[Joint House and Senate Hearing, 111 Congress]
[From the U.S. Government Publishing Office]
S. Hrg. 111-372
THE FEDERAL STATISTICAL SYSTEM
IN THE 21st CENTURY: THE ROLE
OF THE CENSUS BUREAU
=======================================================================
HEARING
before the
JOINT ECONOMIC COMMITTEE
CONGRESS OF THE UNITED STATES
ONE HUNDRED ELEVENTH CONGRESS
FIRST SESSION
__________
JULY 21, 2009
__________
Printed for the use of the Joint Economic Committee
U.S. GOVERNMENT PRINTING OFFICE
54-938 WASHINGTON : 2010
-----------------------------------------------------------------------
For sale by the Superintendent of Documents, U.S. Government Printing
Office Internet: bookstore.gpo.gov Phone: toll free (866) 512-1800; DC
area (202) 512-1800 Fax: (202) 512-2104 Mail: Stop IDCC, Washington, DC
20402-0001
JOINT ECONOMIC COMMITTEE
[Created pursuant to Sec. 5(a) of Public Law 304, 79th Congress]
HOUSE OF REPRESENTATIVES SENATE
Carolyn B. Maloney, New York, Chair Charles E. Schumer, New York, Vice
Maurice D. Hinchey, New York Chairman
Baron P. Hill, Indiana Edward M. Kennedy, Massachusetts
Loretta Sanchez, California Jeff Bingaman, New Mexico
Elijah E. Cummings, Maryland Amy Klobuchar, Minnesota
Vic Snyder, Arkansas Robert P. Casey, Jr., Pennsylvania
Kevin Brady, Texas Jim Webb, Virginia
Ron Paul, Texas Sam Brownback, Kansas, Ranking
Michael C. Burgess, M.D., Texas Minority
John Campbell, California Jim DeMint, South Carolina
James E. Risch, Idaho
Robert F. Bennett, Utah
Nan Gibson, Executive Director
Jeff Schlagenhauf, Minority Staff Director
Christopher Frenze, House Republican Staff Director
C O N T E N T S
----------
Members
Hon. Carolyn B. Maloney, Chair, a U.S. Representative from New
York........................................................... 1
Hon. Kevin Brady, a U.S. Representative from Texas............... 3
Witnesses
Vincent P. Barabba, Director From 1973-1976 and 1979-1981,
Chairman, Kings County Ventures and the Market Insight
Corporation, Capitola, CA...................................... 5
Barbara Everitt Bryant, Director From 1989-1993, Research
Scientist Emeritus, Ross School of Business, University of
Michigan, Ann Arbor, MI........................................ 6
Martha Farnsworth Riche, Director From 1994-1998, Principal,
Farnsworth Riche Associates, Trumansburg, NY................... 8
Kenneth Prewitt, Director From 1998-2001, Carnegie Professor of
Public Affairs and Vice President for Global Centers, Columbia
University, New York, NY....................................... 10
Charles Louis Kincannon, Director From 2002-2008, Vice President,
Board of Directors of Capitol Hill Village, Washington, DC..... 11
William F. Eddy, John C. Warner Professor of Statistics, Carnegie
Mellon University, Pittsburgh, PA, Chair, Committee on National
Statistics, National Academy of Sciences, Washington, DC....... 25
Andrew Reamer, Fellow, Brookings Institution, Metropolitan Policy
Program, Washington, DC........................................ 26
Submissions for the Record
Prepared statement of Representative Carolyn B. Maloney, Chair... 34
Prepared statement of Representative Kevin Brady................. 34
Prepared statement of Vincent P. Barabba......................... 35
Prepared statement of Barbara Everitt Bryant..................... 37
Prepared statement of Martha Farnsworth Riche.................... 39
Prepared statement of Kenneth Prewitt............................ 40
Prepared statement of Charles Louis Kincannon.................... 42
Prepared statement of William F. Eddy............................ 44
Prepared statement of Andrew Reamer.............................. 46
Prepared statement of Linda A. Jacobsen.......................... 49
THE FEDERAL STATISTICAL SYSTEM
IN THE 21st CENTURY: THE ROLE
OF THE CENSUS BUREAU
----------
TUESDAY, JULY 21, 2009
Congress of the United States,
Joint Economic Committee,
Washington, DC.
The committee met, pursuant to call, at 12:59 p.m., in Room
2203, Rayburn House Office Building, The Honorable Carolyn B.
Maloney (Chair) presiding.
Representatives present: Maloney, Hinchey, Cummings,
Snyder, Brady, and Burgess.
Senators present: Klobuchar
Staff present: Gail Cohen, Nan Gibson, Colleen Healy,
Annabelle Tamerjan, Andrew Wilson, Chris Frenze, and Robert
O'Quinn.
Chair Maloney. The committee will come to order.
Before we begin opening statements, I would like to thank
all of our witnesses for agreeing to testify. I believe it is
the first time in history we have had five former Census
Directors before the committee.
Due to the House schedule, we may not have time to hear the
testimony of the second panel of witness, but, if that happens,
I will ask unanimous consent to place the prepared testimony of
Drs. Eddy, Reamer, and Jacobsen into the record; and we will
reschedule them for a later time.
The Chair recognizes herself for 5 minutes.
OPENING STATEMENT OF THE HONORABLE CAROLYN B. MALONEY, CHAIR, A
U.S. REPRESENTATIVE FROM NEW YORK
I would like to welcome our two distinguished panels of
witnesses and thank them for agreeing to testify today on the
role of the Census Bureau in the 21st century. This is the
first in a series of hearings the Joint Economic Committee will
hold to examine the state of the Federal Statistical System.
The data collected by our statistical agencies are vital to
informing policy debates and evaluating the effectiveness of
those policies we put into place.
As we debate health care reform, Census data tells us that
46 million Americans are without health insurance.
Unemployment, family income, poverty--the numbers we see in the
headlines every day--they are our motivation for making
policies and writing legislation. Federal statistics are a
bargain, costing between $10 and $25 per person each year, but
the information gleaned is invaluable.
I cannot stress enough how heavily policymakers on this
committee and at all levels of government rely on the data
produced by the Census Bureau as we weigh policy decisions. The
data enables us to evaluate whether or not a policy is
achieving the goals that we intended.
We begin today by focusing on the Census Bureau, the
country's largest principal statistical agency. The Census
Bureau is most well known for its role in conducting the
national Census every 10 years. Beyond that, the Bureau
conducts the annual American Community Survey and many other
surveys that provide key information on other economic and
demographic subject areas. The Bureau's population estimates
determine congressional districts and drive how we allocate
funding for millions of dollars in Federal aid.
There is no doubt of the Bureau's significance and the
importance of the work it does. But I am concerned when I see
that the new Director of the Census Bureau, Dr. Robert Groves,
was confirmed by the Senate just last week, 6 months into the
new administration and less than a year--261 days according to
the countdown on the Bureau's Web site--before Census Day 2010
on April 1, 2010.
The decennial Census, the largest peacetime mobilization of
government workers, takes place every 10 years, but the
leadership changes every 4 years with a new administration. But
statistical agencies like the Census Bureau should be absent
political pressures so that the data remains unbiased and
objective.
Today we will hear from former Census Directors who
combined have over 20 years experience, spanning five different
administrations. Yet we find ourselves in the same peril today
as in previous decades with the Bureau, like some heroine tied
to the railroad tracks. Given that we have a wealth of
expertise and knowledge in conducting the Census and we know
how important sound data is to policymaking, I am interested to
hear your perspectives on how to avoid flirting with disaster
every decade. I would like to hear your practical suggestions
of how we can avoid ending up in the same predicament in 2020.
I have introduced legislation, which you have all endorsed,
to give the Census Bureau independent status, similar to the
National Science Foundation and NASA. Other Federal statistical
agencies, such as the Bureau of Labor Statistics and Energy
Information Administration, are part of another executive
branch agency, but the director is appointed for a fixed term.
In order to be relevant to policy decisions, the major
ongoing surveys conducted by the Census Bureau need regular
review, updating, and sometimes complete redesign depending on
economic, social, and technological changes. The Bush
administration era cuts to our statistical and scientific
infrastructure budgets have undermined our ability to evaluate
the effectiveness of our policies.
We must impress upon those around us the value of the
Federal Statistical System and challenge lawmakers and
departments to support the system with resources and ensure
that the statistical agencies maintain a strong position of
independence.
I yield back the balance of my time, and I now recognize
Mr. Brady for as much time as he may consume.
[The prepared statement of Representative Carolyn B.
Maloney appears in the Submissions for the Record on page 34.]
OPENING STATEMENT OF THE HONORABLE KEVIN BRADY, A U.S.
REPRESENTATIVE FROM TEXAS
Representative Brady. Thank you, Madam Chairwoman. Thank
you for holding this important hearing. I want to join you in
welcoming both panels of witnesses testifying before us.
There is broad bipartisan agreement on the importance of
impartial, accurate, and timely economic data. We also need to
ensure that the Federal economic statistics fully reflect the
growing importance of service industries and exports to
America's economy.
The Census Bureau publishes a number of economic statistics
but also collects and compiles data for other agencies. For
example, Census Bureau personnel are engaged in collecting the
data used for Bureau of Labor Statistics' household surveys
from which the unemployment rate is derived.
In recent and early hearings before the committee, I have
voiced my concern about some of the rosy economic assumptions
put in place by the administration which I think will end up
masking much higher deficits than the economy or the Congress
is prepared to accept. I worry as we sit here that perhaps the
administration is attempting to defer the release of its
midsession budget update until after Congress votes on the
massive health care reform bill. I think that is a huge
mistake. Congress should be fully informed of the financial
condition of this country as we vote on a one to two trillion
dollar commitment on health care.
And I also believe, as others do on this committee, that we
ought not have any potential influence of politics in the
Census Bureau, period. Many of us were alarmed earlier to see
about reports that the White House was seeking to directly
oversee the Census Bureau in connection with the 2010 Census.
I continue to believe the political and ideological groups
such as ACORN shouldn't have anything to do with any process
leading up to the decennial census. The Census Bureau is an
important national resource, and the statistical integrity must
be protected from potential political pressures.
I conclude with this. One of the questions I am going to
pursue of our panelists today is how do we make the census even
more accurate and more meaningful in this new economy? How do
we measure better innovation and its role on economic growth
and productivity? National income and products account don't
always adequately measure it. How do we look at expenditures
for technological research and development, brand equity, human
capital, organizational efficiency, all which can help measure
a new economy? We haven't had those statistics and data in the
past, but using your expertise and your thoughts, how do we
more accurately measure the key indicators that are so
important to the economy that may not have even been worth a
whisper a decade or two ago? You have got some insights there I
am anxious to hear about.
[The prepared statement of Representative Kevin Brady
appears in the Submissions for the Record on page 34.]
Madam Chairman, thank you for hosting this.
Chair Maloney. Well, thank you; and I would like to be
associated with your remarks on having more information on how
we could track innovation and the economy. I think that is a
very wise focus that you put forward.
I also share your concern of the independence of the Census
Bureau and welcome my colleague to look at legislation I have
introduced that would create an independent Census Bureau
totally separate from any political influence so that they
could have their budget and plan and go forward appropriately.
We have bipartisan support, and I hope the gentleman will give
it his studied review.
We are extremely honored to have five former Directors of
the Census with us today, whom I will now introduce:
The Honorable Vince Barabba is the Chairman of Kings County
Ventures and Market Insight Corporation. He twice served as
Director of the Census Bureau and is the only person in history
to have been appointed to that position by U.S. Presidents of
different political parties, serving as Director from 1973 to
1976 under Presidents Nixon and President Ford and again from
1979 to 1981 under President Carter.
The Honorable Barbara Bryant joined the faculty of the Ross
School of Business at the University of Michigan in 1993. She
was Director of the U.S. Bureau of the Census from 1989 to
1993. She was appointed by President George H.W. Bush and
confirmed by the Senate as the first woman to head the Census
Bureau in 200 years of census taking.
The Honorable Martha Riche consults, writes, and lectures
on demographic changes and their effects on policies, programs,
and products; and she was a founding editor of American
Demographics, the Nation's first magazine devoted to
interpreting demographic and economic statistics for corporate
and public executives. Dr. Riche served as Director of the U.S.
Census Bureau between 1994 and 1998 under President Clinton.
The Honorable Kenneth Prewitt is the Carnegie Professor of
Public Affairs and Vice President for Global Centers at
Columbia University. Dr. Prewitt served as the Director of the
Bureau from 1998 to 2001 under President Clinton. His most
recent book is The Hard Count: The Political and Social
Challenges of Census Mobilization.
The Honorable Charles Louis Kincannon is the Vice President
of the Board of Directors of Capitol Hill Village, a nonprofit
organization that supports seniors who prefer to remain in
their homes as they age. Mr. Kincannon was nominated as
Director of the Census Bureau by President Bush; and he served
from 2002 to 2008, the longest-serving director since the
Eisenhower administration.
We thank all of you for your service and for being here
today.
Mr. Barabba, if you will begin for 5 minutes, and then we
will go down the line, and we hope to get everybody's testimony
in before we are called for roughly 30 votes, I think. That is
just an astronomical amount of votes. So we will not be coming
back once we are called for votes. So let us try to get moving.
Thank you all for coming.
STATEMENT OF THE HONORABLE VINCENT P. BARABBA, DIRECTOR FROM
1973-1976 AND 1979-1981, CHAIRMAN, KINGS COUNTY VENTURES AND
THE MARKET INSIGHT CORPORATION, CAPITOLA, CA
Director Barabba. Thank you.
I have had a long interest in this subject----
Chair Maloney. Pull your mic to you.
Director Barabba [continuing]. How is that? Is that better?
I have had a long interest and an example of that interest
is found in a 2002 presentation I gave at the 100th anniversary
of the Census Bureau, and it was titled: The Next 100 Years,
Starting Today.
In that presentation, I chose not to focus on new
technologies but instead on the need for a more fundamental
change; and that is because few statistical agencies are
equipped or authorized to comprehensively assess what society
needs to know, the question that you were raising. To do so
would require a dialogue across many functions and the special
interests that need to use that information. The Census Bureau
has at times come close to accomplishing this.
I suggested an improved dialogue about form, accuracy, and
cost in both time and money between those who decide what
information needs to be collected and those who do the
collecting. I stated that a continued improvement in this area
was needed for at least two reasons: The first, it is no longer
sufficient to address the societal issues from a limited
perspective of a functional policy organization such as labor,
commerce, health, education, and so forth; and, second, because
our society now faces increasing complexity at an accelerating
rate of change, government can no longer predict and then
prepare for the future. Instead, we must now use information to
sense and respond and at times to anticipate and to lead.
I would like to use an analogy of solving a jigsaw puzzle
to explain why changes that have occurred in our society have
contributed to the need for a new system of government
statistics. In the mechanistic mindset of the government
statistics, the industrial age encouraged us to think about
addressing problems in government and business as if we were
solving a jigsaw puzzle. Solving a jigsaw puzzle is relatively
simple, because you can assume all the pieces of the problem
that are needed are in the box. Each of the pieces will
interact with only a few other pieces and do it in a very
specified way, and there is only one correct solution. And if
you could confirm it, all you have got to do is make sure all
the pieces are in the puzzle. And if you are not sure of that,
you can look on the cover of the box and the single solution is
there for you to see.
We don't face problems like that anymore.
This solve-the-puzzle metaphor fit reasonably well for most
of the issues we faced during the early part of the 20th
century and represented to a great extent the way things were
thought of at many public and private agencies and,
unfortunately, taught at many colleges and universities.
We now operate in an environment that has a constantly
changing process of relationships and components. Today, it is
more like managing the elements of a molecular structure than
solving a jigsaw puzzle. Depending on how the elements of a
molecule interact, particularly when external positive and
negative forces are imposed, we can end up with an entirely
different outcome than we expected.
In a presentation at the Census Bureau, I referenced an
experience that I think has direct bearing on the topic of
today and particularly the legislation that you propose. During
the annual budget process, a Commerce Department budget analyst
decided to reduce the Department's current budget problem by
eliminating a sensitive agriculture item from the Census Bureau
budget. As might be expected, particularly since that census is
mandated by the Congress, the Department of Agriculture
protested and appealed to Congress to transfer the census of
agriculture to their Department.
While most everyone in government, including me, was
focusing on who should collect the information, Jim Bonnet, who
would become one of my most constructive critics, pointed out
that society needed to know and understand both the specifics
and the interactions of the agricultural system. In essence,
what are the inputs--the seeds, the fertilizer, the machinery?
Then what happens after it is grown? How does it go through
processing? How does it go through transportation? How does it
go through commodities? How does it go through wholesaling? How
does it go through retail? And then how does it get to the
consumer table?
And he pointed out that the Census Bureau, doing all of the
economic censuses that relate to all those functions, should
take that data and relate the census of agriculture to it so
that you can see the impact of these different aspects of the
entire system of agriculture, rather than just the measurement
of a particular form of our economy.
I think the Census Bureau has tried in the past to do that,
and it has to some extent been successful, but it is very hard
to do that when you are sitting inside of a functional entity
whose interests are more narrow than the broad interests of the
entire economy.
