CENSUS 2000: LEGAL ISSUES RE DATA FOR REAPPORTIONMENT AND REDISTRICTING

CRS Report for Congress
Census 2000: Legal Issues re: Data for
Reapportionment and Redistricting
Updated March 12, 2001
Margaret Mikyung Lee
Legislative Attorney
American Law Division


Congressional Research Service The Library of Congress

Census 2000: Legal Issues re Data for
Reapportionment and Redistricting
Summary
The release of the 2000 Census data for the apportionment of the House of
Representatives among the States and the release of the state redistricting data to the
States as required by P.L. 94-171 has renewed the decennial debate over several
issues. First, the debate over the use of sampling to adjust the decennial population
census data was not completely resolved by the U.S. Supreme Court decision that a
federal statute prohibits the use of sampling to adjust the decennial census for the
purposes of apportionment of the House of Representatives among the States. The
Court did not hold that the adjustment of census data for other purposes, such as
intrastate redistricting, was prohibited. Additionally, last year, the Census Bureau
indicated that it would likely release both adjusted and unadjusted sets of data. In the
wake of these developments, there were legislative efforts in many States to specify
which data are to be used in intrastate redistricting. Controversy surrounded
preclearance under the Voting Rights Act for some state laws. Meanwhile, the
transition between Administrations resulted in a policy change concerning the release
of adjusted census data for redistricting. Last year, the Secretary of Commerce
promulgated a rule which delegated the decision to adjust the official decennial census
figures to the Director of the Census Bureau and required the release of adjusted
figures where the Secretary chose not to release them as the official redistricting data
despite the recommendation of the Executive Steering Committee for Accuracy and
Coverage Evaluation Policy (ESCAP) to adjust the data. This year, the Secretary of
Commerce under the new Administration rescinded that rule and took back the
authority to decide to adjust state redistricting census data. The City of Los Angeles,
joined by other municipalities, filed a suit challenging the anticipated change in policy
on the grounds that proper rule-making procedures were not followed. The suit relied
on the assumption that the ESCAP would recommend to adjust. However, on March
1, 2001, the Acting Director of the Census Bureau and ESCAP recommended not to
adjust redistricting data; accordingly, on March 6, 2001, the Secretary announced the
release of the unadjusted data. Since this appeared to render the Los Angeles suit
moot, the City Attorney for Los Angeles announced that he would amend the
complaint to add new allegations that the Secretary’s decision is arbitrary and
capricious and violates the Census Act, which requires the Secretary to use statistical
sampling, where he considers it feasible, for purposes of the census other than
apportionment of the House of Representatives. Congress has the constitutional
authority to determine issues of census methodology, including whether or not the
release of adjusted data suitable for intrastate redistricting purposes is feasible.
Second, Utah has filed a suit challenging the apportionment of the House of
Representatives on the grounds that expatriate Utahans, specifically missionaries for
the Church of Jesus Christ of the Latter-Day Saints, should have been included in the
population count for Utah. Such an inclusion would have meant that Utah would
have gained a congressional seat which went to North Carolina instead. Although the
U.S. Supreme Court held previously that the Secretary of Commerce had the
discretion to include overseas federal personnel in the apportionment census, it did
not address the issue of whether other expatriates should be included as well, once the
decision was made to include one segment of the expatriate population. Congress,
however, has the authority to legislate census methodology with regard to the
inclusion or exclusion of expatriates and categories of expatriates.



Contents
Release and Use of State Redistricting Census Data......................1
Background ................................................ 1
State Legislative Activity......................................5
Administrative Activity.......................................8
Congressional Activity.......................................11
Utah v. Evans.................................................11
Additional Reading.............................................12



Census 2000: Legal Issues re Data for
Reapportionment and Redistricting
The release of the 2000 Census data for the apportionment of the House of
Representatives among the States and the release of the state redistricting data to the
States as required by P.L. 94-171 has renewed the decennial debate over several
census-related issues. First, the perennial statistical adjustment issue, although
decided by the United States Supreme Court with regard to the apportionment of the
House of Representatives among the States, remains active with regard to intrastate
redistricting census data. The decision of the Secretary of Commerce to release
unadjusted 2000 census data for intrastate redistricting purposes was seen as an
unequivocal victory by opponents of statistical adjustment,1 but does not seem to have
quelled the determination of proponents of statistical adjustment to obtain release of
adjusted data.2 Second, Utah has brought a lawsuit challenging the apportionment,
alleging that expatriate missionaries for the Church of Jesus Christ of the Latter-Day
Saints should have been included in the population count for Utah.
Release and Use of State Redistricting Census Data
Background
On January 25, 1999, in Department of Commerce v. U.S. House of
Representatives,3 the United States Supreme Court held that the Census Act4
prohibits sampling in the census for the apportionment of the House of
Representatives, but declined to decide whether sampling would also be a violation
of the census clause of the U.S. Constitution. This decision was the culmination of
two lawsuits which had been brought to challenge the plans of the Census Bureau to
use sampling in the 2000 census. Opponents of sampling claimed victory and
promised to focus on improving the traditional headcount through methods such as
expanded outreach to undercounted groups and the use of administrative records.
But proponents of sampling, including the Clinton Administration, noted that the
decision did not determine the constitutionality of sampling and also did not hold that


