on Baylor bottoms.
I ~~~~~~~~~~~~-73-

STORL = transport the harvested seed (from the HARS
activity) to leased bottoms.

HNSP = harvest natural seed production.

HNMOP = harvest naturally occurring market oysters.

TAXC = tax collection.

TFUND = transfer funds from one year to another.

Rows

RBF = firm Baylor grounds.

RBS = soft Baylor grounds.

RL = leased bottoms.

FRS = fresh shell.

DRS = dredge shell.

NSPC = natural seed production.

NMOP = natural market oyster production.

BT = VMRC's repletion budget.

GOAL = the prespecified level of market oyster harvest.

TR = tax revenue.

TST = transfer transported seed to a public seed-transplant
activity

TSTL = transfer transported seed to a private
seed-transplant activity.

TSH = transfer seed harvested seed to a seed transport activity.

THT = transfer seed produced seed from repleted bars to
a seed harvest activity.

-74-

I
I
I
I
Appendix A, Part 2
The Linear Programming Model:
Estimation of Technical Coefficients

I
I
I
I

I
I
I
I
I
I
* -75-
I

Appendix A, Part 2

I ~~~~~~~~The Linear Programming Model:
Estimation of Technical Coefficients

Section 1.O: Introduction

* ~~~The decision model developed in Part Ilpermits anevaluation of the
alternative oyster grounds management options available to VMRC.
Underlying each management option is the assumption that the effects of
implementing a given policy or repletion option are known. It is assumed,
for example, that VMRC knows exactly how many bushels of market oysters
will be produced when it plants seed in any given river system. However,
the information base upon which management decisions are formed is
I ~ ~imperfect. An inadequate data base, the absence of an efficient means of
collecting and processing fisheries management information and the lack of
economic and biological modeling contribute toward management uncertainty.
I ~ ~The purpose of this chapter is to provide a detailed treatment of how.the
technical coefficients required for the model developed in Part I were
estimated. Data requirements for objective function constraints and
activities are discussed separately. Included in each section is a list of
data sources used and a treatment of the informational or data problems
associated with each coefficient type.

I ~~Section 2.0: Objective Function Coefficients

The objective function coefficients indicate the total public plus
U ~ ~private cost of undertaking a given primary or secondary activity. These
cost coefficients may be broken down into three general categories; public
grounds repletion costs, private grounds production costs and private
3 ~~harvest costs.

Section 2. 1: Public Grounds Repletion Costs

I ~~~All activities that take place on public oyster rocks have associated
with them a repletion cost and a private harvester cost. Together these
costs make up the total cost of undertaking any public grounds activity.
I ~ ~The private harvester costs, however, will be discussed later as estimation
of these coefficients merits separate consideration.

VMRC contracts the services of private watermen to carry out its
I ~ ~repletion program. Estimation of these cost coefficients under current
repletion program policies is an easy task as the per-bushel costs for
contractual services are readily available from VMRC Annual Reports. To
I ~~put these costs on a per-acre basis it is necessary to multiply the per-
bushel cost of seed or shell planted by the quantity planted per acre. The
public grounds shell planting costs are computed as follows:

* ~~~~~~~~~~~~-76-

Repletion Cost/Acrej
I ~ ~~~~~~(a/bushe~l shellj) x (bushels plantedi',1')
where i = river system
j = shell type (fresh or dredge shell)
k = bottom type*
Public shelling costs, therefore, depend on the river system shelled, the
I ~ ~shell type and the quality of the growing bottom. Public grounds seed
transplanting costs may be calculated in similar fashion by the following
* ~~formula:

Seed Planting Cost/acrej
(s/bushel for seedh) x (bushels planted**)
where i = receiving river system
h = system of seed origin

U ~~~Using these formulae, the repletion cost of any specific public
grounds repletion-harvest activity can be determined. The objective
function coefficient for any of these activities is simply the sum of the
repletion cost and the cost of harvesting the resulting oysters.
The repletion costs are reported in tables from 1983 VNRC repletion
cost data sheets (unpublished VMYRC records). These recorded costs include
the cost per bushel of acquiring, transporting and planting shell or seed
by shell type, seed, and river system. Unfortunately these costs are not
broken down into their individual components.

The inability to consider purchasing, transportation and planting
costs separately masks the fact that each repletion activity is itself a
I ~ ~process composed of several subactivities. Redefined in this way it can be
seen that there exists alternative shell or seed sources and their
associated transportation costs, transportation technologies, and shell or
I ~ ~seed planting technologies. If it were possible to examine each of these
costs separately it would enable the introduction of a greater level of

I~~~~~~~~~----------
*Two classes of bottom are considered in the model, firm and soft bottom.
Bottoms in which grab samples contain at least 50% shell or shell fragments
I ~ ~in volume are defined as being firm (Haven et. al. 1981). Firm bottoms
require planting of 7,000 bushels of shell per acre to make them suitable
for oyster culture. All other bottoms are classed as soft and require
I ~ ~10,000 bushels of shell per acre to make oyster culture possible.
**VNARC personnel recommend planting between 550 and 750 bushels of seed per
acre depending on the size and the spat count (number of spat per bushel)
of the seed. An average of 650 bushels per acre was chosen for the model
to allow for this variation in seed quality.

I ~~~~~~~~~~~~-77-

detail into the model. The model would be able to determine not only the
I ~ ~shell type to plant, where to plant it, and how much to plant, it would
also determine the least costly way of purchasing, transporting and
planting the shell.