To address the improvements, I think we need to act a
little bit more--including--summing up here fast--to just say
that we need to really think more systemically. And the issue
of a systemic thinking is to create a whole that is greater
than a sum of the parts. And I think we can do that by some of
the things that you suggested in this legislation, that we can
create a Federal Statistical System that is of greater value to
society than the sum of each of the individual statistical
agencies which it encompasses.
[The prepared statement of Vincent P. Barabba appears in
the Submissions for the Record on page 35.]
Chair Maloney. Thank you very much.
Dr. Bryant.
STATEMENT OF THE HONORABLE BARBARA EVERITT BRYANT, DIRECTOR
FROM 1989-1993, RESEARCH SCIENTIST EMERITUS, ROSS SCHOOL OF
BUSINESS, UNIVERSITY OF MICHIGAN, ANN ARBOR, MI
Director Bryant. Chair Maloney, Mr. Brady, the Census
Bureau got off to a roaring start for the 21st century by
implementing the American Community Survey. As of 2010, the
American Community Survey, the acronym ACS, provides new data
on the characteristics of the U.S. population every year. Prior
to that, we had to wait every 10 years.
ACS implementation frees the decennial census from a lot of
social and economic questions and leaves the questionnaire with
only the eight questions needed for reapportionment and
redistricting, the legislative and congressional purposes of
the census.
In the rest of this century, I envision that the Census
Bureau will build upon its past as the originator of data
processing, of computerization of the TIGER geographic system,
which is the basis for computer mapping every block in the
country, of reaching and counting hard-to-reach people, and as
a source of continuous improvement in capturing data faster and
more accurately. I think we can count on the Census Bureau to
be a continuing source of innovation, new products, and
processes. That is if we structure it for the future and not
for the past.
When he figured out how to massively count large numbers
with punch cards for the 1890 Census, Herman Hollerith, a
Census employee, didn't envision today's computer industry.
More recently, we who watched the computer mapping of every
block in the Nation for 1990 didn't envision that this, coupled
with communications satellites, would become the start of a
large GPS industry.
Amazingly, the Census Bureau is an enormous bureaucracy
with all the negatives that implies but also a hotbed of
innovation. The challenge for this century is to keep it that
way, although we do not know what all the changes and the
innovations will be.
Let me speak to two ways you might change structures to
make the Census Bureau more nimble and less bureaucratic as we
move through these next 91 years.
First, recognize that the Census Bureau operates on a 10-
year cycle, not a 2-year or 4-year or 6-year political cycle.
Within the decade are two 5-year cycles for the economic
censuses. As it faces its largest project, the decennial
census, the last Census was always conducted by a prior
administration. The only institutional know-how for how to
count a population is in the career employees at the Census
Bureau, not at the Commerce Department in which the Census
Bureau resides.
This cycle also means that every 20 years, as occurred so
recently, the Director of the Census is a Presidential
appointee of a President inaugurated in January of the year
ending in 9. That Director is not in office--and I was one of
them--in time to have any role in the planning of the census
which he or she directs.
The solution to this is, obviously, to make the term of the
Census Director a 5-year term, half of the decennial cycle,
starting in the years ending in 1 or 2 or 5 or 6 or 7; and,
that way, a Director coming in at mid-decade would be in for
the ramp-up to the decennial census and for the immediate
dissemination of the data from it.
My second recommendation is to flatten the bureaucracy by
removing the Census Bureau from the Department of Commerce.
Since leaving the Census Bureau, I have spent 16 years at
the University of Michigan Business School working on an
economic indicator, the American Customer Satisfaction Index.
The current mantra for customer satisfaction, for getting
closer to the customers and users is to flatten organizational
structures. Successful corporations are doing this to be
profitable. Unfortunately, some corporations from my own State
of Michigan, however, are learning this lesson too late and
only now, after bankruptcy, changing their structures to be
leaner and more responsive.
Structures that were very successful for the 20th century
may not be necessarily working in the 21st. The Census Bureau
is a bureaucracy under a bureaucracy. It is a large
organization that reports to another large one. The Department
of Commerce is on a 4-year cycle, not a 10-year one. The Census
Bureau paper work goes through not one but two levels of
approval in the Commerce Department. Every response to a letter
from you in Congress gets delayed that way.
I have a 2-minute talk on trying to get computer
interviewing into the Census Bureau and how it was axed at the
budget level. I don't have 2 minutes.
In conclusion, my two recommendations are a 5-year term for
the Census Director and remove the Census Bureau and make it an
independent agency like the National Science Foundation.
[The prepared statement of Barbara Everitt Bryant appears
in the Submissions for the Record on page 37.]
Chair Maloney. Thank you so much.
Dr. Riche for 5 minutes.
STATEMENT OF THE HONORABLE MARTHA FARNSWORTH RICHE, DIRECTOR
FROM 1994-1998, PRINCIPAL, FARNSWORTH RICHE ASSOCIATES,
TRUMANSBURG, NY
Director Riche. Chairman Maloney, thank you very much for
this opportunity.
I also want to make the case that the Department of
Commerce, which currently has line responsibility for the
Census Bureau, has an inherent management conflict.
Like every Cabinet agency, the Department is made up of
many offices and bureaus; and they all are vying for Federal
funds each year. None of them has that constitutionally
mandated responsibility that you spoke about earlier. As a
result, as each census approaches, the Bureau's request for
funds for the census jumps from hundreds of millions of dollars
to many billions of dollars; and that throws the Department of
Commerce budget off track every decade.
This year, for instance, the Bureau is requesting more than
$7 billion for the decennial census account. That is up from
just $512 million 3 years ago. It is a hard pig to move through
the python, the Commerce Department.
So the census inevitably causes conflict between the Bureau
and the Department, because the Department, quite naturally,
has quite different priorities. When I was Director of the
Census Bureau, NASA--excuse me--NOAA wanted a brand-new weather
satellite. It was certainly a needed satellite. We are not
complaining about it, but it was a real conflict. The result of
these kinds of conflicts is that the Department tends to defer
important census activities often until it is too late to
undertake them efficiently.
There are three other issues I would like to address.
The first one is content. It takes--as Vince pointed out,
the inventory of statistics and demographic measures needs
constant update, but it takes about 20 years between perceiving
a need and actually getting the data on the street, and that is
if all goes well.
Only the Federal Government can collect official
statistics. Only the Federal Government has the authority and
the resources to get the job done. But the policy questions
that call for the kinds of complex data that Mr. Barabba
highlighted tend to be asked by different agencies, not by the
Department of Commerce.
Developing these complex measures effectively requires
regular advisory input from stakeholders, statistical
professionals, and measurement experts, as well as the
oversight from Congress, your Government Accountability Office
and, in the executive branch, the Office of Management and
Budget, which has the coordinating responsibility for both
collecting Federal statistics and for measurement burdens. This
task calls for constant listening and communications
activities, requiring direct access going in both directions.
My experience as a former director is that the Department of
Commerce too often seeks to shield the Census Bureau from some
of these conversations and in the process it ends up isolating
the Bureau.
My second point is about resources. As you know, they are
always limited. They need to be addressed in the context of
statistical priorities.
I could tell a story to Mr. Brady, if he is interested,
about what happens when you have some of--for some of the
issues that you are concerned about, trying to get those things
funded inside the current system that we have now, rather than
a bigger one.
And the final point I would like to make is independence.
As we have all said, the decennial census is very political.
That is the point of it. For that matter, all government
statistics are political. The word itself means measures used
for governance. So the issue at hand is how to maintain the
Census Bureau and other statistical agencies' independence in
pursuit of accurate data. We need a set of regular processes to
build on transparency, collaboration with other measurement
agencies, and regular reporting, at fixed times, not subject to
manipulation. And not subject to political appointee, many who
are extremely well meaning, but these are complex issues and
people often go wrong.
Finally, successful measurement depends on willing
respondents. Federal statisticians have very little control
over Americans attitudes about surveys and censuses created by
people with varying motives and varying expertise. I think that
increases the value to the Census Bureau of advertising,
outreach, and stakeholder relationships, as well as innovative
data collection methods. It also heightens the value of an
untroubled reputation for guarding confidentiality, especially
as technology and security concerns challenge those standards.
I think those results would be much more achievable if the
Bureau were independent.
[The prepared statement of Martha Farnsworth Riche appears
in the Submissions for the Record on page 39.]
Chair Maloney. Thank you so much.
Dr. Prewitt.
STATEMENT OF THE HONORABLE KENNETH PREWITT, DIRECTOR FROM 1998-
2001, CARNEGIE PROFESSOR OF PUBLIC AFFAIRS AND VICE PRESIDENT
FOR GLOBAL CENTERS, COLUMBIA UNIVERSITY, NEW YORK, NY
Director Prewitt. Thank you so much.
Let me emphasize that I speak today as a private citizen,
and that is because I do have a consultantship right now with
the Department, but obviously my testimony has not been
reviewed by the Department or the Census Bureau. I should say
that when I was a Census Bureau Director my testimony was
always reviewed by the Commerce Department, and I sometimes was
surprised what I found myself saying to the Congress. It was
different from what I intended to say. And I make that point
because that is exactly what some of us are trying to talk
about.
The other thing I would say is specifically to Mr. Brady's
question with respect to innovation. This is what a very
enormously talented, creative organization can be. It lives and
breathes innovation in measurement. That is its professional
identity, its professional ambition.
And the difficulty, exactly as Martha Riche just said, is
the decennial is so big and so important to our society that
everybody gets very excited about it, including, of course, the
United States Congress, for understandable reasons, and then
forgets the Census Bureau; and it goes--it just goes into a
quiescent period for 5 or 6 or 7 years before the attention
builds back up.
That is exactly the period of time when you will be doing
this innovation, and every one of us can testify to how
difficult it is to create a morale, a staffing structure, and a
budget to do the kind of work that you would have the Bureau do
and which it is quite capable of doing.
Let me speak from a different point of view on the issue of
independence. I think it is sad but true that the country has
dug itself into a very large hole about the so-called political
manipulation issue. On the floor of Congress, a distinguished
Senator said during Robert Groves' hearing just last week, by
overcounting here, undercounting there, census manipulation
could take place for sole political gain.
Fine. The problem was it was said casually. In previous
versions of this sentence, people sort of thought about it and
said, wait a minute, do I actually want to say the Census
Bureau could manipulate the numbers for political gain? And
here it was just kind of mentioned in passing without notice.
That political hole that we put ourselves into starts with
the Democrats in the 1980s when they brought a case to force
the Census Bureau to report the data the Census Bureau itself
did not think was ready to report. And it has continued through
that period, as all of us who are veterans of the 2000 census
know.
And I guess my concern about independence is I am desperate
for us to get out of that political hole, and I think no
stronger signal of that is available to us than the
independence bill. The fixed term, yes, but the independence
bill is a stronger signal. And I really urge the Congress,
after we get through the 2010, to think hard about whether we
want to keep repeating this debate that we had yet again over
Dr. Groves' nomination for something he did 20 years ago and so
forth.
The census numbers are political. They are supposed to be
political. They start a political process of redistricting, of
appropriations, redistricting, and then elections and so forth.
But the starting point has not to be political. It has to be
nonpartisan, scientific. By creating an independent agency,
what you are saying this is a piece of science; and we treat it
like science. Once the numbers come out, then we can argue
about them and debate about them and they can be put on the
Republican side or the Democratic side and so forth. That is
the nature of our democracy and that is healthy.
But the starting point itself should not be politicized,
and we have not politicized it. The 2000 census was--I am
certain was the most scrutinized census in our history; and I
have actually done a lot of work on the history of the Census
Bureau, so I can say that with some confidence. I testified
more than 20 times in less than 2 years. That is a lot of times
to be brought down here to say this, that, and the other and so
forth.
The GAO was extremely active around the 2000 census. The IG
was active. There were eight formal advisory committees paying
attention to what we were doing and so forth, and some of you
will remember there was a special Census Monitoring Board with
its own budget and own staff that was deliberately put in place
by the United States Congress to search for manipulation,
political misuse, and so forth. No documentation has ever been
put on the table that the Census Bureau, in terms of what it
controls, has been subjected or has engaged in anything that
could remotely look like political use of the information.
So, yes, the Congress can say, here is the budget; here is
what we need information on. Once you make that decision, then
the Census Bureau scientifically, professionally has got to
design a census and execute it.
Thank you.
[The prepared statement of Kenneth Prewitt appears in the
Submissions for the Record on page 40.]
Chair Maloney. Thank you, Dr. Prewitt.
And Mr. Kincannon.
STATEMENT OF THE HONORABLE CHARLES LOUIS KINCANNON, DIRECTOR
FROM 2002-2008, VICE PRESIDENT, BOARD OF DIRECTORS OF CAPITOL
HILL VILLAGE, WASHINGTON, DC
Director Kincannon. Thank you, Chairwoman Maloney. It is a
pleasure to join my distinguished colleagues here in testifying
before your committee, and I thank you for inviting me to
appear.
The Census Bureau is central to the Federal Statistical
System and to statistics to help policymakers make sound
decisions. It is a key producer of economic statistics broadly
construed to cover businesses, establishments, persons, and
households. It produces about 70 percent of the hard figures
that go into making up the GDP. It produces about half of the
principal economic indicators as defined by the OMB. There is
in my testimony a list of 16 indicators, and I would just
mention six of those very quickly: construction put in place,
housing vacancies and homeownership, new residential
construction, new residential sales, our data collection
contribution to the BLS release on employment and unemployment,
and on their preparation and release of the CPI.
If those are not in the headlines governing and trying to
direct what we do to try to solve our economic problems, I
cannot think what series would be.
Census figures also steer about $375 billion of Federal
domestic assistance each year, according to the Brookings
Institution.
Other nations also view the Census Bureau as a central
player. The French Finance Ministry surveyed statistical
practices in other countries about 5 years ago at the request
of their Minister. This report noted that the Federal
Statistical System relies on service provisions and financial
transfers between agencies, as some producers are largely
dependent on others for collecting data.
And another quote: The Census Bureau plays a central role
in this respect, as even large agencies, such as the Bureau of
Labor Statistics or the Bureau of Economic Analysis, depend
heavily on its collection network.
The French report also noted that the great strength of the
Federal Statistical System was the timeliness of its data and
the closeness of producers to users. However, as a
decentralized system, it requires coordination, which, as they
noted, depends on seven people at the OMB. The institutional--
this is a quote. ``The institutional framework of producers of
official statistics remains a strong limitation on
coordination. It results in redundancies of tasks such as
keeping registers providing sampling bases for surveys, in
difficulties in the sharing of micro data, and in
classifications and concepts that are not always consistent
between various products and institutions.''
The Census Bureau will remain central to policymakers
needing statistics for informed decisions, and the decennial
census and the American Community Survey are a very towering
contribution to this process.
What do we need to do to ensure the census meets the
Nation's needs?
We need to recognize--and this is not news to you by this
time in the hearing. We need to recognize the long lead time
needed to develop, select, and apply modern technology to
agency work. This is true of the Census Bureau, just as it is
of NASA or NOAA. The current arrangements in the executive
branch fail to meet fully that goal of 2010 as we all know.
We need to recognize the long planning cycle, more than 10
years, for the decennial census, even setting aside
technological matters.
The Census Bureau itself must be organized to deal with
this. It needs continuity of leadership, which to me implies a
long and probably fixed term of service for the director to
connect responsibility for planning with that of production and
outcome. We need to pay special attention to the role and the
person holding the deputy director post, which has a particular
strategic, structural role in the organization.
I will illustrate. In the 1990s census cycle, we made
significant technological progress. A prime example already
mentioned is the TIGER system of automated digitized maps and
address registers. It was developed in cooperation with the
U.S. Geological Survey. This replaced a paper and paste pot
system of producing maps for census figures with a modern
system. It could not have been done without sustained
leadership and a willingness to accept some risk in change.
There was one deputy director during that period and three
census directors, but that together provided that kind of
continuity.
What else is needed?
We need to ensure independence and integrity of the
planning process.
We need to ensure staffing is purpose-based, including SES
appointment authority being the responsibility of the director
of this multibillion dollar agency with 8,000 employees, not
counting temporary census workers. Now it is exercised by an
official with more modest responsibility.
We need to place the census budget cycle in an environment
that is not hostile. I don't mean that there are enemies of the
census budget in the Commerce Department. I don't mean that at
all. I mean there are severe natural conflicts that work
against the census budget and its off-beat rhythm. And you have
heard many examples of that.
Madam Chair, I thank you for your invitation and for the
entire hearing.
[The prepared statement of Charles Louis Kincannon appears
in the Submissions for the Record on page 42.]
Chair Maloney. I thank all of the panelists for your
insightful testimony.
I have introduced legislation to give the Census Bureau
independent status and remove it from the Commerce Department,
similar to the National Science Foundation and NASA. Other
Federal statistical agencies, such as the Bureau of Labor
Statistics, are part of another executive branch agency, but
the director is appointed for a fixed term.