1See press releases of the Chairman of the House Subcommittee on the Census, Rep. Dan
Miller, Game. Set. Match. at [http://www.house.gov/danmiller/census/mar1_01.html] (last
visited March 8, 2001) and Miller Applauds Secretary’s Final Decision at
[http://www.house.gov/danmiller/census/mar6_01.html] (last visited March 8, 2001).
2Mae M. Cheng, Despite Undercount, Census Data to Stand; Adjustment advocates vow
legal challenges, NEWSDAY, March 7, 2001, at A6.
3525 U.S. 316, 119 S.Ct. 765 (1999).
4Codified as amended at 13 U.S.C. §§ 1 et seq.

sampling was prohibited for purposes other than apportionment of the House of
Representatives among the States. Because the Court stated in one part of its opinion
that section 195 of the Census Act “requires [the use of] statistical sampling in
assembling the myriad demographic data that are collected in connection with the
decennial census,”5 supporters of adjustment argued that sampling techniques were
not only permissible, but were required, in the taking of the census for the purposes
of intrastate redistricting and federal funding allocations. They indicated an intention
to seek the use of sampling techniques in population counts used for intrastate
redistricting and funding allocation formulas.
Other case law arguably supports the use of figures other than the official data
for apportionment of the House of Representatives among the States. The Federal
Constitution does not require the use of federal decennial census data for intrastate
congressional and state legislative redistricting; it only provides for the use of census
data for apportionment among the states, not for redistricting and reapportionment
within them. Federal courts have held that States are not required to use federal
census data, adjusted or unadjusted, for redistricting, and therefore the Bureau of the
Census itself has no duty or requirement to provide the states with adjusted data for
redistricting purposes. In the 1969 Supreme Court decision in Kirkpatrick v. Preisler6
involving Missouri’s congressional redistricting plan, the Supreme Court, while
invalidating the plan, nevertheless indicated that the use of projected population
figures was not per se unconstitutional and that States may properly consider such
statistical data if such data would have a high degree of accuracy (however, the7
Court also stated that the federal decennial census data were the best data available).
In Senate of the State of California v. Mosbacher,8 in which the state senate was
suing for the release of adjusted data after the Bureau decided not to adjust the
official 1990 census data, the U.S. Court of Appeals for the Ninth Circuit noted that
if a State knows that census data is underrepresentative of the population, it can and
should utilize non-census data, in addition to the official count, for redistricting,9 but
the court also held that the Secretary of Commerce had no affirmative duty under the
Census Clause of the Federal Constitution (Art. 1, § 2, cl. 3) and the federal census
statutes to assist the State by providing adjusted census data.10 However, in Assembly


5525 U.S. at 339, 119 S.Ct. at 777.
6394 U.S. 526 (1969).
7See also Dixon v. Hassler, 412 F.Supp. 1036, 1040-41 (W.D. Tenn 1976), aff’d sub nom.
Republican Party of Shelby County v. Dixon, 429 U.S. 934 (1976); Exon v. Tiemann, 279
F. Supp. 601, 608 (D. Neb. 1967).
8968 F.2d 974 (9th Cir. 1992).
9968 F.2d at 979, citing Garza v. County of Los Angeles, 918 F.2d 763, 772-73 (9th Cir.

1990), cert. denied, 498 U.S. 1028 (1991).


10968 F.2d at 979 (but Judge Pregerson, dissenting, argued that by refusing to disclose the
adjusted data, the Secretary may have impermissibly interfered with the state senate’s duty
to redistrict under the Federal Constitution and the Voting Rights Act).

