Congressional Record, Volume 143 Issue 133 (Tuesday, September 30, 1997)

[Congressional Record Volume 143, Number 133 (Tuesday, September 30, 1997)]
[House]
[Pages H8216-H8229]
From the Congressional Record Online through the Government Publishing Office [www.gpo.gov]

DEPARTMENTS OF COMMERCE, JUSTICE, AND STATE, THE JUDICIARY, AND RELATED
AGENCIES APPROPRIATIONS ACT, 1998

The SPEAKER pro tempore. Pursuant to House Resolution 239 and rule
XXIII, the Chair declares the House in the Committee of the Whole House
on the State of the Union for the further consideration of the bill,
H.R. 2267.

{time} 1755

In the Committee of the Whole

Accordingly, the House resolved itself into the Committee of the
Whole House on the State of the Union for the further consideration of
the bill (H.R. 2267) making appropriations for the Departments of
Commerce, Justice, and State, the Judiciary, and related agencies for
the fiscal year ending September 30, 1998, with Mr. Hastings of
Washington in the chair.
The Clerk read the title of the bill.
The CHAIRMAN. When the Committee of the Whole House rose on Friday,
September 26, 1997, amendment No. 16 by the gentleman from Georgia [Mr.
Barr] had been disposed of and section 616 was open to further
amendments.
Are there further amendments to this section of the bill?
Mr. ROGERS. Mr. Chairman, I move to strike the last word to discuss
the evening schedule.
Mr. Chairman, the first order of business on the consideration of
this bill is the matter dealing with the census. Under the unanimous-
consent agreement of last week, debate time on this amendment was
limited to 80 minutes.
On this side of the aisle, I do not anticipate any extraneous
motions, in which case, if the other side could agree to that, we could
have 80 minutes where Members would be able to attend

[[Page H8217]]

to other business while the debate on this matter proceeds.
I wonder if the gentleman from West Virginia [Mr. Mollohan] would
like to discuss that. If so, I will yield.
Mr. OBEY. Mr. Chairman, would the gentleman from Kentucky [Mr.
Rogers] renew his motion? We could not hear it.
Mr. ROGERS. I did not have a motion. What I had attempted to do was
to try to explain to the Members that the first order of business now
is the consideration of the census matter, which under the unanimous
consent of last week, the debate time is limited to 80 minutes.
If there are no extraneous motions intervening during that period of
time on either side, Members can feel free to attend to other business
during that period of time without fear of a vote.

{time} 1800

I think I can assure the body that there will not be such motions on
this side, and if we can have that assurance from that side, Members
could have 80 minutes.
Mr. OBEY. Mr. Chairman, will the gentleman yield?
Mr. ROGERS. I yield to the gentleman from Wisconsin.
Mr. OBEY. With all due respect, Mr. Chairman, I cannot give that
assurance on this side because I intend to make one of the motions
myself.

Amendment Offered by Mr. Mollohan

Mr. MOLLOHAN. Mr. Chairman, I offer an amendment.
The CHAIRMAN. The Clerk will designate the amendment.
The text of the amendment is as follows:

Part II amendment printed in House Report 105-264 offered
by Mr. Mollohan:
In the first paragraph under ``DEPARTMENT OF COMMERCE--
Bureau of the Census--periodic censuses and programs'' strike
``Subject to the limitations provided in section 209, for''
and insert ``For''.
Strike section 209 and insert the following:
Sec. 209. None of the funds made available in this Act for
fiscal year 1998 may be used by the Department of Commerce to
make irreversible plans or preparations for the use of
sampling or any other statistical method (including any
statistical adjustment) in taking the 2000 decennial census
of population for purposes of the apportionment of
Representatives in Congress among the States.
Sec. 210. (a) There shall be established a board to be
known as the Board of Observers for a Fair and Accurate
Census (hereinafter in this section referred to as the
``Board'').
(b)(1) The function of the Board shall be to observe and
monitor all aspects of the preparation and implementation of
the 2000 decennial census (including all dress rehearsals) to
determine whether the process has been manipulated in any way
so as to bias the results in favor of any geographic region,
population group, or political party, or on any other basis.
(2) In carrying out such function, the Board shall give
special attention to the design and implementation of any
sampling techniques and any statistical adjustments used in
determining the population for purposes of the apportionment
of Representatives in Congress among the several States.
(3) The Board shall promptly report to the Congress and the
President evidence of any manipulation referred to in
paragraph (1).
(c)(1) The Board shall be composed of 3 members as follows:
(A) 1 individual appointed by the President.
(B) 1 individual appointed jointly by the Speaker of the
House of Representatives and the President pro tempore of the
Senate.
(C) The Comptroller General of the United States.

The members appointed under subparagraphs (A) and (B),
respectively, shall be former Presidents or others of similar
stature.
(2) Members shall not be entitled to any pay by reason of
their service on the Board, but shall receive travel
expenses, including per diem in lieu of subsistence, in
accordance with sections 5702 and 5703 of title 5, United
States Code.
(d)(1) The Board shall have an Executive Director who shall
be appointed by the Board and paid at a rate not to exceed
level IV of the Executive Schedule.
(2) The Board may appoint and fix the pay of such
additional personnel as it considers appropriate, subject to
the provisions of chapter 51 and subchapter III of chapter 53
of title 5, United States Code.
(3) Subject to such rules as may be prescribed by the
Board, the Board may procure temporary and intermittent
services under section 3109(b) of such title 5, but at rates
for individuals not to exceed the daily equivalent of the
maximum annual rate of pay payable for grade GS-15 of the
General Schedule.
(4)(A) Upon request of the Board, any personnel of an
agency under subparagraph (B) may be detailed to the Board,
on a reimbursable basis or otherwise, to assist the Board in
carrying out its duties.
(B) The agencies under this subparagraph are the General
Accounting Office, the Congressional Research Service, and
the Congressional Budget Office.
(e)(1) Notwithstanding any provision of title 13, United
States Code, or any other provision of law, members of the
Board and any members of the staff who may be designated by
the Board under this paragraph shall be granted access to any
data, files, information, or other matters maintained by the
Bureau of the Census (or received by it in the course of
conducting a decennial census of population) which they may
request, subject to such regulations as the Board may
prescribe in consultation with the Secretary of Commerce.
(2) The regulations shall include provisions under which
individuals gaining access to any information or other matter
pursuant to paragraph (1) shall be subject to sections 9 and
214 of title 13, United States Code.
(f) The Board shall transmit to the Congress and the
President--
(1) interim reports, as least semiannually, with the first
such report due by August 1, 1998; and
(2) a final report not later than August 1, 2001.

The final report shall contain a detailed statement of the
findings and conclusions of the Board with respect to the
matters described in subsection (b), together with any
recommendations regarding future decennial censuses of
population.
(g) Of the amounts appropriated to the Bureau of the Census
for each of fiscal years 1998 through 2001, $2,000,000 shall
be available to the Board to carry out this section.
(h) To the extent practicable, members of the Board shall
work to promote the most accurate and complete census
possible by using their positions to publicize the need for
full and timely responses to census questionnaires.
(i) The Board shall cease to exist on September 30, 2001.

The CHAIRMAN. Pursuant to House Resolution 239, the gentleman from
West Virginia [Mr. Mollohan] and a Member opposed will each control 40
minutes.
Who seeks time in opposition?
Mr. HASTERT. I do, Mr. Chairman.
The CHAIRMAN. The gentleman from Illinois [Mr. Hastert] will control
40 minutes.
The gentleman from West Virginia [Mr. Mollohan] is recognized for 40
minutes.
Mr. MOLLOHAN. Mr. Chairman, I yield myself such time as I may
consume.
Mr. Chairman, I rise to offer an amendment to the bill.
I would first like to thank the distinguished chairman of the
Committee on Rules, the gentleman from New York [Mr. Solomon], and the
distinguished ranking member, the gentleman from Massachusetts [Mr.
Moakley], for making the Mollohan-Shays amendment in order. It was the
fair thing to do.
Mr. Chairman, this is a bipartisan amendment offered jointly with my
colleague from Connecticut [Mr. Shays]. I want to take this opportunity
to thank him and the many other Members on both sides of the aisle,
especially the gentleman from Ohio [Mr. Tom Sawyer] and the gentlewoman
from New York [Mrs. Carol Maloney], who have worked so hard in support
of this amendment.
Mr. Chairman, the Constitution requires that we take a census of the
entire population of the United States every 10 years. That means we
count everyone, rich people, poor people, rural, urban, all races.
We are increasingly having a problem doing this count accurately. The
error rate skyrocketed in 1990 to include 26 million people with an
undercount of 1.6 percent of the population, and if we do not do
something, Mr. Chairman, it is estimated that in 2000 the undercount
will continue to climb. That is a lot of men, women, and children that
will be left out of our Nation's family, just left out, Mr. Chairman, a
lot from the inner city, a lot of the very rural, a lot of poor folks
just left out of the count.
We can do something about this by building on sampling methods which
have been a part of the census for the last 50 years. The Census Bureau
wants to employ sampling, not only in this Democratic administration,
but going back to President Bush's administration when Barbara Bryant,
Republican appointed director of the 1990 census, started working to
increase the use of sampling in the census. She says now, Mr. Chairman:
``I am very much in favor of the plan the Census Bureau has. It builds
work that I started on back in 1990.''
Well, these plans and recommendation are good. It is also good that
this bill contains $381 million to plan and

[[Page H8218]]

run tests next spring for what could be the most accurate census in our
Nation's history.
But there is a very bad provision in this bill, the Hastert
substitute which calls for a constitutional review of sampling, and
during that review, this provision kills sampling by prohibiting the
Census Bureau from spending any money on sampling planning. If the
Census Bureau cannot spend money planning for sampling, then we cannot
use sampling in the 2000 census; it is just that simple.
Now, Mr. Chairman, the amendment the gentleman from Connecticut [Mr.
Shays] and I offer removes the Hastert prohibitions and replaces them
with the most reasonable language contained in the Senate-passed bill
which lets the Census Bureau test scientific sampling methods so long
as they are not irreversible. And our amendment goes one step further.
We propose to create a board of advisors for a fair and accurate
census. This body would be made up of three individuals, one appointed
by the President, one jointly appointed by the Speaker and the
President pro tem of the Senate, and third, the Comptroller General.
The first two appointments shall be former Presidents or men and women
of similar stature. The main purpose of the board would be to observe
and monitor all aspects of the preparation and the implementation of
the 2000 census to assure the process is not in any way manipulated.
Mr. Chairman, those who object to sampling use three main arguments
which I think can be soundly refuted. In their first arguments,
opponents of sampling cite the Constitution. They assert that the
Constitution requires an actual head count of the population. However,
separate opinions issued by the Department of Justice under Presidents
Carter, Bush, and Clinton, bipartisan in nature, all concluded that the
Constitution permits the use of sampling and statistical methods as a
part of the census.
Stuart M. Gerson, assistant attorney general, Civil Division, in the
Bush administration, concluded in a July 1991 memorandum to the
Commerce Department's attorney general that the meaning of the term
``enumeration of the Constitution'' is, quote, more likely found in the
accuracy of census-taking than in the selection of any particular
method. Continuing, he says, nothing indicates any additional intent on
the part of the Framers to restrict for any time, for all time, the
manner in which the census is conducted, end of quote.
Additionally, on this issue of constitutionality of sampling, Mr.
Chairman, Federal courts have uniformly upheld the use of sampling. For
example, in the City of New York v. Department of Commerce, a 1990
case, the court concluded that, quote, because article 1, clause 2,
requires the census to be as accurate as practicable, the Constitution
is not, is not, a bar to statistical adjustment.
In their second argument, Mr. Chairman, opponents of sampling say
that it is bad science. Quite the opposite. The experts and
statisticians disagree. After the 1990 census, the Congress asked,
because of the bad count, the Congress asked the National Academy of
Sciences what could be done to make sure that every person in our
country is counted in the 2000 census, unlike the 1990 census. And the
National Academy of Sciences recommended sampling, a greater use of
sophisticated sampling techniques.
Further, the National Research Council, the American Statistical
Association, and the General Accounting Office all have endorsed the
use of sampling, the increased use of sampling, in the census.
Barbara Bryant, again, census director under none other than
President Bush, had the following to say in a recent letter to Speaker
Newt Gingrich:
In the long run, our Nation is best served by accuracy. Sample
surveys to estimate those who will not or cannot be counted in the 2000
census after the Census Bureau has made every reasonable and good-faith
effort to voluntarily enumerate them will increase the accuracy of the
census.
Mr. Chairman, in their third argument, opponents of sampling say that
the Commerce Department will politicize the results of the census.
While I do not in any way share this view, its nature makes it
impossible to refute through fact or expert opinion. It can only be
refuted through a guarantee of careful oversight, and that is precisely
what the Mollohan-Shays amendment does with the board of advisers for a
fair and accurate census; it assures oversight.
Mr. Chairman, having refuted the three most used arguments against
sampling, only one remains: Fear, the fear that using sampling will
affect the political makeup of the House of Representatives. The real
manipulation going on today is the Republicans' majority attempt to
control funding to prevent the Census Bureau from using the one
technique all the experts say will yield the most accurate census. And
why are they doing this? By their words, it is, they indicate, that it
is because they are afraid of what will happen if every person in this
country is counted, afraid they may lose seats in the Congress. I do
not agree with that view. It is a false fear.
But in any event, let me remind my colleagues that the purpose of the
census is to count the people of our Nation, not to ensure that any
political party controls the Congress. We should strive toward accuracy
and let the political chips fall where they may. To quote the recent
commentary in a Business Week magazine, Census 2000, Math, Not
Politics, Please, end of quote.
Mr. Chairman, I would like to close by reaching out to my Republican
colleagues, perhaps some from States that had a large undercount in the
1990 census. We cannot pass this amendment without them. Join us in
fashioning a census where we count all women, all men, and all
children, where we do not leave out four or five or six million inner
city, rural, and poor folks. Let us take advantage of this historic
opportunity in a bipartisan way to have the best census ever.
Vote for the Mollohan-Shays amendment.
Following are excerpts from decisions of several Federal courts which
have considered the issue of the constitutionality and legality of use
of sampling and statistical adjustment in the census, and from legal
memoranda by senior Justice Department officials from both Republican
and Democratic administrations.