I would like to ask each of the panelists if you could
answer in yes or no, just to get it on the record, do you think
true independence of the Census can be achieved without
removing it from the Commerce Department? And I just would like
to go down with yes, no, and then come back and let you give an
example of why you take the position that you take. Starting
with you, Mr. Barabba.
Director Barabba. The question was, can you do it without
removing it from the Commerce Department? I don't think so.
Chair Maloney. Dr. Bryant.
Director Bryant. No, I don't think so.
Chair Maloney. Dr. Riche.
Director Riche. No.
Chair Maloney. Dr. Prewitt.
Director Prewitt. No.
Chair Maloney. Mr. Kincannon.
Director Kincannon. No.
Chair Maloney. Would anyone like to elaborate on why they
think it is impossible to be independent under the Commerce
Department? I open it to the floor.
Director Barabba. If I could, it is interesting that the
Bureau has an incredible reputation even though it has been
sitting inside the Department of Commerce. But that is a
function of the people there.
I think one of the things that is not really understood
about the Bureau is that its employees are as much engaged in
their professional organization, representing the various
interests of society as any agency I know. And they are more
respectful of their peers in the different departments,
different societal organizations as they are relative to the
oversight of the Department of Commerce. So no one inside the
Census Bureau will do anything that somebody in government asks
them to do that is wrong or not efficient if they have to go
face their colleagues and their professional organizations and
say I agreed to that. And I think that has been an important
aspect of the Bureau that is not really fully understood.
Chair Maloney. Thank you.
As a former Census Director, Dr. Bryant, what are the pros
and cons you see of removing it from Commerce?
Director Bryant. Well, I think that the big pro is getting
it out of the Department of Commerce. Commerce is on a 4-year
cycle and a very new-political-appointee, 4-year cycle, and you
can't plan a 10-year process that way.
The cons. There are many talented people in the Commerce
Department--there are many people that have given great help to
the Census Bureau, but it is very up and down, irregular, and a
lot of lack of understanding.
I mentioned computer-assisted interviewing. It was already
in the commercial, private, and academic sectors. When we tried
to get budget to improve it at the Census Bureau, the Commerce
Department didn't understand it and for several years deleted
it from our budget. Without a good Deputy Secretary who knew
how to shuffle some of our other R&D money around I don't think
we would have had it to this day. It has been much used. We
were even behind--and they kidded us--a much smaller
Netherlands Bureau of Statistics, and they came in and
consulted for us on how we could catch up.
Chair Maloney. Dr. Riche, the pros and cons of--do you see
the pros and cons of removing it from Commerce?
Director Riche. I don't actually see any cons. As Vince
pointed out, there are almost no relationships below the
director and a few officials with the Department of Commerce.
The Bureau does function independently. It does come down to
the issues of budget and to the issues of communication, who
the Bureau is allowed to talk to and when. That is what,
really, Commerce tries to manage.
Some of us--I think everybody here has been pretty
successful in doing end runs on budget to OMB, which does
understand priorities for Federal statistics and doesn't
usually need to be educated.
But I don't mean to reproach and say any bad things. I was
very fortunate to work with a very good team of people in
Secretary Brown's Department of Commerce. But they just have a
different set of priorities, a different set of incentives.
Chair Maloney. My time has expired.
Mr. Brady.
Representative Brady. First, I wish my wife was half as
agreeable as this panel is to your bill here today. By the way,
who picked these guys?
You know, the question I am going to ask about innovation,
as we do, sort of following on what the chairman asked. Just
removing yourself from the current times, but do you--as former
Census Directors, do you continue to worry about or do you have
a fear about political manipulation at the outset of census
taking?
But, secondly, what can we do to highlight and gather the
data that can more accurately measure innovation in our
economy?
I know that the National Science Foundation developed a
series of questions with existing business research and
development survey for the census that hasn't been incorporated
yet, but it seems to me we have gathered some parts of
innovation but left other huge gaps in statistics and data that
can be very helpful.
I will just open it up to the panel. Why don't we start on
the other end and go down with your thoughts on either of those
issues.
Director Kincannon. On measuring innovation, Secretary
Gutierrez did establish a high-level group to consider and
discuss that and make some recommendations. Those
recommendations, that came close to the time that I was leaving
or even after that, but it must have produced some useful work.
I know the BEA gained some benefits from their recommendations.
Measuring innovation is difficult. Everybody knows it is
good. Everybody knows it has benefits. And you can see it is
sort of like the wind blowing. You can see the leaves shake,
but exactly which wind and where it came from requires NOAA
quite a lot of money to figure out. So I think it is something
that requires a lot of work and might not be principally an
assignment of the Census Bureau but a more analytical agency to
then identify what needs to be measured, and the Census Bureau
takes that over.
With regard to political interference, I have a long
experience at the Census Bureau at the--either as a political
appointee or in the level below so that you have a lot of
contact. I have not seen any bold or clear-cut attempt to say
change a number. I have heard some sincere wishes that the
numbers could be different but not a foolish directive to
change them. Because everybody understands pretty easily that
can't be defended.
I did not see any interference while I was director in
finishing up the process of the analysis of the coverage
measurement work for the 2000 census. We reached an
independent, career-based decision. We did not have a process
that worked to adjust census results.
There were a lot of people nervous about that in the
Commerce Department, but Don Evans was not one of them. He
understood the technical work that we were doing and satisfied
himself that we would come to a sound scientific conclusion and
let us alone for 3 years to work all through that.
Representative Brady. Thank you, sir.
Director Prewitt. Yes, Mr. Brady.
I would point out, as an example of what I consider to be
gross political interference, the National Academy of Science
has published something called Principles and Practices of
Statistical Agencies. It is widely read. It is like the Bible
for statistical agencies. It makes a very strong argument, of
course, that statistical products have to be the product of the
statistical agency.
And if you were to find out that the Bureau of Labor
Statistics brought the unemployment rate down to the Secretary
of Labor and said what do you think about this rate and the
Secretary of Labor said, well, my gosh, I wish it were a little
lower or a little higher, you would be outraged and you should
be outraged.
If the--any GDP number, any statistical number that sort of
describes--if the number of uninsured were first brought to the
Health Secretary and said what do you think about this and
before you knew about it, before the press knew about it, you
would be outraged and should be outraged.
In 2001, the Census Bureau--in fact, it also happened in
1991, but under Dr. Bryant. But they were under court order to
bring the major statistical product, the decennial census, down
to the Department of Commerce and let the Department of
Commerce Secretary decide about that number. They were under
court order.
In 2000, this was repeated; and they were not under court
order. The Secretary of Commerce simply said that, with respect
to this statistical procedure, that the results that we brought
down--and he would invite in his own experts to tell him
whether it was right or--what the Census Bureau was doing was
right or wrong.
That is political manipulation. I think that is a blight on
the history of statistical agencies in the United States, and I
wish it hadn't happened. I am very sad, and neither of the
people around this table were in a position to do anything
about it, but I thought it was very unfortunate that the Census
Bureau complied with that instruction. I think they should have
said no.
Chair Maloney. The gentleman's time has expired.
Mr. Cummings.
Representative Cummings. Thank you, Madam Chair.
Thank you all for your testimony.
Our Nation is facing some very difficult circumstances now
with one out of nine folks in my State of Maryland facing
foreclosure. And according to some testimony of our HUD
Secretary the other day, we are not--we are dealing with
foreclosures, but we are dealing with them on the basis of 100,
maybe 300,000. We need to be in the millions of addressing
them. That means a lot of people are going to be displaced.
Do you all see any special problems with that as far as
getting an accurate count? And how--I mean, if you had to give
advice to the Census Director, what would you--what would that
advice be? We already have a lot of problems getting every
single person counted. I am just curious.
Somebody speak up. Somebody.
Director Barabba. I would just say that the Census Bureau
has a very extensive program of dealing with the homeless and
people without a specific residence. I think your observation
is correct; and if I had to bet, somebody over there is
thinking about right now about how they are going to have to
expand that program.
Representative Cummings. Anybody else?
Director Kincannon. It is true that if you are not living
at an address it is harder to count you, whatever the situation
is. There are special steps to be taken. There are procedures
in place to handle the displacement of people who are still
displaced in Louisiana and those areas that were affected by
Katrina, and the same kind of procedures can be used to try to
track the people who have been evicted from their homes. Many
will go to another place of abode, but if they don't, it is
difficult to do.
Director Bryant. I will second that, that the Census Bureau
has--in my time, it was Hurricane Andrew; in your time, it was
Hurricane Katrina. The Census Bureau does a great job of
tracking people. I mean, they will go to the neighbor's house
and say, do you know what happened to, and follow them to the
ends of the Earth if they have to.
Representative Cummings. Dr. Riche.
Director Riche. I would just add this reinforces the need
for some of the basic census processes to be strong.
One of those needs is outreach and communication to let the
people know the census is being taken. If you are not in your
home, it is still important that you be counted and that we
have employees, as we try to do, who are conducting the census
who are from the communities and are trusted individuals and
will gain cooperation.
Representative Cummings. In April, 2008, the Census Bureau
announced it would drop plans to use handheld computers to help
count Americans for the 2010 census, which would have added an
estimated $3 billion to the cost for the census. Additionally,
in June, 2008, the Government Accounting Office reported that
the 2010 census will still cost between $13.7 billion and $14.5
billion. What other cost-saving measures would any of you
recommend? And would you have recommended those?
Director Barabba. If I could? I have often thought it would
be really interesting if there was a contest among the Members
of the Congress, based on a formula that would be developed, of
who could get the highest return on the mail-out, mail-back
census form. And that would create a conversation at the
congressional level in each district that would put every
Congressman and woman on notice that their job was to reduce
the cost of the census by getting people to fill out the census
form and mail it back. I think if you could do that, you could
save a considerable amount of money.
Representative Cummings. Anybody else?
Yes, sir.
Director Kincannon. I would like to go back to one of Mr.
Barabba's suggestions earlier on, that is, a serious national
conversation through some mechanism about the kinds of
information we collect, the cost of doing it, and the detail
that is needed.
For example, a great deal that drives up the cost of the
decennial census is the requirement for block-level data for
use in redistricting. A satisfactory, fair way of redistricting
with slightly elevated geography would save a considerable
amount of money. And collecting information about industries
that are no longer major industries but continue to be
collected, that could also address some economies.
Representative Cummings. Mr. Prewitt.
Director Prewitt. Just quickly underline this one on block-
level data. Block-level data are basically not needed to make
sense of this country with respect to redistricting. And if the
Congress were willing to--block-level data are notoriously
flawed for all kinds of reasons. They are extremely difficult
to get that number right down to the block level. And a lot of
redistricting presumes that number is accurate, and it is
really not, and we all know around this table that it is not.
So I think the idea of a higher level of geography for even
redistricting makes a lot of sense. So you are not down to five
people. You are down to 50 people or even 500 people. It is not
a big deal with congressional districts of 750,000 people. So
that is a serious thing.
But, Congressman Cummings, there are lots of other ways.
And one of the things that is going to happen in this country--
and we are not prepared for it. This country is producing
enormous amounts of information, not from survey instruments.
They are producing it by administrative records; and they are
producing it by swipe data, digital data. And it is a--talk
about an innovation, a really serious innovation in census
taking, we are going to have to create a way to collect that
information and use it for the purposes that we use the census
now.
It is extremely important to get information by going out
and knocking on doors. When a lot of that information exists
already in administrative records and in all kinds of other
sources--enormous problems, privacy, confidentiality, et
cetera, et cetera. But they are solvable. They are not solvable
if you don't have a strong scientific agency that reports not
to the Department of Commerce but reports, like the National
Science Foundation does, to a board of scientific and
technical, statistical experts. That is the kind of agency
which produces the kind of conversation that would allow us to
use information which is practically free. And yet we don't
have a way to do so.
Chair Maloney. The gentleman's time has expired.
Mr. Burgess is recognized for 5 minutes.
Representative Burgess. Thank you, Madam Chair.
Dr. Bryant, I apologize if I was out of the room and you
have already answered this. But you made the comment that the
census should be removed from the Department of Commerce. Did I
understand you correctly?
Director Bryant. I think we all did, as a matter of fact.
Representative Burgess. Where would it go?
Director Bryant. The model would be the National Science
Foundation, which would be a free-standing agency.
Representative Burgess. But it would not be the White
House?
Director Bryant. No, no, it would not be the White House.
Representative Burgess. Let me stay with you. This is
really a question that I would like to hear from several of
you.
My home State of Texas has a situation where we have a lot
of people in our State who don't have the benefit of a Social
Security number or any of the other accoutrements of
citizenship. So taking into account undocumented workers or
people who are unwilling to fill out a census survey, how have
the previous census takers dealt with that consideration?
Director Bryant. Over the past three censuses--I am going
on to 2010 now--we have had an enormous outreach program with
the local organizations. You are talking really a lot about the
undocumented, Hispanic population; and we have worked a lot
with Latino organizations, nonprofits, to have them convince
people it is safe to be answered by the census. However, I
think with all the immigration attacks right now on the
undocumented, this is going to be one of the biggest challenges
of the 2010 census.
Representative Burgess. What--but, historically, what were
you able to do to overcome that?
Let me take a step back, and I will just ask another
question. It is--maybe my understanding is not correct, but it
is important to count individuals even though they may not be
citizens; is that right?
Director Bryant. Yes. The constitutional mandate is to
count everyone resident in the country. Because, after all,
everyone resident in the country uses the resources of local
government and State governments.
Representative Burgess. In the past, what have the--in the
previous census-taking efforts, what has been done to mitigate
that problem?
Director Bryant. Well, as I say, it has been by outreach,
to have these nonprofit organizations that work with them.
Social agencies and things like that try and convince the
people that it is safe to be counted, that the Census Bureau is
never going to trade the information with your employer or
anybody else. But that is a very hard sell; and that is
probably the biggest challenge always to census taking, is
those who don't understand why the census is important or are
fearful of government and having them counted.
Representative Burgess. That was going to be my next
question to you. Because in my area of Texas there is--but--you
may not have noticed, but our congressional approval ratings
are not very high right now, and there is a great and growing
mistrust of the government. And things that have happened in
the past 12 months--the stock market meltdown, the bailout
bill, the stimulus bill--all these--the cap and trade bill--all
of these things have added to that anxiety that is out there.
So how--perhaps, Dr. Prewitt, I should ask you. You were in
charge in the 2000 census, so you have had more recent
experience. You have got two groups. One group of people who is
fearful that they might be identified and placed out of the
country and one group of people who just is fearful of the
government. How do you overcome that?
Director Prewitt. Not easily. I spent a lot of time in
Texas in 2000 down in the border especially. Enormously
important partnership with the Catholic Church.
Representative Burgess. I am on the other border up by
Oklahoma, and you know how much trouble that is.
Director Prewitt. That is true. But by far the most
important partner in the 2000 census with respect to this
population group was the Catholic Church. And also social work
agencies. You have to find the trusted voices.
I am a government bureaucratic. I can go make a speech all
day long about the confidentiality and they--but if they are
trusted voices--and that is what we did. We thought we did very
well.
I think Barbara is quite correct. I think it is going to be
tougher in 2010 than it was in 2000.
Representative Burgess. Mr. Kincannon, you may have a
better answer to this, having recently left the Department. Are
there any new strategies in place to deal with this?
Director Kincannon. Well, yes, there are new strategies,
including mailing bilingual questionnaires in areas where there
are a significant proportion of people who speak Spanish only
at home. We are trying that one language now. We will see
whether we can expand that use--I still say ``we''--but we will
see whether they can expand that use in the future. But in
tests that we did, that increased response.
I don't know that that deals with undocumented workers. I
grew up in Corpus Christi. I know the problem you are talking
about.
Representative Burgess. What about the other segment that
just simply does not trust the government? And I will tell you
they are large, and they are growing, and they are vocal. They
are on talk radio almost every afternoon, if you want to go
listen to them. We can find them on the Internet. And they are
concerned and legitimately concerned. They don't want to answer
anything but name, rank, and serial number and even that they
will only divulge with some stress.
Chair Maloney. The gentleman's time has expired.
Mr. Kincannon can respond.
Director Kincannon. I will just say the work through what
we call our partnership program with local grassroots leaders,
church leaders, Protestant or Catholic, doctors and health care
providers, barbers, shopkeepers and so on to make sure that
they--those people who speak to and are listened to by the
people who may be afraid of the government has been very
helpful in improving it. As a matter of fact, in the 2000
census, the census counted more people than we estimated were
in the country. That showed that the administrative and
estimated data on immigrants was lower than real immigration,
and I think we documented that pretty well. Ken did a good job.
Chair Maloney. Mr. Hinchey in recognized for 5 minutes.
Representative Hinchey. Thank you very much, Madam
Chairman.
And thank you very much. It is very interesting to see
clearly the importance of what you are doing. I very much
appreciate your being here, and I appreciate the opportunity to
listen to what you said and the response that you have given to
the questions.
The issue of independence seems to be one issue that is
significantly important, moving out of the place where you are
now. I don't know if you want to make any additional comments
on that, how that would come about, how effective it would be
how the changes might occur. But, briefly, I am interested in
hearing something about that issue of independence.
Director Bryant. I think it is interesting that we
represent 25 years, almost 30 of census taking here. We were
appointed by different administrations, and we are all in
agreement on this issue, that independence is necessary for the
Census Bureau.