of the State of California v. U.S. Department of Commerce,11 the same court affirmed
a lower court’s decision requiring that computer tapes containing statistically adjusted
data from the 1990 census be released to the Assembly under the Freedom of
Information Act and noted that the “states are not obliged to use official census data
when drawing their state legislative or congressional districts.”12 In a similar case,
Florida House of Representatives v. U.S. Department of Commerce,13 the U.S. Court
of Appeals for the Eleventh Circuit held that the statistically adjusted data was exempt
from disclosure under the Freedom of Information Act.
In Young v. Klutznick, a suit by the city of Detroit to get an adjustment of the
alleged undercount in the 1980 census data, in dicta, the appeals court stated that the
state legislature is not required by the federal Constitution to use census data supplied
by the Bureau for congressional redistricting, but could use adjusted population
figures when redistricting between decennial censuses, as long as the adjustment is
thoroughly documented and applied in a systematic manner.14
In City of Detroit v. Franklin,15 the city sought an adjustment of the alleged
undercount in the 1990 census data and argued that the causation analysis by the
Court of Appeals in the case involving the 1980 census had been overruled by the16
holding in Karcher v. Daggett. The city argued that in Karcher the U.S. Supreme
Court held that the apportionment clause imposes an obligation on states to use only
the official population count as determined by the Bureau in redistricting. This
argument is probably based on the sentence "[a]dopting any standard other than
population equality, using the best census data available . . . would subtly erode the
Constitution's ideal of equal representation"17 and the fact that the Court considered
the census data the only reliable indication of the districts' relative population levels.18
The District Court in City of Detroit held that the plaintiffs misconstrued Karcher and
that it did not hold that states must use census figures in redistricting and did not
overrule Young. Rather, the Supreme Court had "merely reiterated a well-established
rule of constitutional law: states are required to use the `best census data available' or
`the best population data available' in their attempts to effect proportionate political19
representation."
In the cases above, except for Kirkpatrick v. Preisler, where the courts held that
the States were not required to use federal census data for intrastate redistricting


11968 F.2d 916 (9th Cir. 1992).
12968 F.2d at 918, n. 1, citing Burns v. Richardson and Young v. Klutznick, discussed below.
13961 F.2d 941 (11th Cir. 1992).
14652 F.2d 617, 624 (6th Cir. 1981).
15800 F. Supp. 539 (E.D. Mich. 1992).
16462 U.S. 725 (1983).
17462 U.S. at 731 (citing Kirkpatrick v. Preisler, 394 U.S. 526, 532 (1969)).
18462 U.S. at 738.
19800 F. Supp. at 543 (quoting also from Kirkpatrick v. Preisler, 394 U.S. 526, 528 (1969)
("the best population data available")).

activities, the plaintiff States had chosen to use the total population based on federal
census figures as the basis for redistricting activities and sued to obtain the census
data as adjusted by the Census Bureau. Federal courts have also considered cases
where state legislatures did not use federal decennial census data or even total
population data as the basis for redistricting activities. In Kirkpatrick, the state
legislature apparently performed rather haphazard adjustments and projections based
on total population and the Court found that the legislature had not justified its20
methodology. In Burns v. Richardson, the Supreme Court held that in state
legislative redistricting cases the Constitution "does not require the States to use total
population figures derived from the federal census as the standard" of measurement.
The Court noted that in earlier cases it was careful to leave open the question of what
population basis was appropriate in redistricting activities, even though in several
cases total population figures were in fact the basis for comparison when determining
whether the Equal Protection Clause of the Constitution had been violated. The
Court recognized that in a particular case, total population might not be the
appropriate basis for redistricting plans. In the Burns case, Hawaii had used the
number of registered voters as the basis for redistricting the state senate. The Court
found that the redistricting plan “satisfies the Equal Protection Clause only because
on this record it was found to have produced a distribution of legislators not
substantially different from that which would have resulted from the use of a
permissible population basis.”21 Hawaii was found to have a unique situation, wherein
the significant number of tourists, military personnel, and other transient population
segments distorted the distribution of actual state citizens. The redistricting plan that
would have resulted from a total population basis would not have reflected the true
state population distribution as accurately as a state citizen population basis. Since
a registered voter population basis was the closest approximation of a state citizen
population basis, the use of the registered voter population basis was deemed
consistent with the Equal Protection Clause. However, the Court was careful to note
that the ruling in the Burns case did not establish the validity of the unique
redistricting population basis for all time or circumstances.22 Although the federal
decennial census figures need not be used as the basis for redistricting activities, any
alternate figures used must be shown to be the best data available or to be justified by
particular circumstances as resulting in a more accurate redistricting plan than one
based on federal decennial census total population figures.
Since, under the Federal Constitution, the States arguably can and should use
data other than the official apportionment census data in their own redistricting
process if they know the other data to be the best available data, one must look at


20384 U.S. 73, 91 (1966).
21384 U.S. at 93.
22See also MacGovern v. Connolly, 673 F. Supp. 111 (D. Mass. 1986) (court upheld state
redistricting scheme which entailed use of data from a decennial state census held every 10
years beginning in 1975 and refused to order a new scheme based on “inapposite” 1980
federal census data); Klahr v. Williams, 313 F. Supp. 148 (D. Ariz. 1970) (court held invalid
congressional and state legislative redistricting plans based, inter alia, on a population
estimate formula “converting 1968 voter registration to 1960 census on a proportionate basis”
which did not truly represent the population, but ordered the plan used anyway because no
better alternative was feasible before the next election).