United States Court of Appeals for the Sixth Circuit:
``Although the Constitution prohibits subterfuge in
adjustment of census figures for purposes of redistricting,
it does not constrain adjustment of census figures if
thoroughly documented and applied in a systematic manner.''

Young v. Klutznik, 652 F.2d 617, 625 (6th Cir. 1981)
____

United States District Court for the Eastern District of
New York: ``This Court concludes that because Article I,
section 2 requires the census to be as accurate as
practicable, the Constitution is not a bar to statistical
adjustment.''

City of New York v. U.S. Dept. of Commerce, 739 F.Supp. 761,
767 (E.D.N.Y. 1990)
____

United States District Court for the Southern District of
New York: ``It appears to the Court that this language [in
the Constitution] indicates an intent that apportionment be
based on a census that most accurately reflects the true
population of each state.''
____

``Consequently, the Court finds defendants' constitutional
and statutory objections concerning the impropriety of
employing statistical adjustments to compensate for the
undercount without merit.''

Carey v. Klutznik, 508 F.Supp. 404, 415 (S.D.N.Y. 1980)
____

United States District Court for the Eastern District of
Michigan: ``It is unthinkable to suggest, that, when the
allocation of federal resources and the apportionment of
Congressional Representatives rest upon an accurate census
count, and when the Census Bureau itself knows that there is
an undercount, which heavily disfavors Blacks and minorities,
and when a method can be found to correct that undercount,
that the words `actual enumeration' in the Constitution
prevent an adjustment to obtain a more accurate figure than
the actual headcount.''

Young v. Klutznik, 497 F.Supp. 1318, 1333 (E.D. Mich 1980)
____

United States District Court for the Eastern District of
Pennsylvania: ``It may be that today an actual headcount
cannot hope to be an accurate reflection of either the size
or distribution of the Nation's population. If so, it is
inconceivable that the Constitution would require the
continued use of a headcount in counting the population.
Therefore, the Court holds that the Constitution permits the
Congress to direct or permit the use of statistical
adjustment factors in arriving at the final census results
used in reapportionment.''

City of Philadelphia v. Klutznick, 503 F.Supp. 663, 679
(E.D.Pa. 1980) (emphasis in original)
____

United States Court of Appeals for the Second Circuit:
``Reading sections 141 and 195 [of

[[Page H8219]]

the Census Act] together in light of their legislative
history, we conclude that Congress intended the Secretary (a)
to conduct an actual enumeration as part of the decennial
census, and (b) in lieu of a `total' enumeration to use
sampling and special surveys `whenever possible'.
Accordingly, we conclude that a statistical adjustment to the
initial enumeration is not barred by the Census Act and
indeed was meant to be encouraged.''

City of New York v. U.S. Department of Commerce, 34 F.3d
1114, 1125 (2d Cir 1994) (citations omitted)
____

Stuart Gerson, Assistant Attorney General (Civil Division)
in the Bush Administration (Legal Opinion for Commerce Dept.,
July 9, 1991): ``Though the conclusion is not entirely free
from doubt, it does appear the Constitution would permit a
statistical adjustment if it would contribute to an accurate
population count.''
____

Stuart Gerson, Assistant Attorney General (Civil Division)
in the Bush Administration, (Legal Opinion for Commerce
Dept., July 9, 1991): ``By directing the conduct of an
`actual Enumeration' for use in subsequent congressional
apportionments, the Framers replaced the `conjectural ratio'
used in the initial apportionment, with a more permanent and
precise standard. Nothing in the constitutional debates or
any other historical records, insofar as we are aware,
indicates any additional intent on the part of the framers to
restrict for all time--except by constitutional amendment--
the manner in which the census is conducted. Rather, the
thrust of the `actual Enumeration' language appears to be
simply that the decennial census should represent an accurate
counting of the population `in such manner as [the Congress]
shall by Law direct'.''

* * * * *

``In sum, the essence of enumeration, as the term is both
generally and constitutionally understood, is more likely
found in the accuracy of census taking rather than in the
selection of any particular method, i.e., a headcount.''
____

Walter Dellinger, Assistant Attorney General in the Clinton
Administration (Memorandum for the Solicitor General, Oct. 7,
1994): ``Accordingly, we conclude that the Constitution does
not preclude the [Census] Bureau from employing technically
and administratively feasible adjustment techniques to
correct undercounting in the next decennial census.''
____

Walter Dellinger, Assistant Attorney General in the Clinton
Administration (Memorandum for the Solicitor General, Oct. 7,
1994): ``These discussions [at the constitutional convention]
make clear that, in requiring an `actual' enumeration, the
Framers meant a set of figures that was not a matter of
conjecture and compromise, such as the figures they had
themselves provisionally assumed. An `actual' enumeration
would instead be based, as George Mason put it, on `some
permanent and precise standard'. There is no indication that
the Framers insisted that Congress adopt a `headcount' as the
sole method for carrying out the enumeration, even if later
refinements in the metric of populations would produce more
accurate measures.''
____

John M. Harmon, Asst. Attorney General (Office of Legal
Counsel) in the Carter Administration, (Memorandum dated
Sept. 25, 1980): ``In sum, the position that the Constitution
prohibits any statistical adjustment is not supportable--not
as a matter of semantics, Framers' intent, or Supreme Court
case law.''
______

The American Statistical Association Report of the Census Blue Ribbon
Panel

executive summary

In order to improve the accuracy and to constrain the costs
of the Decennial Census for the year 2000 the Census Bureau
is planning to make increased use of scientific sampling when
conducting the Census. Critics have questioned the Bureau's
intent to make greater use of sampling. Their criticism may
be based upon a misunderstanding of the scientific basis of
the Census Bureau's sampling plans. The President of the
American Statistical Association appointed this panel and
charged it with considering this aspect of the Bureau's plans
and the criticisms of them. In our statement, we point out
that sampling is an integral part of the scientific
discipline of statistics and explain how its use can be an
appropriate part of the methodology for conducting censuses.
Congress directed the Bureau of the Census to develop plans
for the 2000 Decennial Census that (1) reduce the undercount,
particularly the differential in the undercount across
population groups, and (2) constrain the growth of costs.
Because sampling potentially can increase the accuracy of the
count while reducing costs, the Census Bureau has responded
to the Congressional mandate by investigating the increased
use of sampling. An additional benefit of sampling is that
its appropriate use can also reduce the response burden on
the population. We endorse the use of sampling for these
purposes; it is consistent with best statistical practice.

background

The Bureau of the Census is planning to improve coverage
and constrain the costs of the Decennial Census for the year
2000 by making greater use of scientific sampling. Sampling
is not new to the Census; it has been used for decades in
compiling the Census. The Census Bureau has employed sampling
to monitor and improve the quality of interviewers' work, to
reduce respondent burden by asking some questions of only a
sample of households, to estimate the number of vacant
housing units, and to evaluate the completeness of the
Census's coverage of the population. In addition, for the
year 2000, the Census Bureau's plans include sampling
households that do not respond to the mail questionnaire and
are not reached in initial interviewer follow-up. This is a
procedure known as sampling for ``non-response follow-up.''
The Census Bureau also plans to use sampling to account for
the remaining small percentage of households that cannot be
counted in the enumeration. This procedure is referred to as
``integrated coverage measurement.'' This increased use of
sampling has been criticized; however, we believe the critics
may have misunderstood the scientific basis of the Census
Bureau's sampling plans.
Plans for the 2000 Census have been developed in response
to a dual Congressional mandate to the Bureau. First, the
Census Bureau is charged with improving the population count
by reducing the undercount (which increased from 1.2% of the
population in 1980 to 1.8% of the population in 1990) and, in
particular, with reducing or eliminating the differentially
higher undercount of some groups, such as Africa-Americans
and Hispanics. Second, the Census Bureau is charged with
constraining the cost of the 2000 Census (census costs
escalated sharply between 1970 and 1990, even after allowing
for inflation and population growth). In carrying out this
dual mandate from the Congress, the Census Bureau has
considered a variety of procedural and technical improvements
to the 2000 Census and has developed plans to use sampling
for non-response follow-up and for integrated coverage
measurement. The Bureau has also created and consulted with a
number of advisory groups and has sought the advice of
several National Academy of Science panels.
As the Decennial Census draws nearer, Congress has been
monitoring the Bureau's planning process more closely. The
Bureau's proposed additional uses of sampling have created
some controversy within Congress. Several recent actions, as
well as proposed legislation, would affect the Bureau's
ability to use sampling in the 2000 Census.
Two bills have been introduced in Congress that would
restrict the role of sampling in the 2000 Census. One bill,
HR3558, sponsored by Congresswoman Carrie Meek (D-Florida),
states that ``the Bureau shall attempt to contact every
household directly (whether by mail or in person), and may
use sampling as a substitute for direct contact in a
particular census tract only after direct contact has been
made with at least 90 percent of the households in such
tract.'' This bill reflects concern about the Census
Bureau's proposed plan to begin the use of sampling for
non-response follow-up when 90 percent of the households
have been enumerated in each county (counties are usually
larger and more diverse geographic areas than are census
tracts). The other bill, HR3589, sponsored by Congressman
Thomas Petri (R-Wisconsin), states that Title 13 of the
U.S. Code shall be amended to add the following: ``In no
event may sampling or other statistical procedures be used
in determining the total population by states . . . for
purposes of the apportionment of Representatives in
Congress among the several States.'' This bill would
prohibit the use of any sampling to determine population
counts used for congressional apportionment. This
effectively prevents the use of sampling for any purpose
other than collection of demographic or economic data
through the ``long form.''
In June, the House Committee on Government Reform and
Oversight prepared a report that recommended against sampling
in the Census either to complete the field work or to correct
the undercount. The committee has not yet considered or voted
on the report. In early August, the Senate Committee on
Appropriations approved a report to accompany the Fiscal Year
1997 Commerce Department funding bill that would prohibit the
Census Bureau from preparing to use sampling in the Decennial
Census. The full Senate is expected to consider the bill in
September.
This statement has been composed by a panel appointed by
the President of the American Statistical Association to
consider the Census Bureau's plans to increase the use of
sampling in the conduct of the next Census. The purpose of
this statement is to point out that sampling is an integral
part of the scientific discipline of statistics and to
explain briefly how its use can be an appropriate part of the
methodology for conducting censuses.

statement

Uses of and the Scientific Basis for Sampling
Sampling is used widely in science, medicine, government,
agriculture, and business because it is the fundamental basis
for addressing specific questions in these arenas. Sampling
is a critical tool for reducing uncertainty; it is possible
to draw conclusions from a scientific sample of empirical
observations with specific levels of confidence in our
conclusions. Statistics, a branch of applied mathematics, is
a rigorous discipline based upon centuries of development of
the principles of probability and the empirical study of
their applications. The use of sampling combined with the
mathematics of