Director Barabba. I would add that if you think about Title
13 of the U.S. Code and what it allows this agency of
government to do, if you--if that law did not exist today and
the Census Bureau did not have its reputation, I doubt if that
legislation could be written today. Because it--fundamentally,
you have given the right--you have given the authority to an
agency of government to do things that some don't want to do.
I think the Census Bureau has earned a position in the
minds of a lot of people that it has the capability of
operating as an independent agency. And I would say that, since
Title 13 is always subject to review, it might be one way of
avoiding somebody coming through the back door and affecting
what might be considered one of the valued aspects of our
government, which is the ability to know what is going on in a
nonpartisan way.
Representative Hinchey. I think that is absolutely--I am
sorry.
Director Prewitt. Just quickly, Mr. Hinchey. Specifically
to those kinds of things. I think the model has to be a
scientific agency, like a NASA, like an NSF, like NIH. If you
think about statistical data, it is part of the scientific
infrastructure of the society. That is how we analyze and
understand our society. So conceptualize it as science, and I
think that will be the most important argument you can put
before your colleagues.
Representative Hinchey. Yes.
Director Riche. I would just add that checks and balances
have turned out to be very useful in our government from its
beginnings. And so setting such an agency up independently
would be--it still has to report someplace. It has to get its
money from someplace. Thinking about those checks and balances,
the stakeholders, obviously, the Congress with the power of the
purse and the professional community, setting up that kind of
board oversight reporting, that is what would be really needed.
Representative Hinchey. Well, thanks very much.
I think it is very important. I think most people, if they
were asked, they would probably say that you are independent,
that you function that way, but that is not the case. Have you
experienced in any way any political influence on the kinds of
ways in which you operate and the results that come about as a
result of that operation? Any negative interaction in any way?
Director Riche. I think every director comes into office
with the knowledge that there will be pushes on you coming from
someplace, and we all have that talk with ourselves as to how
we are going to react to it.
Ken pointed out the existence of the Code of Practices for
all statistical agencies. In my own case, I had to take that
Code when a new team came in the Department of Commerce, yellow
highlight, offering it as a gift, presentation, so on and so
forth, to educate people. And there were certainly occasions
when I had to say, no, no, you can't have this data. That would
be from another branch of the Commerce Department.
There are always things like that that we all confront.
Representative Hinchey. Yes, sir.
Director Kincannon. Another matter that is not necessarily
political interference and doesn't deal with exactly a task at
hand, but consistent interference with personnel appointments,
including, well, mainly, therefore, career officials, since
that is the nature of the Census Bureau; blocking communication
with OMB and with the Hill, which harms the Census Bureau's
ability to serve and to inform and doesn't necessarily have any
partisan motive but a bureaucratic motive. Perhaps the failure
to notify the Secretary of a decision by OMB to disapprove a
needed portion of the budget initiative but one which OMB was
going through, the little kabuki dance that it goes through,
but expected the Census Bureau to appeal. They failed to inform
the Secretary of that need to appeal, and so we didn't appeal
it. And it made it a more difficult task to recover from. And
the likelihood----
Well, I am not going to go into that.
Chair Maloney. The gentleman's time has expired.
Representative Hinchey. Unfortunately, the time is up.
Chair Maloney. Senator Klobuchar.
Senator Klobuchar. Thank you very much, Madam Chair. Thanks
for holding this important hearing.
And I did love hearing for all the Presidents that you
worked for and the bipartisan nature of this panel and the
independent nature of this panel, and I truly appreciate that.
And I think everyone knows by law that they have to do the
census. I am not sure everyone in our country knows how
important the census is for things like how money is divided
and also congressional seats, something my State cares a lot
about right now as we are kind of on the cusp of retaining or
not retaining a congressional seat.
Yesterday, the Census Bureau actually reported that, in the
2008 Presidential election, my State of Minnesota led the
Nation in turnout with 75 percent. So we like participation,
and hopefully that will transcend with the census as well.
I was wondering--just a few issues. First of all, there has
been concerns about--raised about the privacy and
confidentiality of personal information that is shared as part
of the census. Could you explain, any of you, one or two of
you, how the Census Bureau protects the personal information
that is shared during the census and what precautions are
taken?
Director Kincannon. If I may, the law expressly forbids
disclosing any individual information. In the case of persons,
that extends for 72 years. And that is long enough mostly for
it to become less sensitive, and mainly it is people doing
ancestral research that would look at it then.
The Census Bureau--this is an important way of our life.
The individual data are not shared within the Census Bureau
unless there is a need to know and use those data. For very few
people is there a need to see individual data, and that need is
protected and is examined closely before access is allowed.
We try to secure data physically, both--and electronically
in the Census Bureau and its branch offices and in the laptops
that enumerators use and so on.
So there are very strong steps made, and the culture of
protecting that is seen as a part of our contract with the
people for their willingness to respond to our questions.
Senator Klobuchar. Thank you.
Director Barabba. I have a story that demonstrates how deep
it is into the Bureau.
During the 1980 census, someone broke into one of the local
census offices. I think it was in Colorado. Because it was a
Federal office, the FBI came in to investigate. The Census
Bureau employee informed the FBI agent they could not see the
records because they were not sworn employees of the Census
Bureau, but there was about to be a little fist fight.
Senator Klobuchar. Did you intervene?
Director Barabba. It got to my office, and I called the
Director of the FBI, and I explained to him the situation. He
said, well, we have got to do our job, too. I said, would you
mind if we swore your agents in as employees of the Census
Bureau for this investigation? He said, that seems like a
reasonable solution; and so they were sworn in.
So they were under the rules of the Census Bureau as far as
confidentiality and privacy when they were performing that--but
it was down to the Census Bureau office, this attitude that is
pervasive in the Bureau. It is the one thing that I think if
you ask anybody what is the most important thing the Bureau
does, it is keep privacy and the trust of the American people.
Senator Klobuchar. Very good.
And then, just secondly, a little more question specific to
my State, and that is that we have a population--it gets a
little cold in Minnesota in the winter. So we have some people
that go south for a month or two. We also have a lot of college
students in our State. Could you talk about just the unique
challenges of trying to reach those groups of people? Dr.
Bryant.
Director Bryant. I think one thing that is different
between the American Community Survey and the census itself,
the census counts where you are on April 1st or where you lived
most of the year around that time. The ACS will now give us
some data that helps with the winter/summer snowbird vacationer
population by saying, well, here is the population estimate for
Scottsdale, Arizona, versus Minnesota at a particular time.
Incidentally, Minnesota had the highest return of mail
questionnaires in the 1990 census. Probably we can all say that
for our censuses.
Senator Klobuchar. Okay. Very good. I guess that is why I
am here. Thank you.
Chair Maloney. Thank you.
That concludes our first panel. We have time for the second
panel.
Representative Brady. Just 2 seconds. One, is there a limit
on how many times the Senator can brag about Minnesota on this
panel? Because, really, she is at that--there must be a census
limit.
Do you think we could have the panelists in writing give us
their thoughts on if the block level census data isn't accurate
or needed, what would be the next level that would be
appropriate and why? I think that would be interesting for us
to know as a committee.
Although I would say without the block level census data it
would deny State legislatures the ability to draw those
compact, commonsense districts that they do each 10 year
period. But I think that would be very helpful information, if
you don't mind.
Chair Maloney. I think that would be very helpful. If
anyone would like to comment to that question.
Director Prewitt. I want to make sure that when I say the
data are flawed, they are flawed for two reasons. One, it is
extremely difficult to get exactly the right address on Ms.
Watkins.
The other reason--and it goes back to the confidentiality/
privacy issue. Block-level data are what we call scrambled,
which is to say, on a given block, let us say, there is a black
male between 25--when you actually see the census data, there
is no black male who is 25. There may be an Hispanic woman who
is 25 and on some other block where there is actually an
Hispanic woman that is 47. On that block, the age, the gender,
and the races are scrambled so that when we get up to a high
level, we have got exactly an accurate picture, statistical
picture of that level, census track level, block cluster level,
what have you.
We do that so there can't be identification. The
redistrictors actually believe this race ethnicity data that
they have got at the block level. We know it is not there. So,
in some complicated way, we are misleading the population to
believe that there is a false precision in these data.
That is why I think it is so important, and we ought to
sort of say it is not precise. You are using it as if it is
precise exactly for creating the districts so that they have
the right number of Democratic and Republican voters and
ethnicities and so forth.
Representative Brady. What is the better level?
Director Prewitt. It depends on the variable. But I would
think you can easily get away with block clusters or census
track data for redistricting and you will not mess up the
fundamental premises of democracy.
Chair Maloney. Thank you.
And this panel, we thank you very much.
Chair Maloney. We will call the second panel.
I would like now to introduce the second panel:
Dr. William Eddy is the Chair of the Committee on National
Statistics. He is the John C. Warner Professor of Statistics,
Machine Learning, and Biological Sciences at Carnegie Mellon.
He is a fellow of the American Statistical Association and the
Institute of Mathematical Statistics. In addition to serving on
the Committee on National Statistics, Dr. Eddy has been a
member of several CNSTAT panels and committees; and he holds a
PhD from Yale University in Statistics.
Dr. Andrew Reamer is a fellow in the Brookings
Institution's Metropolitan Policy Program. At Brookings, Dr.
Reamer manages the Federal Data Project, which seeks to
increase the availability and accessibility of detailed,
accurate, up-to-date Federal statistics relevant to
metropolitan areas. Currently, he is President of the
Association of Public Data Users and Chair of the Bureau of
Labor Statistics Data Users Advisory Committee. He received a
Masters of City Planning and a PhD in Economic Development and
Public Policy from MIT Department of Urban Studies.
Thank you both, gentlemen, for coming.
Chair Maloney. Dr. Eddy, you are recognized for 5 minutes.
STATEMENT OF WILLIAM F. EDDY, JOHN C. WARNER PROFESSOR OF
STATISTICS, CARNEGIE MELLON UNIVERSITY, PITTSBURGH, PA, CHAIR,
COMMITTEE ON NATIONAL STATISTICS, NATIONAL ACADEMY OF SCIENCES,
WASHINGTON, DC
Mr. Eddy. Thank you, Madam Chairman.
My remarks today are going to address the findings and
recommendations in two reports issued by the Committee on
National Statistics, CNSTAT as we call it, that are both
relevant to the governance of the Census Bureau and the
usefulness of the data it provides. The two reports are one
that has been referred to previously, namely Principles and
Practices for a Federal Statistical Agency--and I am going to
talk in some detail about this report--and the second concerns
the American Community Survey, which is replacing the long form
in the census, Using the American Community Survey: Benefits
and Challenges.
The Purple Book, as it is known, is a report that was
developed by the committee in 1992; and, starting in 2001, we
have issued a new addition every 4 years as a new
administration came into office. The report lists four basic
principles that a statistical agency should follow and 11
practices that it should follow.
The next four principles are: one, that the agency must be
in a position to provide objective information that is relevant
to issues of public policy; two, have credibility with those
who use its data and information; three, have a strong position
of independence within the government; and, four, the trust of
those whose information it obtains.
I want to elaborate on the principle concerning
independence, since it was mentioned by all of our--the
previous panel members.
The report does not directly speak to structural
independence, which is what they were talking about. It refers
to the independence of control. Now, obviously, that can be
affected through structure, but the report does not actually
address the structural question directly.
The characteristics that we relate to a strong position of
independence are that a statistical agency should have
authority for professional decisions over the scope, content,
and frequency of data compilation, analysis, and publishing.
It should have authority for selection and promotion of
professional, financial, and operational staff.
It should have recognition by policy officials outside the
statistical agency of its authority to release statistical
information, including press releases and documentation without
prior clearance.
It should have an authority to control information
technology systems for data processing and analysis in order to
securely maintain the integrity and confidentiality of data and
reliably support timely and accurate production of key
statistics.
It should have authority for the head and qualified staff
to speak about the agency's statistics before Congress, with
congressional staff, and before public bodies.
It should adhere to fixed schedules in the public release
of important statistical indicators to prevent even the
appearance of manipulation of release dates for political
purposes.
It should maintain a clear distinction between statistical
information and policy interpretations of such information by
the President, the Secretary, and others in the executive
branch.
And its dissemination policy should foster regular,
frequent release of major findings from an agency's statistical
programs to the public via media, Internet, and other means.
We have not undertaken a formal evaluation of the Census
Bureau vis-a-vis these criteria. But I note that, as I think
Lou Kincannon mentioned, the Department of Commerce has not
always respected important aspects of statistical agency
independence, such as authority for selection and promotion of
staff.
I want to repeat that the report does not address the issue
of structure, of the organizational placement of the Census
Bureau. I should say personally I would advocate the creation
of an independent scientific agency such as the National
Science Foundation or NASA.
One of the steps that could be taken to strengthen the
agency head's independence would to be have him or her
appointed for a fixed term by the President with approval of
the Senate, as is the case with the heads of the Bureau of
Labor Statistics and the National Center for Education
Statistics. Today, some statistical agencies are headed by
senior executive career officials, some have presidentially
appointed heads with fixed terms, and some, including the
Census Bureau, have presidentially appointed heads that lack
fixed terms and serve at the pleasure of the President.
Chair Maloney. Please bring your comments to a close. Your
time has expired.
Mr. Eddy. Okay.
I was going to just briefly say that the American Community
Survey is a very important step forward for the Census Bureau
and the data that they collect.
[The prepared statement of William F. Eddy appears in the
Submissions for the Record on page 44.]
Chair Maloney. Thank you. Thank you.
Dr. Reamer.
STATEMENT OF ANDREW REAMER, FELLOW, BROOKINGS INSTITUTION,
METROPOLITAN POLICY PROGRAM, WASHINGTON, DC
Mr. Reamer. Madam Chair, thanks for the invitation to speak
today.
Census Bureau data are essential to the effective
functioning of our Nation's democracy, public policy at all
levels of government, and our $14 trillion economy. The return
on the Nation's $2 billion average annual investment in the
Census Bureau--and that figure includes the decennial census--
is almost infinite. At the most fundamental level, the Nation
could not operate without this agency.
However, the Census Bureau is not yet a 21st century
statistical agency. I believe its offerings need to reflect
more fully three new realities: first, major changes in the
Nation's economic structure; second, the potential for the
Census Bureau to provide data to enable more informed,
effective public and private decision making across the Nation;
and, third, opportunities for new data products afforded by
advances in information technology.
Regarding the changing economic structure, while the Census
Bureau has long-standing, frequent data collections on
manufacturing, it does not yet have the same data collection
effort for service industries that now dominate the Nation's
economy. Until fiscal year 2009, the Census Bureau sought but
has been unable to obtain the $8 million needed to regularly
survey the finance, insurance, and real estate industries. The
lack of these data has meant that the gross domestic product
estimates are not as accurate as they might be and, as a
result, macroeconomic policymakers at the Federal Reserve and
elsewhere have been making policy without the best possible
data.
In a similar vein, the Census Bureau does not have a
regularly collected survey of residential finance. In light of
the causes of the current recession, policymakers need an
accurate current picture of housing finance markets. Efforts
are under way to address this data gap, including a proposed $3
million biannual multifamily residential finance survey.
I am pleased to see that, due to recent and likely
appropriations, the Census Bureau's economic statistics are
poised to be more reflective of our 21st century economy.
However, as that economy is never static, going forward, strong
and stable funding is needed to ensure that the Census Bureau's
data products fully capture current realities and policymakers'
needs.
Regarding more informed and effective decision making, with
the advent of the Web-based data access, State and local
governments and millions of private businesses across the
Nation can more easily and quickly incorporate census data in
their decision making, with the potential to improve the
outcomes of trillions of dollars in investment. These data
users are primarily interested in data at the subnational
level--States, metros, counties, cities, neighborhoods--and
they will be rewarded soon with the first annual publication of
the American Community Survey data at the neighborhood level.
However, the ACS sample, 3 million households a year, is
too small to provide reliable estimates at the neighborhood
level. So, for 2011, I encourage the Census Bureau to request
and Congress to approve a larger sample, 3 percent of the
Nation's households. As our Nation's economic health is a
function of the competitiveness of our regional economies,
policymakers at the Federal, State, and local level need a full
understanding of the performance and the structure of these
economies.
Detailed, accurate economic data at the metropolitan level
are particularly important; and I encourage the Census Bureau
to publish metro-level data on research and development, on
innovation, foreign trade, place-to-place migration, and
business starts, expansions, reductions, and closures.
Regarding technically innovative data products, the Census
Bureau has been in the forefront of efforts to develop new data
products to take full advantage of information technology
advances. However, it has had difficulty getting funds to fully
exploit these possibilities. A case in point is the Local
Employment Dynamics program, which describes how firms and
people move through the economy over space and time, giving a
look under the hood of the economy, the hires and fires, where
people live in relationship to where they work.
LED has been in existence for over a decade, with an
appropriation of only $2 million a year to keep LED afloat. The
Census Bureau has needed to draw another $6\1/2\ million from
discretionary funds and reimbursable work. As a result, LED has
been limited in its value to macroeconomic and regional
policymakers.
