4 F.3d 1367

The CITY OF DETROIT, A Municipal Corporation; and Coleman
A. Young, Individually and as Mayor, City of
Detroit, Plaintiffs-Appellants,
v.
Barbara H. FRANKLIN, Secretary of Commerce of the United
States; and Barbara E. Bryant, Director of the
United States Bureau of the Census,
Defendants-Appellees.

No. 92-2113.

United States Court of Appeals,
Sixth Circuit.

Argued June 21, 1993.
Decided Sept. 22, 1993.

Ulysses W. Boykin, III (briefed), Lewis, White & Clay, Robert A. Sedler (argued), Detroit, MI, for plaintiffs-appellants.

Thomas W. Millet, David M. Glass, U.S. Dept. of Justice, Michael Jay Singer (briefed), U.S. Dept. of Justice, Appellate Staff, Civil Div., Mark B. Stern (argued), U.S. Dept. of Justice, Civil Div., Washington, DC, for defendants-appellees.

Before: KENNEDY and NORRIS, Circuit Judges; and ENGEL, Senior Circuit Judge.

KENNEDY, Circuit Judge.

1

Plaintiffs-Appellants City of Detroit and its mayor, Coleman A. Young, appeal the District Court's grant of summary judgment in favor of defendants, Director of the United States Bureau of the Census ("Census Bureau") and Secretary of Commerce, in this action challenging the results of the 1990 decennial census. Plaintiffs contend that the Census Bureau and Department of Commerce breached their constitutional obligation to make a tabulation of the population of each state and of each city and sub-unit within a state, that as accurately as possible includes all persons reasonably capable of being determined to reside therein. They assert that the census is deficient because (1) the manner in which the Census Bureau conducted the 1990 census in the City of Detroit resulted in a serious miscount (i.e., undercount of minorities) in that large numbers of persons actually resident in the City of Detroit were missed and so were not included in the official population count and (2) the Secretary of Commerce refused to make a statistical adjustment to the census headcount that would correct the disproportionate undercount of blacks that occurred in Detroit. The plaintiffs claim that as a result of this breach, the individual plaintiff's voting power was diluted relative to other parts of Michigan when the unadjusted census data was used for redistricting purposes and that the City of Detroit will lose a fair share of federal funds allocated on the basis of the census figures. Plaintiffs seek an injunction to compel the responsible parties to make an appropriate statistical adjustment for the racially differential undercount in the 1990 official population count of the cities and sub-units of the State of Michigan or, alternatively, to make a new count of certain areas. For the reasons set forth below, we affirm the grant of summary judgment.

I. Statement of the Case

A. Census Background

2

The Constitution requires Congress to conduct a decennial census for the purpose of determining the number of representatives in the United States House of Representatives to which each state is entitled. Article I, Sec. 2, cl. 3, of the Constitution provides that Representatives "shall be apportioned among the several States ... according to their respective Numbers," which requires, by virtue of section 2 of the Fourteenth Amendment, "counting the whole number of persons in each State." The number of persons in each State is to be calculated by "actual Enumeration," conducted every 10 years, "in such Manner as [Congress] shall by Law direct." U.S. Const., Art. I, Sec. 2, cl. 3. Congress, in turn, has delegated responsibility for taking the census to the Secretary of Commerce. The Bureau of the Census, an agency within the Department of Commerce, actually conducts the census.

3

The Census Bureau began to prepare for the 1990 decennial census in 1983. During six years of planning and preparation, the Bureau engaged in extensive consultations with data users, performed a series of "formal tests of alternative procedures and questionnaire content," and conducted a number of "dress rehearsal" censuses. While the parties may disagree on the quality of the census counts achieved in 1990 and the efficacy of the procedures used to conduct the census, the manner in which the census was conducted is largely undisputed.