each State’s laws to determine whether the States themselves require the use of
official federal decennial census data in the redistricting processes. Although most
States prescribe a redistricting procedure by statute for state legislative redistricting,
many do not have a statutory procedure for congressional redistricting. The state
legislatures in such States conduct the congressional redistricting as they decide on
an ad hoc basis after a federal decennial census. This means that often in such States
there is no explicit statutory requirement to use official federal decennial census data
for congressional redistricting, although there may be such an explicit requirement for
state legislative redistricting. To the extent that a State’s own laws do not explicitly
require the use of official federal decennial census data for intrastate redistricting, the
State is free to use any other data. Even if a State’s laws require the use of official
federal decennial census data, it is unclear what a reference to official federal
decennial census data would mean, if the Federal Government released two official
sets of data. This issue was also considered during the oral arguments in the census
sampling cases.23 If the Secretary of Commerce transmits an official, second, adjusted
data set, that data arguably could still be considered official federal decennial census
data, even if it is not the data used for apportionment of the House of
Representatives. One should note that the Court’s holding on standing for the
plaintiffs in Department of Commerce v. U.S. House of Representatives indicates that
a majority of the Court considers the references to official federal decennial census
data to be a reference to the apportionment data.24
Because of the absence of sufficiently clear and explicit statutory guidelines
concerning the appropriate data to be used in intrastate congressional and state
legislative redistricting activities at the time of the decision in Department of
Commerce v. U.S. House of Representatives, in the wake of that decision there was
a flurry of state legislative activity concerning the type of federal decennial census
data to be used in intrastate redistricting. Although there has been no congressional
activity concerning state redistricting census data as yet, there is a potential role for
Congress in determining what data will likely be used by the States.
State Legislative Activity
Since the 1999 Supreme Court decision concerning sampling, a majority of the
States have considered legislation concerning the type of data to be used for intrastate
redistricting. So far, four States – Alaska, Arizona, Colorado, Kansas, and Virginia25
– have actually enacted legislation requiring that unadjusted federal decennial census
data be used as the basis for redistricting activities. No States have enacted legislation
requiring the use of adjusted data for redistricting. The vast majority of state


23Oral Argument Transcript, found at 1998 WL 827383 on Westlaw (oral argument of
Michael A. Carvin on behalf of the appellees in No. 98-564).
24525 U.S. at 332-4, 119 S. Ct. at 774-5.
25In Alaska, S.B. 99, Ch. 18 of the 1999 Acts, was enacted on May 11, 1999. In Arizona,
H.B. 2698, Ch. 47 of the 1999 Laws, was enacted on April 22, 1999. In Colorado, S.B. 206,
Ch. 170 of the 1999 Laws, was enacted on May 7, 1999. In Kansas, S.B. 351, Ch. 148 of
the 1999 Laws, was enacted on May 12, 1999. In Virginia, H.B. 1486, ch. 884 of the 2000
Acts, was enacted on April 9, 2000.

legislative activity appears to have been in the form of resolutions noting the
apportionment purpose of the census, the Supreme Court decision, and the possible
unconstitutionality of using adjusted figures for intrastate redistricting; calling on the
Census Bureau to conduct the census in a manner consistent with the Supreme Court
ruling, i.e., without sampling; opposing the use of adjusted data for state redistricting;
requesting or demanding the transmission of unadjusted data for state redistricting;
and urging Congress to take any necessary steps to ensure a fair and legal census. A
few States had resolutions urging the Federal Government to use modern, scientific
techniques, i.e., sampling, in providing the States with P.L. 94-171 redistricting data.
There appears to have been very little state legislation which would require the use
of adjusted data in the redistricting process.
Currently, under the Voting Rights Act of 1965 and its regulations,26 all or parts
of sixteen States are considered “covered jurisdictions” required to submit any
proposed changes in voting, election or redistricting laws to either the U.S. Attorney
General/Department of Justice or to the U.S. District Court for the District of
Columbia [District Court] for preclearance through an administrative finding or
declaratory judgment, respectively.27 Alaska, Arizona, and Virginia are among those
sixteen States. Alaska and Arizona submitted their laws to the Department of Justice
[DOJ] for preclearance in 1999. Virginia submitted its law to the District Court for
preclearance in 2000. Under the Act and its regulations, the DOJ has sixty days to
respond to a submission. If it requests further information, a new sixty-day period
starts when it receives the new information from the State. Further requests do not
restart the clock. If the State receives a notice of preclearance or no response from
the DOJ within the sixty-day period, the law is considered precleared. If the DOJ
finds either that the law negatively impacts voting rights or that it is unclear whether
there is a negative impact, it is supposed to notify the State of an objection to
preclearance. Either way, the decision of the DOJ regarding preclearance is
unreviewable. However, a declaratory judgment can still be sought from the District
Court. Neither the administrative finding of the DOJ nor the declaratory judgment
of the District Court regarding preclearance precludes litigation challenging the
constitutionality of the law being considered. Preclearance simply constitutes a
preliminary finding permitting the law to be implemented or to take effect in the first
place.
The preclearance process for all three states was delayed until the release of the
2000 census redistricting data. The DOJ requested further information from both
Alaska and Arizona, but decided not to issue any decision, pending the release of the
data. The DOJ apparently did not explicitly communicate such a policy to Arizona
which, after a longer than usual time period for an initial preclearance decision had
elapsed, chose to withdraw its law from the preclearance process in February 2000,
which means that the law cannot take effect. In July 2000, the DOJ informed Alaska
in writing that, until the release of redistricting data, it could not evaluate whether the