However, full funding for LED is on the horizon. The
administration has requested $13.7 million to expand and
stabilize the program. The House and the Senate Appropriations
Committee have approved the requested funding, and I encourage
the full Senate to approve it as well.
In conclusion, the Census Bureau has the potential to
transform how the Nation conducts its work--at a little
additional cost to the taxpayer. For the Census Bureau to
become a 21st century statistical agency, it must understand
and effectively respond to user data product needs, take full
advantage of opportunities offered by cutting-edge information
technologies, and have the support of the Commerce Department,
OMB, and Congress to obtain stable funding. The Joint Economic
Committee can play a valuable role in ensuring that these steps
are taken so that the Bureau can achieve its potential.
I thank the committee for your attention.
[The prepared statement of Andrew Reamer appears in the
Submissions for the Record on page 46.]
Chair Maloney. Thank you very much.
I just have one question to our panelists. Thank you for
your testimony. Do you have any specific advice for the new
Census Director who was just sworn in?
Mr. Eddy. I do not.
Mr. Reamer. I think one is to speak regularly, loudly,
clearly that the American people can have confidence in the
decennial census, specifically, in the Census Bureau in
general, to address some of the concerns made by the previous
panel. And as the decennial census is the major focus of
attention for the next year, he should not forget the other
aspects of the Census Bureau, which are vital to keeping the
Nation running politically and economically.
Chair Maloney. Thank you very much, and the Chair
recognizes Mr. Hinchey for 5 minutes.
Representative Hinchey. Thanks very much, Madam Chair.
I appreciate this hearing. It is very interesting, quite
frankly, much more than I anticipated.
Chair Maloney. Actually, the census is fascinating, really.
Representative Hinchey. And the testimonies have been very
fascinating as well.
Dr. Eddy, I think that you were about to talk a little bit
more about the American Community Services; is that correct?
Mr. Eddy. The American Community Survey, yes.
Representative Hinchey. Which is used by the Census Bureau
and which is something that, frankly, although it is not hidden
as a secret or anything, but, nevertheless, it is not very
widely known. So maybe you can tell us a little bit more about
that and what you think about it, how effective it is, and how
it could be more effective if it were to be used in that way.
Mr. Eddy. Well, the single most important thing to
understand about the American Community Survey is that it was
conceived as a way of reducing the burden of the census.
Because the census since--I don't know--perhaps 1940, has
collected something called the long form, which in the early
days asked questions like how many toilets and how many
telephones do you have and most recently has asked about your
electric bill and various other things that are sort of
difficult to answer and, as a result, the long form in the
census didn't get very good data.
So this survey was substituted. It is a very complex
survey. I couldn't possibly begin to explain it to you. But,
essentially, every month a number of households are interviewed
for all of the questions that would appear on this long form.
And over a period of time we then develop a picture of the
whole country by the sampling mechanism of the survey.
The difficulty is we have sort of only just begun; and, in
fact, next year will be the first year that the 5-year averages
will be released. These provide data down to geographic units
of about 20,000 population, and so one of the losses with the
American Community Survey is we don't have as fine geographic
detail as we used to get with the census. But we get much finer
temporal result detail, because we gather the data basically
every month.
Representative Hinchey. So----
Mr. Eddy. I guess I should add there is a wonderful
potential for money saving. I would expect, by 2020, the Census
Bureau will have figured out how to reduce the cost of the
decennial census because they don't have to collect this
information.
Representative Hinchey [continuing]. Do you think the
American public should be made more aware of the American
Community Survey?
Mr. Eddy. Absolutely. I think they are, in a very indirect
way. I would guess once a week USA Today has a front-page
article about some change in the demographics of the country
that come from that kind of data.
Representative Hinchey. There are a number of countries
around the world, as I understand it--for example, Ireland,
Japan, Australia, New Zealand--that carry out census every 5
years, rather than every 10 years the way we do. Do you think
this is something that we should be thinking about?
Mr. Eddy. Not at all, particularly given the innovation of
the American Community Survey. I think the need for anything
more often than every 10 years is not there.
Representative Hinchey. Any cost-benefit analysis that has
been done on that?
Mr. Eddy. I don't believe there is, but I think it is clear
that it would be less expensive to do it every 10 years than
every 5 years.
Representative Hinchey. No question about that, yes. But is
there any----
Mr. Eddy. I am suggesting that you wouldn't actually gain
that much additional information.
Representative Hinchey [continuing]. Okay. Dr. Reamer.
Mr. Reamer. I agree with that.
Representative Hinchey. Dr. Reamer, you were saying that
there should be a larger percentage of national households----
Mr. Reamer. In the American Community Survey sample.
Representative Hinchey [continuing]. Could you talk a
little more about that, why you think and what the benefits
would be?
Mr. Reamer. Dr. Eddy was saying the ACS replaces what was
called the long form, and the long form went to one out of six
households in the country. In 2000, the long form went to 17
percent of the households. When the ACS was developed, the
original plan was to sample 3 percent of the households every
year so that over 5 years you would hit 15 percent.
As a budget-saving measure a decade ago, the 3 percent was
changed to 3 million households. So the Census Bureau is
surveying 3 million households every year, but the population
is increasing. The result is that the percentage of households
being surveyed is falling, and we are now well under 3 percent.
So I think for the 5-year data coming out, the sample size
is more like 12 percent of the households rather than the 17
percent we got in the long form. So the result is, for the
really small areas like the neighborhoods, the data are less
reliable; and we need a slightly higher sample, moving from 3
million to 3 percent which would be today about 3\1/2\ million
households, to get a more reliable sample.
Representative Hinchey. I think that is very important. We
ought to know more about what is going on in this country with
regard to the families, how they are operating, how they are
not operating, how the effects of the various circumstances,
particularly the economic circumstances, are affecting them. I
think that that is very, very important. And the idea that you
need to cut the budget here is pretty silly, because that is
only a tiny, tiny, tiny fraction of the overall budget.
Mr. Reamer. It is very tiny. The budget for the ACS is
about $200 million a year. So increasing the sample would not
increase the budget that much.
Representative Hinchey. One other thing----
Chair Maloney. The gentleman's time----
Representative Hinchey [continuing]. I just wanted to ask
one last thing.
Do you know anything about ACORN, ACORN as part of the 2010
census partner? Do you think anything about that?
Mr. Reamer. No.
Representative Hinchey. Thanks.
Chair Maloney [continuing]. The Chair recognizes Mr.
Snyder, who has joined us. If he would like to make a statement
or ask a question. We appreciate your presence, your work on
this committee.
Mr. Snyder. Madam Chair, I apologize. I would have been
here earlier. I applaud you for all your efforts. You have
spent quite a number of years on this topic, and I appreciate
you.
I don't want to ask any questions.
Chair Maloney. Thank you.
I do want to note that Dr. Linda A. Jacobson was unable to
join us, but her testimony will be made part of the official
record.
[The prepared statement of Linda A. Jacobson appears in the
Submissions for the Record on page 49.]
Chair Maloney. I want to thank all of our witnesses today
for being here and talking about how we can strengthen the
already invaluable role the U.S. Census Bureau plays in policy
making.
The hearing is adjourned.
[Whereupon, at 2:37 p.m., the committee was adjourned.]
SUBMISSIONS FOR THE RECORD
Prepared Statement of Representative Carolyn B. Maloney, Chair, Joint
Economic Committee
Good afternoon. I'd like to welcome our two distinguished panels of
witnesses and thank them for agreeing to testify today on the role of
the Census Bureau in the 21st century.
This is the first in a series of hearings the Joint Economic
Committee will hold to examine the state of the federal statistical
system. The data collected by our statistical agencies are vital to
informing policy debates and evaluating the effectiveness of those
policies we put into place.
As we debate health care reform, Census data tells us that 46
million Americans are without health insurance.
Unemployment, family income, poverty--the numbers we see in the
headlines every day--they are our motivation for making policies and
writing legislation. Federal statistics are a bargain, costing between
$10 and $25 per person each year, but the information gleaned is
invaluable.
I cannot stress enough how heavily policy makers on this committee
and at all levels of government rely on the data produced by the Census
Bureau as we weigh policy options. The data enables us to evaluate
whether or not a policy is achieving the goals we intended.
We begin today by focusing on the Census Bureau, the country's
largest principal statistical agency. Census is most well known for its
role in conducting the national census every 10 years. Beyond that, the
Bureau conducts the annual American Community Survey, and many other
surveys that provide key information on other economic and demographic
subject areas. The Bureau's population estimates determine
congressional districts, and drive how we allocate funding for millions
of dollars in federal aid.
There is no doubt of the Bureau's significance and the importance
of the work it does. But I am concerned when I see that the new
Director of the Census Bureau, Dr. Robert Groves, was confirmed by the
Senate just last week--six months into the new administration, and less
than a year--261 days according to the countdown on the Bureau's
website--before Census Day 2010 on April 1, 2010.
The decennial census--the largest peacetime mobilization of
government workers--takes place every ten years, but the leadership
changes every four years with a new Administration. But statistical
agencies like the Census Bureau should be absent political pressures so
that the data remains unbiased and objective.
Today we will hear from former Census Directors who combined have
almost twenty years experience spanning five Administrations. Yet, we
find ourselves in the same peril today as in previous decades, like
some heroine tied to the railroad tracks. Given that we have a wealth
of expertise and knowledge in conducting the census and we know how
important sound data is to policymaking, I am interested to hear your
perspectives on how to avoid flirting with disaster every decade. I
would like to hear your practical suggestions of how we can avoid
ending up in the same predicament in 2020.
I have introduced legislation, which you have all endorsed, to give
the Census Bureau independent status, similar to the National Science
Foundation and NASA. Other federal statistical agencies, such as the
Bureau of Labor Statistics and Energy Information Administration, are
part of another Executive branch agency but the director is appointed
for a fixed term.
In order to be relevant to policy decisions, the major ongoing
surveys conducted by the Census Bureau need regular review, updating,
and sometimes, complete redesign depending on economic, social, and
technological changes. The Bush administration era cuts to our
statistical and scientific infrastructure budgets have undermined our
ability to evaluate the effectiveness of our policies.
We must impress upon those around us the value of the federal
statistical system and challenge lawmakers and departments to support
the system with resources and ensure that the statistical agencies
maintain a strong position of independence.
______
Prepared Statement of Representative Kevin Brady, Senior House
Republican
It is a pleasure to join in welcoming both panels of witnesses
testifying before us this morning. There is broad bipartisan agreement
on the importance of impartial, accurate, and timely economic data. We
also need to ensure that federal economic statistics fully reflect the
growing importance of service industries and exports to GDP.
The Census Bureau publishes a number of economic statistics but
also collects and compiles data for other agencies. For example, Census
Bureau personnel are engaged in collecting the data used for the Bureau
of Labor Statistics (BLS) household survey, from which the unemployment
rate is derived.
Unfortunately, a review of the data published by the Census Bureau
as well as other statistical agencies shows that the economy remains in
a severe recession and that the Administration's optimistic economic
forecast is not consistent with the facts. This rosy economic forecast
means that the Administration's projections of budget deficits and debt
are significantly understated, misleading policy makers as they
consider trillions of dollars of additional federal spending related to
the health insurance proposal and other programs.
This may explain why the Administration is attempting to defer the
release of its midsession budget update until after Congress votes on
the health insurance proposal, a measure that the Congressional Budget
Office says will boost budget deficits as well as health care costs.
Congress should not further undermine the financial position of the
federal government and should fully consider the implications of how
far off the Administration's economic assumptions have been.
For example, last January top Administration economists projected
that the unemployment rate would not exceed 8.0 percent if the stimulus
were enacted, but this rate is currently 9.5 percent and will probably
be above 10 percent by the end of the year. Similarly, the
Administration projected that GDP will decline by 1.2 percent in 2009,
less than half as much as forecast by the Blue Chip consensus. Clearly,
the stimulus is not having the positive impact assumed in the
Administration's forecast.
Gross domestic product declined 5.5 percent in the first quarter of
2009. According to the Blue Chip consensus, the economy is forecast to
decline 1.8 percent in the second quarter of 2009, and then increase by
1.0 percent in the third quarter and 1.9 percent in the fourth quarter.
Consumption spending increased 1.4 percent in the first quarter.
More recent monthly data show that consumption spending slipped in
March and April and edged up 0.2 percent in May. Many households are
under severe financial pressures from heavy debt burdens and the lower
values of their homes and equity investments. Temporary additions to
disposable incomes from the stimulus are not significantly boosting
consumer spending, which is unlikely to be a driving source of economic
recovery. As households continue to pay down debt for the next several
years, consumption growth will likely be constrained.
Instead, higher business investment will be needed to return to
healthy economic growth. However, business investment has collapsed in
recent quarters. One important component of business investment,
equipment and software spending, dropped 33.7 percent in the first
quarter of 2009. The prospect of higher taxes and federal spending,
more intrusive regulations, and higher inflation in the future all
undermine the likelihood of a strong rebound in business investment
needed for adequate economic growth.
I would also suggest that the potential influence of politics in
the Census Bureau should be curtailed. Many of us were alarmed earlier
this year by reports that the White House was seeking to directly
oversee the Census Bureau in connection with the 2010 Census. I
continue to believe that political and ideological groups such as ACORN
should have nothing to do with any process leading up to the decennial
census. The Census Bureau is an important national resource and its
statistical integrity must be protected from potential political
pressures.
______
Prepared Statement of The Honorable Vince Barabba, Director From 1973-
1976 and 1979-1981
Thank you for inviting me to participate in this hearing. This is a
topic for which I have had a long and deep interest. As an example, in
2002, I addressed my concerns at the 100th Anniversary of the Census
Bureau in a presentation appropriately titled for today's hearing: THE
NEXT 100 YEARS . . . Starting Today.
During that presentation I pointed out few statistical agencies are
either equipped or authorized to comprehensively assess what society
needs to know because such an assessment would require a dialogue
across the many functions and special interests that will use that
information in their attempt to serve society. I pointed out that what
was needed was an open discussion between those who determine what they
need to know and those who collect it about the form, accuracy, and
cost (in both time and money) of the information required. I stated
that continued improvement in this area was needed for at least two
reasons:
First, it is no longer sufficient to address societal issues
from the limited perspective of functional policy organizations
such as labor, commerce, health, and education.
Second, government can no longer ``predict and prepare'' for
the future. The fact that our society faces an increasing
complexity and an accelerating rate of change now requires
government to use information to ``sense and respond'' and at
times ``anticipate and lead.''
I have used two metaphors to portray fundamental changes that have
occurred which have--and continue to--required us to design a new
system of government statistics:
The first metaphor is the jigsaw puzzle. The mechanistic mind-
set of the industrial age encouraged us to think about
addressing problems in government and businesses as if we were
solving a jigsaw puzzle. When one starts a jigsaw puzzle, one
knows how many pieces one is supposed to have, and the chances
are that they are all there. Each of the parts will interact
with only a small portion of the other parts. If any of us had
trouble trying to complete the puzzle, there is a picture on
the box that reveals the single ultimate solution. This solve
the puzzle metaphor fit reasonably well for most of the issues
we faced during the early part of 20th Century--and
represented, to a great extent, the way things were thought of
at many public and private enterprises and taught at many
colleges and universities.
The second is a molecular structure of interacting elements. In
the latter part of the 20th century, business and societal
challenges became far more complex. On a daily basis, we saw
(and are seeing) the impact of this increasing complexity and
accelerating rate of change on our daily lives.
We now operate in an environment consisting of constantly
changing processes, relationships and components . . . more
like the elements in a molecular structure than a jigsaw
puzzle. Depending on how the elements of a molecule interact,
particularly when external positive and negative forces are
imposed, we can end up with an entirely different outcome than
we expected.
In the presentation at the Census Bureau I referenced an experience
I had during my first tenure at the Census Bureau that relates to this
issue and which is very relevant to the topic of this hearing.
During the annual budget development process a Commerce Department
budget analyst had decided to reduce the Department's current budget
problem by eliminating the Census of Agriculture item from the Census
Bureau's budget. As might be expected, particularly since that Census
is mandated by the Congress, the Department of Agriculture protested
and appealed to the Congress to transfer the Census of Agriculture to
their department.
While almost everyone in government was focusing on who should
collect the information, Jim Bonnen, who would become one of my most
constructive critics, pointed out that society needed to know and
understand both the specifics and interactions of the agricultural
system that started with the growing of agricultural products and ended
with putting them on consumers' tables. This meant we needed to
integrate the data and information collected from the inputs (that is,
seed, fertilizer, machinery, etc.) through agricultural production,
commodity assembly, initial processing, further manufacturing,
wholesaling, retailing, transportation, and eventually to consumer
consumption.
With that systems view in mind, he suggested the Census Bureau
commit resources to identifying and integrating the different pieces of
food sector statistics scattered throughout several economic censuses
and surveys and relate them to the agricultural census. In essence, Jim
suggested we align our statistical practices around the user's needs
and not the existing functional organizational structure designed to
collect information. Although we have made some improvements in this
area, we still face similar issues because of the increased level of
complexity and accelerating rate of change that has occurred since that
time. As an example, who at that time would have expected an energy
crisis and global warming that would encourage the use of corn based
ethanol which eventually impacted the availability and eventually the
price of corn?