4

First: the Census Bureau compiled an address control file ("ACF")--a computerized list of housing units to be counted. In constructing the list, the Census Bureau relied primarily on commercial mailing lists. The Census Bureau obtained its "Vendor List" for Detroit from ADVO-System, Inc., a large volume mailer of third-class bulk mail. After obtaining the "Vendor Lists," the Census Bureau conducted four precensus update operations to enhance the completeness and accuracy of the ACF. In the summer and fall of 1988, the United States Postal Service conducted an Advance Post Office Check ("APOC") in which it reviewed the addresses on the ACF to identify those that were undeliverable or incorrect. Then, the census enumerators canvassed census blocks looking for and adding census housing units not on the address list (the "Precanvass"). In Detroit, the Precanvass consisted of a complete recanvass of the entire city. The Census Bureau then conducted a "Precensus Local Review," in which the Bureau sent units of local government, including the City of Detroit, maps of the census blocks within their jurisdictions and preliminary counts of the number of housing units and group quarters contained therein. Where a local government asserted inaccuracies in the Census Bureau's data, census enumerators were sent to the challenged block to recheck the number of housing units.1 Finally, in March 1990, the United States Postal Service conducted a second review of the ACF (the "Casing Check") in which addresses on the updated lists were once again reviewed for deliverability and accuracy.

5

Second: census questionnaires were mailed to each housing unit. In March 1990, most of the housing units in Detroit contained in the ACF were mailed the census questionnaires which were to be mailed back to the Census Bureau (the "Mail Out/Mail Back" procedure). However, for certain low income/high density public housing projects designated by the City of Detroit, the questionnaires were delivered directly by census enumerators to the housing units to which they had been addressed. The Census Bureau also sent enumerators to a variety of other places thought unlikely to produce accurate results (e.g., blocks with boarded up housing; boarding houses; nursing homes; dormitories; rectories; convents; hospitals; YMCA/YWCA). The Census Bureau also embarked on an extensive follow-up campaign in which second mailings were sent to households that failed to return the initial form and in districts with especially low return rates, the Bureau remailed census forms to all residents. City of New York v. Department of Commerce, 822 F. Supp. 906 (E.D.N.Y.1993).

6

Third: The Census Bureau conducted a variety of supplemental enumeration efforts. The principal procedure used by the Census Bureau to count residents of housing units that did not return the completed questionnaires was the "Nonresponse Follow-up." For housing units that failed to return questionnaires, a census enumerator was assigned and directed to make up to six attempts to contact a household member and complete the questionnaire. Additionally, census enumerators personally visited all units reported as vacant or identified for deletion to verify their occupancy status. The Census Bureau also surveyed parole and probation officers to locate individuals (on parole or probation) who might have been missed in the original count.

7

The Census Bureau also mounted an extensive "Were you Counted?" campaign, in which they asked units of local government, the media, and community-based organizations to encourage people who believed they had been missed during the census enumeration to call one of the Bureau's toll-free numbers or report data for their household on special forms. Based upon the City of Detroit's participation, the Bureau received 88,525 "were you counted" forms that covered a total of 155,412 individuals. The processing of these forms (both with field verification and without) resulted in the addition of 47,109 people to the population of Detroit, enough to put the City over the one-million mark.2

8

In August 1990, the Census Bureau conducted a "Housing Coverage Check" in which they recanvassed select blocks based upon evidence generated from a computerized search of census questionnaires that identified communities or blocks displaying a low count.

9

Additionally, the Census Bureau conducted a "Postcensus Local Review" in which they again provided units of local government with updated maps of the census blocks in their communities and with data showing the preliminary results of the census enumeration. The Census Bureau sent enumerators to recanvass census blocks challenged by local governments. At this stage, the City of Detroit challenged the Census Bureau's count of housing units in 5,613 of the City's 13,364 census blocks. Thereafter, the Census Bureau conducted a postcensus recanvass covering more than 3,800 of the census blocks within Detroit.

B. Results of the 1990 Census

10

Despite the massive efforts of the Census Bureau, "it is undisputed that the 1990 Census was not--and could not realistically be--successful in its goal of achieving an exact count of the nation's population." City of New York, at 913. See also Tucker v. United States Department of Commerce, 958 F.2d 1411, 1412 (7th Cir.) ("The census takers try to account for everybody in America, but of course they don't succeed...."), cert. denied, --- U.S. ----, 113 S. Ct. 407, 121 L. Ed. 2d 332 (1992). As with previous censuses, the 1990 census resulted in a net national undercount (of about five million people), with blacks and other minorities undercounted to a greater degree than non-Hispanic whites. See Tucker, 958 F.2d at 1412 ("There is reason to believe ... that the undercount is not randomly distributed, but instead is concentrated in the poor, among whom blacks and Hispanics are disproportionately represented...."). Specifically, according to the figures derived from the Post-Enumeration Survey ("PES") that was conducted by the Census Bureau after the census headcount, the black undercount rate on the national level was 4.8%, compared to a 1.7% undercount rate for whites.3

