Civ. A. No. 80-71330.
September 25, 1980.
Goodman, Eden, Millender Bedrosian by Ernest Goodman, James A. Tuck, Detroit, Mich., Special Counsel, for the City of Detroit.
George W. Crockett, Jr., Joseph N. Baltimore, Detroit, Mich., Robert A. Sedler, Wayne State University School of Law, Detroit, Mich., for plaintiffs.
Alice Daniel, Asst. Atty. Gen., Dept. of Justice, Washington, D.C. Mark C. Rutzick, William Z. Elliott, Sheila Lieber, Dept. of Justice, Washington, D.C. for defendants.
John J. Gunther, Executive Director, Stephen C. Chapple, Gen. Counsel, U.S. Conference of Mayors, Washington, D.C. Jim Leach, Member of Congress, First District of Iowa, Washington, D.C. for amicus curiae.
Can a headcount of the persons in the United States, made on the 1st of April, 1980, and unadjusted for an acknowledged undercount, be used for the reapportionment of the United States House of Representatives, and the distribution of Federal funds among states and sub-state units of government? That is the question in this case. The answer is clearly no.
Since 1790, with the exception of 1970 when nearly five million people were imputed and added to the headcount, the population of the United States, and its various governmental units has been determined by an unadjusted headcount, made by various means. In the past several censuses, this has resulted in an undercount of the population. This undercount has been the greatest among minorities-particularly blacks and Hispanics-and particularly blacks and Hispanics living in large urban areas. In 1960 and 1970, four times as many blacks were not counted as whites, and in 1970 this occurred even after the imputation of nearly five million people. In 1980, it is predicted that the same differential undercount will occur. Additionally, it is estimated that 8 percent of the Hispanics will not be counted. This lawsuit seeks to require an adjustment of that undercount.
The plaintiffs are Coleman Young, individually and as Mayor of the City of Detroit, and the City of Detroit. The defendants are Philip M. Klutznick, Secretary of Commerce, and Vincent P. Barabba, Director of the Census. The United States Conference of Mayors has filed a brief amicus in support of the plaintiffs, and Congressman Jim Leach of Iowa has filed one in support of defendants.
It is conceded that the way in which the Bureau of the Census goes about determining the number of people who reside in the United States on a particular day in each decade results in missing a great number of people. A disproportionate number of these individuals are blacks and minorities. In 1970, by the Census Bureau's own estimate, nearly 10.2 million persons were not personally contacted during the count. This represented 5 percent of the population. The Bureau estimated that 1.9 percent of the whites and 7.7 percent of the Blacks were not counted. In 1970, of this 10.2 million, 4.9 million who were not actually counted were included in the official count by "imputation" and allocated among the states for purposes of apportioning congressional seats. Still, approximately 5.3 million people were not included in the census figures.
Plaintiffs claim that, unless there is an adjustment for the undercount, individuals who live in urban areas with high black populations and those who live in areas with high Hispanic populations will be deprived of their right of equal representation in the House of Representatives of the United States, and in the State Legislature of Michigan, and will lose substantial sums of money, in that more than 100 federally funded programs are tied in one way or another to census figures.
The constitutional theory of plaintiffs' claim is that Article 1, Section 2, Clause 3 of the United States Constitution, providing for the taking of a decennial census, requires that the official decennial population count reflect as accurately as is reasonably possible the true population of the states, cities, and other governmental sub-units within each state. Plaintiffs claim, and it is not seriously disputed, that an official population count, using only raw, unadjusted census data, does not accurately reflect the actual population of the states, and governmental units within the states. Furthermore they claim-and this is not substantially disputed-that about four times as many blacks as whites are undercounted in urban areas, and an equal number of Hispanics are undercounted in the areas where they are located throughout the United States.
Defendants, on the other hand, contend that since 1790 the decennial census of the United States has consisted of a straightforward headcount of all persons residing within the United States. They claim that the goal of the census is to be as accurate as reasonably possible, but that there is always inherently less than absolute accuracy since a relatively small number of persons are missed in the census process. They say that since the first census was taken in 1790, these persons have not been counted as part of the population base used for apportionment of members of the House of Representatives among and within the states, and that Article 1, Section 2, Clause 3 of the Constitution requires this result, since, by its terms, apportionment can be based only on persons who are actually enumerated in the census.