[[Page H8220]]

probability provide the basis for drawing scientific
inferences from observations. Without this basis, confirming
or rejecting scientific theories would be impossible.
Specific areas that use statistical sampling extensively
include auditing, market research, quality assurance,
approving new drugs, and medical testing. For example,
physicians use a sample of blood drawn from a patient to draw
conclusions about all the blood in the patient's body. A full
census of a patient's blood is not possible, and a small
sample is fully adequate to measure the concentration of a
specific chemical in the patient's blood system. Sampling
permits observations to be made efficiently, economically,
and fairly. Without sampling, we would not have quality
control in our industries, soil testing in agriculture, or
most of the national statistics on which the nation depends.
Well-designed samples are used to draw accurate conclusions
in many applications. The specific design of a sample in a
particular setting depends on the particular problem being
addressed. In complex situations such as the census, the
detailed sample designs require careful analysis by people
skilled and experienced in census taking.
Using Sampling to Improve the Population Count
The appropriate use of sampling can improve the count of a
population. The basic idea underlying this conclusion is that
some parts of the population will be easier to count and some
more difficult. After an effort has been made to reach all
households, some number of households will not have been
reached; little is known about these households. Well-
designed sampling to obtain information about them can reduce
what would otherwise be a differential undercount between the
easier to count and harder to count groups in the population.
The attachment to this statement briefly explains the
underlying logic of how sampling can improve population
counts and also reduce costs.
In fact, every census is, in some sense, a sample, since
everyone cannot be reached. Some countries, more
authoritarian than ours, have ordered all people to remain in
their homes all day on Census Day until the police or the
army have come to count them. In democratic countries,
however, everyone cannot be reached and counted. Those who
have been counted amount to a sample of the total population,
but this is not a sample based on probability theory because
the reasons for missing information in the census are not
understood. A probability based sample design, as planned
by the Census Bureau, permits inferences to be drawn about
the entire population with a specified level of
confidence. The discipline of statistics largely focuses
on reducing uncertainty through the use of sampling and
other statistical techniques that permit inferences to be
drawn about those missing in a sample. Thus, scientific
probability sampling is broadly applicable to census
taking.
In addition, sampling can reduce the burden on respondents
to the census. Just as it is not necessary to impose on the
medical patient the burden of withdrawing all the blood to
measure the platelet count, it is not necessary to count
every household and every person in the country in order to
draw conclusions about the country. Careful design and
execution of probability sampling can permit samples to
generate data and precise inferences in which we can have
considerable confidence. Indeed, the ability to employ
sampling is perhaps the single most important element in the
government's effort to reduce the burden it imposes on the
population from which it collects statistics.
Conclusion
Congress directed the Bureau of the Census to develop plans
for the 2000 Decennial Census that (1) reduce the undercount
particularly the differential in the undercount across
population groups, and (2) constrain the growth of costs.
Because sampling has the potential to increase the quality
and accuracy of the count and reduce costs, the Census Bureau
has responded to the Congressional mandate by investigating
the increased use of sampling. An additional benefit of
sampling is that its appropriate use can also reduce the
response burden on the population. The use of sampling for
these purposes is consistent with sound statistical practice.
______

Congressional Research Service,

Library of Congress,

Washington, DC, September 29, 1997.

To: Honorable Carolyn B. Maloney, Attention: David McMillen
From: American Law Division
Subject: Questions re Legislative Provision for Expedited
Judicial Review of Use of sampling and statistical
Adjustment in Year 2000 Census

This memorandum is in response to your request for our
consideration of four questions dealing with the
implementation and likely impact of language added to H.R.
2267, the Commerce, Justice, State, and Judiciary
Appropriations Bill. By the terms of the Rule granted the
bill by the Committee on Rules, H. Res. 239; H. Rept. 105-
264, the provision, set out in the cited report, was adopted
upon the adoption of the Rule.
Briefly stated, the provision Sec. 209 of H.R. 2267,
authorizes ``[a]ny person aggrieved'' by the use of a
statistical method of determining population in connection
with the year 2000, or later, census, to bring a civil action
for declaratory, injunctive, and other appropriate relief
against the use of the method on the ground that it is
contrary to the Constitution or statute. The definition of an
``aggrieved person'' for purposes of the section is stated to
be any resident of a State whose congressional representation
or district ``could'' be changed by the use of a statistical
method, any Representative or Senator, or either House of
Congress. The action authorized is to be heard and determined
by a three-judge district court, pursuant to 28 U.S.C.
Sec. 2284. Expedited appeal direct to the Supreme Court of
any decision by the district court is provided for under
specified deadlines for filing.
A significant provision, subsection (b), states that ``the
use of any statistical method in a dress rehearsal or similar
test or simulation of a census in preparation for the use of
such method, in a decennial census, to determine the
population for purposes of the apportionment or redistricting
of members in Congress shall be considered the use of such
method in connection with that census.''
Under subsection (d)(2), no appropriated funds may be used
for any statistical method, in connection with the decennial
census, once a judicial action is filed, until it has been
judicially determined that the method is authorized by the
Constitution and by act of Congress.
Three of your questions relate to the likelihood of a
Supreme Court decision, using the expedited procedure, either
by the time of the beginning of the 1998 census dress
rehearsal (approximately March 15, 1998) or prior to the
census in 2000. Inasmuch as the date of the decision in any
such case depends substantially on the filing date of the
suit, and the beginning of the running of any period of
expedition, we cannot even guess whether a Supreme Court
decision would be likely before either event. Certainly, the
date of the start of the dress rehearsal, if it is March 15,
1998, is less than six months from now, much less from the
time of enactment of the provision, if it is enacted, and
from the time a statistical method is tested, if that is
sufficient to confer standing. Thus, we can be confident that
a decision by March 15, 1998, is highly unlikely. A decision
by the beginning of the start of the 2000 census is certainly
possible, if a suit may be filed early enough. However, as we
indicate below, it is doubtful that anyone would have
standing by then, even in light of the section, to bring an
action.
We can indicate, from the time line of past cases,
especially those where Congress has provided especially for
judicial review and expedited consideration, that the courts
are enabled to proceed promptly and in less time than with
respect to the ordinary case. For example, the most recent
case was handled very expeditiously. Raines v. Byrd, 117
S.Ct. 2312 (1997). Congress in 1996 enacted the Line-Item
Veto Act, which went into effect on January 1, 1997. The
following day, six Members of Congress filed suit. The
District Court handed down its decision on April 10, 1997.
Pursuant to the statute's authorization, an appeal was filed
in the Supreme Court on April 18, the Court granted review on
April 23, and, even though the argument period for the Term
had run, special oral argument was entertained on May 27, and
the decision by the Supreme Court was rendered on June 26.
Thus, the time from filing in the District Court to the
issuance of a decision by the Supreme Court was less than
seven months, although we must observe that the decision was
based on the lack of standing by the Members, perhaps a less
difficult issue than the question on the merits. Nonetheless,
the time frame was significant.
Other cases could be cited. For example, in Bowsher v.
Synar, 478 U.S. 714 (1986), testing the constitutionality of
certain features of the Gramm-Rudman-Hollings law, the
Balanced Budget and Emergency Deficit Control Act of 1985,
the courts moved promptly, again acting within a
congressionally-enacted provision for expedited judicial
review. The President signed the bill into law on December
12, 1985, and suit was filed the same day. A three-judge
district court was impaneled, and a decision was issued on
February 7, 1986. An appeal was filed in the Supreme Court on
February 18, review was granted on February 24, oral argument
was held on April 23, and the Court's decision was issued on
July 7.
The time line was thus about seven months.
One may assume, therefore, that a suit, properly brought,
challenging the use of some form of statistical adjustment,
could be processed within a relatively brief time, perhaps
within seven months and perhaps within a briefer period.
However, that assumption is of little importance, because the
substantial question, the hard issue, turns on what party has
standing to bring such a suit; that is, when is a suit
``properly brought''?
That the use of statistical methods, of samplings and
adjustments, is not a frivolous question is evident. The
argument is whether the Constitution in requiring an ``actual
Enumeration,'' Art. I, Sec. 2, cl. 3, mandates an actual
counting or permits some kind of statistical analysis to
enhance the count; the further argument is whether Congress,
in delegating to the Secretary of Commerce its authority to
conduct the census ``in such Manner as [it] shall by Law
direct,'' has by instructing him to take ``a decennial census
of the population . . . in such form and content as he may
determine . . .'', 13 U.S.C. Sec. 141(a), supplied him with
sufficient authority to supplement or to supplant the

[[Page H8221]]