26The Act is codified as amended at 42 U.S.C. §§ 1971, 1973 – 1973bb-1; its regulations are
at 28 C.F.R. part 51.
27These sixteen states are Alabama, Alaska, Arizona, California, Florida, Georgia, Louisiana,
Michigan, Mississippi, New Hampshire, New York, North Carolina, South Carolina, South
Dakota, Texas, and Virginia.

Alaskan statute would negatively effect the voting rights of minority voters in future
redistricting plans and so could not make a preclearance decision. In Virginia v.
Reno,28 its suit for declaratory judgment of preclearance, Virginia claimed – (1) that
its new law does not require preclearance because federal law requires the use of
unadjusted figures anyway and the new law continuing the status quo does not
constitute a change in redistricting laws and practices requiring preclearance; (2) that
the new law does not have the purpose nor will have the effect of abridging the right
to vote of minority voters; and (3) that the Attorney General’s plans to use adjusted
figures in evaluations of redistricting plans violates federal law (another claim dealt
with a part of the law unrelated to redistricting census data). A three-judge panel of
the District Court granted a motion by the defendants to dismiss the claims as not ripe
for review, ruling that none of these claims was ripe, since no final Administrative
decision concerning the release of adjusted figures had yet occurred. The court noted
that if only unadjusted figures were released, any ruling regarding the need for
preclearance would be a prohibited advisory opinion. It could not evaluate any
negative effect on voting rights unless and until adjusted figures were released so that
it could compare the results of using one data set versus the other. Finally, until a
Census Bureau decision concerning the release of data was made, there could be no
plans by the Attorney General to use a particular data set for preclearance evaluations29
of redistricting plans.
On March 6, Secretary of Commerce Donald L. Evans announced that he had
decided to send the actual 2000 census enumeration data, unadjusted by statistical
methods, to the States for the purpose of redistricting, in accordance with the
recommendation of the Acting Director of the Census Bureau, William G. Barron,
and the Executive Steering Committee for Accuracy and Coverage Evaluation Policy
(ESCAP).30 As mentioned above, the District Court ruling on the Virginia statute
noted that if only unadjusted figures were released, any ruling regarding the need for
preclearance would be a prohibited advisory opinion. Since unadjusted data will be
the official P.L. 94-171 redistricting data, the remaining issues in the Virginia case are
moot. Presumably, the DOJ will likely grant preclearance to the Alaskan statute,
since the officially released data and the data required by the statute will be the same,
although the decision of the Secretary appears to have rendered all the state statutes
mandating use of unadjusted data superfluous. Nevertheless, proponents of


28117 F. Supp. 2d 46 (D.D.C. 2000); aff’d, 531 U.S. ___ (U.S. Jan. 8, 2001) (No. 00-862).
29The Attorney General announced that the DOJ would evaluate redistricting plans according
to data released pursuant to the Public Law 94-171 program, even if these data were not used
in drawing the plan. Use of other data in the plan would not be per se invalid, but unless other
data was shown to be more accurate and reliable, the DOJ would consider Public Law 94-171
data to be the population measure for preclearance purposes. 66 Fed. Reg. 4900, 5413 (Jan.

18, 2001).


30See press release of the Secretary at [http://www.census.gov/Press-Release/www/date.html]
(last visited March 8, 2001); and recommendation of the Acting Director of the Census
Bureau and ESCAP at 66 Fed. Reg. 14004 (March 8, 2001) and also at
[http://www.census.gov/dmd/www/EscapRep.html] (last visited March 8, 2001).

adjustment still seek the release of adjusted data, apparently for redistricting as well
as federal funding allocation purposes.31
Administrative Activity
As noted above, on March 1, 2001, the ESCAP made a recommendation not to
adjust the census redistricting data and the Secretary of Commerce made the final
decision not to adjust on March 6, 2001. The ESCAP report indicated that the
ESCAP was not yet able to resolve to its satisfaction certain discrepancies between
the data resulting from the adjustment sampling survey and other accuracy analyses
and needed more time to study the matter. Pursuant to 13 U.S.C. § 141 (commonly
referred to as P.L. 94-171), the Administration is required to provide decennial census
figures for intrastate redistricting activities to the States participating in the P.L. 94-
171 program prior to April 1, 2001. Given the statutory deadline, the ESCAP could
not recommend adjustment of the data. Last year the Clinton Administration had
announced the intention of releasing two sets of figures, unadjusted and adjusted.32
Concerns about insulating the decision on whether to adjust the redistricting
census figures from the “taint” of partisan politics led the Commerce Department, on
June 13, 2000, to propose a rule (1) establishing a process by which a committee of
senior career officials (ESCAP) of the Census Bureau would advise the Director of
the Census Bureau regarding a decision to adjust census data and (2) delegating33
authority for this decision from the Secretary to the Director. Additionally, if a