The case came to trial in late August, and testimony was taken for several days. Briefs were filed, and final argument was held on September 10, 1980.
At the outset, it must be noted that the decennial census is a process, or in the words of the Supreme Court, an event. Gaffney v. Cummings, 412 U.S. 735, 93 S.Ct. 2321, 37 L.Ed.2d 298 (1973). The taking of a decennial census is authorized and directed by the express language of the Constitution in Article 1, Section 2, Clause 3, which, in pertinent part, provides:
"Representatives and direct Taxes shall be apportioned among the several states which may be included within this Union, according to their respective Numbers . . . The actual Enumeration shall be made within three Years after the first Meeting of the Congress of the United States, and within every subsequent Term of ten Years, in such Manner as they shall by Law direct . . ."
The very concept of a decennial census arose out of the attempt of the framers of the Constitution to preserve the equality of vote for the House of Representatives. The adoption of the proposal for the taking of the census occurred on the heels of the "Great Compromise" that gave birth to the United States Constitution. As the Supreme Court pointed out in Wesberry v. Sanders, 376 U.S. 1, 13, 14, 84 S.Ct. 526, 533, 11 L.Ed.2d 481 (1964):
"The debates at the Convention made at least one fact abundantly clear: that when the delegates agreed that the House should represent `people' they intended that in allocating Congressmen the number assigned to each State should be determined solely by the number of the State's inhabitants. The Constitution embodied Edmund Randolph's proposal for a periodic census to ensure `fair representation of the people, an idea endorsed by Mason as assuring that `numbers of inhabitants' should always be the measure of representation in the House of Representatives."
A long line of cases rooted in the Constitution and its history has established that the right to vote is one crucial to the preservation of our democratic society, and that such right cannot be abrogated. In Reynolds v. Sims, 377 U.S. 533, 555, 84 S.Ct. 1362, 1378, 12 L.Ed.2d 506 (1963) the Supreme Court stated:
See: Ex Parte Yarbrough, 110 U.S. 651, 4 S.Ct. 152, 28 L.Ed. 274 (1883); United States v. Mosley, 238 U.S. 383, 35 S.Ct. 904, 59 L.Ed. 1355 (1914); Gomillion v. Lightfoot, 364 U.S. 339, 81 S.Ct. 125, 5 L.Ed. 2d 110 (1960); Gray v. Sanders, 372 U.S. 268, 83 S.Ct. 801, 9 L.Ed.2d 821 (1962).
"The right to vote freely for the candidate of one's choice is the essence of a democratic society, and any restrictions on that right strike at the heart of representative government. And the right of suffrage can be denied by a debasement or dilution of the weight of a citizen's vote just as effectively as by wholly prohibiting the free exercise of the franchise."
Thus the necessity for an accurate population count among and within states is inexorably tied to fair apportionment of congressional seats. That is what this case is all about. The issue is as simple as that.
A threshold issue is that of standing. In ruling upon defendant's motion to dismiss filed earlier in this case, Judge Guy of this Court determined that the plaintiff had standing. Young v. Klutznick (No. 80-71330, May 29, 1980). In that opinion, he stated, inter alia (pages 14 15):
"While the original concern of the `capital framers' was with representation of one state vis-a-vis another, the reapportionment cases clearly extend the requirement to equal representation within the state. As the Court noted in Wesberry v. Sanders, 376 U.S. [1, 84 S.Ct. 526, 11 L.Ed.2d 481] (1963): `[I]t would defeat the principles solemnly embodied in the Great Compromise-equal representation in the House for equal numbers of people-for us to hold that, within the states, legislatures may draw the lines of congressional districts in such a way as to give some voters a greater voice in choosing a Congressman than others. The House of Representatives, the Convention agreed, was to represent the people as individuals and on a basis of complete equality for each voter.'
"Logically, these principles can be extended to that branch of the Federal Government which is responsible for collecting the figures upon which districting is based."
Defendants, in their post-trial brief, contend that Judge Guy suffered from the misconception that states were not free to adjust the census data, and that the states were, in fact, obligated to use that data. They rely on that portion of Judge Guy's Opinion where he states at page 17:
"If it were possible for the State Legislature to adjust census figures before engaging in congressional and state legislative redistricting, then, of course, the argument plaintiffs would have would lie before the Legislature of the State of Michigan."