11

In response to litigation filed by the States of California and New York in the Second Circuit, the Census Bureau agreed to conduct a comprehensive review of the undercount problem. On July 15, 1991, the Secretary concluded that an adjustment should not be undertaken because:

12

while adjustment by the best method available would increase the census totals, it would not significantly alter the apportionment of seats in the House of Representatives among the states, in part because there is overcounting as well as undercounting. After the dust settled, [Michigan's] representation would be unchanged, although California and Arizona would pick up a few seats at the expense of Pennsylvania and Wisconsin. Federal grant allocations might not be much affected either. Moreover, any attempt to make a statistical adjustment to the mechanical headcount would, by injecting judgmental factors ... open the census process to charges of political manipulation. And while a statistical adjustment for the undercount would undoubtedly improve the accuracy of the nationwide census total, there is no consensus among statisticians and demographers that it would make the state and district census totals--the level at which the adjustment would actually affect representation and funding--more accurate.

13

Tucker, 958 F.2d at 1413 (citations omitted).

14

Subsequent to this decision, plaintiffs filed suit in the United States District Court for the Eastern District of Michigan arguing that defendants violated the Constitution by undercounting the population of the City of Detroit. In their complaint, plaintiffs alleged two reasons for this undercount. First, despite elaborate safeguards, the Census Bureau made certain enumeration errors that were caused by circumstances peculiar to Detroit; and second, that the 1990 census undercounts minority populations generally and minority groups comprise a significant segment of Detroit's population. Plaintiffs sought a declaratory judgment that the official population of the City of Detroit for the 1990 decennial census, as reported by the Census Bureau, seriously underreported the actual population of the City of Detroit; that the Constitution and census laws require the Census Bureau to undertake affirmative efforts to correct the miscount and to make a statistical adjustment for the racially differential undercount or, alternatively, recount parts of the city door-to-door.

15

On August 14, 1992, the District Court granted the defendants' motion for summary judgment. The court found that plaintiffs lacked standing to challenge the undercount pursuant to this Court's ruling in Young v. Klutznick, 652 F.2d 617 (6th Cir.1981), cert. denied, 455 U.S. 939, 102 S. Ct. 1430, 71 L. Ed. 2d 650 (1982); that plaintiffs failed to state a constitutional violation of a cognizable right; and that, in any event, plaintiffs had failed to demonstrate a genuine issue as to any material fact regarding the government's good-faith efforts to count the population of Detroit to the maximum extent practicable. This timely appeal followed.

II. Standard of Review

16

This Court reviews a District Court's grant of summary judgment de novo, making all reasonable inferences in favor of the non-moving party. EEOC v. University of Detroit, 904 F.2d 331, 334 (6th Cir.1990). Summary judgment is appropriate only when there is no genuine issue of material fact and the moving party is entitled to summary judgment as a matter of law. Celotex Corp. v. Catrett, 477 U.S. 317, 322-23, 106 S. Ct. 2548, 2552, 91 L. Ed. 2d 265 (1986). The inquiry is "whether the evidence presents a sufficient disagreement to require submission to a jury or whether it is so one-sided that one party must prevail as a matter of law." Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 251-52, 106 S. Ct. 2505, 2512, 91 L. Ed. 2d 202 (1986).

III. Standing

A. Dilution of Voting Power

17

The City of Detroit and Mayor Young do not assert that alleged errors in the census have caused Michigan to be underrepresented in the House of Representatives. Instead, plaintiffs claim that the City of Detroit has been disproportionately undercounted relative to other (predominately white) areas of the state because the Census Bureau certified an official population count that is not adjusted for the differential undercount of blacks. Consequently, when the federal court created voting districts for the State of Michigan based upon the unadjusted 1990 census figures, the voting strength of the Detroit residents, such as Mayor Young, was diluted.4

18

As a threshold matter, this Court must determine whether the plaintiffs have standing to challenge the conduct of the defendants. In order for plaintiffs to establish the requisite standing, they must demonstrate "at an irreducible minimum" (1) that they personally have suffered an "injury-in-fact;" (2) that the injury can be traced to the challenged conduct of the defendants (causation); and (3) that the injury is likely to be redressed by a favorable judicial decision. Lujan v. Defenders of Wildlife, --- U.S. ----, 112 S. Ct. 2130, 2136, 119 L. Ed. 2d 351 (1992).