I~~~Sci 2 .2: Piate Grounds Prdcto Costs

Before describing how the private planter production costs were
estimated it is first necessary to discuss the assumptions that were
required to motivate private grounds market oyster production. It is
assumed that private planters seek to maximize profit. In so doing,
leaseholders minimize their costs by selecting the most cost-effective
combination of activities required to produce market oysters. Recall that
grondsmarket oyster production be profitable. Economic theory of the
firm, however, predicts that the individual will only produce a positive
amount of output if at least all operating costs are recovered in the short
run and at least all fixed plus operating costs are recovered in the long
I ~ ~run. Given that the model considers a ten year planning period for public
and private decisions, it is assumed that no private grounds production
will occur unless long run economic returns to private grounds market
N ~ ~oyster production is greater than or equal to zero. In order to guarantee
that private grounds market oyster production is positive the assumption is
made that the price of market oysters is always at least as great as the
unit cost of producing market oysters. Similarly, for private watermen
working the public grounds, it is assumed that the price of market oysters
is always sufficient to cover all harvest costs incurred by the harvester.
The costs associated with private grounds market oyster production are now
discussed. As previously stated, harvester costs will be presented in a

* ~~~The costs associated with private grounds oyster production considered
in the objective function, consist of the cost of acquiring, transporting,
and planting seed along with the cost of harvesting the market oysters.
The model considers production of market oysters in a two step process by
including two separate activities. One for purchasing and transporting
seed oysters, and another for planting seed and harvesting market oysters.

I ~~~Each seed purchase and transport activity defines a unique combination
of the river system from which seed is transported and the receiving river
system. The objective function coefficient for each activity is determined
U ~ ~by the price of the seed plus the cost of transporting the seed. These
coefficients are estimated by determining the price of seed in each seed
producing river system and the transportation cost of bringing the seed to
the planting site. VMRC seed and market oyster production and price data
were used to determine the price of seed in seed producing river systems.
Little information, however, is available to estimate the cost of
transporting seed. According to records of a Rappahannock river planter
obtained by Haven et. al (1981), transportation costs ranged between 25 and
50 cents per bushel. Unfortunately no insight is offered as to the

1 ~~~~~~~~~~~~-78-

distances corresponding to these costs.

In order to establish a relationship between transportation costs and
distance, the 25 to 50 cent cost interval is assumed to provide upper and
lower bounds to per bushel transportation costs. The purpose here is to
provide a means of assigning transport costs based on relative distance
between seed origins and receiving systems. In this respect the 25 to 50
cent interval is being used as an indexing mechanism. Using five cent
increments, relative transportation costs were assigned to each seed
purchase and transport activity. These assignments are made on the basis
of the geographical location of a receiving system relative to how far away
it is from a specific seed origin when compared to the closeness of other
potential receiving river systems competing for seed from the same source.
For example, the cost assigned to transporting seed from the James to the
Rappahannock is $.40. The cost of transporting seed assigned to the York
from the James is $.25. Here shipping seed from the James to the York is
less costly than shipping seed from the James simply because the York is
closer to the seed source. The opposite is true when shipping seed from
the Great Wicomico to the Rappahannock and the York, ($.30 and $.45
respectively).

The transport costs are not inflated because they are based only on
relative distance between any two river systems and not on actual shipping
cost data. If these shipping costs were inflated, only the magnitude of
the cost would change. The relative difference in transport cost between
any two river systems as compared to any other two river systems, however,
would remain unaltered. The objective function coefficient for each seed
purchase-to-transport activity is left on a per-bushel basis and may be
calculated with the following formula:

Purchase-to-Transport Cost/bu i =
Seed Pricej + transport cost from system j to system
where: i = receiving system
j = seed origin

The second step in the production process is the seed planting and
market oyster harvesting activity. Once again a discussion of harvesting
costs will be put aside momentarily. Seed planting costs simply involve
the costs of putting the seed on leased growing bottoms. For simplicity it
is assumed that a uniform planting technology is employed by all private
planters. Seed planting costs can then be assumed to be the same across
all private planters and river systems. Seed planting costs are reported
by Haven et. al. (1981) as being $.30/bu (1975 dollars). When inflated to
1983 dollars (using the CPI), and multiplied by a fixed per acre planting
rate, the cost of planting seed on any one acre of leased bottom is
$412.5*.

*According to Haven et. al. (1981), planting rates on private grounds range

-79-

The objective function coefficient for a one acre unit of any private
I ~ ~seed-transplant-harvest activity is the sum of the seed planting cost and
the cost of harvesting the resulting market oysters. The following formula
can be used to determine the objective function coefficient for each
private seed-transplant-harvest activity:
Seed-Transplant-Hlarvest Cost/acre =

Seed Planting Cost/acre + (Harvest cost/bu.) x

3 ~~~~(bushels harvestedi). **

where: i = river system

I ~~~Estimation of private production costs is made difficult by the lack
of published information on what these costs are. The records of the
Rappahannock grower mentioned previously constitute the only source of
I ~~published quantitative data which was made available. While these records
are better than none at all, they do not represent a sample of the cost
conditions faced by other planters in the Rappahannock or any other river
system. The lack of published information on private leaseholder's oyster
production techniques also presents a problem in that there is no way to
ensure that all relevant cost information has been included in the
production cost estimates incorporated in the model. For example, during
I ~~the normal production process some amount of shell is lost due to siltation
or removal during harvesting. Substrate maintenance costs are, therefore,
likely to be incurred. It was not possible to determine under the current
information base what these costs might be. A better information base
would permit more accurate and reliable estimation of private market oyster
production costs.

between 500 and 1,000 bushels per acre. Planting rates chosen by private
I ~ ~planters depend on seed size and the firmness of the growing bottoms.
Bottom type for all leased bottoms is assumed to be the same across all
planters and river systems. An average planting rate of 750 bushels per
3 ~ ~acre-was chosen for the model to account for varying seed size and to
reflect a bottom quality which is assumed to be representative of most
leased acreage.

**Note that the seed planting cost per acre has already been assumed
constant. It will be shown later that per bushel harvest cost is also
assumed constant. The value of this coefficient varies, therefore, only
3 ~~with the number of bushels of market oysters harvested from a given acre of
leased bottom in a given river system.

* ~~~~~~~~~~~~-80-

Section 2.3: Oyster Harvesting Costs

The harvest of market oysters takes place on both public and private
grounds. It is assumed that only hand tongs are used on public grounds and
I ~ ~all private planters use an oyster dredge. It is necessary, therefore, to
estimate a cost coefficient for each gear type. Both hand tong and dredge
harvest costs were estimated by creating a budget which was estimated on an
annual cost per harvest season basis. Items included in each budget are
vessel and gear maintenance, fuel, wages, food, and amortized vessel and
gear principle plus interest costs.

I ~~~~Table A.9 presents the format that was used to form each budget. The
per bushel harvest cost for each gear type was estimated by dividing total
annual harvest cost by average annual landings for a representative
I ~ ~harvester. This coefficient is assumed to be constant for all harvesters.
It is now possible to substitute this per bushel harvest cost coefficient
into the following formula to obtain the objective function coefficient for
each activity previously discussed in this section.

Objective Function Value i,k=~

I ~~~$/activity unit ~+ (harvest cost/buj) x

(bushels harvestedi,k)
I ~ ~~where: i = activity type
j = gear type
k = river system

The principle source of data for compiling the harvest budgets is the
Chesapeake Bay Harbors Study (1961). This study, conducted by the Corps of
Engineers in .1961 surveyed watermen to determine various harvest related
I ~ ~expenditures. While the study does not construct budgets for different
gear types, it does provide insight into the cost items associated with
harvesting oysters for several gear types and vessel sizes. Also included
in the survey is information relating to hours and days worked per year and
average daily catches. The more recent work by Haven et. al. (1981), The
Oyster Industry of Virgini: Its Status, Problems and Promise, was also
I ~ ~used to confirm and supplement the find ings reported in the the Corps
study.

Even with these sources, one basic difficulty arises. While technical
information on vessel, gear, fuel and labor costs is not difficult to
obtain, information on average daily catch by gear type remains elusive.
The problem arises because daily catches vary depending on where oysters
I ~ ~are being harvested, how many men are in the boat and even with the
harvester himself. The most complete information which is contained in the
Corps study reports a range of 20-300 bushels per day for a medium dredge
and 5-100 bushels per day for hand tongers. These ranges apply to the
entire survey area which includes only the Southern half of the Virginia
portion of the Chesapeake Bay. More reliable harvest cost coefficients