To address many of the improvements that this committee is seeking
will require an appreciation of thinking and acting in a more systemic
way. Russell Ackoff defines a system as ``any entity, conceptual or
physical, which consists of interdependent parts.'' Conversely, ``a
system is a whole that cannot be divided into independent parts.'' Each
element of the system must rely on and interact with the rest of the
system if the enterprise as a whole hopes to succeed. Problems are best
solved not by breaking them up into functional bits, but by carrying
them into the next larger system and solving them through integrative
mechanisms. In short, we want to create a whole that is more valuable
than the sum of its parts.
In my mind the proposed legislation is a potential first step to
address the information needs of those who establish policy and laws.
If implemented properly it could serve as a basis for the creation of a
Federal Statistical System that is of greater value to society than the
sum of each of the individual statistical agencies which it
encompasses.
______
Prepared Statement of Barbara Everitt Bryant, PhD, Director From 1989-
1993
Chair Maloney, members of the Joint Economic Committee, I am
Barbara Everitt Bryant. I was Director of the Census Bureau from 1989
to 1993 and of the 1990 census. You have asked me to talk about how I
perceive the role of the Census Bureau in this century.
The Census Bureau got off to a roaring start for this 21st century
by implementing the American Community Survey. As of 2010, the American
Community Survey--acronym ACS--provides new data on the characteristics
of the U.S. population every year with enough interviews to report on
even the smallest communities. Prior to the ACS, the nation had to wait
every 10 years for a decennial portrait of who we are and how we live.
The American Community Survey was envisioned in the 1960s. It was
finally researched and designed in the 1990s and in this decade. ACS
was a long time coming but worth the wait. Its implementation frees the
decennial census of many social and economic questions and leaves the
2010 questionnaire with only the eight questions needed for
reapportionment and redistricting, the Constitutional and legislative
purposes of the census.
In the rest of this century, I envision that the Census Bureau will
build upon its illustrious past as the originator of data processing,
of computerization, of the Tiger geographic system which is the basis
for computer mapping of every block in the nation, and as the source of
continuous improvements in capturing data faster and more accurately. I
think we can count on the Census Bureau to be a continuing source of
innovation, new products and processes. That is, if we structure it for
the future and not for the past. When he figured out how to do
massively large counts with punch cards for the 1890 census, Herman
Hollerith--a census employee--didn't envision today's computer
industry. More recently, we who watched the computer mapping of every
block in the nation get implemented for the 1990 census didn't envision
that this, coupled with communication satellites to transmit the data,
would become the start of a large GPS industry. Amazingly, the Census
Bureau is an enormous bureaucracy with all that implies, but it is also
a hotbed of change. The challenge for this century is to keep it that
way although we do not know what all the changes will be.
Let me speak to two ways you might change structures to make the
Census Bureau nimble, and less bureaucratic as we move into the next 91
years of this 21st century.
1. recognize the 10-year cycle in which the census bureau operates and
make its director's term of office 5 years
First, recognize that the Census Bureau operates on a 10-year
cycle, not a 2-year, 4-year, or 6-year political cycle. Within the
decade are two five-year cycles for the Economic Censuses. As it faces
its largest project, the decennial census, the prior census was always
conducted by a prior administration. The only institutional how-to-do
it memory for census taking rests in career employees at the Census
Bureau, not at the Commerce Department in which the Census Bureau
resides. This cycle also means that every 20 years--as has occurred so
recently--the Director of the Census Bureau is a Presidential appointee
of a President inaugurated in January of the year ending in ``9.'' That
Director is not in office in time to have any role in the planning of
the census which he or she is charged to direct. The Senate just
confirmed a new Director last week, seven months before the start of
the 2010 census. Twenty years ago, I was not in office until December
7, 1989, with the census to start in early 1990. Questionnaires for
which I would have implemented easier-to-use formats were already
rolling off the printing presses.
The solution to the inherent difficulties of a 10-year cycle is to
make the term of the Census Director a 5-year term, half of that cycle,
starting in the years ending in 1 or 2 and 6 or 7. That way a Director
coming to office in mid-decade could fully participate in the ramp up
to the decennial census and the first dissemination of data from that
census.
2. flatten the bureaucracy
My second recommendation is to flatten the bureaucracy by removing
the Census Bureau from the Department of Commerce. Since leaving the
Census Bureau, I have spent 16 years at the University of Michigan in
its business school working on an economic indicator, the American
Customer Satisfaction Index, or ACSI. The current mantra for customer
satisfaction, for getting close to the customer, is to flatten
organizational structures. Successful corporations are doing this to be
profitable and get repeat customers. Unfortunately, some corporations
from my own state of Michigan learned this lesson too late and are only
now, after bankruptcy, changing their structures to be leaner and more
responsive to customers. Structures that were very successful in the
20th century don't necessarily work in the 21st.
The Census Bureau is a bureaucracy under a bureaucracy. It's a
large organization that reports to another large one, the Department of
Commerce. Commerce is not geared to a 10-year cycle but to a four-year
one. Commerce has many other large organizations under it--the National
Oceanic and Atmospheric Administration, the National Institute of
Science, and Technology, the Patent Office, and others. The Commerce
Department has a large load of responsibilities to deal with of which
the Census Bureau is only one. When the Census Bureau was small, it was
logical that it be under a larger department. Today the Census Bureau
is the largest statistical organization in the federal government and
could be more responsive to its customers if it were a free-standing
scientific organization like the National Science Foundation.
Who are the Census Bureau's customers? First, there is the
Congress, then the other statistical organizations for which it
conducts major surveys such as the Current Population Survey (BLS), the
Housing Survey (HUD), the Health Interview Survey (HHS), the crime
survey (DOJ). State and local governments depend on census data for
decision making. Finally, the American public is both a customer and a
data supplier. The Census Bureau depends upon the confidence and good
will of the public for interview responses which become its data. The
Census Bureau needs to be flexible in communicating to these customers.
Census Bureau paperwork goes through not one, but two levels of
approval in the Commerce Department at an under-secretariat level and
again at the secretary level. Every response to a letter from you in
Congress, every press release, and every major decision must be vetted,
and often is edited, at the Department of Commerce. This delays
responses and leaves customers, in turn, thinking that the Census
Bureau has something to hide because response is so slow.
The Department of Commerce is not a statistical organization. Its
personnel, for the most part, do not understand what the Census Bureau
does or needs. Its own budget is dominated in census years by that of
the Census Bureau and once the census is completed, Commerce does not
understand why the Census Bureau starts immediately planning for the
next census, and needing money to do so.
When I became Director I found an organization that was way behind
the academic and private sector survey research organizations, and even
the much smaller Statistics Netherlands, in implementing computer-
assisted-telephone-interviewing, or CATI, for surveys. I had been using
such interviewing methodology for several years at the medium-size
market research company from which I had come. Once the 1990 census was
over, I made moving the Census Bureau away from paper and pencil
interviewing to computer interviewing a priority. The Census Bureau,
which had been an early 20th century leader in computerization, was
lagging in what was proving to be a cost-saving and accuracy-improving
technology. Since the Census Bureau's budget request is within the
Commerce Department's budget that goes to OMB, budgeters at Commerce
deleted our requests for funding for research and development of
computer-assisted interviewing two or three years in a row. After all,
their thinking in the years following the census was that it was time
for other Commerce agencies to get larger budget shares. The Census
Bureau shouldn't need new money. Finally, at the Census Bureau we had
to rearrange R&D budgeting, borrowing from existing programs to find
the money to bring in a panel of experts from academic and private
sector survey organizations, from Statistics Canada, and from
Statistics Netherlands to assess our computerized interviewing
situation and give us guidance on how to play catch-up. Ultimately, the
Census Bureau caught up and by the mid-1990s, using software adapted
from the University of California, Berkley, and from the Netherlands,
all survey interviews were computerized with the charges built into
survey costs. The computer-assisted interviewing was much used for
follow-up on non-respondents in the 2000 census and will be in 2010.
But the Census Bureau might not have had it if it was still waiting to
get funds approved by Commerce personnel who didn't understand what
CATI was and why it was the methodology of the future.
conclusion: two recommendations
In conclusion, my two recommendations for structuring the Census
Bureau to be successful in the 21st century and to serve its customers,
including the other statistical agencies are:
1. Make the term of the Census Director a 5-year term
2. Make the Census Bureau an independent agency removed from
the Commerce Department.
______
Prepared Statement of Dr. Martha Farnsworth Riche, Director From 1994-
1998
Chairwoman Maloney and members of the Joint Economic Committee,
thank you for providing this opportunity to testify on my experiences
as a director of the U.S. Bureau of the Census. In my testimony I want
to make the case that the Department of Commerce, which currently has
line responsibility for supervising the activities of the Bureau, has
an inherent management conflict. I believe the other former directors
of the Bureau who are here today share this view, regardless of the
political affiliation of the Administration in which they served.
Like every cabinet agency, the Department of Commerce is made up of
many offices and bureaus, all vying for federal funds each year to
perform their responsibilities. But none of these other branches has a
constitutionally mandated responsibility to conduct the nation's
largest peacetime mobilization of money and manpower every 10 years--
the decennial census.
As each census approaches, the Census Bureau's annual request for
funds jumps quickly from hundreds of millions of dollars to many
billions of dollars. That circumstance alone throws the entire
Department of Commerce budget off track every decade. This year the
Bureau's budget request includes more than $7 billion dollars for the
forthcoming census. Three years ago, before the final decennial ramp-up
began, the Bureau received $512 million for Census 2010.
The census thus inevitably causes conflict between the Census
Bureau and the Department of Commerce, which has to manage an overall
departmental budget according to quite different priorities. For
instance, during my tenure as director of the Census Bureau, the
National Oceanographic and Atmospheric Administration (NOAA) wanted a
new weather satellite just at the time we were starting to ramp-up our
funding requests for Census 2000. So budget conflicts are inevitable,
and one result is that Commerce tries to defer important Census
activities, often until it is too late to undertake them efficiently or
effectively.
There are three additional issues that I would like to address:
1. Content: The inventory of statistics and demographic measures
needs to be constantly updated to reflect changing needs, but it takes
about 20 years between perceiving an important need and getting the
data on the street . . . if all goes well.
Only the federal government can collect ``official'' statistics.
Only the federal government has the resources and the authority to get
the job done. But policy questions that call for new general-purpose
data tend to be asked by different agencies, not the Department of
Commerce.
For instance, low-skilled American workers are now in competition
with low-wage workers around the world. Policymakers are looking for
measures of education, occupations, and incomes across the work life,
not just at a point in time, to probe for ways to improve the outlook
for Americans whose economic well-being is stagnant at best.
Developing such complex measures effectively requires regular
advisory input from stakeholders, statistical professionals, and
measurement experts, as well as oversight from Congress, its Government
Accountability Office (GAO), and, in the executive branch, the Office
of Management and Budget (OMB), which has coordinating responsibility
for federal statistics and measurement burdens.
This task calls for constant listening and communications
activities, requiring direct access, in both directions. My experience
as a former Director is that the Department of Commerce too often seeks
to shield the Census Bureau from some of these conversations and in the
process ends up isolating the Bureau instead.
2. Resources: Resources are always limited, so the Census Bureau's
resources need to be addressed in the context of statistical
priorities. As I indicated earlier, right now the Census Bureau is
contained within a cabinet-level department that has its own
priorities, and a multi-agency appropriations sub-committee with an
even broader focus.
I believe that this calls for situating the Census Bureau in a
resource context that is focused on producing federal information, and
thus in a position to prioritize effectively. This is especially
important given the development process for producing new measures, and
the ongoing evolution of measurement techniques and technology.
3. Independence: The decennial census is very political; that's the
point of it. For that matter, all government statistics are political:
the word ``statistics'' means ``measures of state,'' or metrics used
for governance.
The issue at hand is how to maintain the Census Bureau and other
statistical agencies' independence in pursuit of accurate data. We need
a set of regular processes built on transparency, collaboration with
other measurement agencies and professionals, and regular reporting,
and that are not subject to political appointees, no matter how well
intentioned.
Finally, successful measurement depends on willing respondents.
Federal statisticians have very little control over respondent
attitudes created by other actors, with varying motives and expertise.
This increases the value to the Census Bureau of advertising, outreach,
and stakeholder relationships, as well as innovative data collection
methods. It also heightens the value of an untroubled reputation for
guarding confidentiality, especially as technology and security
concerns challenge standards for maintaining respondents' privacy.
I think these results would be much more achievable if the Census
Bureau were an independent agency.
This concludes my testimony.
______
Prepared Statement of The Honorable Kenneth Prewitt, Director From
1998-2001
Secretary of Commerce Robert A. Mossbacher, 1991
``. . . the choice of the adjustment method selected by the
Census Bureau officials can make a difference in apportionment,
and the political outcome of that choice can be known in
advance. I am confident that political considerations played no
role in the Census Bureau's choice of an adjustment model for
the 1990 census. I am deeply concerned, however, that
adjustment would open the door to political tampering with the
census in the future.''
Chairman of the Republican National Committee Jim Nicholson, 1997
``The Clinton Administration is implementing a radical new way
of taking the next census that effectively will add nearly four
and one-half million Democrats to the nation's population. This
is the political outcome of a controversial Executive decision
to use a complex mathematical formula to estimate and `adjust'
the 2000 census / . . . /.''
Senator Richard C. Shelby, 2009
``By overcounting here, undercounting there, [census] manipulation
could take place for sole political gain.''
I have no interest in rehashing the political debate over the use
of sampling (to be technically correct, the statistical methodology of
dual system estimation) but do draw your attention to the tone of these
three quotations.
In 1991, the Secretary's language was cautious; he was careful to
say that political considerations could come into play, not that they
had.
In 1997, the language is declarative. They had come into play.
In 2009, the language assumes political manipulation almost matter
of fact.
Although the first quote here listed is dated 1991, the
politicization of ``sampling'' was initiated more than a decade
earlier, when the Census Bureau was taken to court by the City of
Detroit, the City of New York, and New York State. This was census-
taking by litigation, as the big city mayors and a state governor tried
(unsuccessfully) to overrule the statistical and scientific judgment of
the Census Bureau.
This three-decade long political mess was authored by both
parties--perhaps a rare instance of bipartisanship.
It is seriously worrisome that in high political circles, and in
the media, it is suggested that the nonpartisan, professionally
managed, scientifically grounded Census Bureau can easily choose a data
collection methodology that would favor one political party over
another.
To state this worry in the simplest of terms:
The fundamental premise of our representative democracy--that it is
fair--starts with the longest running applied science project in the
nation's history: counting the American people. An unfair census--
counting population groups or geographical regions at less than or more
than their share of the total population--biases all subsequent steps
in our representative democracy.
The suggestion that the census would deliberately tamper with our
democracy is a heavy charge.
Policy as well as democracy is at stake.
As currently practiced, the design, implementation and evaluation
of public policy cannot take place without a robust federal statistical
system. Hundreds of programs and laws rest on particular statistical
products. The collection of federal statistics in health, crime,
education, housing, and much more cannot take place without a robust
decennial census.
If the decennial census is thought to be easily manipulated for
political gain, it becomes just one more feature of partisan politics.
It loses both its majesty and its practicality.
I have no argument with partisan politics; no argument with a
strong contest to win elections; no argument with the politics of
policy-making; and most emphatically, no argument with the role of
statistical information in political debate.
But to pull census-taking into the world of partisan politics is to
weaken it. A weakened census weakens our democracy; it weakens our
policy process.
there is no evidence of political manipulation?
The taking of the 2000 census was more scrutinized than any in
history. As Director, I testified before Congress nearly two dozen
times between late 1998 and mid-2000. There were numerous GAO
investigations of census preparation and process. The IG was active.
There were eight formal advisory committees, all with an interest the
conduct of the census.
In addition, there was an eight-member Census Monitoring Board--
unique in census history. This bipartisan Board, working with its own
staff and $3m budget, was specifically appointed to guard against
``political'' influence.
In the millions of words written about the 2000 census, many of
them about statistical adjustment, political influence was often hinted
out, but never documented.
To state this most emphatically. No evidence has been presented
that what is under the control of the Census Bureau itself--collecting,
processing, and reporting statistical information--has been politically
manipulated.
As I concluded five years ago, after reflecting on my Directorship
of the 2000 Census:
Although the many-headed and seemingly endless scrutiny of the
census occupied management time that might otherwise have focused on
the job at hand, we welcomed its contribution to an open and
transparent census. The unprecedented oversight was a consequence of
the polarized partisan battles over census design, with its sub-text
that the Census Bureau could have a partisan agenda. This charge was
groundless and even silly. An agency said to have ``failed'' in 1990
was, a few years later, suspected of being so clever and competent that
it could design a census able to shift seats from one party to another
a number of years in the future. We could answer this accusation only
by complete transparency.