19

In Young v. Klutznick, 652 F.2d 617 (6th Cir.1981), cert. denied, 455 U.S. 939, 102 S. Ct. 1430, 71 L. Ed. 2d 650 (1982), these same plaintiffs challenged the 1980 census alleging precisely the same injury (dilution of voting power and loss of federal funds due to the census undercount). Unlike the case at hand, however, the suit in Klutznick was brought prior to the redistricting of Michigan's representative districts. This Court concluded that the plaintiffs lacked standing because they failed to satisfy the causation element. The Court held that dilution of plaintiffs' voting power was not caused by the Census Bureau's differential undercounting of minorities, but rather by the redistricting decisions of the Michigan state legislature. Specifically, this Court stated that "[i]f the Michigan legislature may adjust for the undercount, whether the injuries feared by plaintiffs do in fact arise does not depend upon defendants' actions. An independent third party, the Michigan state legislature, would play a necessary role in determining the effects of the census upon plaintiffs." Id. at 624. The Court went on to find that the state legislature was not constitutionally compelled to use the Bureau's census data.

20

Relying on Klutznick, the District Court found that the plaintiffs in this case likewise lacked standing because:

21

[T]he Census Bureau neither creates state and federal legislative districts nor prescribes legislative spending. Those duties are reserved to the Michigan state legislature and Congress. Therefore, the conduct of intervening third parties, not the actions of the defendants, are directly responsible for plaintiffs' claimed injuries.

22

City of Detroit v. Franklin, 800 F. Supp. 539, 542 (E.D.Mich.1992). Plaintiffs attempt to distinguish this case from Klutznick on the ground that in Klutznick the Michigan legislature had not yet redistricted so it was uncertain whether the legislature would, in fact, rely on the census data. Here, however, it is undisputed that the federal court, which redistricted Michigan's representative districts after the 1990 census, relied upon the unadjusted census data. Austin v. Good, 800 F. Supp. 557 (E. & W.D.Mich.1992) (three-judge court). This Court's decision in Klutznick, however, makes clear that if the Michigan legislature is not constitutionally compelled to use the Bureau's census data when redistricting, then its decision (whether to use the data or not) is an independent act breaking the chain of causation between the challenged actions of the Census Bureau (undercounting and failing to adjust for the undercount) and the injury to the plaintiffs. Therefore, the dispositive issue for this Court is whether the Michigan legislature (or the federal court which acted in its place) is constitutionally compelled to use the unadjusted headcount reported by the Census Bureau.

23

In Wesberry v. Sanders, 376 U.S. 1, 84 S. Ct. 526, 11 L. Ed. 2d 481 (1964), the Supreme Court construed the Apportionment Clause of the Constitution as requiring "as nearly as is practicable one man's vote in a congressional election [to be] worth as much as another's." Id. at 8, 84 S.Ct. at 530. Over time, the Supreme Court has refined the "nearly as is practicable" standard of Wesberry. In Kirkpatrick v. Preisler, 394 U.S. 526, 530-31, 89 S. Ct. 1225, 1229, 22 L. Ed. 2d 519 (1969), the Court held that this standard requires a "good faith effort to achieve precise mathematical equality." Then, in Karcher v. Daggett, 462 U.S. 725, 732, 103 S. Ct. 2653, 77 L. Ed. 2d 133 (1983), the Court "reaffirmed its dedication to the principle of exact mathematical equality [in determining the population of a state's electoral districts] and rejected the argument of the state of New Jersey that variations in population that are smaller than the margin of error embodied in the census figures should be regarded as de minimis." Good, 800 F. Supp. at 560.

24

In Karcher, the Supreme Court stated that "the census data provide the only reliable--albeit less than perfect--indication of the districts' 'real' relative population levels." Id. at 738, 103 S.Ct. at 2662. The Court further stated that "because the census count represents the 'best population data available,' it is the only basis for good-faith attempts to achieve population equality." Id. (citation omitted). We agree with the District Court's interpretation that:

25

The Court in Karcher did not hold that the states must use census figures to reapportion congressional representation. The Supreme Court 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 political representation. Nothing in the constitution or Karcher compels the states or Congress to use only the unadjusted census figures. Karcher does not overrule Young [v. Klutznick].