Further, on oral argument, defendants contended that 13 U.S.C. § 141(c) demonstrates that Congress did not impose a mandatory duty on the Bureau of Census to develop census data for regional areas within a state, and that the Constitution does not impose such a duty.
The standing issue involves a two-part inquiry: has the plaintiff alleged an injury in fact, economic or otherwise, and is there a substantial likelihood that relief requested will redress the injury claimed? See Duke Power Co. v. Carolina Environmental Study Group, 438 U.S. 59, 98 S.Ct. 2620, 57 L.Ed.2d 595 (1978). In Duke Power Co. the Court said at page 72, 98 S.Ct. at page 2630:
"The essence of the standing inquiry is whether the parties seeking to invoke the court's jurisdiction have `alleged such a personal stake in the outcome of the controversy as to insure that concrete adverseness sharpens the presentation of the issues upon which the court so largely depends for illumination of difficult constitutional questions' . . . As refined by subsequent reformulation, this requirement of a `personal stake' has come to be understood to require not only a `distinct and palpable injury' to the plaintiff . . . but also a `fairly traceable' causal connection between the claimed injury and the challenged conduct."
Here, in essence, defendants claim that the injury complained of by plaintiffs is not fairly traceable to defendants' acts or omissions.
This Court is not persuaded by this argument. Wesberry v. Sanders (supra) applied the constitutional standard for apportionment of representation among the several states to Congressional redistricting within the states. In so doing, the Court enunciated the following standard, (376 U.S. pages 7 8, 84 S.Ct. page 530):
"We hold that, construed in its historical context, the command of Art. 1 § 2 that Representatives shall be chosen `by the People of the several States' means that, as nearly as is practicable, one man's vote in a congressional election is to be worth as much as another's."
In the context of Congressional apportionment among the states, it is clear that the census serves as a vehicle through which a system of representation, based upon conjecture, is rejected, and equality of representation is, in a substantial measure, achieved. It would be anomalous to conclude that the framers of the Constitution intended only to insure equality of representation among the states, but not within the states, and intended to allow the states to independently select some other means to establish equality. The principal purpose of the census is to insure fair representation of the people.
During the Constitutional Convention, Edmund Randolph observed that the states were too interested to properly conduct the census. He said, "The census must be taken under the direction of the General Legislature. The states will be too much interested to take an impartial one for themselves." 1 Farrand, Records of the Federal Convention 580 (Yale Univ.Press 1911). Therefore, the obligation of overseeing the census was relegated to Congress as an exclusively federal function. His observation is no less true today. Just as the states have a strong interest in the number of seats in the House, so, too, the states have a strong interest in the relative strength of the vote between different and often times competing geographical areas within the states.
Moreover, the Supreme Court has implied that, in the first instance, the states must rely on census data for the purpose of Congressional redistricting. In Kirkpatrick v. Preisler, 394 U.S. 526, 89 S.Ct. 1225, 22 L.Ed.2d 519 (1968) the State of Missouri argued that its apportionment plan must be based upon considerations of population shifts subsequent to the 1960 census. The redistricting was done in 1967, and the census had been taken in 1960. The Supreme Court rejected this argument. It said: (page 535, 89 S.Ct. page 1231):
"We recognize that a congressional districting plan will usually be in effect for at least 10 years . . . Situations may arise where substantial population shifts over such a period can be anticipated. Where these shifts can be predicted with a high degree of accuracy, States that are redistricting may properly consider them."
The Court held that, since it could not be demonstrated that population trends in Missouri had been thoroughly documented and applied throughout the state in a systematic manner, Missouri was required to redistrict in 1967 solely on the basis of the 1960 census figures.
It follows that Missouri will be required to redistrict in 1982 on the basis of the 1980 census figures. So will Michigan and every other state in the Union. Article 1, Section 2, Clause 3 of the Constitution so commands, so that there be a proper and federally certified base for redistricting in accordance with strict mathematical equality. It is clear that the framers of the Constitution could not have intended that the census provide the standard for apportionment among the states while simultaneously intending that a different and possibly conflicting assessment of state population govern apportionment within the state.
In Prigg v. The Commonwealth of Pennsylvania, 16 Pet. 539, 618, 619, 41 U.S. 539, 618, 619, 10 L.Ed. 1060 (1842), the Supreme Court observed that, although the power to apportion representatives amongst the several states is not expressly granted to the Congress, such power logically flows to Congress as a result of its express authorization to provide for an enumeration of population every ten years. Here too, the requirement that the states employ the official population count, as established by the decennial census has "always been acted upon as irresistibly flowing from the duty positively enjoined by the Constitution."