In fact, neither the culture nor the competencies of the Census
Bureau are suited to advancing a partisan agenda. The professional
statistical community--inside and outside the government--is the
bureau's peer community, and the bureau would not jeopardize its high
standing among its peers for a short-term political purpose. Of even
greater importance, the Census Bureau has the confidence of the
American public--a confidence indispensable for public cooperation with
its large complement of largely voluntary statistical surveys and
studies (see note, end of chapter). To risk public trust and
cooperation for a one-time political outcome would be an act of
institutional suicide.
Even if its culture were to allow it, the bureau does not have the
competence to predecide partisan outcomes. There is no expertise in the
bureau on trends in voting behavior or in the fine art of drawing
election lines. To deliberately influence partisan outcomes, the bureau
would need to bring to bear such expertise as it decided on
methodologies several years in advance of when census results are going
to be used for redistricting.
These factors notwithstanding, the concern that the Census Bureau
could be subjected to partisan influence was in the air. Active
cooperation with the oversight process was the only means available to
the bureau to answer this concern. In the end, all the oversight
processes, advisory groups, and public watchdogs failed to find
partisan intention in the design or conduct of the census. Given the
scope of the monitoring effort and the number of groups intent on
finding partisan bias, that is powerful evidence that there simply was
none to be found. [From Kenneth Prewitt, Politics and Science in Census
Taking (Russell Sage Foundation & Population Reference Bureau)]:
What was in the air in 2000 is in the air today. We are near the
precipice where the refutable presumption is partisan bias.
we need to get rid of this presumption?
It would be silly to claim that there are no politics associated
with census-taking. I have joined with many scholars in documenting
endless instances of political considerations surrounding the census,
starting with the infamous three-fifths rule written into the
Constitution in 1787--a counting rule that rewarded slave-owning states
with more than a dozen ``extra'' congressional seats and electoral
college votes. This slave-bonus sent Jefferson, Madison, Monroe and
other southerners go the White House.
But if the census itself is political in this broad sense, census-
taking is a different matter. It must rest on the best scientific
principles available. That of course must be true for the collection of
all federal statistics. A recent Symposium to this effect was held at
the National Academy of Sciences, and co-sponsored by the American
Association of Political and Social Sciences. Among the speakers were
all three of the Academy Presidents and the current Science Advisor to
the President. It was repeatedly stressed that federal statistics are
science in the first instance, and only then available for program and
policy purposes.
I strongly believe that an institutional reform could help to
establish the scientific integrity and independence of census-taking,
and have urged that reform since I left the Census Bureau Directorship
in 2001. Here I cite from a 2003 publication, titled Politics and
Science in Census Taking (Russell Sage Foundation & Population
Reference Bureau):
A much needed reform could help further insulate the Director from
the political battles of the moment. At present the Director has no
fixed term, but serves at the pleasure of the President. Representative
Carolyn B. Maloney, formerly senior Democrat on the House Census
Oversight Subcommittee, has introduced a bill (H.R. 1571), which would
set a five-year fixed term for the Director. If a fixed term were to
start in a year ending in ``7'' or ``2'', no President could dismiss
the Director in mid-census--as I was when President Bush came to
office. This would signal that the Census Directorship is a scientific
rather than political position, as is the case for the head of other
statistical agencies such as the Census Bureau of Labor Statistics and
also for the Director of the National Science Foundation and of the
National Institutes of Health. These too are presidential appointments,
but all with fixed terms. In fact, among all high level presidential
appointees with scientific responsibilities, the Census Bureau Director
is unique in not having a fixed term.
A more ambitious reform, and one that I urge, would be to make the
Census Bureau an independent agency, reporting directly to the
President. It might then have a prestigious and bi-partisan national
board, similar to that of the National Science Foundation. This would
insulate it from the sometimes short-sided partisan fights than can so
easily capture congressional debate.
These institutional reforms are not all that is needed, but I know
of no better way to begin the long process of ridding our political
discourse of the casual assumption that the Census Bureau could, and
even would, be complicit in a political effort. It took three decades
to dig this unfortunate hole; it may take three decades to dig
ourselves out of it. The starting point, in my view, is to position the
Census Bureau as a scientific agency, obviously subject to
congressional oversight--just as is true of other independent agencies
such as NSF and NIH, but one in which census-taking itself rests on
rigorous scientific principles fixed on only one goal: provide the
country with the best statistical products possible.
______
Prepared Statement of Charles Louis Kincannon, Director From 2002-2008
Chairwoman Maloney, it is a pleasure to join my distinguished
predecessors in testifying before your committee. Thank you for
inviting me to appear.
As the topic implies, the Census Bureau is central to the Federal
Statistical System and to statistics to help policy makers make sound
decisions. It is a key producer of economic statistics, broadly
construed to cover not only businesses and establishments but also
persons and households. It produces about 70% of the hard figures used
to estimate GDP. It produces about half of the Principle Economic
Indicators stipulated by the Office of Management and Budget.
Principle economic indicators produced by the Census Bureau:
1. Advanced Monthly Sales for Retail and Food Services
(monthly)
2. Advance Report on Durable Good (monthly)
3. Construction Put in Place (monthly)
4. Housing Vacancies and Home Ownership (quarterly)
5. Manufacturers' Shipments, Inventories, and Orders (monthly)
6. Manufacturing and Trade: Inventories and Orders (monthly)
7. Manufacturing and Trade: Inventories and Sales (monthly)
8. Monthly Wholesale Trade (monthly)
9. New Residential Construction
10. New Residential Sales
11. Quarterly Financial Report: Manufacturing, Mining, and
Trade
12. Quarterly Financial Report: Retail Trade
13. Retail E-Commerce (quarterly)
14. U.S. International Trade in Goods and Services (monthly)
Plus:
15. Census collects the survey data that measure monthly
employment and unemployment, a principle economic indicator
released by the BLS
16. Census also collects household consumption data that is a
major ingredient for the Consumer Price Index, another
principle economic indicator released by the BLS.
It cooperates especially with BEA, BLS and health, education, and
social service agencies. It plays a strong role in the Interagency
Committee on Statistical Policy. Census figures steer about $375
billion of Federal domestic assistance each year, according to the
Brookings Institution.
Other nations also view it as a central player. About five years
ago, the French Finance Ministry surveyed statistical operations and
practices at the request of the Minster. This report noted that the FSS
relies on ``service provisions and financial transfers between
agencies, as some producers are largely dependent on others for
collecting data.'' ``The Census Bureau plays a central role in this
respect, as even large agencies, such as the Bureau of Labor Statistics
or the Bureau of Economic Analysis, rely heavily on its collection
network.''
The French report also noted that a great strength of the FSS was
timeliness of data and closeness of producers to users. However, as a
decentralized system, it requires coordination, which depends on seven
persons at the OMB. ``However, the institutional framework of producers
of . . . official statistics remains a strong limitation on
coordination. It results in: redundancies of tasks such as keeping
registers providing sampling basis for surveys; in difficulties (mainly
legislative) in the sharing of microdata; and in classifications and
concepts that are not always consistent between various products or
institutions.''
The Census Bureau will remain central to the FSS and to
policymakers needing economic statistics for informed decisions. The
Decennial Census, including the American Community Survey, is a
towering contribution to this process.
What do we need to do to ensure the census meets the Nation's
needs?
Recognize the long lead time to develop, select, and
apply modern technology to all agency work. This is true of the Census
Bureau as of NOAA or NASA. The current arrangements in the Executive
Branch failed to meet fully that goal for 2010.
Recognize the long planning cycle (more than 10 years)
for the Decennial Census, beyond technological needs.
The Census Bureau must be organized to deal with this. It
needs continuity of leadership, which implies a long term of service
for the director to connect responsibility for planning to that of
production. We need to pay special attention to the role and person
holding the deputy director post, which has a strategic structural role
in the organization.
Let me suggest an illustration. In the 1990 Census cycle, we made
significant technological progress. A main example is the TIGER,
developed in cooperation with the U.S. Geological Survey. This replaced
a paper and paste pot system of producing maps for census takers with a
modern, digitized system. This could not have been done without
sustained leadership, and a willingness to accept some risk in change.
I was deputy director throughout the 10-year planning and execution
cycle. There were three directors in this period.
What else is needed?
Ensure independence and integrity of the planning
process.
Ensure staffing is purpose-based, including SES
appointment authority being the responsibility of the Director of this
multibillion dollar agency with 8,000 employees, not counting the
temporary census workforce.
Place the census budget cycle in an environment that is
not ``hostile.'' I do not mean there are enemies of the census budget
in the Commerce Department. I mean there are severe natural conflicts
that work against the census budget and its offbeat rhythm.
Madam Chairwoman, I thank you for your invitation and this entire
hearing.
______
Prepared Statement of William F. Eddy
Good morning. My name is William Eddy, John C. Warner Professor of
Statistics at Carnegie Mellon University and chair of the Committee on
National Statistics of the National Research Council. The Research
Council is the operating arm of the National Academy of Sciences, the
National Academy of Engineering, and the Institute of Medicine of the
National Academies, chartered by Congress in 1863 to advise the
government on matters of science and technology.
My remarks today will address findings and recommendations in two
of the Committee's reports, both of which are relevant to the
governance of the U.S. Census Bureau and the usefulness of the data it
provides. The two reports are the fourth edition of Principles and
Practices for a Federal Statistical Agency, issued by the Committee in
2009, and Using the American Community Survey: Benefits and Challenges,
issued by a panel of the Committee in 2007. Both reports are available
on the web site of The National Academies Press, www.nap.edu.
By way of introduction, the Committee on National Statistics was
established at the National Research Council in 1972 at the
recommendation of the President's Commission on Federal Statistics to
improve the statistical methods and information on which public policy
decisions are based. The Committee carries out studies at the request
of government agencies on statistical programs and methods. It also
addresses the statistical policy and coordinating activities of the
federal government, which are essential in a highly decentralized
statistical system. Support for the Committee's work is provided by a
consortium of federal agencies through a grant from the National
Science Foundation. Support for the Committee's Panel on the American
Community Survey was provided by the U.S. Census Bureau.
what makes for an effective statistical agency? the importance of
independence
A major activity of the Committee to strengthen the federal
statistical system is its signature white paper, Principles and
Practices for a Federal Statistical System, known as P&P or the
``purple book.'' The Committee first issued P&P in 1992 in response to
queries on what constitutes an effective statistical agency. Since
2001, the Committee has updated and reissued P&P every 4 years so that
new cabinet appointees and others could be provided with a current
edition. P&P has been widely cited and used by congressional and
executive agencies including GAO and OMB.
The fourth edition lists four principles and eleven practices. The
four principles are that a statistical agency must: (1) be in a
position to provide objective information that is relevant to issues of
public policy, (2) have credibility with those who use its data and
information, (3) have the trust of those whose information it obtains
(including households and businesses), and (4) have a strong position
of independence within the government. The practices include, among
others, a commitment to quality and professional practice and an active
program of methodological and substantive research.
I want to elaborate on the fourth principle of a strong position of
independence because it is relevant to many of the debates about
governance of the U.S. Census Bureau and the decennial census. The
Committee states in P&P that, without the credibility that comes from a
strong degree of independence, users may lose confidence in the
accuracy and objectivity of a statistical agency's data, and data
providers may become less willing to cooperate with agency requests,
thereby undermining the agency's ability to carry out its mission to
provide relevant, accurate, timely, and impartial statistics to serve
all sides in the policy debate, as well as researchers, private and
public sector planners, the media, and the general public. Of course,
statistical agency independence is always exercised within a broad
framework of departmental, OMB, and congressional oversight.
Characteristics related to a strong position of independence are
that a statistical agency has the following:
Authority for professional decisions over the scope,
content, and frequency of data compiled, analyzed, or published within
the framework set by its authorizing legislation. Most statistical
agencies have such broad authority, limited by budgetary constraints,
departmental requirements, OMB review, and congressional mandates.
Authority for selection and promotion of professional,
technical, and operational staff, including senior executive career
staff.
Recognition by policy officials outside the statistical
agency of its authority to release statistical information, including
accompanying press releases and documentation, without prior clearance.
Authority to control information technology systems for
data processing and analysis in order to securely maintain the
integrity and confidentiality of data and reliably support timely and
accurate production of key statistics.
Authority for the statistical agency head and qualified
staff to speak about the agency's statistics before Congress, with
congressional staff, and before public bodies.
Adherence to fixed schedules in public release of
important statistical indicators to prevent even the appearance of
manipulation of release dates for political purposes.
Maintenance of a clear distinction between statistical
information and policy interpretations of such information by the
president, the secretary of the department, or others in the executive
branch.
Dissemination policies that foster regular, frequent
release of major findings from an agency's statistical programs to the
public via the media, the Internet, and other means.
The Committee has not undertaken a formal evaluation of the Census
Bureau vis-a-vis these aspects of a strong position of independence.
However, I note that the Department of Commerce has not always
respected important aspects of statistical agency independence for the
Bureau, such as authority for selection and promotion of staff.
Regarding the organizational placement of the Census Bureau or
other statistical agencies, P&P takes no position as such. A variety of
organizational structures can work. However, P&P makes clear that a
statistical agency should be separate from the law enforcement,
regulatory, and policy-making parts of a department. Moreover, steps
that can usefully strengthen a statistical agency head's independence
include that the head be appointed for a fixed term by the President,
with approval by the Senate, as is the case with the heads of the
Bureau of Labor Statistics and the National Center for Education
Statistics. For a fixed term, it is desirable that it not coincide with
the presidential term so that professional considerations are more
likely to be paramount in the appointment process. It is also desirable
that a statistical agency head have direct access to the secretary of
the department or the head of the independent agency in which the
statistical agency is located. Such access allows the head to inform
new secretaries about the appropriate role of a statistical agency and
present the case for new statistical initiatives to the secretary
directly. Similarly, it is desirable for a statistical agency to have
its own funding appropriation from Congress and not be dependent on
allocations from the budget of its parent department or agency, which
may be subject to reallocation.
Today, some statistical agencies are headed by senior executive
career officials, some have presidentially appointed heads with fixed
terms, and some, including the Census Bureau, have presidentially
appointed heads that lack fixed terms and serve at the pleasure of the
president. Presidential appointment without a fixed term can be
detrimental to the independence of a statistical agency because the
agency head has political visibility but no guarantee against
politically motivated pressure and even dismissal. A recently released
report of a panel of the Committee, Ensuring the Quality, Credibility,
and Relevance of U.S. Justice Statistics, documents the firing of the
director of the Bureau of Justice Statistics in 2005 because the
director refused to alter a statistical press release to suit the
policy views of departmental officials. This situation should never
occur for a statistical agency head, and the report recommends a fixed
term for the director of BJS. A fixed term of office for the Census
Bureau director would also strengthen the independence and reputation
for objectivity of this critically important statistical agency.
the american community survey: major innovation, challenging to use
Since its inception in 1972, panels of the Committee on National
Statistics have produced over 30 interim, letter, and final reports on
the decennial census and related programs, including several reports on
the American Community Survey (ACS). The ACS, which went into full
production beginning in 2005, represents a seismic shift in the
landscape of small-area data on the U.S. population. This shift
promises important benefits to data users in terms of much more timely,
up-to-date, and higher quality information than the sample-based
questions on the decennial census could ever provide. (Some of the
census questions were first asked of a sample of the population in
1940; beginning in 1960, the sample questions were included on a
separate ``long form.'')
The benefits of the ACS can already be seen from the much more
frequent articles in the media about important population changes in
counties and cities--such as changes in the country of origin of
immigrants in some areas--that formerly could only be identified at 10-
year intervals. However, as the comprehensive review of the ACS in the
2007 report on Using the American Community Survey: Benefits and
Challenges indicates, the ACS' continuous design will initially
challenge many small-area data users in federal, state, and local
government agencies, researchers, the private sector, the media, and
the public. These users were accustomed to the point-in-time estimates
from the census long-form sample and must learn how to work with and
interpret the 1-year, 3-year, and 5-year moving-average estimates from
the ACS. Moreover, this learning process is still to occur for users of
data for the smallest geographic areas given that the first 5-year
estimates for areas with fewer than 20,000 people will not be released
until late 2010, representing averages of data collected in 2005-2009.
The ACS without doubt is of great benefit for users of estimates
for large geographic areas, such as states and large cities and
counties. Not only are reliable 1-year average estimates produced every
year from the ACS for such areas, but the 5-year estimates will enable
users to compare estimates for user-defined areas within, say, a major
city, by aggregating the estimates for census tracts and block groups.
It is also undeniable that the sample size of the ACS is at present too
small to provide precise estimates, even averaged over 5 years, for
small counties, cities, and towns. For example, based on the
calculations in our 2007 report, the 5-year average estimate of the
poverty rate does not meet acceptable standards of precision until an
area has about 10,000 people, and the 5-year average estimate of the
poverty rate for school-age children does not reach acceptable
standards of precision until an area has about 50,000 people. We
understand that, historically, the sample size for the ACS represented
a compromise between the size required for precise estimates for small
areas and the budget that was deemed acceptable to the executive and
Congress at the time the ACS was being designed. Additional funding
will be required to increase the sample size sufficiently for precise
small-area estimates.