actual count through statistical methods. The Supreme Court
has reserved decision on both issues. Wisconsin v. City of
New York, 116 S.Ct. 1091, 1101 nn. 9, 11 (1996).
Courts have entertained suits arising out of these and
similar issues. E.g., Wisconsin v. City of New York, supra;
Franklin v. Massachusetts, 505 U.S. 738 (1992); Dept. of
Commerce v. Montana, 503 U.S. 442 (1992). However, all three
cases arose after the actual conduct of or official decision
about a particular action that resulted in actual injury to a
State or to a political subdivision. These cases, and earlier
decisions in the lower courts concerning the 1990 and 1980
censuses, certainly stand for the proposition that polities
have standing to sue to contest actions that have already
occurred and that have injured them. They do little to
advance the inquiry required by Sec. 209.
All citizens, of course, have an interest that the
Constitution be observed and followed, that laws be enacted
properly based on and permitted by the Constitution, and that
laws be correctly administered. However, this general
interest, shared by all, is insufficient to confer standing
on persons as citizens or as taxpayers. Schlesinger v.
Reservists Com. to Stop the War, 418 U.S. 208 (1974); United
States v. Richardson, 418 U.S. 166 (1974). See also Valley
Forge Christian College v. Americans United, 454 U.S. 464,
483 (1982); Allen v. Wright, 468 U.S. 737, 754 (1984); Lujan
v. Defenders of Wildlife, 504 U.S. 555, 560 (1992). Congress
may not overturn this barrier to suit in federal court by
devising a test law suit. E.g., Muskrat v. United States, 219
U.S. 346 (1911) (striking down a statute authorizing certain
named Indians to bring a test suit against the United States
to determine the validity of a law affecting the allocation
of Indian lands, in which the attorneys' fees of both sides
were to be paid out of tribal funds, deposited in the
Treasury).
Standing is one element of the justiciability standard,
which limits Article III federal courts to the decision only
of cases that properly belong within the role allocated to
federal courts. ``[A]t an irreducible minimum,'' the
constitutional requisites under Article III for the existence
of standing are that the party seeking to sue must personally
have suffered some actual or threatened injury that can
fairly be traced to the challenged action of the defendant
and that the injury is likely to be redressed by a
favorable decision. E.g., Allen v. Wright, 468 U.S., 751;
Lujan v. Defenders of Wildlife, supra, 504 U.S., 560;
Raines v. Byrd, 117 S.Ct., 2317-18. ``We have always
insisted on strict compliance with this jurisdictional
standing requirement.'' Id., 2317.
The first element, injury in fact, is a particularly
stringent requirement. ``[T]he plaintiff must have suffered
an `injury in fact'--an invasion of a legally protected
interest which is (a) concrete and particularized, . . . and
(b) actual or imminent, not conjectural or hypothetical.
``Lujan v. Defenders of Wildlife, 504 U.S., 560 (internal
quotation marks omitted). As the latter part of the element
indicates, a party need not await the consummation of the
injury in order to be able to sue. However, as the decisions
combining parts of standing and of Article III ripeness show,
pre-enforcement challenges to criminal and regulatory
legislation will be permitted if the plaintiff can show a
realistic danger of sustaining an injury to his rights as a
result of the governmental action impending; a reasonable
certainty of the occurrence of the perceived threat to a
constitutional interest is sufficient to afford a basis for
bringing a challenge, provided the court has before it
sufficient facts to enable it to intelligently adjudicate the
issues. Buckley v. Valeo, 424 U.S. 1, 113-18 (1976); Duke
Power Co. v. Carolina Environmental Study Group, 438 U.S. 59,
81-2 (1978); Babbitt v. Farm Workers, 442 U.S. 238, 298
(1979); Regional Rail Reorganization Act Cases, 419 U.S. 102,
138-48 (1974). The Court requires, though, particularized
allegations that show a reasonable certainty, an actual
threat of injury. See Renne v. Geary, 501 U.S. 312 (1991);
Lujan v. Defenders of Wildlife, 504 U.S., 564-65 & n. 2.
Critically, in any event, the certainty of injury
requirement is a constitutional limitation, while the factual
adequacy element is a prudential limitation on judicial
review. Regional Rail Reorganization Act Cases, 419 U.S.,
138-48.
Congress is free to legislate away prudential restraints
upon the jurisdiction of the courts and to confer standing to
the utmost extent permitted by Article III. But, Congress may
not legislatively dispense with Article III's constitutional
requirement of a distinct and palpable injury to a party or,
if the injury has not yet occurred, a realistic danger of its
happening. Warth v. Seldin, 422 U.S. 490, 501 (1975); Raines
v. Byrd, 117 S.Ct., 2318 n. 3. Cf. United States v. SCRAP,
412 U.S. 669 (1973), disparaged in Whitmore v. Arkansas, 495
U.S. 149, 159 (1990), asserting that it ``surely went to the
outer limit of the law.'' The Court has firmly held that
Congress, in pursuit of judicial oversight over government
activity in areas of general public interest, areas that
would not support standing in the first instance, may not
enlarge the scope of judicial review by definitionally
expanding the meaning of standing under Article III. Lujan v.
Defenders of Wildlife, 504 U.S., 571-78. ``Whether the courts
were to act on their own, or at the invitation of Congress,
in ignoring the concrete injury requirement described in our
cases, they would be discarding a principle fundamental to
the separate and distinct constitutional role of the Third
Branch--one of the essential elements that identifies those
`Cases' and `Controversies' that are the business of the
courts rather than of the political branches.'' Id., 576.
``[Statutory] broadening [of] the categories of injury
that may be alleged in support of standing is a different
matter from abandoning the requirement that the party
seeking review must himself have suffered an injury.''
Id., 578 (quoting Sierra Club v. Morton, 405 U.S. 727, 738
(1972)).
Turning, then, to the proposed Sec. 209, we must observe
that the precedents strongly counsel that the conferral of
standing, especially in its definitional design of injury in
fact, would be inadequate to authorize judicial review until
the occurrence of the injury, the calculation of population
figures showing the gains and losses of seats in the House of
Representatives.
First, the conferral of standing in subsections (c)(2) and
(3) is likely ineffective. In Raines v. Byrd, supra, Congress
had included in the Line-Item Veto Act authorization for
``[a]ny Member of Congress'' to bring an action to contest
the constitutionality of the Act. The Court held that the
Members seeking to sue had suffered no personal,
individualized injury, only rather an assertion of an
institutional injury to their status as Members, that was
inadequate under Article III. Conceivably, Members
representing a State that lost one or more seats in the House
as a result of statistical re-evaluation of the census
enumeration could suffer the same injury that all residents
of the State incurred, but that injury would be confined as
we discuss below.
Second, while either the House of Representatives or the
Senate may have interests that could be injured by Executive
Branch action, giving either body or both bodies standing to
bring an action, what interest either House could assert in
the reallocation of seats in the House of Representatives is
unclear at best.
Third, Sec. 209(a) authorizes ``[a]ny person aggrieved by
the use of any statistical method . . . in connection with .
. . [a] census, to determine the population for purposes of
the apportionment or redistricting of members of Congress . .
.'' to bring a court action to challenge the
constitutionality of or the statutory basis of the
statistical method. Under Sec. 209(c)(1), an ``aggrieved
person'' is defined to include ``an resident of a State whose
congressional representation or district could be changed as
a result of the use of a statistical method.'' (Emphasis
supplied). By Sec. 209(b), it is provided that ``the use of
any statistical method in a dress rehearsal or similar test
or simulation of a census in preparation for the use of such
method . . . shall be considered the use of such method in
connection with that census.'' (Emphasis supplied). That is,
any person residing in a state that ``could'' lose House
representation as a result of a statistical adjustment of a
census may sue as soon as there is ``a dress rehearsal or
similar test or simulation of a census.''
The case law makes it clear that this authorization, if
enacted, would run afoul of constitutional barriers to
congressional conferral either of standing or of ripeness or
both.
Under Article III, for a litigant to have standing, he must
allege an injury in fact to himself or to an interest; if the
injury has not yet occurred, he must allege a strong basis
for fear that the injury will happen, that there is a real
danger of the injury being felt. The quoted provisions
purport to confer standing far beyond this constitutional
requirement.
To illustrate, when each census occurs, it is the
responsibility of the Bureau of the Census to calculate,
using what is called ``the method of equal proportions,'' 2
U.S.C. Sec. 2a(a), the number of seats, above the one each
State is constitutionally guaranteed, to be allocated to each
State, and the numbers are processed by the Department of
Commerce, which refers them to the President, who has the
responsibility to transmit them to Congress. See generally
Dept. of Commerce v. Montana, 503 U.S. 442 (1992); Franklin
v. Massachusetts, 505 U.S. 788, Wisconsin v. City of New
York, 116 S.Ct. 1091 (1996). The allocation is not final
until the President submits the figures to Congress. Franklin
v. Massachsuetts, 505 U.S. 796-801. It is then that the loss
of a seat or seats is legally final, and it seems clear that
the States losing seats have suffered a cognizable injury,
enabling them to bring suit to challenge at least certain
aspects of the conduct of the census. Id., 801-803.
Whether residents of a State that has lost one or more
seats in the House of Representatives have standing to bring
suit is questionable. Certainly, voters in a State in which
redistricting is not accomplished through the creation of
equally-populated districts have standing to complain about
the dilution of their voting strength. E.g., Wesberry v.
Sanders, 376 U.S. 1 (1964); Darcher v. Daggett, 462 U.S. 725
(1983). And a resident of a congressional district that has
been drawn impermissibly using race has standing to challenge
that districting. United States v. Hays, 515 U.S. 737 (1995).
But in the context of a State losing a House seat, every
resident of that State has a general interest that is shared
by all other residents. It is not a particularized injury in
fact that is what normally confers standing.
Let us, however, assume that residents would have standing.
The injury would not occur until the President transmits the
figures to Congress. Even if one could allege the imminent
likelihood of injury, a realistic danger of injury, that
development is only

[[Page H8222]]

going to mature when the census is completed and the
calculations are made awarding the correct number of seats to
each House. And we hear speak of a challenge to the actual
census.
The challenge, however, authorized by Sec. 209, is to the
use of a statistical method that ``could'' change the result
of the census enumeration. An injury in fact would not occur,
again, until the result is reported to Congress by the
President; an imminent injury in fact could conceivably occur
when the Census Bureau and the Commerce Department utilize a
statistical adjustment that changes the allocation of seats.
But that occurs after the tabulation of the census result and
the utilization of a statistical method that changes the
result of the census count itself.
The Supreme Court has never approved standing premised on
an allegation that a particular governmental action ``could''
cause an injury. Of course, the application of a statistical
method ``could'' work a change in the census, but to which
States and with what results would be extremely speculative
under the best of circumstances.
Moreover, the definition of the ``use of any statistical
method'' to include a test, or dress rehearsal, or simulation
of a census would confer standing that is even further
removed from the occurrence of the event that ``could'' or
``might'' result in an injury. It would be impossible to
point to any result of the conduct of a test or whatever that
might conceivably occasion the loss of one or more House
seats.
Because Congress lacks the power to create a definition of
standing or of the imminent likelihood of injury giving
standing that would infringe the constitutional requirement
of standing--of injury in fact or of the imminent likelihood
of injury--it appears extremely likely that the Supreme Court
would either strike down the provision, cf. City of Boerne v.
Flores, 117 S.Ct. 2157 (1997), or disregard it. Cf. Raines v.
Byrd, supra.
Finally, we must note Sec. 209(e) that purports to
authorize any executive branch agency or entity having
authority to carry out the census to bring a civil action to
obtain a declaratory judgment as to its constitutional and
statutory powers in this regard. It seems doubtful that this
authority could be exercised. It would likely fall under the
principle that no suit may be maintained unless there is
adversity between the plaintiffs and the defendants. See
Muskrat v. United States, 219 U.S. 346 (1911). What
government agencies have to do is to proceed on the basis of
their judgment about their powers, and then they will be
subject to suit challenging that judgment. This subsection
appears to do nothing less than to authorize an agency to
seek an advisory opinion.

Johnny H. Killian,

Senior Specialist,
American Constitutional Law.
Mr. MOLLOHAN. Mr. Chairman, I reserve the balance of my time.
Mr. HASTERT. Mr. Chairman, I yield myself as much time as I may
consume.
Mr. Chairman, I believe that every Member of this House can agree
that we need to conduct the census that includes all Americans and is
free of any partisan manipulation. There are those who say that this no
longer can be accomplished by actually counting Americans. They want to
restore the statistical methods in order to estimate or guess how many
people are in this country. They have thrown up their hands and said an
accurate census cannot be done by counting.
Mr. Chairman, it can be done, and in fact it has been done. Once
again Washington bureaucrats need to listen and learn from folks
outside the beltway.
In testimony before my subcommittee, communities like Milwaukee,
Wisconsin, Indianapolis, and Cincinnati describe how they conducted an
actual count at accuracy levels higher than those the Census Bureau
proposes to achieve with their risky statistical scheme. Census Bureau
Director Riche may not trust her ability to count, but Michael Morgan
in Milwaukee proved he knew how to do it.
Mr. Chairman, census sampling is a bad idea, but there is a more
fundamental question: Is it legal and constitutional to use sampling
and statistical adjustment to apportion this House among the States? I
believe it is clear that census sampling and statistical adjustments
are both illegal and unconstitutional. In that light, to blindly move
forward with a $5 million census that could well be thrown out by the
Supreme Court would be very foolish.

{time} 1815

Article I, section 2 of the Constitution states that actual
enumeration of the population be conducted every 10 years.
To enumerate means to count, one-by-one. It does not mean that we
should use sampling as a shortcut just because counting might be hard.
Nor does it mean that we should use statistical adjustment to
manipulate the count so that the results are more to someone else's
liking.
The 14th amendment to the Constitution States that ``Representatives
shall be apportioned among the several States according to their
respective numbers, counting the whole number of persons in each
State.'' The 14th amendment does not tell us to use statistics; it
tells us to count.
Title 13 of the United States Code, section 195, states that ``Except
for the determination of population for purposes of apportionment of
Representatives in Congress among the several States, the Secretary
shall, if he considers it feasible, authorize the use of the
statistical method known as sampling.''
Mr. Chairman, the statute is crystal clear. While allowing
statistical methods for nonconstitutionally required purposes, the 1957
statute explicitly maintained an absolute firewall against the use of
statistical methods for reapportionment. This was a wise, bipartisan
precaution designed to prevent the census from deteriorating into a
partisan power grab.
Mr. Chairman, the Congress reaffirmed this firewall once again in
1976 when it passed into law Title 13, section 141 of the United States
Code. This section allows the Secretary broad discretion in the use of
statistical methods for nonapportionment purposes. Let me repeat: for
nonapportionment purposes.
The supporters of census sampling would have us believe that section
141 allows that sampling be used for reapportionment. That is simply
not true. Congress specifically left intact the absolute prohibition on
their use of apportionment purposes established in section 195. If
Congress had intended that sampling be used for reapportionment, they
would have repealed section 195 at that time. They did not.
Mr. Chairman, the law is clear, and I believe that the Justices will
confirm that. The Justices know that actual enumeration means to count.
Listen to what Justice Scalia said during the last census case, and I
quote:

The text of the Constitution, as I read it, does not say
that there will be an estimate of the number of citizens. It
talks about actual enumeration. It doesn't even use the word
``census''. It says actual enumeration.

He added, and I quote,

Adjustment techniques ultimately involve kinds of value
choices and are therefore politically manipulable.

Mr. Chairman, the Justices also know that they will ultimately be
called on to rule on the legality of sampling. In the case that I just
mentioned the city of New York tried to force a statistical adjustment
of the census. The Supreme Court ruled that the Secretary of Commerce
could not be forced to do so. During the oral arguments, Justice Scalia
said that this case will decide whether you must use statistical
estimates and the next one will decide whether you may use it.
Mr. Chairman, the Supreme Court will answer that fundamental question
sooner or later. My language in this bill is designed to make it
sooner. My colleagues on the other side of the aisle should not be
afraid to let the Supreme Court rule. It is our duty as the people's
representatives to see their tax money is spent wisely, not wasted. The
wisest course for Congress today is to take the politics out of the
census and let the Supreme Court decide before billions of tax dollars
are wasted.
Mr. Chairman, the Mollohan-Shays amendment does not protect the
census from political mischief or the taxpayers from fiscal disaster.
The Mollohan-Shays amendment will leave taxpayers wide open to
multibillion dollar boondoggles. Protect the integrity of our census
and the tax dollars of hard-working Americans. Reject the Mollohan-
Shays amendment and allow the Supreme Court to rule.
Mr. Chairman, I reserve the balance of my time.
Mr. MOLLOHAN. Mr. Chairman, I yield such time as he may consume to
the gentleman from Illinois [Mr. Davis].
(Mr. DAVIS of Illinois asked and was given permission to revise and
extend his remarks.)
Mr. DAVIS. Mr. Chairman, I rise in support of the Mollohan-Shays
amendment.
Mr. Chairman, I rise today and join my colleagues in strong support
of the Mollahan-Shays amendment. This amendment is about

[[Page H8223]]

ensuring an accurate count of the 2000 census. The Constitution
requires an accurate count, not a headcount. This amendment would allow
the use of statistical sampling to conduct the 2000 census.
Since 1790, during the first census there was a significant
undercount especially among minorities. Two hundred years later in
1990, it is estimated that the census missed 10 percent of the
population. The Government Accounting Office estimates that as many as
26 million people were missed. Locally, in the State of Illinois, the
undercount was about .98 percent. In Cook County undercount was about
1.6 percent. The city of Chicago undercount was about 2.4 percent.
Furthermore, African-Americans were said to have anywhere from a 5-6
percent undercount; Latinos were about 5 percent; and Asian Pacific
Islanders were about a 3-percent undercount.
The statistics demonstrate that the poor and mainly racial minorities
are seriously missed. Africans-Americans are 7 times as likely to be
missed as Whites. That translates into being--7 times more likely to be
denied resources and representation in Congress, State legislatures,
city councils, county boards and other political subdivision. An
undercount among minorities furthers their deprivation to Federal money
while devaluing their political power. Billions of Federal dollars are
at stake. Governmental agencies often use census data to dole out money
or at least to determine targeted areas for distribution. There are
some 120 federally-funded programs that move approximately $150 billion
a year, which use the census data in their formulation for
distribution.