31Mae M. Cheng, Despite Undercount, Census Data to Stand; Adjustment advocates vow
legal challenges, NEWSDAY, March 7, 2001, at A6.
32Notice in 65 Fed. Reg. 38374, 38394-5 (June 20, 2000) (statement issued by Kenneth
Prewitt, then-Director of the Census Bureau, with a preliminary conclusion that adjustment
is feasible and expected to be more accurate than unadjusted figures; statement adopted by
then-Secretary of Commerce William M. Daley); see also statement of then-Secretary of
Commerce Daley on plan for Census 2000, Wednesday, February 24, 1999, at
[http://www.doc.gov/Opa/Speeches/censusplan.htm] (last visited Feb. 28, 2001).
3365 Fed. Reg. 38370 (June 20, 2000). In 1980, the Secretary of Commerce apparently
delegated the decision to the Director of the Census Bureau. In May 1980, then-Secretary
Philip M. Klutznick directed then-Director Vincent Barabba to “take direct personal charge”
of the process of examining the validity of adjustment methods and the desirability of making
an adjustment:
The culmination of this process should be a decision by the Director of the Census
Bureau on whether and how any statistical adjustment should be made to 1980
census data. . . . Even if there were some basis for an adjustment of the population
count to be used for apportionment of the House of Representatives, I do not
believe that any adjustment can be made prior to the statutory deadline for the
delivery of this information to the President. I do expect, however, that by the end
of the calendar year, or shortly thereafter, you will be prepared to announce a
decision on adjusting the census data for other uses. 45 Fed. Reg. 83110, 83117
(Dec. 17, 1980); 45 Fed. Reg. 69515, 69524 (Oct. 21, 1980).
The Director made the decision to not adjust the census data by the end of that year. 45 Fed.
(continued...)

decision were made to adjust the census data, the unadjusted data would also be
released, simultaneously with the report of the adjusted data, in accordance with P.L.
105-119, § 209(j), 111 Stat. 2440, 2483 (1997). If a decision were made to not
adjust, despite the ESCAP recommendation to adjust, the adjusted data would also
be released. The regulation was silent with regard to releasing adjusted data if
ESCAP recommended the release of unadjusted data. After the consideration of
comments from opponents and proponents of statistical adjustment, on October 6,34
2000, the final rule, 15 C.F.R. part 101, was issued. Apparently in response to some
of the comments, a new paragraph was added, 15 C.F.R. § 101.1(a)(5), which stated
that:
Nothing in this section diminishes the authority of the Secretary of
Commerce to revoke or amend this delegation of authority or relieves the
Secretary of Commerce of responsibility for any decision made by the
Director of the Census pursuant to this delegation. This section shall
remain in effect unless or until amended or revoked by the Secretary of
Commerce.
However, on February 16, 2001, Secretary of Commerce Donald L. Evans
announced a new final rule revoking the delegation of authority to make the
adjustment decision to the Director of the Census Bureau and the regulatory section
concerning the release of two sets of numbers.35 Proponents of adjustment viewed
this as preparation for a decision to not adjust and to release only one set of numbers
for redistricting, as in 1991. Accordingly, the City of Los Angeles, joined by other
municipalities, filed a suit on February 21, 2001, indirectly challenging the anticipated
policy by alleging that the new rule improperly revoked the previous rule in a manner
inconsistent with the requirements of the Administrative Procedure Act, 5 U.S.C. §§36
551 et seq. The plaintiffs alleged that the new rule was substantive and therefore
subject to a notice and comment period before it could be promulgated; that the new
rule results in a decision-making process governed by politics, not science; that the
new rule could result in a prohibition on the release of adjusted census data; and that
the plaintiffs would be irreparably harmed by all the above. The defendants argued,
inter alia, that the revocation of the earlier rule was a procedural, not substantive, act
and therefore not subject to notice and comment; that under the revoked rule, the
decision was in the hands of the Director of the Census Bureau, who is a political
appointee, as is the Secretary, and not a career Census Bureau professional; and that
the plaintiffs’ claims of irreparable harm were speculative and the claims of
deprivation of access to adjusted data were not ripe, because no decision had been
made concerning the release of adjusted data. The federal District Court for the
Central District of California denied a temporary restraining order to the plaintiffs and


33(...continued)
Reg. 82872-82885 (Dec. 16, 1980). For the decision in 1990, see text accompanying notes

39-41.