26

City of Detroit, 800 F. Supp. at 543 (citation omitted). See also Senate of State of California v. Mosbacher, 968 F.2d 974, 979 (9th Cir.1992) ("If the State knows that the census data is underrepresentative, it can, and should, utilize noncensus data in addition to the official count in its redistricting process."); Tucker, 958 F.2d at 1419 ("Not only because population shifts over a ten-year period, but also because population and voting population are not the same, states are not required to use census figures for the apportionment of their legislatures") (citation omitted). If figures other than the census count are the best population data available, the Supreme Court did not, in Karcher, bar their use. Consequently, plaintiffs cannot premise standing on their underrepresentation claim.

B. Funding

27

Plaintiffs do, however, have standing to challenge the defendants' actions based upon their claim that the census undercount will result in a loss of federal funds to the City of Detroit. It is well established that when considering whether a complaint provides the plaintiffs with standing, "both the trial and reviewing courts must accept as true all material allegations of the complaint in favor of the complaining party." Warth v. Seldin, 422 U.S. 490, 95 S. Ct. 2197, 45 L. Ed. 2d 343 (1975). Here, plaintiffs allege that:

28

The Congress has established numerous programs of grants-in-aid to states, cities and other sub-units that are based on the population of the states, cities, and sub-units, as contained in the official population count prepared and certified by the Bureau in connection with the decennial census and/or the mid-year census authorized under 13 U.S.C. sec. 141(d)(e). Such programs, under which the City of Detroit receives substantial amounts of federal funds, include Community Development Block Grants, Rental Housing Rehabilitation, Emergency Shelter grants, and Urban Mass Transit Capital and Operating Assistance. As a consequence, the amount of federal funds to be received by plaintiff, CITY OF DETROIT, in the years following the 1990 decennial census will be affected substantially by the official population count of the City of Detroit that is prepared and certified by the Bureau.

29

Joint App. at 11-12. Defendants argue, and the District Court found, that plaintiffs cannot premise standing on their claim that Detroit's access to federal funds will be impaired by the alleged undercount of the City's population because Congress is not required to allocate federal funds to states and localities on the basis of population. FAIR v. Klutznick, 486 F. Supp. 564, 569 n. 9 (D.D.C.1980).

30

It is undisputed, however, that many federal programs do disburse funds based upon population figures as reported in the decennial census. See 13 U.S.C. Sec. 141(e)(1); Tucker, 958 F.2d at 1415 ("there is no doubt that, as a matter of fact, the allocation of state and federal funds is heavily influenced by census figures"). Consequently, several courts have reasoned that although "the Census Bureau and the Department of Commerce are not in charge of distribution of federal funds, the Bureau's actions significantly affect the distribution of funds and therefore satisfies the requirements for injury in fact." State of Texas v. Mosbacher, 783 F. Supp. 308 (S.D.Texas 1992). In Carey v. Klutznick, 637 F.2d 834 (2d Cir.1980), rev'd on other grounds, 653 F.2d 732 (2d Cir.1981), the city and mayor of New York challenged the manner in which the Census Bureau conducted the 1980 census in the State of New York, claiming the census had resulted in an undercount and a subsequent loss of federal funds. In conferring standing, the Second Circuit stated:We agree with the court in City of Camden v. Plotkin, 466 F. Supp. 44, 47-51 (D.N.J.1978), that citizens who challenge a census undercount on the basis, inter alia, that improper enumeration will result in loss of funds to their city have established both an injury fairly traceable to the Census Bureau and a substantial probability that court intervention will remedy the plaintiffs' injury. The present case is distinguishable from FAIR v. Klutznick ..., where no plaintiff alleged that his or her own state or congressional district would benefit if the relief sought were granted. The individual plaintiffs in this case have alleged concrete harm in the form of dilution of their votes and decreased federal funds flowing to their city and state, thus establishing their standing.