In 1990, children made up only one-fourth of the population but
accounted for 52 percent of the undercount. The children, the most
vulnerable people in our society have been denied representation and
valuable resources because of this significant undercount.
This amendment simply seeks to ensure that each and every individual
is counted without regard to color, wealth, or status. This amendment
protects both the urban and rural dweller.
If the primary goals of the upcoming census are to reduce cost and to
eliminate the differential undercount, then let's take the politics out
of the census. The real issue is how to get the most accurate count and
the real answer is sampling.
Statistical sampling and estimation techniques have been proposed as
a means to finish the undercount for the 10 percent that are the
hardest to reach--the hardest to find, the left out, the hopeless and
helpless, traditionally minorities and the poor. This is not the first
time that sampling has been used in the census. This approach has also
been endorsed by expert panels of the National Academy of Sciences, the
American Statistical Association, the Commerce Department's Inspector
General, the GAO and various other professional organizations.

As a matter of fact, three separate panels convened by the National
Academy of Sciences have recommended that the Census Bureau use
sampling in the 2000 census to save money and improve accuracy. The
commerce IG has said that sampling and statistical methods are the only
way to eliminate the historic, disproportionate undercount of people of
color and the poor.
Ten percent of the count in 1990 was wrong. The Census Bureau will
make an unprecedented effort to count all Americans directly. Sampling
is scientific, not guessing.
Conducting the most accurate census must be the goal for the 2000
census, that goal cannot be met without the use of sampling. We owe it
to ourselves and we owe it to the American people.
Therfore, I urge my colleagues to join me in support of this
amendment that would allow for the use of statistical sampling.
Mr. MOLLOHAN. Mr. Chairman, I yield such time as he may consume to
the gentleman from Missouri [Mr. Clay].
(Mr. CLAY asked and was given permission to revise and extend his
remarks.)
Mr. CLAY. Mr. Chairman, I thank the gentleman for yielding, and I
rise in support of the Mollohan-Shays amendment.
Mr. Chairman, no one honestly or seriously disputes that the 1990
census undercounted the population. Nor does anyone honestly or
seriously dispute the fact that minority populations, blacks and
Hispanics especially, as well as rural residents and children were
disproportionately undercounted.
Though my colleagues on the other side of the aisle will try to
confuse the issue, there is no debate at all within the scientific
community that the use of statistical sampling would improve the
accuracy of the census.
So what is this debate about? Some have contended that statistical
sampling may be a means by which the census would be intentionally
distorted. The sponsors of this amendment have dealt with that concern
by crafting an amendment that, among other things, provides assurances
that sampling will be conducted in a scientific, non-partisan manner.
So what are the real concerns? Well, Mr. Chairman, it is blatantly
obvious to me that those who oppose sampling fear that their own
political power would be threatened by an accurate census. And, rather
than contest for political power out in the open, they prefer a system
that denies millions of Americans the representation they are due under
our Constitution.
In the end, what this debate is about is whether we reject the view
that some people may as well be invisible and whether we will abide by
the principle of one man-one vote. I urge my colleagues to support the
Mollohan-Shays amendment.
Mr. MOLLOHAN. Mr. Chairman, I yield 2 minutes to the gentlewoman from
New York [Mrs. Maloney], who is the ranking minority member on the
Subcommittee on Government Management, Information, and Technology of
the Committee on Government Reform and Oversight.
Mrs. MALONEY of New York. Mr. Chairman, sending the census sampling
issue before the Supreme Court certainly sounds like a righteous
compromise, but beware of a wolf in sheep's clothing. The Supreme Court
will decide in favor of sampling, but while we are waiting as long as a
year, the stalling will kill sampling for the 2000 census. Indecision
will become the decision. Missing the Census Bureau deadlines for as
long as a year means certain death for a fair and accurate census.
There has been a great deal of misinformation that has been bandied
about, and I would like to set the record straight on the Census
Bureau's plan. What the Census Bureau plans to do will be the largest
peacetime mobilization ever. Ninety percent of the people will be
counted using traditional methods. People will be contacted four times
through the mail. They will be contacted by phone for the first time.
Community outreach will include forms that are in post offices, stores,
churches, malls, and TV ads are in the works.
Then the Bureau will begin to knock on doors, but we know that many
of these doors will remain shut because people do not open their doors
to strangers, they are not there, they are at work. And only for the
last 10 percent, for those people who could not be reached by mail,
phone, a knock on the door, or through the media, only for that last 10
percent will statistical sampling be used.
Mr. Chairman, we know that some people are more likely to be missed
than others. They are our Nation's poor, our Nation's minorities. They
are the people who most need to be heard and who are most often
silenced. The use of sampling is the civil rights issue of the 1990's.
There are hundreds of professional organizations, community groups,
editorial boards across the country, experts, who all endorse sampling.
The Mollohan-Shays amendment will give people the simple right to the
representation that they deserve.
I urge my colleagues to do what is right for all of their
constituents. Make sure they can count on us not to count them out in
the year 2000 census. Vote for the Mollohan-Shays bipartisan amendment.
Mr. Chairman, I include for the Record data from the Congressional
Research Service in support of my position. The CRS report says that
the Hastert amendment will just block forward-going of an accurate
census.
Congressional Research Service,

The Library of Congress,

Washington, DC, September 29, 1997.
To: Honorable Carolyn B. Maloney, Attention: David McMillen.
From: American Law Division.
Subject: Questions re Legislative Provision for Expedited
Judicial Review of Use of sampling and statistical
Adjustment in Year 2000 Census.
This memorandum is in response to your request for
consideration of four questions dealing with the
implementation and likely impact of language added to H.R.
2267, the Commerce, Justice, State, and Judiciary
Appropriations Bill. By the terms of the Rule granted the
bill by the Committee on Rules, H. Res. 239; H. Rept. 105-
264, the provision, set out in the cited report, was adopted
upon the adoption of the Rule.
Briefly stated, the provision Sec. 209 of H.R. 2267,
authorizes ``[a]ny person aggrieved'' by the use of a
statistical method of determining population in connection
with the year 2000, or later, census, to bring a civil action
for declaratory, injunctive, and other appropriate relief
against the use of the method on

[[Page H8224]]

the ground that it is contrary to the Constitution or
statute. The definition of an ``aggrieved person'' for
purposes of the section is stated to be any resident of a
State whose congressional representation or district
``could'' be changed by the use of a statistical method, any
Representative or Senator, or either House of Congress. The
action authorized is to be heard and determined by a three-
judge district court, pursuant to 28 U.S.C. Sec. 2284.
Expedited appeal direct to the Supreme Court of any decision
by the district court is provided for under specified
deadlines for filing.
A significant provision, subsection (b), states that ``the
use of any statistical method in a dress rehearsal or similar
test or simulation of a census in preparation for the use of
such method, in a decennial census, to determine the
population for purposes of the apportionment or redistricting
of members in Congress shall be considered the use of such
method in connection with that census.''
Under subsection (d)(2), no appropriated funds may be used
for any statistical method, in connection with the decennial
census, once a judicial action is filed, until it has been
judicially determined that the method is authorized by the
Constitution and by act of Congress.
Three of your questions relate to the likelihood of a
Supreme Court decision, using the expedited procedure, either
by the time of the beginning of the 1998 census dress
rehearsal (approximately March 15, 1998) or prior to the
census in 2000. Inasmuch as the date of the decision in any
such case depends substantially on the filing date of the
suit, and the beginning of the running of any period of
expedition, we cannot even guess whether a Supreme Court
decision would be likely before either event. Certainly, the
date of the start of the dress rehearsal, if it is March 15,
1998, is less than six months from now, much less from the
time of enactment of the provision, if it is enacted, and
from the time a statistical method is tested, if that is
sufficient to confer standing. Thus, we can be confident that
a decision by March 15, 1998, is highly unlikely. A decision
by the beginning of the start of the 2000 census is certainly
possible, if a suit may be filed early enough. However, as we
indicate below, it is doubtful that anyone would have
standing by then, even in light of the section, to bring an
action.
We can indicate, from the time line of past cases,
especially those where Congress has provided especially for
judicial review and expedited consideration, that the courts
are enabled to proceed promptly and in less time than with
respect to the ordinary case. For example, the most recent
case was handled very expeditiously. Raines v. Byrd, 117
S.Ct. 2312 (1997). Congress in 1996 enacted the Line-Item
Veto Act, which went into effect on January 1, 1997. The
following day, six Members of Congress filed suit. The
District Court handed down its decision on April 10, 1997.
Pursuant to the Statute's authorization, an appeal was filed
in the Supreme Court on April 18, the Court granted review on
April 23, and, even though the argument period for the Term
had run, special oral argument was entertained on May 27, and
the decision by the Supreme Court was rendered on June 28.
Thus, the time from filing in the District Court to the
issuance of a decision by the Supreme Court was less than
seven months, although we must observe that the decision was
based on the lack of standing by the Members, perhaps a less
difficult issue than the question on the merits. Nonetheless,
the time frame was significant.
Other cases could be cited. For example, in Bowsher v.
Synar, 478 U.S. 714 (1986), testing the constitutionality of
certain features of the Gramm-Rudman-Hollings law, the
Balanced Budget and Emergency Deficit Control Act of 1985,
the courts moved promptly, again acting within a
congressional-enacted provision for expended judicial review.
The President signed the bill into law on December 12,
1985,and suit was filed the same day. A three-judge district
court was impaneled, and a decision was issued on Feburary 7.
1986. An appeal was filed in the Supreme Court on February
18, review was granted on February 24, oral argument was held
on April 23, and the Court's decisions was issued on July 7.
The time line was thus about seven months.
One may assume, therefore, that a suit, properly brought,
challenging the use of some form of statistical adjustment,
could be processed within a relatively brief time, perhaps
within seven months and perhaps within a briefer period.
However, that assumption is of little importance, because the
substantial question, the hard issue, turns on what party has
standing to bring such a suit; that is, when is a suit
``properly brought''?
That the use of statistical methods, of samplings and
adjustments, is not a frivolous question is evident. The
argument is whether the Constitution in requiring an ``actual
Enumeration,'' Art. I, Sec. 2, cl. 3, mandates an actual
counting or permits some kind of statistical analysis to
enhance the count; the further argument is whether Congress,
in delegating to the Secretary of Commerce its authority to
conduct the census ``in such Manner as [it] shall by Law
direct,'' has by instructing him to take ``a decennial census
of the population . . . in such form and content as he may
determine . . .'', 13 U.S.C. Sec. 141(a), supplied him with
sufficient authority to supplement or to supplant the actual
count through statistical methods. The Supreme Court has
reserved decision on both issues. Wisconsin v. City of New
York, 116 S.Ct. 1091, 1101 nn. 9, 11 (1996).
Courts have entertained suits arising out of these and
similar issues, E.g., Wisconsin v. City of New York, supra;
Franklin v. Massachusetts, 505 U.S. 738 (1992); Dept. of
Commerce v. Montana, 503 U.S. 442 (1992). However, all three
cases arose after the actual conduct of or official decision
about a particular action that resulted in actual injury to a
State or to a political subdivision. These cases, and earlier
decisions in the lower courts concerning the 1990 and 1980
censuses, certainly stand for the proposition that polities
have standing to sue to contest actions that have already
occurred and that have injured them. They do little to
advance the inquiry required by Sec. 209.
All citizens, of course, have an interest that the
Constitution be observed and followed, that laws be enacted
properly based on and permitted by the Constitution, and that
laws be correctly administered. However, this general
interest, shared by all, is insufficient to confer standing
on persons as citizens or as taxpayers. Schlesinger v.
Reservists Com. to Stop the War, 418 U.S. 208 (1974); United
States v. Richardson, 418 U.S. (1974). See also Vallety Forge
Christian College v. Americans United, 454 U.S. 464, 483
(1982); Allen v. Wright, 468 U.S. 737, 754 (1984); Lujan v.
Defenders of Wildlife, 504 U.S. 555, 560 (1992). Congress may
not overturn this barrier to suit in federal court by
devising a test law suit. E.g., Muskrat v. United States, 219
U.S. 346 (1911) (striking down a statute authorizing certain
named Indians to bring a test suit against the United States
to determine the validity of a law affecting the allocation
of Indian lands, in which the attorneys' fees of both sides
were to be paid out of tribal funds, deposited in the
Treasury).
Standing is one element of the justiciability standard,
which limits Article III federal courts to the decision only
of cases that properly belong within the role allocated to
federal courts. ``[A]t an irreducible minimum,'' the
constitutional requisites under Article III for the existence
of standing are that the party seeking to sue must personally
have suffered some actual or threatened injury that can
fairly be traced to the challenged action of the defendant
and that the injury is likely to be redressed by a
favorable decision. E.g., Allen v. Wright, 468 U.S., 751;
Lujan v. Defenders of Wildlife, supra, 504 U.S., 560;
Raines v. Byrd, 117 S.Ct., 2317-18, ``We have always
insisted on strict compliance with this jurisdictional
standing requirement.'' Id., 2317.
The first element, injury in fact, is a particularly
stringent requirement. ``[T]he plaintiff must have suffered
an `injury in fact'--an invasion of a legally protected
interest which is (a) concrete and particularized, . . . and
(b) actual or imminent, not conjectural or hypothetical.''
Lujan v. Defenders of Wildlife, 504 U.S., 560 (internal
quotation marks omitted). As the latter part of the element
indicates, a party need not await the consummation of the
injury in order to be able to sue. However, as the decisions
combining parts of standing and of Article III ripeness show,
pre-enforcement challenges to criminal and regulatory
legislation will be permitted if the plaintiff can show a
realistic danger of sustaining an injury to his rights as a
result of the governmental action impending; a reasonable
certainty of the occurrence of the perceived threat to a
constitutional interest is sufficient to afford a basis for
bringing a challenge, provided the court has before it
sufficient facts to enable it to intelligently adjudicate the
issues, Buckley v. Valeo, 424 U.S. 1, 113-18 (1976); Duke
Power Co. v. Carolina Environmental Study Group, 438 U.S. 59,
81-2 (1978); Babbitt v. Farm Workers, 442 U.S. 289, 298
(1979); Regional Rail Reorganization Act Cases, 419 U.S. 102,
138-48 (1974). The Court requires, though, particularized
allegations that show a reasonable certainty, an actual
threat of injury. See Renne v. Geary, 501 U.S. 312 (1991);
Lujan v. Defenders of Wildlife, 504 U.S., 564-65 & n. 2.
Critically, in any event, the certainty of injury
requirement is a constitutional limitation, while the factual
adequacy element is a prudential limitation on judicial
review. Regional Rail Reorganization Act Cases, 419 U.S.,
138-48.
Congress is free to legislate away prudential restraints
upon the jurisdiction of the courts and to confer standing to
the utmost extent permitted by Article III. But, Congress may
not legislatively dispense with Article III's constitutional
requirement of a distinct and palpable injury to a party or,
if the injury has not yet occurred, a realistic danger of its
happening. Warth v. Seldin, 422 U.S. 490, 501 (1975); Raines
v. Byrd, 117 S.Ct., 2318 n. 3. Cf. United States v. SCRAP,
412 U.S. 669 (1973), disparaged in Whitmore v. Arkansas, 495
U.S. 149, 159 (1990), asserting that it ``surely went to the
outer limit of the law.'' The Court has firmly held that
Congress, in pursuit of judicial oversight over government
activity in areas of general public interest, areas that
would not support standing in the first instance, may not
enlarge the scope of judicial review by definitionally
expanding the meaning of standing under Article III. Lugan v.
Defenders of Wildlife, 504 U.S., 571-78, ``Whether the courts
were to act on their own, or at the invitation of Congress,
in ignoring the concrete injury requirement described in our
cases, they would be discarding a principle fundamental to
the separate and distinct constitutional role of the Third
Branch--one of the essential elements that identifies those
`Cases' and `Controversies' that are the business of the
courts rather than of the political