3465 Fed. Reg. 59713 (Oct. 6, 2000).
35This rule was published and took effect on Feb. 23, 2001. 66 Fed. Reg. 11231 (Feb. 23,

2001).


36City of Los Angeles v. Evans, No. CV 01-01671GAP (MCX) (C.D. Cal.).

set a hearing for March 5, 2001, concerning a preliminary injunction,37 but in light of
the recommendation of the Acting Census Director and the ESCAP report and
recommendation, the hearing was cancelled, while Los Angeles city officials and
statisticians pondered the ESCAP report.
Secretary Evan’s decision not to release adjusted data appeared to render the
lawsuit moot since the original rule promulgated on October 6, 2000, did not require
the release of adjusted data if the ESCAP recommended against it. The plaintiffs
appeared to have no standing since they could not show they were injured by the
Secretary’s action in revoking the original rule. However, in the wake of Secretary
Evan’s decision, the City Attorney of Los Angeles, Jim Hahn, announced that he
would pursue further legal action to obtain the adjusted census data by amending the
complaint to add new allegations that the Secretary’s decision is arbitrary and
capricious and violates section 195 of the Census Act, 13 U.S.C. § 195, which
requires the Secretary to use statistical sampling, “if he considers it feasible,” for38
purposes of the census other than apportionment of the House of Representatives.
In some ways, this seems like a repeat of litigation following the 1990 census.
For the 1990 census, the Secretary originally decided to not adjust the census data.39
After this decision resulted in lawsuits challenging the decision, the Secretary planned
to make the adjustment decision personally according to guidelines pursuant to a
court order in the main adjustment lawsuit at that time.40 Ultimately, following a
process pursuant to these guidelines, the Secretary decided to not adjust.41 The
litigation over this decision culminated in the U.S. Supreme Court decision in
Wisconsin v. City of New York.42 The Supreme Court held that the Secretary of
Commerce had acted, within his broad powers over the census, in a manner consistent
with the constitutional goal of equal representation in deciding not to adjust the 1990
census data, in light of the fact that there was no clear consensus among experts that
the adjusted data were superior to the unadjusted data. Although the Supreme Court
did consider the “goal of equal representation” to be important, it did not find that the
goal was violated by the use of unadjusted data. In Wisconsin, the Court did not
decide whether Congress had constrained the Secretary’s authority to decide not to


37Mae M. Cheng, L.A. Loses Round in Census Case; But suit, which NYC may join, is still
alive, NEWSDAY, Feb. 25, 2001, at A14.
38See press release of the Los Angeles City Attorney, Hahn declares that census legal fight
will continue, at [http://cityofla.org/atty/pressrel.htm] (last visited March 8, 2001).
39Clifford D. May, U.S. Rejects Pleas to Adjust 1990 Census for Undercount, NEW YORK
TIMES, Oct. 31, 1987, at A1, col. 5.
40See 55 Fed. Reg. 9838 (Mar. 15, 1990) (in response to a comment that the Secretary should
delegate final decision-making authority about adjustment to the Director, as in 1980, the
Commerce Department responded that the responsibility for the conduct of the census, under
the law, belonged to the Secretary, who would exercise the responsibility as he deemed most
appropriate); 54 Fed. Reg. 51002 (Dec. 11, 1989).
4156 Fed. Reg. 33582 (Jul. 22, 1991).
42517 U.S. 1, 116 S.Ct. 1091 (1996).

adjust the census since the issue was not raised by the parties to the case.43 As noted
above, in Department of Commerce v. U.S. House of Representatives, the Court
found that section 195 of the Census Act required the Secretary to use statistical
sampling techniques, if he considered it feasible, for purposes other than
apportionment of the House of Representatives. Proponents of sampling interpret this
as meaning that adjustment must be used for intrastate redistricting data, if feasible.
The issue to be raised by the City of Los Angeles, that the Secretary refused to adjust
although adjustment was feasible, has not been decided before, but the
recommendation of the Acting Census Director and the ESCAP would seem to
constitute a finding within the Census Bureau and the Commerce Department that
adjustment, within the statutory timetable for reporting redistricting data, was not
feasible.
Congressional Activity
Although the adjustment debate has moved to the state legislatures and the
federal courts for the moment, there is a possible congressional role should Congress
choose to intervene. Congress could legislatively require that the Census Bureau
make adjusted data available, whether or not it is the official data transmitted through
the P.L. 94-171 program, just as it has already required the release of unadjusted data
under P.L. 105-119, as noted above. Although Congress has not explicitly required
States to use federal decennial census data in congressional redistricting, it could
arguably do so under the same constitutional powers which give Congress the
authority to establish other redistricting guidelines if it chooses, Art. I, § 2, cl. 1,
which provides that the Members of the House of Representatives shall be chosen by
the People and Art. I, § 4, cl. 1, giving Congress the authority to determine the times,
places and manner of holding elections for Members of Congress. Where it is not
clear that one data set is more accurate than the other and the constitutional goal of
equal representation is not implicated, arguably, Congress could require that a
particular type of data, unadjusted or adjusted, be used in congressional redistricting.
However, it could not do so with regard to the redrawing of state legislative or
municipal districts, which remain the prerogative of the States as long as no
constitutional voting rights are violated. H.R.941, introduced by Rep. James Clyburn
on March 8, 2001, would require the use of adjusted census data in the administration
of any law of the United States under which population or population characteristics
is used to determine the amount of benefits received by State or local governments,
but would not use adjusted data for congressional redistricting or apportionment of
the House of Representatives.
Utah v. Evans
On January 10, 2001, the State of Utah filed suit against the Secretary and
Department of Commerce alleging that the defendants had unlawfully excluded
overseas missionaries of the Church of Jesus Christ of the Latter-Day Saints in
violation of the Census Clause (Art. I, § 2, cl. 3), the Free Exercise Clause of the First
Amendment, the Equal Protection Clause of the Fifth Amendment, and the Due