Table A. 9

I ~~~~~~~~~Annual Harvest Cost Budget

I ~~~Variable Costs

Vessel Maintenance

Gear Maintenance ___

3 ~~~~~~Fuel__

Wages

I ~~~~~~Food__

3 ~~~~~~~~~~~Total Variable Costs

Fixed Costs

U ~~~~~~Vessel Principal

* ~~~~~~Gear Principal__

Interest Payment

3 ~~~~~~~~~~Total Fixed Cost___

Total Cost___

I ~~~~~~~~~~~~-82-

would necessitate a new survey or improved methods of catch reporting to
I ~ ~obtain a better understanding of what average daily catches might be by
region and gear type.
Section 1.0.: Estimation of Constraint Coefficients

This section discusses only estimation of the coefficients which
determine the bounds or Right-Hand Side (RHS) for each constraint. All
I ~ ~other coefficients found in the constraint rows will be discussed in the
Activities Coefficients section of this appendix.

Referring to the model tableau, (Figure A.4) it can be seen that the
first three constraints are oyster grounds acreage constraints. The RHS
for the constraints RBFt and RBSt define the total quantity of firm and
soft bottomed Baylor grounds respectively in any given river system. The
I ~ ~RHS coefficient for RLt'defines the total available acreage of leased
bottoms in any given river system. In the most recent survey of the Baylor
grounds VIMS scientists classified each discrete oyster growing area by
size and bottom type. This information is recorded and reported in The
Present and Potential Productivity of th Baylor Grounds in Virginia (Haven
et. al. 1981). Total availability of public grounds by bottom type and
river system was determined by summing the acreage of each discrete area
classified as being at least ten acres in size. The ten ace cut-off was
offiersconsidered feasible to replete. Acreage of privately held oyster
grounds by size of holding is published in Th Oyster Industry of Virginia:
Lt~ Status, Problems Promise (Haven at. al. 1981). The total lease
acreage was estimated by determining the amount of privately held bottoms
I ~ ~in each river system held in units of twenty acres of more in size. Any
acreage held in units less than twenty acres is either riparian or
considered too small to be commercially productive and was, therefore, not
* ~~included in the leased bottom totals.

Following the acreage constraints are two constraints, one for each of
the two different shell types used by VNRC in its repletion program. An
inspection of the RIHS coefficients shown in Figure A.4 for each of these
constraints (FRSt and DRSt ) reveals that they have no upper limit. The
reason for this is a lack of information in regard to the limits of
I ~ ~availability of either reef or fresh shall. There is no way to tell
whether VMRC uses reef shell instead of fresh shell or vica versa because
one is in short supply relative to another. Alternatively, there may other
I ~ ~reasons for choosing one shall type over the other. In some studies, for
example, reef shell has been found to be slightly more efficient in
catching spat than fresh shell (Campbell and Forste, 1978). The lack of an
upper bound avoids this measurement problem as the quantity of either type
of shell used becomes an output of the model solution.
The constraints NSPCt and NMDPt determine the limits of natural
(unaided by man) seed and market oysters produced annually on Baylor
grounds. Annual natural seed production was estimated by averaging seed

* -~~~~~~~~~~~~83-

harvest reported in VMRC Annual Reports and VMRC Oyster Repletion and
Production Data (in pub.) for the years 1976-1983. It was then assumed
that some of that production had been man-induced. The quantity of man-
induced seed oyster production can be calculated by multiplying the number
of acres shelled in the James River seed area in the previous year by the
corresponding seed oyster productivity coefficient*. This estimate of man-
induced seed oyster production was then subtracted from total seed
production to obtain an estimate of naturally produced seed. This quantity
is estimated with the following formula:

Natural Seed t=,Total Seed Harvested t'

(Acres Seed Beds Shelled it-) x (Productivity i).
where: i = river system
t = time

Natural market oyster production was calculated in a similar manner.
Total public ground market oyster production was averaged over the same
eight year period. The 1983 repletion program was chosen as a
representative repletion year. The man-induced market oyster production
was then estimated by summing the product of repleted acres in each river
system and the appropriate market oyster productivity coefficient. Total
man-induced market oyster production was then subtracted from total public
grounds production to obtain an estimate of "natural" market oyster
production. In formula form this coefficient may be expressed as;

Natural Market Oyster Production =

Total Public Grounds Production t

Sum (Repleted Acres j, t-z ) x

(Productivity Coefficient i~j)
where: i = river system
j = type of repletion activity
t = time
z = maturation period ie. for;
j = shell-harvest, z = 3
j = seed-transplant-harvest, z = 2.

The RHS coefficient for the BT row represents the total funds
available to VMRC to carry out its repletion activities. This coefficient
was estimated by averaging VMRC expenditures for repletion purposes over
the years 1979-1982. VMRC Annual Reports were used to obtain the necessary

*The term "productivity coefficient" refers to the total production of seed
or market oysters per unit of a seed or market oyster production activity.
The estimation of these coefficients are discussed in the next section.

-84-

data.

The coefficient for the prespecified harvest goal was determined by
averaging oyster production from both public and private grounds to
U ~ ~determine a total market oyster production goal. This coefficient provides
a harvest goal under current repletion policies aiid budgetary levels. The
initial production goal can be varied and the cost and policy implications
of an increased harvest goal examined. Harvest on public grounds was
averaged over the 1976-1983 period. VMRC production and repletion data
were used for this average. Private grounds market oyster production data
fr om VMRC Annual Reports were only available for the years 1976-1980. The
*~~~sum of average public and private market oyster production was then
calculated to estimate current average annual market oyster production.

3 ~~~The remaining rows in Figure A.4 are transfer rows whose RHS
coefficients are by definition equal to zero. These rows will not be
further discussed.

U ~~Section 4.0: Estimation of Matrix Elements

The following discussion will deal specifically with the matrix
elements of the model tableau. Once again Figure A.4 will be used as a
reference. The matrix elements in the RBF, RBS, and RL rows indicate the
amount of Baylor or lease acreage required to undertake any given activity.
Acoefficient of 1 will be found in the intersection of any one of these
rows and market or seed oyster production columns. The interpretation
being for every unit of a seed or market oyster production activity one
* ~~acre of bottom must be used.

All matrix elements denoted by the letter F represent the quantity of
fresh shell required by employing a unit of any particular oyster
I ~ ~production activity. Similarly the letter D represents the quantity of
dredge shell required per unit of given oyster production activity. An
inspection of these coefficients shows that all of them are associated with
I ~ ~public grounds repletion activities. Estimation of these coefficients is
based, therefore, on shelling rates recommended by VMRC pesonnel.

The next set of matrix coefficients is denoted by the letter B in
U ~ ~Figure A.4. These coefficients define the expenditures required by VMRC to
undertake an acre of any public grounds seed or market oyster repletion
activity. VMRC contracts private individuals to carry out its repletion
a c iiis The value of these services represents the cost or budget
requirement for a particular repletion activity. These costs are reported
in the VMYRC Annual Reports on a total cost per river system basis. Also
reported is the total amount of shell or seed placed in each river system.
The coefficient B was estimated by calculating the per bushel cost of
planting fresh or dredge shell or seed by river system. The per bushel
cost of planting shell or seed was multiplied by their respective shell or
seed per acre planting rates. The resulting product yields the estimated

I ~~~~~~~~~~~~-85-

budget requirement of undertaking an acre of a given repletion activity.
I ~ ~In formula form, this relationship may be expressed as follows:
Bushels/acrer,i = ($/bur,i) x (planting ratej).
I ~ ~~where: i = river system
r = type of repletion activity
j = material planted ie. shell or seed.

The elements denoted by T determine the amount of dollars generated by
initiation of any public grounds market oyster or seed oyster harvest
I ~ ~activity. The coefficient T is a function of the quantity of seed or
market oysters harvested on public grounds and the per bushel price of the
seed or market oyster. The tax revenues come directly from the public
I ~ ~rocks oyster repletion tax which is levied on all oysters harvested on
public grounds with the amount of the tax depending on the per bushel sale
price of the harvested oysters. Tax rates from seed and market oysters
were obtained from Laws of Virginia (1980). The coefficient T was then
estimated by multiplying the appropriate tax rate ($.50 for all market
oysters and $.10 for all seed oysters) by-the number of bushels of seed or
market oysters produced when a given seed or market oyster production
activity was initiated.
The matrix coefficients denoted by SR and SRL indicate bushel per acre
seed planting rates by public-market oyster repletion and private market
oyster production activities respectively. Seed planting rates for public
grounds market oyster repletion activities were determined through personal
interview with VMRC repletion officers. Seed planting rates for private
grounds market oyster production activity were determined by a literature
search to be 750 bushels of seed per acre, (Haven et. al. , 1981; Bailey and
Biggs, 1968).

Initiation of any public or private market oyster production activity
results in a specific level of market oyster harvest which contributes
I ~~toward the satisfaction of the prespecified market oyster harvest goal.
These coefficients are indicated by the letter G in Figure A.4. An
examination of these coefficients shows that they all correspond to either
3 ~~a shell-harvest activity on public grounds or a seed-transplant-harvest
activity on public or private grounds.

Seed-transplant-harvest coefficients were determined by establishing a
relationship between bushels of market oysters produced per bushel of seed
planted. This relationship is easily expressed as a ratio which can be
estimated. According to Haven et. al. (1981) long run trends indicate that