31

Carey, 637 F.2d at 838. Similarly, in City of Willacoochee, Ga. v. Baldrige, 556 F. Supp. 551 (S.D.Ga.1983), the city and its mayor commenced an action challenging the accuracy of the population count of the city as reported in the 1980 decennial census. The plaintiffs claimed that the inaccurate population count would deprive them of their share of the proceeds of two federally-funded programs which distributed funds on the basis of population data. The District Court found that the city had established the necessary standing since the loss of federal funds due to an inaccurate population count "would be a distinct and palpable injury" to the city. Id. at 554. Additionally, in State of Texas, the District Court reasoned that:

32

The State of Texas ... has not asserted through its pleadings a direct injury and therefore, does not have standing as a recipient of federal funds. In Carey ..., the State of New York alleged direct loss due to revenue sharing funds. Although citizens of Texas may suffer a direct injury because of this, the State of Texas will not suffer in this same direct way. Absent an allegation of a direct loss, the State does not show injury-in-fact. The claimed loss of federal funding will only be actionable against the United States if it was brought by plaintiffs who could show an actual direct loss of funds due to an undercount.

33

783 F.Supp. at 314 (footnotes omitted and emphasis added). Taking the allegations of the complaint as true, plaintiffs have established that they were directly affected by the action of the defendants as that conduct affected their city and their funds. See Ridge v. Verity, 715 F. Supp. 1308, 1318 (W.D.Pa.1989).

IV. Census Methodology

34

Having found standing, this Court must next determine whether plaintiffs have stated a violation of any judicially enforceable constitutional right. Article I, Sec. 2, cl. 3 requires that "Representatives ... shall be apportioned among the several states ... according to their respective numbers," and that the "actual enumeration shall be made ... within every subsequent Term of ten years, in such Manner as [the Congress] shall by law direct." Plaintiffs come close to arguing that this provision creates a constitutional right to census accuracy. See Senate of State of California, 968 F.2d at 978 ("The Senate argues that because the Constitution provides for a census based upon an 'actual Enumeration' of the people in this country, there is a right to an accurate count"); Tucker, 958 F.2d 1411 (court rejected argument that decisions mandating equality in legislative districts provide a constitutional foundation for good-faith disputes regarding the accuracy of the census count). Specifically, plaintiffs contend that the apportionment clause requires the Census Bureau to make a tabulation of the population of each state and of each city and sub-unit within a state that as accurately as possible includes all persons reasonably capable of being determined to reside therein.5 Plaintiffs do not claim that the Census Bureau deliberately set out to undercount the City of Detroit in the 1990 decennial census, but rather contend that the way in which the census was conducted in the city resulted in a substantial undercount of the residents of Detroit with its large minority and low income population. A similar argument was considered in Tucker, wherein the Seventh Circuit dealt with an attempt by plaintiff residents to force the Secretary of Commerce to adjust the census data for much the same reasons as those asserted here. In finding judicial review improper, the court said:

35

It might be different if the apportionment clause, the census statutes, or the Administrative Procedure Act contained guidelines for an accurate decennial census, for that would be some evidence that the framers of these various enactments had been trying to create a judicially administrable standard. There is nothing of that sort, and the inference is that these enactments do not create justiciable rights. The Constitution directs Congress to conduct a decennial census, and the implementing statutes delegate this authority to the Census Bureau. There is little more to the statutes--they specify a timetable, and a procedure for translating fractional into whole seats--but they say nothing about how to conduct a census or what to do about undercounts. So nondirective are the relevant statutes that it is arguable that there is no law for a court to apply in a case like this--that you might as well turn it over to a panel of statisticians and political scientists and let them make the decision, for all that a court could do to add to its rationality or fairness.

36

Id. at 1417-18 (citations omitted).

37

The Constitution directs Congress to conduct a decennial census in such manner as they shall by law direct. Congress in turn has delegated this task to the Secretary of Commerce "in such form and content as he may determine." 13 U.S.C. Sec. 141(a). Despite this deferential language, it is clear that if Congress or the Census Bureau failed (altogether) to conduct a decennial census or failed to conduct a good-faith enumeration, or intentionally reduced some group's representation or funding, these claims would be reviewable by a federal court. However, we agree that "merely by directing congressional apportionment in accordance with the decennial census, Article I, section 2, clause 3 does not authorize lawsuits founded on disagreement with the Census Bureau's statistical methodology." Tucker, 958 F.2d at 1418.

38

This is especially true in a case like this, where the Census Bureau's efforts to conduct as accurate a census count as possible are extraordinary. According to one court, the 1990 census is said to be one of the best ever taken in this country because despite our large population, approximately 98 percent of the population was counted. Senate of the State of California, 968 F.2d at 979. Unfortunately, the errors were not distributed proportionately (i.e., larger numbers of minorities were missed than whites). After reviewing the record, it is difficult to envision (at least with respect to the City of Detroit) what more the Census Bureau could have done to further eradicate these errors.