[[Page H8225]]

branches.'' Id., 576. `` `[Statutory] broadening [of] the
categories of injury that may be alleged in support of
standing is a different matter from abandoning the
requirement that the party seeking review must himself
have suffered an injury.'' Id., 578 (quoting Sierra Club
v. Morton, 405 U.S. 727, 738 (1972)).
Turning, then, to the proposed Sec. 209, we must observe
that the precedents strongly counsel that the conferral of
standing, especially in its definitional design of injury in
fact, would be inadequate to authorize judicial review until
the occurrence of the injury, the calculation of population
figures showing the gains and losses of seats in the House of
Representatives.
First, the conferral of standing in subsections (c)(2) and
(3) is likely ineffective. In Raines v. Byrd, supra, Congress
had included in the Line-Item Veto Act authorization for
``[a]ny Member of Congress'' to bring an action to contest
the constitutionality of the Act. The Court held that the
Members seeking to sue had suffered no personal,
individualized injury, only rather an assertion of an
institutional injury to this status as Members, that was
inadequate under Article III. Conceivably, Members
representing a State that lost one or more seats in the House
as a result of statistical re-evaluation of the census
enumeration could suffer the same injury that all residents
of the State incurred, but that injury would be confined as
we discuss below.
Second, while either the House of Representatives or the
Senate may have interests that could be injured by Executive
Branch action, giving either body or both bodies standing to
bring an action, what interest either House could assert in
the reallocation of seats in the House of Representatives is
unclear at best.
Third, Sec. 209(a) authorizes ``[a]ny person aggrieved by
the use of any statistical method . . . in connection with .
. . [a] census, to determine the population for purposes of
the apportionment or redistricting of members of Congress . .
.'' to bring a court action to challenge the
constitutionality of or the statutory basis of the
statistical method. Under Sec. 209(c)(1), an ``aggrieved
person'' is defined to include ``any resident of a State
whose congressional representative or district could be
changed as a result of the use of a statistical method.''
(Emphasis supplied). By Sec. 209(b), it is provided that
``the use of any statistical method in a dress rehearsal or
similar test or simulation of a census in preparation for the
use of such method . . . shall be considered the use of such
method in connection with that census.'' (Emphasis supplied).
That is, any person residing in a state that ``could'' lose
House representation as a result of a statistical adjustment
of a census may sue as soon as there is ``a dress rehearsal
or similar test or simulation of a census.''
The case law makes it clear that this authorization, if
enacted, would run afoul of constitutional barriers to
congressional conferral either of standing or of ripeness or
both.
Under Article III, for a litigant to have standing, he must
allege an injury in fact to himself or to an interest; if the
injury has not yet occurred, he must allege a strong basis
for fear that the injury will happen, that there is a real
danger of the injury being felt. The quoted provisions
purport to confer standing far beyond this constitutional
requirement.
To illustrate, when each census occurs, it is the
responsibility of the Bureau of the Census to calculate,
using what is called ``the method of equal proportions,'' 2
U.S.C. Sec. 2a(a), the number of seats, above the one each
State is constitutionally guaranteed, to be allocated to each
State, and the numbers are processed by the Department of
Commerce, which refers them to the President, who has the
responsibility to transmit them to Congress. See generally
Dept. of Commerce v. Montana, 503 U.S. 442 (1992); Franklin
v. Massachusetts, 505 U.S. 788, Wisconsin v. City of New
York, 116 S.Ct. 1091 (1996). The allocation is not final
until the President submits the figures to Congress. Franklin
v. Massachusetts, 505 U.S., 796-801. It is then that the loss
of a seat or seats is legally final, and it seems clear that
the States losing seats have suffered a cognizable injury,
enabling them to bring suit to challenge at least certain
aspects of the conduct of the census. Id., 801-803.
Whether residents of a State that has lost one or more
seats in the House of Representatives have standing to bring
suit is questionable. Certainly, voters in a State in which
redistricting is not accomplished through the creation of
equally-populated districts have standing to complain about
the dilution of their voting strength. E.g., Wesberry v.
Sanders, 376 U.S. 1 (1964); Darcher v. Daggett, 462 U.S. 725
(1983). And a resident of a congressional district that has
been drawn impermissibly using race has standing to challenge
that districting. United States v. Hays, 515 U.S. 737 (1995).
But in the context of a State losing a House seat, every
resident of that State has a general interest that is shared
by all other residents. It is not a particularized injury in
fact that is what normally confers standing.
Let us, however, assume that residents would have standing.
The injury would not occur until the President transmits the
figures to Congress. Even if one could allege the imminent
likelihood of injury, a realistic danger of injury, that
development is only going to mature when the census is
completed and the calculations are made awarding the correct
number of seats to each House. And we hear speak of a
challenge to the actual census.
The challenge, however, authorized by Sec. 209, is to the
use of a statistical method that ``could'' change the result
of the census enumeration. An injury in fact would not occur,
again, until the result is reported to Congress by the
President; an imminent injury in fact could conceivably occur
when the Census Bureau and the Commerce Department utilize a
statistical adjustment that changes the allocation of seats.
But that occurs after the tabulation of the census result and
the utilization of a statistical method that changes the
result of the census count itself.
The Supreme Court has never approved standing premised on
an allegation that a particular governmental action ``could''
cause an injury. Of course, the application of a statistical
method ``could'' work a change in the census, but to which
States and with what results would be extremely speculative
under the best of circumstances.
Moreover, the definition of the ``use of any statistical
method'' to include a test, or dress rehearsal, or simulation
of a census would confer standing that is even further
removed from the occurrence of the event that ``could'' or
``might'' result in an injury. It would be impossible to
point to any result of the conduct of a test or whatever that
might conceivably occasion the loss of one or more House
seats.
Because Congress lacks the power to create a definition of
standing or of the imminent likelihood of injury giving
standing that would infringe the constitutional requirement
of standing--of injury in fact or of the imminent likelihood
of injury--it appears extremely likely that the Supreme Court
would either strike down the provision, cf. City of Boerne v.
Flores, 117 S.Ct. 2157 (1997), or disregard it. Cf. Raines v.
Byrd, supra.
Finally, we must note Sec. 209(e) that purports to
authorize any executive branch agency or entity having
authority to carry out the census to bring a civil action to
obtain a declaratory judgment as to its constitutional and
statutory powers in this regard. It seems doubtful that this
authority could be exercised. It would likely fall under the
principle that no suit may be maintained unless there is
adversity between the plaintiffs and the defendants. See
Muskrat v. United States, 219 346 (1911). What government
agencies have to do is to proceed on the basis of their
judgment about their powers, and then they will be subject to
challenging that judgment. This subsection appears to do
nothing less than to authorize an agency to seek an advisory
opinion.

Johnny H. Killian,

Senior Specialist,
American Constitutional Law.
____

Congress of the United States,

Washington, DC, September 29, 1997

SUPPORT MOLLOHAN-SHAYS

CRS: Supreme Court Review Won't Happen

Dear Colleague: Last week the Rules Committee changed the
restrictive language on the census in the Commerce, Justice,
State Appropriations bill at the request of Rep. Hastert, to
ban the use of modern statistical methods pending a court
decision. Proponents of the Hastert language argue that they
have provided a compromise, but in reality this is just
another attempt to stop the census from counting everyone.
We have always believed that it is legal to use sampling in
the Census, based on Supreme Court decisions and opinions
from the Justice Department under three Presidents. Because
we take seriously concerns about partisan manipulation of the
census, we support the Mollohan-Shays Amendment setting up a
three-member bipartisan panel to oversee Census 2000. Mr.
Hastert instead proposed a court review. Today we received a
memorandum from the Congressional Research Service responding
to a request to analyze the Hastert language. In short, the
Hastert language will not result in a decision on the
constitutionality of sampling, it will only block the use of
appropriated funds.
The first issue is what lawyers call standing: whether
someone can sue over the use of sampling in the census. In
other words, has someone been injured by a government action,
and can thus use the courts to address that injury. The
Hastert language tries to get around this issue by declaring
in the bill who has standing to sue. Unfortunately, the
Constitution does not allow that. There is a Constitutional
test to determine who has standing in a case, and Congress
cannot bypass that requirement in a law. As CRS said, ``The
case law makes it clear that this authorization, if enacted,
would run afoul of constitutional barriers to Congressional
referral either of standing or of ripeness or both.''
Even if standing were not a constitutional problem for the
Hastert proposal, the Supreme Court has made it quite clear
that a challenge to the census must take place after the
numbers are final. As the CRS report says, ``[W]e must
observe that the precedents strongly counsel that the
conferral of standing, especially in its definitional design
of injury in fact, would be inadequate to authorize judicial
review until the occurrence of injury, the calculation of
population figures showing the gains and losses of seats in
the House of Representatives.''
The CRS memorandum is quite clear that this language will
not work. ``The case law makes it clear that this
authorization, if enacted, would run afoul of constitutional
barriers to congressional conferral either of