the bay-wide average for this ratio is 1: 1 or one bushel of market oysters
harvested for every bushel of seed planted. This ratio does, as one might
expect, vary depending on the receiving system. A lack of data made
quantitative estimation of the seed-market oyster relationship impossible
to determine on a river system by river system basis. This problem was
overcome by asking VMRC repletion officers to rank each river system

1. ~~~~~~~~~~~-86-

relative to the 1: 1 ratio. Based on their personal experience in the field
each river system was ranked according to its relative productivity. In
several instances it was possible to obtain quantitative estimates of the
seed-market oyster relationship for specific river systems. Using these
quantitative estimates and the qualitative rankings, it was possible to
U ~ ~estimate a seed-market oyster ratio for each river system on both public
and private grounds. The actual seed-transplant-harvest production
coefficients were estimated by multiplying the seed-market oyster ratios
for each river system by seed planting rates for public and private
N gr~~ltoughs the 1: 1 ratio claimed by Haven is confirmed by other
U ~~researchers (Bailey and Biggs, 1968) and is quite often cited as being the
industry standard, there exist no attempts in the literature surveyed to
estimate this ratio by different rive-r systems. The problem comes down to
I ~ ~a lack of data. This data problem relates particularly to the
unavailability of bar-specific harvest data. It is known, for example,
where VMRC places its shall but without bar-specific harvest data it is not
I ~ ~possible to determine the quantity of market oysters which were produced as
a result of that repletion effort.
The second market oyster producing activity is the shell-harvest
I ~ ~activity. Ideally a similar ratio comparing bushels of market oysters
harvested to every bushel of shall planted should be used. Unfortunately,
as before, the data necessary to calculate this ratio is lacking and VMRC
I ~ ~personnel were unable to rank or estimate the shell-seed-market oyster
relationship for any of the river systems considered in the model. What is
required is an estimator for this relationship that can serve as a suitable
I ~ ~proxy for the shell-seed-market oyster relationship.
The purpose of shelling oyster bottoms is to ensure that oyster larvae
have a suitable substrate upon which to attach. Shelling oyster bottoms,
if done at the appropriate time, enhances the possibility that a successful
strike of seedling oysters will result. If the assumption is made that an
oyster bar repleted by shelling is at least as productive as an unrepleted
I I~~naturall bar and if it is possible to determine the per acre adult oyster
population density on the unrepleted bar then, it may be assumed that the
adult oyster population on the unrepleted bar provides a lower bound to
I ~ ~mature oyster densities on the repleted bar. Fortunately, mature oyster
per acre population densities are reported by Haven et. al. (1981). In the
manner described above these population density estimates were used as
proxie s for the shell-seed-market oyster relationship.

.1 ~~~~~~~~~~~~-87-

The final coefficient to be discussed here is denoted by SP in Figure
A.4. These coefficients represent the quantity of seed produced by
shelling and acre of seed bed. Once again Haven et. al. (1981) was relied
upon for estimates for these coefficients. The lack of quantitative data
is as much a problem here as it was for the previous two coefficients.
Like the seed-transplant-harvest productivity coefficients, the shell-seed
productivity coefficients involve estimating a ratio. Here the ratio is
bushels of seed oysters produced per 100 bushels of shell planted. Shell-
seed ratios are reported in Haven et. al. (1981) only for the Piankatank
and the Great Wicomico. Both of these river systems were not productive of
seed prior to intensive shelling efforts on the part of VMRC. This unusual
situation allowed Haven et. al. to estimate the shell:seed ratios for these
two systems by making the assumption that all seed produced in these river
systems was attributable to VMRC's shelling activities.

Similar situations, however, do not exist for the James, Corrotoman or
Mobjack Bay areas. Shell:seed ratios for these systems were determined by
ranking all the river systems in terms of their historical production and
importance as a seed area. This ranking (from most to least important) is
as follows: James, Great Wicomico, Piankatank, Corrotoman and Mobjack Bay.
As stated earlier, quantitative estimates of the shell:seed ratio exist for
both the Great Wicomico and the Piankatank rivers (1:5 and 1:7
respectively). The James has been and remains the most important seed area
in the Virginia portion of the Bay. For this reason, the shell:seed ratio
for the James was set higher than either the Great Wicomico or the
Piankatank at 1:4. Neither the Corrotoman nor the Mobjack Bay are
currently productive of seed but could possibly be brought into seed
production if VMRC chose to shell these areas heavily (Haven et. al. 1981).
The seed production potential of these systems is unknown and the
shell:seed ratios for these areas were, therefore, set arbitrarily low at
10:1. The actual shell-seed production coefficient for each river system
was estimated by multiplying the shell:seed ratios by the number of bushels
of shell planted on firm Baylor grounds*.

*The shelling rate for firm bottoms was chosen because it reflects the
minimum amount of shell required to provide sufficient quantities of
exposed shell to ensure satisfactory spat settlement. The shelling rate
for soft bottoms includes additional shelling to firm up the bottom. this
shell becomes buried and is not available as a setting material. As a
result, shell:seed coefficients would be biased upwards because of the
additional non-productive shell.

-88-.

I
I
I
I
|I -Appendix B
Demand For Chesapeake Bay Oysters
I Oral Capps and Leonard Shabman

I
I
I
I
I
I
I
I
I
I
I
I -89-
I

I ~~~~~~~~~~Appendix B

Demand For Chesapeake Day Oysters
Oral Capps and Leonard Shabman

This appendix reports upon an econometric demand model estimated for
Chesapeake Bay oysters. The model structure and estimations were developed
during 1984. During 1985 and 1986 attempts were made to secure -data
suitable for updating the model. However, retail price series are no
longer available and time lags in publication of data for other variables
made model updating infeasible.
Seti j.Q: General Background

The model was estimated using time series data to identify demand
relationships for Chesapeake Bay (both Maryland and Virginia) oysters. No
consideration is given to oyster price differentials by state or by
grounds. There was little evidence that average ex-vessel prices differed
by state, however, there is some evidence that prices may differ by grounds
U ~~type (JLARC, 1983).

The estimation of a demand system independent from the supply sector
is justified by the nature of the commercial oyster harvest industry.
I ~ ~First, the supply is determined completely within the harvest sector; the
retail and wholesale sectors merely accept the given supply, they do not
determine it. Second, the harvest in a particular year is determined by
the costs of harvesting in that year (which the fisherman presumably knows
before harvesting) and the ex-vessel prices received in previous years. It
is assumed that individual fishermen use previous prices to form
expectations of current prices, which they further use to decide (along
* ~~with knowledge of current costs) whether or not to participate in the
season'Is harvest; once they have decided to do so, they harvest as much as
they can, regardless of the price they are receiving for their harvest.
These two factors together imply that demand and supply can be modeled
separately, since the current price has no effect on current supply.

Third, consumer demand for oysters is price-dependent; i. e. , instead
I ~ ~of the market price determining the quantity purchased, with the quantity
purchased variable and the pricer fixed (at a market-clearing level), it is
assumed that in seafood markets, a fixed quantity (determined by the
I ~ ~harvest sector) of the product reaches the market, and this quantity is
priced to clear the market. Thus, the dependent variable in the consumer
(retail) demand equation is not the quantity of oysters purchased, but the
I ~ ~retail price; as in traditional demand theory, however, prices of
substitute (e.g., other meat, poultry, and fish products) and complement
(products, income, and other economic and demographic factors could
* ~~influence the dependent variable.

I -~~~~~~~~~~~~90-

Fourth, and finally, the approach to modeling oyster demand assumes
that retail, wholesale, and ex-vessel prices are all determined
simultaneously; i. e. , each price appears on the right-hand side of the
equations explaining the other prices. This simultaneous determination of
I ~ ~prices is in large part due to the fact that the price equation for each
sector is in reality not a demand equation for that sector but a reduced
form equation for the sector, a combination of the corresponding demand and
supply functions: the wholesale price equation, for example, is a
combination of the equation explaining the demand for wholesale seafood
(which is determined largely by the retail demand, as represented by the
retail price) and the supply of wholesale seafood (which is determined
I ~ ~largely by the cost of producing seafood at the wholesale level as
represented by the ex-vessel price, which is the cost of the raw input used
in the wholesale sector). Together, along with consideration of the model
I ~ ~requirements, these four factors led to development of an oyster demand
model which is developed separately from supply and the demand model
comprises a set of three simultaneous equations, explaining the
3 ~~corresponding retail, wholesale, and ex-vessel prices.

Model requirements and economic theory together were used to determine
the overall structure of each model; i. e. , they determined the types of
I ~ ~variables to be included in each model, as well as how in general these
variables would relate to one another. Final specification of the
variables and functional relationships depended in large part, however,
I ~ ~upon consideration of factors peculiar to the species under consideration,
the importance of the Chesapeake Bay harvest of these species relative to
harvests from other regions, and a number of other economic and
institutional factors. In general, these factors fill in the "holes" that
model requirements and economic theory cannot fill; for example, they
determine whose disposable income should appear in the retail price
3 ~~equations as a determinant of consumer demand.

Oyster markets extend outside of the Chesapeake Bay area; i. e. , the
product is sold not only in the Maryland-Virginia-Washington, D.C. area,