39

Plaintiffs, however, point to several alleged deficiencies in the Census Bureau's procedures. For example, plaintiffs contend that because the Census Bureau based its address control file ("ACF") on an address list it purchased from a commercial vendor, it did not make a good-faith effort to count, as nearly as practicable, all Detroit residents, for many low income people would not appear on such a list. Second, plaintiffs contend that the APOC failed to account for substantial pockets of low income people, because the post office allegedly does not deliver mail to a number of areas in Detroit. Plaintiffs contend that the accuracy and incompleteness of the ACF is demonstrated in part by a telephone survey conducted by Michigan Consolidated Gas Company ("MichCon") which allegedly identified a number of areas in the City of Detroit where people were living but did not receive census forms because their addresses were not on the ACF. As a result, plaintiffs argue that the Secretary should conduct a new door-to-door census of every building in certain parts of Detroit to see if there is anyone who was not counted. The fundamental problem with plaintiffs' argument is that MichCon took the survey before the Census Bureau had completed the census. As the District Court found:

40

[T]he Census Bureau did not finish counting Detroit residents for several months after MichCon made its telephone survey. The Census Bureau still had not finished the Nonresponse Follow-Up.... Nor had the Census Bureau completed the "Were You Counted?" campaign.... Nor had the Census Bureau conducted the Housing Coverage Check.... Finally, the Census Bureau had still not undertaken the Postcensus Local Review....

41

City of Detroit, 800 F. Supp. at 545. Accordingly, we find that plaintiffs have failed to put forth evidence creating any genuine issues of material fact. Despite plaintiffs' self-serving contentions to the contrary, it is clear that defendants have counted the residents of the City of Detroit as accurately as possible. The Census Bureau has not undertaken a door-to-door campaign since the 1960 census and plaintiffs have presented no evidence indicating that such an effort would lead to any more accurate results. Plaintiffs cannot seriously be arguing that a perfectly accurate count of upwards of 250 million people (or nine-plus million in Michigan) was feasible, and nothing in the Constitution compels such an accurate result.

V. Failure to Adjust for the Undercount

42

Plaintiffs' final argument is that, at the very least, a genuine issue of material fact exists as to whether, within the state of Michigan, an adjustment for the racially differential undercount would produce an official population count that, on balance, would be more accurate than an official population count based on the raw unadjusted data alone. Plaintiffs describe this statistical adjustment claim as the "Michigan version" of the national statistical claim litigated in City of New York v. United States Department of Commerce, 739 F. Supp. 761 (E.D.N.Y.1990). Just recently, the court in City of New York upheld the Secretary's decision not to adjust the census data, finding that this decision did not violate the Administrative Procedure Act, the Constitution, any statute, or the stipulation previously agreed to by the parties to the litigation. City of New York v. United States Department of Commerce, 822 F. Supp. 906 (E.D.N.Y.1993).

43

The court framed the question for review as "whether the Secretary's application of the decision guidelines as construed in light of constitutional requirements, to reject the [adjusted] counts is arbitrary and capricious." Id. at 919.6 The court recognized that "[t]he conclusion that the Secretary must provide the most accurate census practicable ... does not lead inexorably to the conclusion that a decision against adjustment is therefore unconstitutional." Id. After reviewing the Secretary's consideration of the applicable guidelines, the court found that the decision not to adjust was not arbitrary and capricious.

44

Throughout his decision not to adjust, the Secretary explained that while the adjustment methodology was relatively successful in reflecting the total national population in absolute numbers, it was seriously flawed in portraying the distribution of the population among the states and localities. The Secretary concluded that there was simply not enough convincing evidence to support a finding that the adjusted counts would lead to greater distributive accuracy than the census counts. As the Secretary stated, "[w]hile we know that some will fare better and some will fare worse under an adjustment, we don't really know how much better or how much worse." Additionally, the Secretary's decision also forecloses plaintiffs' argument that an adjustment could be made for the population of Michigan alone.7 As the District Court concluded:In the same way that a national adjustment would create winners and losers, so too would a statewide statistical adjustment. The Secretary's statements make clear that a statistical adjustment would cause a mere shifting of unavoidable inequities and not a net improvement in accuracy. Plaintiffs have failed to provide any evidence to the contrary.