[[Page H8226]]

standing or of ripeness or both.'' The memorandum goes on to
say ``. . . it appears extremely likely that the Supreme
Court would either strike down the provision, or disregard
it.''
Only the Mollohan-Shays Amendment works towards a fair and
accurate census.
Carolyn Maloney,
Christopehr Sahys,
Members of Congress.
____

Professional Organizations That Have Endorsed the Use of Sampling in
the 2000 Census

National Academy of Sciences Panel on Census Requirements
in the Year 2000 and Beyond.
National Academy of Sciences Panel to Evaluate Alternative
Census Methods.
American Statistical Association.
American Sociological Association.
Council of Professional Associations on Federal Statistics.
National Association of Business Economists.
Association of University Business and Economic Research.
Association of Public Data Users.
Decision Demographics.
Mr. HASTERT. Mr. Chairman, I yield 2 minutes to the gentlewoman from
Ohio [Ms. Pryce].
Ms. PRYCE of Ohio. Mr. Chairman, I rise today in strong opposition to
the Mollohan amendment on census sampling, and in support of the
provision offered by the gentleman from Illinois [Mr. Hastert].
As a former judge I want to stress that sampling is neither a
Republican issue nor a Democratic issue. It is a legal issue and a
constitutional issue which ultimately should and must be settled by the
U.S. Supreme Court, not a politicized commission as proposed by the
Mollohan amendment. By defeating the Mollohan amendment, we will help
clear the way for enactment of the Hastert provision.
Now, here is what the Hastert provision does. First, it recognizes
that the legislative and executive branches have reached an
unresolvable impasse on the subject of sampling and statistical
adjustment. Then it asks the judicial branch to fulfill the role
envisioned for it by the Founding Fathers in the Constitution, and step
in and decide this dispute through the court system. Then it protects
the taxpayer by getting a court decision on the legality of sampling
and statistical adjustment before billions of taxpayer dollars are
spent and potentially wasted.
Now, just like a judge would issue a temporary restraining order to
prevent further harm in a dispute between two private parties, the
Hastert provision would move to protect the taxpayers from potential
harm by putting a temporary hold on funding for sampling while the
court hears the case. Once the Supreme Court has reached a final
decision, the temporary funding hold is removed and the Census Bureau
will be free to spend money in compliance with the law as determined by
the court.
Mr. Chairman, I urge my colleagues to defeat the Mollohan amendment
and to allow the enactment of the Hastert provision. Then we will
count. We will count the poor, we will count the minorities, we will
count all Americans, as is required by the Constitution.
Mr. MOLLOHAN. Mr. Chairman, I am pleased to yield 3 minutes to the
distinguished gentlewoman from Maryland [Mrs. Morella].
Mrs. MORELLA. Mr. Chairman, I thank the gentleman for yielding me
this time.
Mr. Chairman, I rise today in strong support of the Mollohan-Shays
amendment. The Census Bureau needs the full $381.8 million appropriated
in fiscal year 1998 to prepare for the Census 2000. Fencing off all but
$100 million would jeopardize critical components of census
preparation, including the dress rehearsal and the preparation of the
long form.
As Members of Congress, we depend on the accurate information
provided by the census to give us insight into our changing communities
and constituencies. If this amendment is not passed and data is not
collected in Census 2000, we will lose the only reliable and nationally
comparable source of information on our population. Both the private
and public sectors, including State, county and municipal agencies,
educators and human service providers, corporations, researchers,
political leaders, and Federal agencies rely on the census long form.
The Mollohan-Shays amendment is critical if we are to prevent the
mistakes that were made in 1990. I served on the Committee on Post
Office and Civil Service during the 1990 census and I saw firsthand the
mistakes that were made.
According to the GAO, the 1990 census got 10 percent of the count
wrong. Over 26 million people were missed, double-counted, or counted
in the wrong place. Let me quote from the GAO Capping report on the
1990 census, which makes it clear that a straight count will not work.
GAO reported that, ``the current approach to taking the census needs to
be fundamentally reassessed.''
``The current approach to taking the census appears to have exhausted
its potential for counting the population cost-effectively,'' et
cetera.
``Specifically, the amount of error in the census increases
precipitously as time and effort are extended to count the last few
percentages of the population.''
There is, my friends, strong scientific evidence that sampling will
result in the most accurate census possible. The experts agree that
spending more money to go door-to-door will result in errors as large
or larger than 1990, and that the 2000 census will be more accurate for
all congressional districts than 1990, 19 times more accurate for the
Nation.
As a result of the GAO evaluation and bipartisan direction from
Congress, the Census Bureau turned to the National Academy of Science
for advice. The first panel said, ``physical enumeration or pure
`counting' has been pushed well beyond the point at which it adds to
the overall accuracy of the census.''
That panel went on to recommend a census that started with a good
faith effort to count everyone, but then truncate physical enumeration
and use sampling to estimate the characteristics of the remaining
nonrespondents.
Following these recommendations, the Census Bureau announced in
February of 1995 a plan for the 2000 census which makes an
unprecedented attempt to count everyone by mail, followed by door-to-
door enumeration until reaching 90 percent of the households in each
census tract. A sample of households is then used to estimate the last
10 percent.
I know my time has expired. A whole list of scientific organizations
agree with it. It will save money, and it will be an accurate count.
Mr. HASTERT. Mr. Chairman, I yield myself 15 seconds just to inform
the gentlewoman from Maryland that the Census Bureau gets all of the
money that they asked for, it is not fenced off, and so she is
misinformed.
Mr. Chairman, I yield 6 minutes to the distinguished gentleman from
Kentucky [Mr. Rogers], chairman of the subcommittee.

{time} 1830

Mr. ROGERS. Mr. Chairman, I thank the gentleman for yielding me the
time.
Mr. Chairman, I rise in opposition to the Mollohan amendment and in
support of the provisions in the bill regarding the 2000 census. While
I certainly respect and appreciate the efforts of my distinguished
ranking member, the gentleman from West Virginia [Mr. Mollohan], and I
know that his intention is good, his amendment fails to address any of
the real issues surrounding the 2000 census.
My colleagues, this is one of the most important issues that will
come before the Congress. It is the Congress' constitutional
responsibility to ensure that an actual enumeration of the population
is conducted once every 10 years. Those are the words in the
Constitution.
There is no other activity conducted by the Federal Government that
has more of an impact on the daily lives of each and every one of our
constituents. The census is used for everything, from ensuring that our
constituents' constitutional right of one person-one vote is upheld, to
determining how Federal dollars are apportioned to our communities.
Many of us are all too familiar with the consequences of a disputed
census. In 1990, the American taxpayer spent $2.6 billion on the 1990
census. What did we get? A botched census, a census whose results were
litigated for most of the decade, a census whose results will forever
be questioned. We cannot afford another disaster like 1990. But that is
exactly where we are headed if the Congress does not accept its
responsibility

[[Page H8227]]

to ensure that the 2000 census is above reproach.
The administration's plan for the 2000 census represents the most
radical departure from the manner in which the census has been
conducted for the last 200 years. Serious doubts have been raised about
whether the administration is planning a fair census, a legal census, a
constitutional census. Many of us believe the administration plans are
not fair, and that they will not result in a more accurate census.
Why? For starters, we have already seen how dangerous an error-prone
statistical manipulation can be in the census. In 1990, over the
objections of the Census Bureau ``experts'', the Secretary of Commerce
refused to adjust the census numbers using statistics because he
thought they were inaccurate. He was right. Years after the fact the
same Census Bureau ``experts'' discovered their statistically
manipulated numbers had overestimated the number of people missed by
millions, and because of a computer glitch would have mistakenly caused
Pennsylvania to lose a seat in this body.
Just last month, the Census Bureau had to retract their own report
extolling the accuracy of their census plans because a computer glitch
underestimated the error rates. But even more importantly, unlike 1990,
we are not even going to have an actual count of the population. Why?
Because the administration only wants to count 90 percent of us, and
then guess the rest. So how will we ever know what the actual count
was, and how will we ever know if statistical adjustment is more
accurate? The answer is, we never will. The administration expects us
to trust the experts, the same ones that recommended we use faulty
numbers to adjust the 1990 census.
But even more fundamental to this debate is the question of whether
the administration's plans are legal and constitutional. Many of us
believe they are not. We can debate those issues all day and night. It
would not matter, because only the courts can decide that, and the
courts will decide that, one way or the other. The only question is,
when.
Under the bill, we say, have the courts resolve the questions now
before we spend $4 billion on a census that is likely to be held
illegal or unconstitutional. Does the Mollohan amendment address those
questions? No. Even worse, it strikes the very provisions in the bill
that would ensure the courts answer these questions before the fact.
In fact, instead of addressing any of these serious questions
surrounding the census, the Mollohan amendment avoids them entirely,
and instead tries to say that the only concern surrounding the census
is the threat of political manipulation. That is just not the case,
though certainly, given the track record of this administration, I can
understand how people would be so concerned.
Even if it were the only concern, the Mollohan amendment is not the
answer. Why? Because the commission has neither the expertise nor the
power to oversee the administration's complicated, convoluted census
2000.
If Members want to know how well an oversight commission works, we
have a recent example, the Teamsters election. The taxpayers spent $21
million on an oversight board for the Teamsters election, and what was
the result? They threw out the election and they are going to start all
over again, I guess. They are going to ask us to oversee it a second
time. They had better ask us real hard about that. If we need any
evidence about whether an oversight commission can protect the census,
look to the Teamsters. We will spend $4 billion on the census, and then
we will have to start all over again in 2001.
It is the Congress' duty to oversee the census. It is our duty to
ensure that it is fair, that it is legal, and that it is
constitutional. The Mollohan amendment would have us abdicate that
constitutional responsibility.
At a time when the public's faith in the institutions of government
is at an all-time low, we have a duty to ensure that the 2000 census is
above reproach. Make no mistake about it, the very integrity of the
census is at stake here, not to mention a multibillion dollar taxpayer
investment.
Mr. Chairman, I urge rejection of the Mollohan amendment.
Mr. MOLLOHAN. Mr. Chairman, I yield such time as she may consume to
the distinguished gentlewoman from California [Ms. Roybal-Allard].
(Ms. ROYBAL-ALLARD asked and was given permission to revise and
extend her remarks.)
Ms. ROYBAL-ALLARD. Mr. Chairman, I thank the gentleman for yielding
time to me.
I rise in strong support of the Mollohan-Shays amendment.
Mr. MOLLOHAN. Mr. Chairman, I yield 2 minutes to the distinguished
gentleman from Massachusetts [Mr. Frank].
Mr. FRANK of Massachusetts. Mr. Chairman, if what the gentleman who
just spoke wanted to have happen could happen, I would support it. What
he said is look, there is a constitutional question here. Let us,
before anything happens, go to the United States Supreme Court and ask
them to tell us. They will not do it. There is a core principle of
American constitutionalism, which conservatives usually adhere to,
which says they do not issue advisory opinions. The United States
Supreme Court does not decide until there is a case or controversy,
defined repeatedly by Justice Scalia, who was quoted only partially on
one point, as injury in fact.
We recently had an effort to try to get around that by getting an
advisory opinion in effect on the line item veto. The Supreme Court
unanimously said, or almost unanimously said no, you cannot have it.
What the gentleman from Kentucky is asking for is impossible. What he
says is, we will go to court.
But the Supreme Court will not decide it. Standing is a core
conservative principle. I thought the gentleman's amendment was written
by William O. Douglas. I thought William O. Douglas had channeled
himself through to somebody on the other side, because he is the great
liberal justice who says there is a constitutional question, let me at
it, I will handle it. What in fact the conservatives said is, no. You
talk about judicial activism, this is a monument to judicial activism.
This is a constitutional question. We will ask the United States
Supreme Court for an advisory opinion. It will not give it to you. It
requires an injury in fact.
Here is how you define standing. Here is who could bring this
lawsuit. Any resident of a State, resident, not even a citizen, any
resident of a State whose congressional district could, not was, could,
in fact be changed. If you thought that your district might gain under
this, you could go in and get an advisory opinion.
The Supreme Court will not do it. No one familiar with this
jurisprudence thinks remotely that you could force this. If it were
possible, it would be a good way. But remember, we said, we will have
to deal with these first through the electoral process and the
political process, and only after the fact can you go to court. Who
said that? That was done by conservatives to keep the non-elected
judiciary from being too intrusive. What the gentleman's amendment does
is to reverse that principle of judicial restraint.
Mr. HASTERT. Mr. Chairman, I yield 2 minutes and 40 seconds to the
gentleman from Georgia [Mr. Kingston].
Mr. KINGSTON. Mr. Chairman, there is a story of a very learned doctor
of theology, a distinguished minister, who was walking through the park
one day. He sees a guy who is kind of an itinerant of sorts, and he is
reading the Book of Revelations. The doctor of theology says to him, in
a condescending, intellectual way, my good man, ``Do you have any idea
at all of what you are reading in the Book of Revelations?'' To which
the guy said, ``No, I can't say I understand every little bit of it.''
And he says, ``Then sir, why are you reading it?'' He said, ``Because I
know how it ends.''
What I am saying, Mr. Chairman, is I do not believe this is a debate
of pointy-headed intellectual bean-counters. I think this is a debate
about common sense. Here is how I understand this issue. Under the
normal U.S. census procedure, you go to a house. You ask how many folks
live there. Three. You go to the second house. How many live there?
Seven. How many live in the third house? Six. You write down three,
seven, six. You come up with 16.
Now, under the Democratic samplematics, you are doing it a little
more creatively. You go to the first house and count three, to the
second