I ~ ~but elsewhere in the United States. Thus, the retail and, to a lesser but
still significant extent, wholesale prices for Chesapeake Bay oyster
products are set in national markets; retail demand is thus in part
determined by national disposable income, the national average prices of
other goods, and other national economic and demograpdhic variables.
The structure of the ex-vessel market varies considerably, depending
I ~ ~on species. Chesapeake Bay oysters are relatively homogeneous; i. e.,
Maryland-harvested oysters and Virginia-harvested oysters are largely
indistisnguishable. Because of this, ex-vessel prices for oysters
I ~ ~harvested in the two states are nearly identical. Only one ex-vessel price
equation is needed for the oyster model.

I -~~~~~~~~~~~~91-

Section 2.0: Chesapeake Bay Oyster Demand

Because of the interdependency between Maryland and Virginia in oyster
processing and distribution, a single system of demand equations is
developed for Chesapeake Bay oysters. Equations (1) to (3) represent this
demand system. Prices at three separate levels of the oyster industry are
explained: consumer or retail level, intermediate or wholesale and
processor level, and production or ex-vessel level.

RPOY = i(WPOY, C, Y, CPIMPF, POP, TIME) (1)

RPOY = j(RPOY, OYEXVP, QBAY, QSG, IIGS, TIME) (2)

OYEXVP = k(RPOY, WPOY, QBAY, TIME) (3)

where:

WPOY = nominal retail price of oysters (standards) at Baltimore,
Maryland (dollars per pound)
C = total consumption of U.S. oysters including western oysters
and imports (millions of pounds)
Y = nominal total disposable personal income in the United
States (billions of dollars)
CPI = consumer price index of meat, poultry, and fish
(1967 = 100)
WPOY = nominal wholesale price of oysters (standards) at Norfolk,
Virginia (dollars per pound)
QBAY = landings in the Chesapeake Bay (millions of pounds)
QSG = landings in the South Atlantic and the Gulf (millions of
pounds)
IIGS = index of cost for intermediate goods and services (1967 = 100)
OYEXVP = nominal ex-vessel price of oysters, weighted average value
of Maryland and Virginia landings (dollars per pound)
POP = civilian population of the United States (millions)
TIME = time trend.

The relationships among ex-vessel, wholesale and retail prices depend
upon consumer demand, product supply and costs of marketing (Gardner 1975;
Heien 1980). Specifically, retail price influences wholesale and ex-vessel
price, wholesale price influences retail and ex-vessel price, and ex-vessel
price, influences wholesale price. The interdependent nature of oyster
price determination constitutes a simultaneous system in which the
endogenous variables are the market prices at the three levels in the
marketing chain. However, it does not follow from this that prices at the
three levels necessarily change together.

The time divisions used in this analysis are probably the major
contributors to the simultaneity of the determination of price levels.
With data for short time units, a recursive or causal chain system for
price determination might be more appropriate than a simultaneous equation

-92-

system. However, in this case, the data interval corresponds to one year,
3 ~~and hence, the simultaneous system is preferable.

Implicit in this system is a price dependent demand (Fox 1953; Waugh
1964) for oysters. Chesapeake Bay landings in any time period are not
I ~ ~responsive to current ex-vessel prices because harvest effort (labor and
seed planting) responds to price with a time lag. A similar logic is
presumed to apply to landings outside the Bay region which also may respond
to prices with a time lag. Thus, landings and total consumption in any
year are treated as being exogenous when estimating the parameters in the
model. Total consumption is set equal to total U. S. landings plus
* ~~imports.

It is hypothesized that in the retail price equation, the coefficient
on consumption is negative, reflecting the usual inverse relationship to
I ~ ~price. The coefficient on income is hypothesized to be positive,
reflecting the arguement that oysters are normal goods. CPIMPF is an index
used to measure the price of substitute products such as meat, poultry,
I ~ ~fish, and shellfish; As such, it is hypothesized to be positively related
to the retail price of oysters. -Also, similar to income and prices of
substitute products, time and population are hypothesized to be positively
related to retail price. The wholesale price is also expected to be
positively related to retail price.
At the wholesale level, it is hypothesized that Bay landings and South
I ~ ~Atlantic and Gulf landings have inverse impacts on the wholesale price.
Therefore, the coefficients are expected to be negative. The
unavailability of time-series data on wholesale cost components led to the
use of the index of intermediate goods and services (IIGS) as a proxy
H ~~variable for marketing costs other than raw materials. IIGS is an index
devloedby the USDA to measure trends in marketing costs for the food
processing sector. As such, it can be assumed to represent costs in the
oyster processing sector. Ex-vessel prices, which are the raw material
costs, are expected to be positively related to wholesale price. Retail
price is also hypothesized to be positively related to wholesale price.
The coefficient on the time trend variable is hypothesized to be positive,
reflecting a general upward price trend over the period.

In the ex-vessel price equation it is expected that landings will be
inversely related to price. Coefficients on wholesale price and retail
price are expected to be positive. The time trend variable is expected to
be negative to reflect the fact that nominal ex-vessel prices over the time
period did not rise with general inflation trends.
The demand model was estimated using annual data for 1960 to 1980.
I ~ ~Collecting historical data for each of the variables identified in the
model specification step was required to estimate the parameters of the
equations. Fisheries-specific information on prices was collected from the
I ~ ~Shellfish Market Review and the Fishery Statistics of the United States,
both published annually (or more frequently) by the National Marine

I ~~~~~~~~~~~~-93-

Fisheries Service, a division of the U.S. Department of Commerce. Data on
national income, interest rates, consumer price indices, and other national
economic and demographic variables were obtained from various Department of
Commerce and Department of Agriculture publications. In general, annual
data for the 1960-1981 period were obtained.

Ordinary least squares (OLS) could not be used to estimate the
parameters of the demand sub-models because of the simultaneous structure
of these submodels. In the situation where two or more variables are
jointly determined in a set of structural equations, as is the case here
(where the retail price determines the wholesale price, and the wholesale
price determines the retail price, etc.) OLS produces coefficient estimates
that are biased (i.e., the expected value of the parameter estimate is not
the true parameter value). To overcome this bias, the two-stage least
squares (2SLS) estimation technique was employed to estimate the
parameters. This commonly employed statistical technique produces
coefficient estimates that are assymptomatically unbiased (i.e., unbiased
for very large samples), and that are also less biased for finite samples
than are the corresponding OLS coefficient estimates. However, attempts to
estimate the simultaneous equation system using the traditional methods of
two-stage and three-stage least squares were plagued by deleterious
collinearity problems. The presence of collinearity in the first stage and
the second stage regressions was confirmed by examination of the singular
value decomposition of the data matrix, the variance-decomposition
proportions, and the variance inflation-factors (Belsley, Kuh, and Welsch,
1980). According to Belsley, Kuh, and Welsch, strong variable
intercorrelations, which ultimately lead to degradation of structural
parameter estimates, exist in the case of condition indices in excess of
30, variance-decomposition proportions in excess of 0.5, and variance
inflation factors in excess of 10. These various measures not only provide
reference points to determine the seriousness of the collinearity problem
but also pinpoint the variables that share in the collinearity. Also,
these measures constitute generalizations of the traditional detection
devices of collinearity, namely, the use of pairwise correlation
coefficients of the data matrix and the use of eigenvalues and eigenvectors
of the matrix of correlation coefficients. Not surprisingly, due to the
nature of the data, almost all the variables in the simultaneous equation
model participate in the collinearity.

To overcome the effects of collinearity in the simultaneous equation
system, this study employs the adaptation of ridge regression with two-
stage least squares (Capps, 1982; Vinod and Ullah, 1982; and Maasoumi,
1980). Although successfully used in economic research to reduce the
effects of collinearity in single-equation applications (Brown, 1973; Brown
and Beattie, 1975; Vinod, 1976; Watson and White, 1976; Moscardi and
deJanvry, 1977; Belongia, 1979), the use of ridge regression in conjunction
with simultaneous systems has rarely been previously attempted. However,
the adaptation of ridge regression with two-stage least squartes provides a
reasonably straightforward method to potentially improve the structural
estimation of simultaneous models.

-94-

In this application, the ridge regression modification occurs in the
second-stage estimations, where the emphasis lies with structural
estimation. The first-stage estimations are oriented towards obtaining
predictions of right-hand side endogenous variables, wherein each
endogenous variable is expressed as a function of all predetermined
variables in the models. Consequently, given the emphasis on prediction
rather than on structural estimation in the first-stage estimations, the
ridge regression modification to circumvent the collinearity problem occurs
onl~y in the second-stage estimations.

I ~~~In brief, the procedure entails the addition of small positive
increments, k-values, to the correlation matrices of the second-stage
equations. The selection of the k-values for the respective equations is
based on the Ridge Trace (Hoerl and Kennard, 1970). The Ridge Trace is a
plot of the structural coefficients versus various k-values. The choice of
the k-values is indicated by the point at which the structural coefficients
begin to stabilize. A draw-back to this criterion rests on the fact that
the k-value selection process is subjective.
The interaction of ridge regression with two-stage least squares
I ~~reduces the effects of collinearity, thereby making it possible to
partition the separate effects of the various factors influencing the
nominal price at each level of the marketing chain. Structural parameter