45

City of Detroit, 800 F. Supp. at 547. Plaintiffs' request for a statistical adjustment for Michigan alone cannot be considered in isolation. It would be impractical for this Court to order (the Secretary to do) such an adjustment without considering the consequences for the other states and other cities. Thus, what plaintiffs' request really comes down to is a nationwide adjustment, the prospect of which has already been rejected. City of New York, 822 F. Supp. 906.

46

This is another case in which "[t]he plaintiffs' claim to a census adjustment invokes no judicially administrable standards. The plaintiffs are not asking us to decree equality. They are asking us to take sides in a dispute among statisticians, demographers, and census officials concerning the desirability of making a statistical adjustment to a census headcount." Tucker, 958 F.2d at 1418. This the Court declines to do.

VI.

47

Accordingly, the order of the District Court granting defendants' motion for summary judgment is AFFIRMED.

1

During this stage, the City of Detroit claimed, on the basis of its Assessor's Files, that there were 436,336 housing units in the city, compared to the 401,279 units listed by the Census Bureau. The Bureau reviewed the City's estimates and also conducted a precanvass review of more than 1,600 of the City's 13,364 census blocks

2

Of the remaining 107,303 individuals who were not added, 86,462 were individuals whom the Bureau had already counted while the other 20,841 were individuals whose existence the Bureau could not verify

3

The official population count of the City of Detroit, as reported by the Secretary of Commerce, was 1,027,974, representing 777,916 blacks and 250,058 non-blacks (giving the City of Detroit a black population of 75.7%). The black population for the State of Michigan totalled 1,291,706 out of a total population of 9,295,297 (giving the State of Michigan a black population of 13.9%)

Moreover, applying the percentages of the national Post-Enumeration Survey, Michigan's undercount would total 108,708, with the undercount for Detroit totalling 37,026.

4

As plaintiffs explain:

[S]ince Michigan, with an officially reported population count of 9,295,297, is entitled to 16 seats in the House of Representatives, each Congressional district within the State of Michigan should have approximately 580,956 persons. The officially reported population count of the City of Detroit is 1,027,974, so that, as reported, the City of Detroit would embrace 1.769 Congressional districts. If an adjustment for the racially differential undercount had been made in accordance with the figures derived from the [PES], Michigan's total population would be approximately 9,404,000.... With an adjusted population count of 1,065,000 according to the [PES], instead of an official population count of 1,027,974, the City of Detroit would embrace 1.811 Congressional districts instead of 1.769. Thus, the use of the unadjusted official population count to determine Congressional representation within the State of Michigan means that the residents of the City of Detroit have proportionately less representation in the House of Representatives than the residents of all or virtually all of the other cities and sub-units in Michigan.

5

Plaintiffs' argument is similar to that raised in Tucker, in which Illinois residents claimed that the Census Bureau violated the apportionment clause of the Constitution as that provision "implicitly command[s] the Bureau to make whatever adjustments in the raw census totals are necessary to make the census the best feasible estimate ... of the American population." Tucker, 958 F.2d at 1415

6

The court distinguished its case from Franklin v. Massachusetts, --- U.S. ----, 112 S. Ct. 2767, 120 L. Ed. 2d 636 (1992), wherein the Supreme Court held that the Secretary's acts in conducting the census and reporting the counts to the President were not "final" for purposes of challenging apportionment, because the Secretary simply reports the results to the President, who in turn transmits the apportionment for each state in the House of Representatives to the Clerk of the House. The City of New York case, however, like the case at bar, involved a challenge to the counts as used for intra-state redistricting and federal fund allocation, neither of which requires the Secretary to transmit the counts to the President. Thus, the Secretary's reporting of the counts for these purposes is a final agency action

7

In its decision, the Secretary points out that:

In attempting to make the actual count more accurate by an adjustment, we might be making the shares less accurate. The shares are very important because they determine how many congressional seats each state gets, how political representation is allocated within states, and how large a "slice of the pie" of federal funds goes to each city and state. Any upward adjustment of one share necessarily means a downward adjustment of another. Because there is a loser for every winner, we need solid ground to stand on in making any changes. I do not find solid enough ground to proceed with an adjustment.

Adjustment Decision, 56 Fed.Reg. 33,582, (July 22, 1991).