43517 U.S. at 19, note 11.

Process Clause of the Fourteenth Amendment (Amend. XIV, § 2) of the Federal
Constitution; of the Administrative Procedure Act (5 U.S.C. §§ 701 et seq.); of the
Religious Freedom Restoration Act (42 U.S.C. §§ 2000bb et seq.); of 2 U.S.C. § 2a;44
and of the Census Act (13 U.S.C. §§ 1 et seq.). On March 5, 2001, District Court
Judge Dee Benson issued an order denying the plaintiffs’ request for a three-judge
panel pursuant to 28 U.S.C. § 2284, finding that the suit did not involve a challenge
to a redistricting statute. Thus, it may take longer for the suit to reach a final
resolution, since a decision by a three-judge panel could have been appealed directly
to the U.S. Supreme Court, whereas now a decision would have to be appealed to the
U.S. Court of Appeals for the 10th Circuit before any potential hearing before the U.S.
Supreme Court. The House has delayed certification of the disputed seat until after
a District Court decision on March 20; North Carolina has joined the Federal
Government in opposing the suit by Utah.
With regard to census methodology, the leading relevant decision is that of45
Franklin v. Massachusetts, in which the United States Supreme Court upheld the
decision of the Secretary of Commerce to include and allocate overseas federal
employees in the census data for the apportionment of the House of Representatives,
which resulted in a loss of one congressional seat for Massachusetts. The Court held
that there was no final agency action reviewable under the Administrative Procedure
Act and that the allocation of overseas federal employees to their home states was
consistent with the “usual residence standard” of other censuses and furthered the
constitutional goal of equal representation. However, the issue of distinguishing
between overseas federal employees and other expatriate U.S. citizens by including
the former in the census and excluding the latter was not before the Court and was
not decided by it.
Congress could have a role by legislatively mandating the inclusion of all
expatriate U.S. citizens in the decennial census. One bill has been introduced by
Representative Maloney, H.R. 680, to provide funds for a special census of
expatriates and to urge the Census Bureau to plan for the inclusion of expatriates in46
the 2010 census. According to Franklin v. Massachusetts, the decision to include
overseas federal employees in the 1990 census appears to have been in response to
legislative activity in the late 1980s.
Additional Reading
CRS Info Pack, IP537C, Census 2000: background and issues, updated as needed,
by George Walser. Includes copies of the CRS Reports Census 2000: New
Legislative Proposals (RS20123), Census 2000: The Sampling Debate
(RL30284), and House Apportionment Following the 2000 Census (98-135).


44State of Utah v. Evans, No. F-2-01-CV-23:B (D. Utah 2001).
45505 U.S. 788 (1992).
46In the 106th Congress, S.Con.Res. 38 and H.Con.Res. 129, not adopted, expressed the sense
of Congress that all expatriates should be included in the 2000 census.

CRS Report RL30182, Census 2000: sampling as an appropriations issue in the

105th and 106th Congresses, by Jennifer D. Williams.


CRS Report RL30284, Census 2000: the sampling debate, by Jennifer D. Williams.
CRS Report 93-1060 GOV, Congressional redistricting: Federal law controls a
State process, by David C. Huckabee.
CRS Report 94-89 GOV, Decennial census coverage: the adjustment issue, by
Jennifer D. Williams.
CRS Report 97-871 A, Sampling for Census 2000: a legal overview, by Margaret
Mikyung Lee.
CRS Report RL30047, Sampling for Census 2000: Department of Commerce v.
United States House of Representatives and its ramifications, by Margaret
Mikyung Lee.
CRS Report 91-736 A, The Voting Rights Act of 1965: A Legal Overview, by L.
Paige Whitaker.