I ~ ~estimates of the equations are exhibited in Table B. 1. The standard errors
of the coefficients of each equation are placed in parentheses below the
coefficients. .Although conventional tests of significance are not exactly
applicable to parameters obtained from estimating simultaneous equation
I ~ ~models, the estimated structural parameter is judged to be significantly
different from zero when the ratio of the parameter estimate to the
* ~~associated estimate of standard error is greater than two.

All estimated coefficients in the model have signs consistent with
prior theoretical expectations. The model explains approximately 99 percent
of the variation in the retail and wholesale prices of oysters and over 90
U ~~percent of the variation in ex-vessel price. The conventional goodness-of-
f it statistic, R2, for each of the three equations is the square of the
* ~~Pearson product-moment coefficient of actual and predicted prices.

Durbin (1957) and Malinvaud (1970) have suggested that the
conventional single-equation Durbin-Watson statistic be used to check for
I ~ ~serial correlation of disturbances in the simultaneous equation setting.
The appropriate number of degrees of freedom is (k,T) where K is the number
of predetermined variables used in the first-stage estimations and T is the
I ~ ~number of observations. For this application, K = 8, and T = 21. The
Durbin-Watson statistics are 0. 7703 for the retail price equation, 1.2658
for the wholesale price equation and 1.0085 for the ex-vessel price
equation. The null hypothesis of no autocorrelation is not rejected for
any of the equations.

I -~~~~~~~~~~~~95-

Table B.1. Demand Estimation: Chesapeake Bay Oyster Industry

Retail Price Wholesale Price Ex-Vessel Price
(RPOY) (WROY) (OYEXVP)

Intercept -0.6186 0.3393 0.5927*
(0.3584) (0.2140) (0.1415)

RPOY 0.1720* 0.1348*
(0.1732E-01) (0.2010E-01)

WPOY 0.6250* 0.3714*
(0. 5564E-01) (0.4706E-01)

OYEXVP 0. 5434*
(0.8665E-01)

QBAY -0.5195E-02 -0.1332E-01*
(0.4910E-02) (0.5219E-02)

QSG -0.8241E-02
(0.4325E-02)

IIGS 0.1605E-02*
(0.3574E-03)

TIME 0.1700E-01* 0.1290E-01* -0.1870E-01*
(0.2445E-02) (0.2977E-02) (0.4181E-02)

C -.07191E-03
(0.3702E-02)

Y 0.4650E-03*
(0.3288E-04)

CPIMPF 0.5237E-02*
(0. 7357E-03)

POP 0.3971E-02*
(0.1578E-02)

R2 0.9908 0.9881 0.9197
D-W 0.7703 1.2658 1. 0085
U2 0.6647 0.7621 1. 1532
K ~~~0. 10 0. 05 0. 05

*Indicates significance.

-96-

Overall, the Theil U2 statistics for the market price equations
indicate that the model is unequivocally better than the naive, no
extrapolation model. The U2 statistic for the no extrapolation model is
unity. While no rigorous test has been developed to judge whether the
~~difference between two U2 coefficients is statistically significant, all
but one of the U2 coefficients of the model are much lower than the U2
coefficients of the naive no extrapolation model. The range of the U2
coefficients for the retail, wholesale, and ex-vessel price equations range
from 0. 6647 to 1. 1532.
In summary, the retail price of oysters is responsive to the wholesale
I ~ ~price; income; the general price level for meat, poultry, fish and
shellfish; population; and the time trend. The key determinants of the
wholesale price of oysters are the retail price, the ex-vessel price, the
time trend, and marketing costs. The ex-vessel price of oysters in the Bay
is responsive to Bay landings, the retail price, the wholesale price, and

I ~~~~The impact of exogenous variables on prices at each market level are
determined from the analytically derived, reduced form equations. These
are reported in Table B.2. It is the analytically derived reduced forms
I ~ ~multipliers which are used in the price simulation for the primary analysis
of this report.

I~~~~~~~~~~~~-7

Table B. 2

Analytically Derived Reduced Form Equations
for Chesapeake Bay Oyster Demand

RPOY = 0.124665 - 0.0120494 QBAY - 0.00798671 QSG + 0.00155547 IIGS

+ 0.0236945 TIME - 0.000890018 C + 0.000575522 Y

+ 0.00648174 CPIMPF + 0.00491484 POP

WPOY = 0.790295 - 0.0192791 QBAY - 0.0127787 QSG + 0.00248876 IIGS

+ 0.0107112 TIME - 0.000273468 C + 0.000176836 Y

+ 0.00199159 CPIMPF + 0.00151014 POP

OYEXVP = 0.869411 - 0.0221045 QBAY - 0.00582263 QSG + 0.001134 IIGS -

0.115278 TIME - 0.00022154 C + 0.000143257 Y + 0.00161342

CPIMPF+ 0.00122339 POP

-98-

I
I
I
I
I Appendix C
|I~~~~~ ~~The Budget Simulator

I
I
I
I

I
I

I
I
I
I
I -99-
l

Appendix C

The Budget Simulator

Section 1. 0: Budzet Format

I ~~~The net returns figures for private planters and public grounds
harvesters shown in Section 6 of Appendix A, Part 2 were estimated with a
budget simulator. The budget simulator employs a spreadsheet program to
I ~ ~evaluate changes in net returns associated with a policy change. The
spreadsheet program allows the use of a 'fixed-format budget in which
formulae are imbedded for each budget item. The individual components for
I ~ ~each budget item are listed outside of the budget format. Table C. 1
illustrates the spreadsheet format for the private planter budget.

Lines 1 through 25 of Table C. 1 comprise the budget portion of the
spreadsheet. The budget formulae are listed in column 2 of Table C. 1. The
budget components in each formula are identified by a row and column
nmber. For example, the formula in row 4 column B is 1xB33. This means
tha the cost per bushel of planting seed is I multiplied by the value that
appears in row 33 column D. Lines 31 to 41 are the names and values of the
budget items. These values are choice values that set the parameters under
I ~ ~which net returns are to be computed. Given these values, the spreadsheet
program performs all the computations indicated in the budget. The
resulting computed budget for the selected parameter values shown in Table
C. 1 is shown in Table C. 2. The data sources for the parameter values shown
in this appendix are discussed in Section 2. 3 of Appendix A, Part 2.
The impact on net returns under any given management policy may easily
be assessed by adjusting the budget parameters to reflect the conditions
that would exist under the new policy.

3 ~~~In a similar fashion to that described above, budgets for public
grounds tongers and dredgers can be formulated. The budget format and the
a sample computed budget for a representative public grounds tonger is
3 ~~presented in Tables C.3 and C.4 respectively. Likewise, Tables C.5 and C.6
present the budget format and sample computed budget for a representative
dredge harvester.

I -~~~~~~~~~~~100-

Table C. 1

Budget Spreadsheet Format for a Representative
Private Grounds Planter

|I A II B
1 Seed Planting Cost/bu.
2
3
4 Seed Planting Cost 1 x B33
5 Seed Purchase Price 1 x B34
6 Seed Transport Price 1 x B35
7 Seed Tax Rate 1 x B36
8
9 Total per Bushel Cost B4 + B5 + B6 + B7
10
11
12 Cost per Acre
13
14 Total Planting Cost B9 x B39
15 Harvest Cost B38 x B37
16
17 Total Cost B14 + B15
18 Average Cost (B14/B15) x (B38 x B41)
19
20 Returns Per Acre
21
22 Bushels Harvested 1 x B38
23 Pounds of Meats Yielded B38 x B41
24
25 Total Returns B23 x B40
26 Net Return B25 - B17
27 Average Net Return B26/B23
28
29
30
31 Budget Parameters
32
33 Planting Cost per bu. .55
34 Seed Cost 2.38
35 Seed Transport .3685
36 Seed Tax .1
37 Harvest Cost 1.602
38 Productivity 750
39 Bushels of Seed Planted 750
40 Ex-Vessel Price/lb. .85889
41 Convert Bu. to Lbs. 4.5919

-101-

Table C. 2

Per Acre Returns to Fixed Costs and Management
For A Representative Private Planter

Seed Planting Cost/bu.

Seed Planting Cost .55
Seed Purchase Price 2.38
Seed Transport Price .3685
Seed Tax Rate .1

Total per Bushel Cost 3.3985

Cost per Acre

Total Planting Cost 2548.875
Harvest Cost 1201.5

Total Cost 3750.375
Average Cost 1.08898277402

I ~~~~~~~~~Returns Per Acre

Bushels Harvested 750
Pounds of Meats Yielded 3443.925

Total Returns 2957.95274325
Net Return -792.42225675
Average Net Return -.230092774015

-102-

Table C.3

Spreadsheet Budget Format for a Representative
Public Grounds Hand Tonger

|I ~ F 1i G
1 Annual Hand Tonger Budget
2
3
4 Nomlhal Variable Costs
5 Vessel Maintenance 1 x G26
6 Gear Maintenance 1 x G27
7 Fuel G28 x G29 x G30
8 Wages G31 x G32
9 Miscellaneous 1 x G39
10
11 Total Nominal Variable Cost G5 + G6 + G7 + G8 + G9
12 Real Variable Cost (1983 $) (Gll/G37) x G38
13
14
15 Returns
16 Annual Catch, (Pounds) G33 x G36 x G30
17 Total Revenue G16 x G35
18 Net Revenue G17 - G12
19 Repletion Taxes Paid G33 x G34 x G30
20 After-Tax Net Revenue G18 - G19
21
22
23
24 Budget Parameters
25
26 Vessel Maintenance 85
27 Gear Maintenance 35
28 Fuel Price, ($/Gallon) .5
29 Fuel Usage, (Gallons/Day) 2.2
30 Days Fished 75
31 Wage Rate, ($/Hour) 3
32 Hours of Labor Hired 113.11
33 Daily Catch, (Bushels) 20
34 Market Oyster Repletion Tax .5
35 Ex-Vessel Price .85889
36 Coversion Rate (Pounds/Bushel) 4.8833
37 CPI 1961 .896
38 CPI 1983 2.984
39 Miscellaneous Expenses 0

-103-

Table C.4

Annual Returns to Fixed Costs and Management for a
Representative Public Grounds Hand Tonger

Nominal Variable Costs
Vessel Maintenance 85
Gear Maintenance 35
Fuel 82.5
Wages 339.33
Miscellaneous 0

Total Nominal Variable Cost' 541.83
Real Variable Cost (1983 $) 1804.48741071

Returns
Annual Catch, (Pounds) 7324.95
Total Revenue 6291.3263055
Net Revenue 4486.83889479
Repletion Taxes Paid 750
After-Tax Net Revenue 3736.83889479

-104-

Table C.5

Spreadsheet Budget Format For A
Public Grounds Dredger

| ~J l ~ K
1 Annual Dredge Budget
2 Nominal Variable Costs
3
4 Vessel Maintenance 1 x K24
5 Gear Maintenance 1 x K25
6 Fuel K26 x K27 x K28
7 Wages K29 x K30
8 Food 1 x K31
9 Miscellaneous 1 x K32
10
11 Total Nominal Variable Cost K4 + K5 + K6 + K7 + K8 + K9
12 Real Variable Cost (1983 $) (Kll/K37) x K38
13
14
15 Returns
16 Annual Catch (Pounds) K33 x K34 x K28
17 Total Revenues K16 x K35
18 Net Revenues K17 - K12
19 Repletion Taxes Paid K33 x K36 x K28
20 Net After-Tax Returns K18 - K19
21
22
23 Budget Parameters
24 Vessel Maintenance 1800
25 Gear Maintenance 350
26 Fuel Price, ($/Gallon) .5
27 Fuel Usage, (Gallons/Day) 14
28 Days Fished 75
29 Wage Rate, ($/Hour) 3
30 Hours Hired (Year) 1187.5
31 Food Expense 675
32 Miscellaneous 0
33 Daily Catch 160
34 Coversion Rate (Pounds/Bushel) 4.8833
35 Ex-Vessel Price .85889
36 Market Oyster Repletion Tax .5
37 CPI 1961 .896
38 CPI 1983 2.984

-105-

Table C. 6

Annual Returns to Fixed Costs and Operator Labor
3 ~~~~~~~for a Public Grounds Dredge Harvester

Nominal Variable Costs

U ~~~~~~Vessel Maintenance 1800
Gear Maintenance 350
Fuel 525
I ~~~~~~Wages 3562.5
Food 675
I ~~~~~~Miscellaneous 0

Total Nominal Variable Cost 6912.5
Real Variable Cost (1983 $)23021. 09375

Returns
Annual Catch (Pounds) 58599.6
I ~ ~~~~~Total Revenues 50330. 610444
Net Revenues 27309. 516694
Repletion Taxes Paid 6000
I ~ ~~~~~Net After-Tax Returns 21309. 516694

I~~~~~~~~~~~~~16

U
I
I
I
I
I Appendix D
~~~~~I ~Contribution of the Oyster Industry to the
Virginia Economy and Tax Base
I
I
I
I
I
I
I
I
I
I
I
I -107-
I