[[Page H8228]]

house and count seven, and at the third house you go to the drugstore
and get yourself a Coca-Cola, and you sample about 20 people there.
Then, depending on how many you need, you say, in total, we got maybe
15 to 25 people, depending on how many the folks need back in the
office, and that is the count.
Now, let us say that is how this thing works, in layman's terms, so I
can understand it. Now think about it in other potential applications.
We may want to take a second look at this as Members of Congress. What
would be some other potential sampling applications?
How about balancing your checkbook? No problems with overdrafts. How
about adjusting your income taxes; you know, sending it to the IRS, and
when they start complaining, there is a lot of IRS passion going on
these days, you can say, ``Hey, look, I just used sampling to send you
what I owed you.''
That has often handicapped us. I will just say that a lot of people
sample on their golfing already. On the SAT, for those Members with
teenaged kids trying to get into college, sample up the SAT score,
1,500. Speeding tickets: ``Officer, I was going about 100, but I was
sampling. Just give it to me at 55.'' That is what this is about.
Mr. Chairman, the 14th Amendment of the United States says it real
easy for someone like me and a lot of other folks, that counting the
whole number of persons in each State is the way to do your sampling.
Mr. MOLLOHAN. Mr. Chairman, I am pleased to yield 2 minutes to the
distinguished gentleman from North Carolina [Mr. Watt].
Mr. WATT of North Carolina. Mr. Chairman, I thank the gentleman for
yielding me the time.
Mr. Chairman, I do not believe there is a Member of this House who
over the last 5 years has risen in defense of the United States
Constitution more than I have. I honestly would tell the Members if I
thought statistical sampling was unconstitutional, regardless of the
political consequences. I would be rising in support of the
Constitution, in defense of the Constitution.
I think this whole constitutional argument is a bogus argument,
however, and it fails to read the entire sentence in Article I, Section
2, clause 3 of the Constitution, because that section of the
Constitution requires an actual enumeration, but then it goes on to
say, ``. . .in such manner as the Congress shall by law direct.'' And
all of these gentlemen who have gotten up and talked about requiring a
head count seem to be ignoring the second part of the sentence.
Every single Justice Department that has opined on this issue, the
Bush Justice Department, the Carter Justice Department, the Clinton
Justice Department, have all said that statistical sampling is fine
under the Constitution. Every single court that has addressed this
issue has said that statistical sampling is acceptable under the
Constitution.

{time} 1845

The Federal District Court, Eastern District of New York, said it is
no longer novel or in any sense new law to declare that statistical
adjustment of the census is both legal and constitutional because
article I, section 2, requires the census to be as accurate as
practical. The Constitution is not a bar to statistical sampling. This
is a bogus argument that my colleagues are using. Statistical sampling
is constitutional.
I rise in support of the amendment.
Mr. HASTERT. Mr. Chairman, I yield 2 minutes to the gentleman from
Iowa [Mr. Latham], a member of the subcommittee, who is well familiar
with bean counters.
Mr. LATHAM. Mr. Chairman, I guess being in the soybean business, we
do count a few beans there.
But I think we have to look at what this debate really is all about.
We are talking about the census, but really what it gets down to is
money and power. It really gets down to the debate of whether we want
those things distributed in a fair and honest manner or if we want
someone possibly with political motivation to guess at where those
things go.
No. 1, with the money, as everyone here knows, and I do not know if
the folks at home know that where the Federal dollars are distributed
is based on the count, would we rather have an actual real count to
know that we are getting our share of Federal dollars or would we like
a bureaucrat here in Washington to guess at it?
As far as power, it has to do with how many Representatives we have
from our States. If our State is kind of on the bubble here as to
whether we are going to lose a seat or gain a seat, do we want that
determined by an actual real count or do we want a bureaucrat here in
Washington to make that determination for us and mute our voices? It is
simply wrong to go that route.
I do not necessarily say that there is going to be politics involved
in this census or this guessing that we are proposing do here, but let
us look at the record. Has this administration politicized any other
departments in government? Look at the FBI. There are 900 files of
private citizens for political reasons in the White House today. They
brought in over a million citizens last year for the election and did
not check the background, for political reasons, of 180,000 of them.
There are 30,000 convicted felons in this country because they
politically wanted to get more people registered to vote.
Would they politicize the census? What do my colleagues think? We
need an honest, fair, real, legal, and constitutional census, and that
means to count real people.
Mr. MOLLOHAN. Mr. Chairman, I yield myself 45 seconds to respond to
the gentleman, if he would stay at the podium.
I would just like to assure the gentleman, that is precisely the
reason. That is the one argument against the census that cannot be
refuted by fact, because it is based upon suspicion. That is why we
created this oversight board, which is composed of former Presidents,
people who have absolute credibility, to give the census credibility,
because this kind of a debate that the gentleman just engaged in, in
and of itself, is the greatest underminer of public confidence.
Also, with regard to the efficacy of sampling, our own Speaker
Gingrich must have believed in the efficacy of sampling because on
April 30, 1991, he wrote, in part, to the Secretary of Commerce, I
quote, I respectfully request that the census numbers for the State of
Georgia be readjusted to reflect the accurate population of the State
so as to include the over 100,000 which were not previously included.
Mr. Chairman, I yield 2 minutes to the distinguished gentleman from
Texas [Mr. Stenholm].
(Mr. STENHOLM asked and was given permission to revise and extend his
remarks.)
Mr. STENHOLM. Mr. Chairman, though much of the debate on correcting
the undercount of the census is centered around the number of people
not counted in urban areas, as one who represents a very rural
district, I want to highlight the fact that people in rural areas are
being missed as well. In fact, some of our rural areas are undercounted
to a greater degree than the entire country.
According to the Census Bureau, the net undercount for the Nation in
1990 was 1.6 percent, while rural areas were undercounted at a rate of
5.9 percent. I want to emphasize that accuracy is critical. Let there
be no disagreement on that as we prepare for the 2000 census. The
Census Bureau should form early and active partnerships with State and
local governments so that these governments will have an early
opportunity to review census address lists and maps for their area.
This amendment will remove the restrictive language included in the
bill and allow the Census Bureau to continue to plan for the 2000
census. Their proposal, which is supported by scientists and
statistical experts, should improve accuracy and save costs.
It is fascinating to sit here and listen to colleague after colleague
argue against the best science available. I have taken to this well day
after day after day, arguing that we should use the best science
available, whether we are talking about environmental issues, food
safety issues, or census issues. But tonight in this debate, we are
being selective as to which science we should use. I find this a
fascinating argument to listen to.
I am convinced, absolutely convinced, that statistical sampling is
the best method to get an accurate census,

[[Page H8229]]

and I urge my colleagues to listen to this debate and to listen to
those who are saying that only some science is good and we will be
selective in which we choose to agree to. Statistical scientists say
that sampling will help us get an accurate count. Is that not what we
all should really be for?
I urge my colleagues to support the Mollohan-Shays amendment.
Mr. HASTERT. Mr. Chairman, I yield 1 minute to the gentleman from
Florida [Mr. Miller]
Mr. MILLER of Florida. Mr. Chairman, I rise in opposition to the
amendment offered by the gentleman from West Virginia and in opposition
to the use of sampling.
I am a former statistics professor. I taught statistics at both the
undergraduate and graduate level at several universities. I have
respect for sampling, but sampling is used when you do not have enough
time or money. What you really want to have is census information,
statistics. When you use sampling, you have bias. You have nonsampling
bias, and you have sampling bias.
In my first lecture on statistics both at the graduate level and the
undergraduate level, I used to use this book, still available to buy in
the book store. It is ``How To Lie With Statistics.''
Statistics can be manipulated in a variety of ways that can be
legitimately defended. I do not trust statistics. I teach my students
to be suspicious of statistics, to be cautious of the use of
statistics. I used to make the statement, tell me the point you want me
to prove, and I will prove it with statistics, because it can be done.
I know all the statisticians say sampling is great. Statisticians
would not have a job if we did not have sampling. That is what
statistics is based on. Statisticians are biased to start with.
I think we are doing a good job. What we need to do is do a good
census. Dr. Riche is moving in that direction. Let us look at the
examples of what took place in Milwaukee and what took place in
Cincinnati. We can do a good census. Let us do the job right and not
play around with sampling.
Mr. MOLLOHAN. Mr. Chairman, I yield 1\3/4\ minutes to the gentlewoman
from Florida [Mrs. Meek].
(Mrs. MEEK of Florida asked and was given permission to revise and
extend her remarks.)
Mrs. MEEK of Florida. Mr. Chairman, first of all, I do not trust
statistics any more than the rest of my colleagues. But I trust even
less the belief that everyone is going to be counted fairly.
If we look at the history of this, we have never had an accurate
count. The under-count has been shown more in African Americans than it
has in any other group. Do we want this repeated? Then we are sending a
message that we do not want a fair census count.
This country does not look like it did in 1990. You better look
around and see that it is different. You see more minorities. There
will be even more. So you may as well learn that you have to count them
accurately. You cannot count them accurately by the kinds of
enumeration that you are doing or that you expect to do.
So it tells me that the issue is that because you know there are more
of them than there are of you, that you do not want an accurate count.
They are going to be there. They are going to be under the bridges.
They are going to be in the homeless shelters. There are going to be
people who do not return those things to the census.
All I am saying to you is, it is fruitless, it is crazy, it is a
waste of money, but you would rather do that politically and for power
than to go to a sampling which the Mollohan amendment is asking us to
do. You would rather take that useless method because you do not want
to count everybody. You want to go back to the time when there was a
serious undercount.
It will repeat itself. It was in 1990, as you see from this chart. It
is going to be in the year 2000, because you are going to insist on
counting every head.
Mr. Chairman, they cannot enumerate and count every head because they
are not going under the bridges, they are not going on the highways and
byways of this country to find these little people and count them. If
that is the way you want it, then you will not support the Mollohan
amendment.
I support the Mollohan amendment because it is fair. African-
Americans will be counted. It has got to be done.
Mr. HASTERT. Mr. Chairman, I yield 2 minutes to the distinguished
gentleman from Illinois [Mr. Hyde].
(Mr. HYDE asked and was given permission to revise and extend his
remarks.)
Mr. HYDE. Mr. Chairman, this is a fascinating debate. I listened to
my good friend, the gentleman from Texas [Mr. Stenholm], talk about the
scientists. I do not think you have to be a scientist, rocket or
otherwise, to read the plain language of the Constitution: ``The actual
enumeration,'' those are not tough words, ``shall be made within 3
years after the first meeting of the Congress.''
And then a constitutional scholar, the gentleman from North Carolina
[Mr. Watt], brought in the entire text. He said, ``in such a manner as
they,'' meaning Congress, ``shall by law direct.''
Well, you cannot by law amend the Constitution. You cannot pass a
statute and erase the first three words of article I, ``the actual
enumeration.''
It is a stretch to ask us to trust the sampling of the population to
an administration that has shown, at best, a reckless disregard for the
letter and the spirit of the law.
It goes beyond the Constitution. We have a statute. Title 13, section
195, says, ``Except for the determination of population for purposes of
apportionment of Representatives in Congress among the several States,
the secretary shall, if he considers it feasible, authorize the use of
the statistical method.'' It specifically excludes counting by sample,
by guess, a determination, ``for the purposes of apportionment.''
We want to count everybody. If they are under the bridges, go down
there and count them. You are getting paid to count them. Why is that
less accurate than guessing how many people are under the bridge? Your
administration does not exactly wear a T-shirt saying, ``trust me,''
and engender an awful lot of confidence to have you count how many
people there are and where they are and what the districts shall be in
the next 10 years.
The CHAIRMAN. The Committee will rise informally in order that the
House may receive a message.
The SPEAKER pro tempore (Mr. Miller of Florida) assumed the chair.

____________________