Our 2007 report strongly supported the ACS but noted that the
transition for users and the Census Bureau would be challenging. The
continuous design of the ACS, in which data are collected, every month,
is essential for a smooth field operation, but it does pose problems
for users of interpreting estimates that are averages over 12, 36, or
60 months. It can also make it difficult to introduce new and revised
questions to meet changing needs. The report urged support for the ACS.
It recommended that the Census Bureau make sufficient funding of the
ACS one of its top priorities and that the Bureau seek funding, not
only for data collection and production, but also for ongoing programs
of methodological research and evaluation and user outreach and
education. Strong research and user education programs are essential
for the ACS to fulfill its mission to provide relevant, useful, and
accurate small-area information and to improve the survey in the future
as experience is gained with its benefits and challenges.
I thank the Joint Economic Committee for this opportunity to
testify and will be happy to respond to any questions the members may
have.
______
Prepared Statement of Andrew Reamer
Chairwoman Maloney, Vice Chairman Schumer, Congressman Brady,
Senator Brownback, and members of the Joint Economic Committee, I am
pleased to speak to you today about the role of the Census Bureau in a
21st century federal statistical system.
Census Bureau data are essential to the effective functioning of
our nation's democracy, public policy at all levels of government, and
our $14 trillion economy. For example, congressional apportionment and
redistricting; federal macroeconomic and regional economic development
policies; the annual distribution of a half trillion dollars in federal
funds; the enforcement of the Voting Rights Act and the Fair Housing
Act; state road-building and emergency planning; the placement of
public schools and community health centers; and business startup,
location, and investment decisions all rely on Census Bureau
statistics. At the most fundamental level, the nation could not operate
without this agency.
Relative to the enormity of the political and economic impacts, the
size of the Census Bureau operation is very small. Outside the
Decennial Census, Census Bureau operations cost in the range of $500
million annually; averaged over a decade, the cost of the 2010 Census
operation is about $1.5 billion a year. The return to the nation on
this investment in the Census Bureau is nearly infinite.
However, the Census Bureau is not yet a 21st century statistical
agency. While the bureau has made substantial, innovative advances in
improving the value of its data offerings, I believe these offerings
need to more fully reflect three new realities:
major changes in the nation's economic structure,
the potential for Census Bureau data to enable more
informed, effective nonfederal public and private decision-making
across the nation, and
significant opportunities for new data products and
techniques afforded by large-scale advances in information technology.
changing economic structure
In the last half-century, the nation's economic structure has
undergone dramatic changes, from one based on manufacturing, large
corporations, physical labor, and little international trade to one
reliant on services, entrepreneurship, knowledge workers, and global
markets.
The Census Bureau's statistical programs need to more fully capture
the essential components of our 21st century economic structure. While
the Census Bureau has highly regarded, long-standing, frequent data
collections for manufacturing activities (consistent with the nation's
economic base in the 1950s and 1960s), it does not yet have the same
level of data collection efforts for the service industries that now
dominate the nation's economy.
For a number of years, the Census Bureau has sought, but was unable
until FY2009 to obtain, the $8 million needed to survey the finance,
insurance, and real estate industries on an annual and quarterly basis.
The lack of these data has meant that the Bureau of Economic Analysis'
quarterly estimates of Gross Domestic Product have not been as accurate
as they might have been, particularly in times of major economic
reversal. As a result, macroeconomic policymakers at the Federal
Reserve, the Treasury Department, the Office of Management and Budget,
and the Council of Economic Advisers have been making determinations on
the basis of other than the best possible data.
In a similar vein, the Census Bureau has not had a regularly
collected survey of residential finance. Typically following the
decennial census, the last Residential Finance Survey (RFS) was carried
out in 2001; the Bush Administration discontinued planning for a 2011
RFS due to budget constraints. Even if the RFS were revived, a once-a-
decade assessment of the workings of the nation's residential finance
markets is far too infrequent in light of the impact of those markets
on the national economy and, in particular, their role in catalyzing
the current recession. It is essential that policymakers and analysts
have a current, accurate picture of the structure and flows of housing
finance markets. Efforts are underway to address this data gap,
including proposed $3 million funding for the Department of Housing and
Urban Development (just approved by the House Appropriations Committee)
to reimburse the Census Bureau to conduct a biennial multi-family
residential finance survey.
Due to recent FY2009 and likely FY2010 appropriations, the Census
Bureau's economic statistics are poised to become more reflective of
our 21st century economic structure and markets. However, as economic
structure and markets are never static, strong and stable future
funding is needed to ensure that the Census Bureau's data products
fully capture current realities and so meet policymaker needs.
more informed, effective decision-making
The advent of Web-based data access has allowed the Census Bureau's
customer base to expand exponentially. State and local governments and
millions of private businesses, from Wal-Mart to home-based
entrepreneurs, can far more quickly and easily incorporate census data
into their analyses and decision-making processes than was so just 15
years ago. The potential exists, then, for federal statistical agencies
in general, and the Census Bureau in particular, to enable
significantly improved public and private decision-making regarding the
allocation of trillions of dollars--generating very substantial
economic benefits at minimal taxpayer cost.
Nonfederal data users are primarily interested in current, reliable
demographic and economic data on states, metro areas, counties, cities
and places, and neighborhoods. In this regard, the Census Bureau is
about to take a major step forward with its upcoming first-time
publication of annually updated small area American Community Survey
(ACS) data. Data users will have access to current five-year averages
for areas as small as census tracts and block groups, replacing the
traditional once-a-decade, nearly always outdated, long-form decennial
data.
However, the Census Bureau is facing a significant issue in that
the ACS sample size, fixed at 3 million households annually, is
increasingly too small to provide reliable small area estimates. To
approach the accuracy of the decennial long form data, the ACS needs to
survey at least 3 percent of households annually (about 3.5 million
households at the moment). For the nation to obtain the full benefit of
the ACS, I strongly suggest that the Census Bureau request, and the OMB
and Congress approve, funds to support an annual 3 percent sample.
Web tools that integrate small area data from the Census Bureau and
nonfederal sources, such as state and local governments, greatly
increase the capacity for improved data analysis and decisionmaking.
For instance, census data populate an ever-growing number of national
and local community indicator websites that provide a detailed picture
of the socioeconomic conditions of local areas, down to neighborhoods.
To facilitate this important use, the Census Bureau should explore
means of providing direct, current data feeds to data intermediaries
through a distributed data system. The bureau's innovative DataFerrett
tool could be the foundation of such an effort.
The Census Bureau should take steps to better meet the needs of
federal, state, and local economic development organizations for
detailed, current subnational statistics. Historically, the Census
Bureau has viewed the nation's macroeconomic policymakers as the
primary customers for its economic statistics. This orientation
developed at a time, in the late 1940s and 1950s, when national
economic policy was concerned primarily with managing the economic
cycle through fiscal and monetary policy. The nation's economic
geography was thought to be highly stable--Detroit would always be the
car-making center and Pittsburgh would always make steel. International
competitiveness was not an issue.
Since 1980, however, due to globalization, increased capital
mobility, and faster technological change, the nation's economic
geography has been in a continual state of flux. Over the last three
decades, many once-stable regions have experienced economic shocks; as
experience makes clear, no region can take its economic base for
granted.
Today, our nation's economic health is very much a function of the
international competitiveness of its regional economies. Consequently,
federal, state, and local policymakers need a full understanding of the
economic structure of, change in, and flows among the nation's regions
and the world. State and regional economic development organizations
are actively preparing and implementing economic adjustment strategies;
the value of such strategies depends on good data that reflect
economic, not political, boundaries.
Hence, detailed, accurate economic data on metropolitan areas,
which provide the large majority of the nation's GDP, are quite
important. Based on existing data collections, for example, the Census
Bureau could publish much needed metropolitan-level data on research
and development; innovation; foreign trade in goods; place-to-place
domestic migration; and business starts, expansions, reductions, and
closures, by industry.
Historically, the Census Bureau has been relatively insulated from
the broad array of its data users. In light of the substantial
potential for improved public and private decision-making, I encourage
the Census Bureau to more actively seek to develop relationships with
representatives of a wide array of users in order to ascertain how it
can best meet their needs. Experience suggests that trade and
professional associations of important data users (e.g., National
Association of Counties, Council for Community and Economic Research,
National Association for Business Economics, National Retail
Federation, Association of Public Data Users) would be useful channels
for this purpose.
technically innovative data products
Advances in computer hardware and software are allowing all
statistical agencies to explore and develop new data products and
methods, to the nation's benefit. The Census Bureau has been in the
forefront of this innovative activity, including:
Very large administrative datasets that replace the need
for more expensive surveys. Under its traditional confidentiality
strictures, the Census Bureau operates the Statistical Administrative
Records System (StARS) that maintains a variety of federal, state, and
private databases.
Dynamic data that describe how firms and people move
through the economy over time and space, giving us a ``look under the
hood'' of the economy. The Census Bureau has a number of data programs
that create dynamic data: on hires and fires and where people live in
relation to where they work (Local Employment Dynamics [LED] Program);
on firm change over time, by age and firm size (Business Dynamics
Statistics); and on establishment births, deaths, expansions, and
contractions (Statistics of U.S. Businesses). Dynamic data have the
potential to be a very powerful new tool for federal and regional
economic policy.
Synthetic microdata that allow the analysis of individual
records without betraying confidentiality--greatly expanding the
potential for understanding the patterns of local economic activity,
with positive implications for public policy. IT advances have resulted
in traditional public use microdata sets being more vulnerable to
possible breeches of confidentiality. One means to address this problem
is by creating synthetic microdata that generate true statistics (for
example, mean, median, and frequency distribution). The Census Bureau's
LED Program has been in the forefront of this area of work, using
synthetic data to map where people live in relation to where they
work--data useful for economic and workforce development,
transportation planning, and emergency planning.
Modeled estimates that reduce the need for large surveys.
By working with administrative records and existing surveys, the Census
Bureau has been able to estimate income, poverty, and insurance
coverage for small areas, enabling more accurate distributions of
federal funds, among other uses.
As these innovative efforts are inexpensive and have substantial
benefits, the Census Bureau has been eager to pursue them. However,
despite the low cost and high potential, the bureau has had difficulty
in obtaining the funds needed to fully exploit the possibilities.
A case in point is the LED program, which works with business and
wage records from state unemployment insurance systems. LED has been in
existence for over a decade, but with a congressional appropriation of
only $2 million; to keep the program afloat, the Census Bureau has had
to draw another $6.5 million from discretionary funds and reimbursable
work, primarily for the Department of Labor. Consequently, LED has been
limited in its geographic coverage and policy impact.
However, full funding for LED may be on the horizon. In its FY2010
budget, the Administration requested $13.7 million to expand and
stabilize the program. As part of the proposal, LED would provide
nationwide coverage, giving federal macroeconomic policymakers a
valuable new tool to assess economic dynamics. Further, the bureau
proposes to add new functions, such as a job-to-job flows tool that
would allow analysts to track the industry, geographic location, and
wages of a group of workers over time. With this tool, for example, LED
would be able to determine the current employment situation of workers
who were in a regional industry that recently experienced substantial
restructuring (such as the Manhattan financial industry, the Detroit
auto industry, or the southern California residential construction
industry). Such a tool would have great value for federal and state
workforce policy. The House and the Senate Appropriations Committee
have approved a Census Bureau budget that includes the requested
funding for LED. I encourage the full Senate to approve this funding as
well.
conclusion
As valuable as it is today, the Census Bureau has the potential to
transform how the nation conducts its work, at little additional cost
to the taxpayer. For the Census Bureau to fully become a 21st century
statistical agency, several conditions need to be met. The Census
Bureau must understand and effectively respond to the data product
needs of its diverse customer base. In doing so, it should take
complete advantage of opportunities offered by cutting-edge information
technologies. And it must have the support of the Commerce Department,
OMB, and Congress to obtain the stable funding necessary to sustain the
programs that meet those needs. As the one congressional committee with
an overview of the nation's economic statistical system, the Joint
Economic Committee can play a valuable role in ensuring that these
steps are taken so that the bureau can achieve this potential.
I thank the committee for your attention and welcome your
questions.
______
Prepared Statement of Dr. Linda A. Jacobsen
Good afternoon, Chairwoman Maloney and members of the Committee.
Thank you for the opportunity to testify today before the Joint
Economic Committee. I will share my perspective on the benefits and
challenges of using the Census Bureau's American Community Survey.
Without a doubt, the American Community Survey (ACS) is
fundamentally changing the way we collect and use data to assess the
nation's population and housing. While the traditional census long form
collected detailed socioeconomic data just once a decade, the ACS is a
continuous survey that provides updated demographic, economic, and
housing data every year. As the pace of change has accelerated in the
U.S., so has the need for timely and reliable data. The ACS has
replaced the census long form to meet that need.
The ACS is already providing substantial benefits to federal
agencies, nongovernmental organizations, and state and local
governments. For example, the Department of Veteran Affairs uses ACS
data to evaluate the need for educational, employment, and health care
programs for veterans, while the Council on Virginia's Future relies on
ACS data to monitor annual trends in the travel time to work. At PRB,
we have used ACS data to track changes in the number, location, and
well-being of children in immigrant families. \1\ We have also used the
ACS to produce a database and wall chart on the U.S. labor force,
including state and metropolitan area estimates of people working in
high-tech and other science and engineering jobs.
---------------------------------------------------------------------------
\1\ See the report at http://www.prb.org/Publications/
ReportsOnAmerica/2009/childreninimmigrantfamilies.aspx.
---------------------------------------------------------------------------
ACS data are also contributing to planning for the 2010 Census. ACS
data from 2005, 2006, and 2007 were used to validate and enhance
population segmentation for the Census 2010 Integrated Communications
Campaign. \2\ The Census Bureau is also using 2005-2007 ACS data on
language spoken at home and English-language ability to select census
blocks that will receive a Census 2010 bilingual English and Spanish
form.
---------------------------------------------------------------------------
\2\ See the report at http://2010.census.gov/2010census/pdf/
C2POMemoNo9.pdf.
---------------------------------------------------------------------------
There are some important differences between the census long form
and the ACS that are essential to understand in evaluating the benefits
and trade-offs in the switch to the ACS. \3\ Foremost is the fact that
the sample size of the current ACS is much smaller than the sample size
of the 2000 Census long form. As a result, ACS data from multiple years
must be combined to provide reliable estimates for geographic areas
with smaller population sizes. The ACS provides 1-year estimates for
areas with populations of at least 65,000, 3-year estimates for areas
with populations between 20,000 and 65,000, and 5-year estimates for
areas with less than 20,000 people. This last group includes small
counties, cities, and towns as well as census tracts and block groups.
The ACS was fully implemented nationwide in 2005, so the first 3-year
estimates for 2005-2007 were released last December. The first 5-year
estimates for 2005-2009 are scheduled for release in 2010.
---------------------------------------------------------------------------
\3\ These are described more fully in the ACS Handbook available at
http://www.census.gov/acs/Downloads/ACSGeneralHandbook.pdf.
---------------------------------------------------------------------------
In 2000, the long form was sent to approximately 18 million
addresses, resulting in 16.4 million final interviews. This represented
about 1 in every 6 households, the same share of households that
received the long form in 1990. In contrast, the ACS is sent to about 3
million addresses each year, resulting in about 2 million final
interviews. When combined over five years, then, the ACS will only be
sent to 15 million addresses, resulting in about 10.5 million final
interviews. This represents only about 1 in every 9 households. Of
course, the number of households in the U.S. continues to increase
every year. Between 2000 and 2007, the number of households increased
by 7 million. While the decennial long form sampled the same proportion
of households in 1990 and 2000, the ACS samples the same number of
households each year. Maintaining a fixed sample size over time
necessarily means that ACS data will be collected from a smaller share
of U.S. households each year.
As a result of the smaller sample size, estimates from the ACS also
have higher levels of sampling variability than estimates from the 2000
Census long form. This means the ACS estimates are less precise or less
reliable, particularly for small geographic areas and population
subgroups. Several evaluation studies have reported that combining 5
years of ACS data did not provide reliable estimates for census tracts
in some counties. \4\ In a recent PRB analysis of 2005-2007 data for 26
states, we found that one-fifth (20 percent) of counties with a
population of 20,000 or more did not have a reliable estimate of the
share of working families that are below 200% of the poverty level.
Although this clearly constitutes a smaller population subgroup, there
were still more than 9.5 million such families nationwide in 2007, and
they are an important group for policy considerations.
---------------------------------------------------------------------------
\4\ For example, see the reports and presentations at http://
www.census.gov/acs/www/AdvMeth/Multi_Year_Estimates/presentations.html.
---------------------------------------------------------------------------
To achieve the objective of fully replacing the long form, the ACS
must provide a comparable scope of reliable data for smaller geographic
areas, including census tracts. The current sample size of the ACS is
the result of funding constraints. Based on their experience to date, a
growing share of data users are calling for an increase in the sample
size of the ACS to improve the reliability of estimates for smaller
geographic areas and subgroups. The ACS has tremendous potential to
provide the timely, detailed data critical for evidenced-based policy
and program design and implementation. Additional funding could
significantly increase the likelihood the ACS will realize this
potential.
Thank you for this opportunity to testify and I will be happy to
answer any questions the Committee may have.