Appendix D

I ~~~~~~Contribution of the Oyster Indusrty to the
Virginia Economy and Tax Base

Sectionl 1~.0: Introduction

This section provides preliminary estimates of the contribution of the
commercial oyster industry to the economy and tax base of Virginia. In
providing these estimates an appropriate approach to regional economic
impact analysis can be demonstrated. The primary purpose of this section
is to provide estimates for the simulation model of this report. A second,
and more long term purpose, is to generally describe the basic framework
for economic impact analyses in order to illustrate basic data needed for
making future improvements to such analyses.
Section _Z.O: Measuring the Regional Economic Impact
of the Oyster Industry
The contribution of an industry (economic sector) to the economy of a
region extends beyond the dollar value of production of that single
industry. By its purchases of labor and materials from other industries
and by its sales to other industries the affect of any one industry on the
economy involves an interdependence with numerous economic sectors. For
I ~ ~example, oyster harvest requires purchase of inputs from sellers of fuel,
labor, and fishing gear. Sales of harvested oysters go to processors,
wholesalers or retailers. All these parties are affected by the fact that
I ~ ~the oyster harvest activity exists within the region. Additionally, income
paid to households employed by the affected sectors of the economy is, in
turn, spent on a wide array of consumer and investment items further
extending the initial effect of the oyster harvest.

In this study, estimates of the impact of the oyster industry on the
Virginia economy are based on a multiplier analysis. Multipliers indicate
I ~ ~the aggregate effects, on the economy of an area (state or region), of some
level (or change in the level) of a specific industry. There are many
multipliers from which an analyst might choose depending upon the
I ~ ~particular measure used to evaluate economic activity--output and value
added are the two measures discussed here.

An output multiplier indicates the change in industry and business
I ~ ~output in the regional economy resulting from a one dollar change in a
specific sector?'s sales to final demand. Sales to final demand are all
sales to buyers outside of the economy being studied plus sales to
I ~ ~households, governments and net capital expenditures within the region's
economy; final demand excludes sales of a sectors output which are used as
inputs by other firms in the regional economy. Thus, if the output
I ~ ~multiplier for the oyster industry in Virginia is 2.0, then ~1.O million of
final demand for oysters results in total output in the states economy of

$2. 0 million. The problem with this multiplier is that total dollar output
includes expenditures made by all affected firms to purchase inputs
necessary for their production. Thus, the output multiplier provides a
gross" rather than "net" income measure.

The value-added/output multiplier indicates the effect of a change in
I ~ ~final demand on the aggregate level of value-added- -the value of output
less the value of inputs used. The concept of value-added is closely
related to the familiar macro-economic measure Gross Domestic Product
(GDP). Value added is a better measure than output of aggregate economic
activity since it excludes purchases made for inputs and, therefore, is a
net" income concept.

The output and value-added/output multipliers for the Virginia oyster
industry, used in this appendix, are based on an existing input-output
I ~ ~model for the Commonwealth. The model was developed by the Regional
Science Research Institute for the use of Virginia Tech. It is based on
the 1972 input-output model of the United States but has been adjusted to
reflect the structure of the Commonwealth's economy (Regional Science
Research Institute, 1981).
Input-output, models portray the economic relationships between the
I ~ ~various sectors in an economy. Since every industry must purchase its
inputs from other industries, a series of relationships ensue which create
complex linkages between almost all sectors. For example, consider fish
I ~ ~harvest. This sector purchases inputs from a rather limited group of'
sectors including the agricultural, forestry and fishing services industry;
the manufactured ice industry; the apparel industry; the ship building and
repairing industry; transportation, and certain other services. However,
I ~ ~these industries must, in turn, purchase inputs from other firms. After a
few rounds of purchases, almost all sectors will be influenced somehwat by
* ~~the original purchases of the fishing industry.

Input-output calculates the level of production required from all
other sectors by a change, for example, in final demand from the oyster
I ~ ~harvest sector. Since each sector is linked to others in a unique manner,
each sector will generate a unique multiplier effect. The original final
demand for a sector-is commonly referred to as the direct effect while the
~~~i mpacts in affected sectors are referred to as indirect effects. The sum
of sect'oral outputs (other than the original final demand) is therefore
referred to as indirect output. Input-output models generate estimates of
* ~~output multipliers by calculating the following ratio:

Output = direct output + indirect output
Multiplier----------------
I ~ ~~~~~~direct output
Input-output models assume that all expenditures for purchased inputs
* ~~by firms are constant proportions of their output. Therefore, the levels
of value-added are fixed as a proportion of the output levels of each

-109-

sector. Thus it is possible to develop a multiplier which indicates the
value added in the economy stemming from a level of final demand in a
particular sector.
Value-added/Output =direct value-added + indirect value-added
- Multiplier---------------------
U ~~~~~~~~~~~direct output

The value-added/output multiplier can be used as an accounting device
I ~ ~to divide existing state GDP among the various economic sectors in the
states economy. Specifically, given final demands for each sector in the
states economy and a sector's value-added/output multiplier, the value
I ~ ~added in the economy derived from each sectors final demand can be
computed. Summing these value-added impacts across all sectors will
closely approximate the GDP for the state. However, it does not follow
I ~ ~that elimination of the industry would reduce state GDP by the calculated
amounts because the resources used in production would be diverted to some
alternative use.. It would only be the case that GDP would fall by the
calculated amount if all labor and capital inputs used in the affected
industries had no value in any other use.

Section 3.0: Virginia' s Commercial Oyster Industry

In Virginia the oyster industry includes the harvest, processing and
distribution of shellfish products. Each of these general activities
I ~ ~requires the purchase of production inputs from other sectors of the
economy and sale of final products to other sectors. These purchase from
and sales to other sectors set in motion a multiplier effect which results
I ~ ~in additional output and value added in the larger economy as a result of
the oyster industry activity. Figure D. 1 is a general depiction of the
flows of oyster products in the Virginia industry. Harvested products move
either to consumers through wholesale and retail distribution channels or
I ~ ~move to processing before being delivered to consumers. In addition the
processing sector may import products harvested outside the state for
* ~~processing.

In order to compute the economic contribution of the oyster industry
to the Virginia economy three types of information must be known: (1)
value and quantity of harvest in Virginia; (2) proportion of harvest
processed in Virginia and imports to Virginia; and, (3) value of processed
product. For purposes of this analysis value and quantity of harvest
information was taken from the simulation model. However no data exist to
I ~ ~determine how much of the harvested product is processed before being
distributed. To make such estimates interviews with persons familiar with
the industry had been conducted. Also, it was assumed for this study that
I ~ ~fifty percent of the oysters processed in Virginia are imported (Virginia
Marine Resources Commission, 1980). It was also assumed that there were no
imports by wholesalers for further distribution. Last values of processed
I ~ ~oysters were not available for recent years and approximations of these
values were made.

-110-

Figure D. I

Product Flows in the Virginia Seafood Industry

[WHOLESALE/RETAIL1 PRI
__________ I PROCESSIN I

III ADDITIONAL IMPORTS OF
[PROCESSING PRIMARY PROCESSED
PRODUCT

ICONSUMER

Key

Harvest: Oyster landings made at Virginia ports for ultimate use by
consumer.

Imports: Oysters in unprocessed condition (or at a stage of preliminary
processing) brought into the state for processing prior to distibution to
wholesale or retail distribution.

Primary Processing: Defined here to include shucking of oysters.

Additional Processing Includes preparation of prepared shellfish products
to decrease consumer preparation requirements (ex. breaded oysters).

Wholesale/Retail: Assembly and transportation to consumers including
restaurant and institutional use.

-111-

Sectionl 4.0: The Multip~lier Effects of Virginia?'s
Oyster Industry

Two distinct multipliers were identified for the Commonwealth's oyster
industry. These include the multiplier effects-from: (1) seafood sold
fresh to households in Virginia and buyers outside the state, and (2)
oysters processed in state. The multiplier effects of these activities are
I ~ ~provided in Table D.I1. Oyster marketed through the fresh market generated
the largest aggregate effect on value-added. This is because of the
relatively high labor intensity involved in harvest. It is important to
I ~ ~keep in mind that these multipliers refer to dollar units and not physical
units. The total value-added in the economy per pound of harvested fish
would be lower for fresh fish than for processed fish since the latter
I ~~would include the value-added of the raw. product plus value-added in
processing.

The use of the multipliers reported in Table D. 1 is illustrated by the
I ~~following general example. Assume 200 pounds of product is harvested in
Virginia with a dockside value of $100. Fifty percent of the landings move
on to processing and the rest is distributed to consumers. Thus $30 of
I ~ ~final demand exists at the harvest level. At the processing sector 100
pounds of raw product harvested in Virginia is processed to provide 75
pounds of processed product (25 pounds are waste). The processed product
I ~~sells for $2.00 per pound yielding a final demand of $150 at the processing
level.

Using the multipliers from Table D. 1, the total effect of these two
I ~ ~final demand levels on the economy can be calculated as shown in Table D. 2.
From Table D.2 it is shown that $100 of landing generated $295.55 of output
3 ~~and $127.95 of value-added in the regional economy.

Section 5. 0: Estimated Impacts of the Oyster Industry

The results for the oyster industry are shown in Table D.3.
Assumptions made for each computation are indicated in footnotes to the
table. Output effects are, of course, substantially larger than value-
added effects from the industry. However, the value-added effects are a
more meaningful measure of the "fnet"? contribution of the seafood industry
to the Virginia economy.

I ~~Section 6.0: Fuur Research Needs

This preliminary economic impact analysis can be improved upon by
I ~ ~improvements in basic data about the seafood industry and by refinements of
the estimated multipliers from the existing input-output model. The
existing input-output multipliers are for aggregated activities and can
only be approximations for the more specific activities considered in this
paper. For example, the multiplier used for oyster harvest was the
forestry/fishery multiplier from the existing model. Estimation of

3 ~~~~~~~~~~~~-112-

multipliers specifically for individual species harvest, processing and
distibuionactivities will require primary survey data from fishermen,
processors and wholesalers on (1) the expenditures made for various
purchased inputs necessary for their production and (2) the volume and
distribution of their sales. Such information can only be gathered from a
sampl~e of repr~esentative local fishermen, processors, and wholesalers.
Thus the cooperation of those whose earn their living from the seafood
~~i ndustry is necessary if more refined estimates than those provided here
are to be developed.

Section 7.0: Indirect Ta Revenue from th OyZster Industry

The value added to the state economy represents a tax base upon which
state general tax revenues are earned. For purposes of this study, only an
approximation of the indirect tax revenues was made. This estimate begins
by assuming a definitional identity between reported measures of value
added from the input output model and the Gross State Product. In fact,
the accounting relationship between GSP and value added is as follows. GSP
is the value of all final goods produced in the State; final goods do not
include products that were used as production inputs into other production
processing. GSP less capital depreciation allowances equals net state
product (NSP) and NSP is an accounting identity with value added from the
input-output analysis.

There are no published state value added or NSP figures. Therefore, a
rough identity of GSP and value added was assumed. Data were collected for
GSP (Cox, 1986) and total state tax revenue United States Department of
Commerce, various years). With these two data series, the percent of GSP
realized in taxes for years 1980-1984 was estimated (Table 4). Applying
the 1980-1984 average percentage of 13. 784% to the estimated state value
added from oyster production, yields the indirect tax estimate reported in

th main text of this report.

Table D. I

Multipliers for the Virginia Oyster Industry
form the Virginia Input-Output Model

Oyster
I ~~~~~~~~Harvest* Process ing**

Output 1.417 1.498

Value-Added/
Output 0.882 0.599

*From Virginia input-output model, sector 9 (fishery and
forestry products).

**From Virginia input-output model, sector 37 (canned and
curred seafood products); assumes 50 percent of raw products

to be processed are imported into the state.

Table D.2

Example of Calculation of Output and Value-Added Effects

Output

Processing .498 * $150 = $224.70

Harvest 1.417 * $ 50 = $ 70.85

Total $295.55

Value-Added

Processing .599 * $150 = $ 83.85

Harvest .882 *$ 50 = $ 44.10

Total $127.95

Table D.3

Economic Impact of the Oyster Industry

Output

Harvest* $986,300 * 1.417 = $1,397,587

Processing** $34,044,006 * 1.498 = $50,997,920

Total $52,395,507

Value Added

Harvest* $986,300 * .882 = $869,916

Processing** $34,044,006 * .599 = $19,030,599

Total $19,900,515

*10 percent of $9,863,000 value of oysters landed is
assumed to be distributed without shucking.

**Assumes one half of oyster processed in Virginia are
imported from elsewhere (primarily Maryland); assumes
40,000 gallons of shucked oysters are sent to additional
processing (breading; stews, etc.); final demand at
primary processing equals $33,057,006 based upon price of
shucked meats of $2.09 per pound; final demand at
additional processing equals $987,000 based upon $2.82
pound for oyster meat after further processing.

-116-

Table 4

Virginia State Tax Revenues as a Percent
of Gross State Product

Virginia GSP Tax Revenue Tax Revenue as
($ Billion) ($ Billion) a % of GSP

1980 59.113 6.712 11.35
1981 65.671 7.883 12.11
1982 71.184 8.643 12.14
1983 77.860 9.098 11.69
1984 87.451 10.018 11.46

-117-