In Dixon v. Hassler, 412 F. Supp. 1036 (W.D.Tenn. 1976) a three-judge District Court held that whether or not Census Bureau estimates effective as of July 1, 1974, could be considered in a Congressional apportionment case, the proof in the case did not show that such estimates were the best evidence of the relevant population, and, therefore, the 1970 decennial census figures must be followed.
It is therefore clear that the Constitution commands that the data taken from the decennial census govern apportionment within the states. Just as apportionment among the states cannot be carried out without federally certified numbers, the redistricting of congressional districts must necessarily take place on the basis of census data. Such redistricting requires federally certified numbers.
All of the 50 states are required to use census data for the purpose of congressional redistricting.
Plaintiffs have alleged injury in fact, both economic and political, if there is no adjustment of the census figures, and clearly there is a substantial likelihood that the relief requested will redress the injury claimed. Thus, plaintiffs have standing to maintain this action.
Defendants again contend, as they did before Judge Guy, that if the Constitution permits a statistical adjustment for apportionment of representatives, then Congress has express constitutional authority to decide whether to require such an adjustment, such decision being an unreviewable political question that this Court cannot resolve. Judge Guy pointed out in his opinion on defendants' motion to dismiss ( Young v. Klutznick, supra) (page 17):
". . . Defendants' counsel conceded in oral argument that the states would not be free to adjust the census data for purposes of congressional districting within the state. That alone would seem to dictate that this court is an appropriate forum that plaintiffs could look to in advancing their argument. If it were possible for the State Legislature to adjust census figures before engaging in Congressional and State Legislative redistricting, then of course the argument plaintiffs would have would lie before the Legislature of the State of Michigan . . . For the purpose of this motion, however, the court must conclude on the basis of the representations of the parties, that, at least insofar as Congressional districting is concerned, the State Legislature has no discretion in terms of adjusting raw census data, and, therefore, this is an appropriate forum in which to address these complaints."
Nothing that has been subsequently heard by this court, either in testimony, briefing, or oral argument, requires a different conclusion.
Defendants argue that the concept of non-justiciability of political questions is founded primarily on the doctrine of separation of powers and the policy of judicial self-restraint. They claim that the term "political question" refers to an issue which, according to the Constitution, must be resolved by means the Constitution has specifically committed to a political department rather than the judiciary. They then argue that Article 1, Section 2, Clause 3 of the Constitution mandates, rather than empowers, Congress to make enumerations, and explicitly requires that Congress determine the manner in which such enumerations are to be made. They claim the Constitution contains a "textually demonstrable commitment" of all census matters to the Congress of the United States, and not to the courts.
Strangely enough, defendants rely upon Baker v. Carr, 369 U.S. 186, 82 S.Ct. 691, 7 L.Ed.2d 663 (1962) in support of their conclusion. They seem to forget that Baker v. Carr specifically overruled Colegrove v. Green, 328 U.S. 549, 66 S.Ct. 1198, 90 L.Ed. 1432 (1946), which held that reapportionment was a political question and not subject to adjudication by the courts. Nonetheless, defendants cite a discussion of six factors in Baker v. Carr which that Court said were characteristic of a political question. The Court said, 369 U.S. on page 217, 82 S.Ct. on page 710:
"Prominent on the surface of any case held to involve a political question is found a textually demonstrable constitutional commitment of the issue to a coordinate political department; or a lack of judicially discoverable and manageable standards for resolving it; or the impossibility of deciding without an initial policy determination of a kind clearly for nonjudicial discretion; or the impossibility of a court's undertaking independent resolution without expressing lack of respect due coordinate branches of government; or an unusual need for unquestioning adherence to a political decision already made; or the potentiality of embarrassment from multifarious pronouncements by various departments on one question."
They claim that any one of these factors, if inextricable from the case at bar, is sufficient to make the question nonjusticiable. They further say that the application of these factors to this case compels such a conclusion.
It is clear that Article 1, Section 2, Clause 3 is not a "textually demonstrable constitutional commitment" to the Congress of the question of whether the Constitution requires that the official population count be adjusted for a black and Hispanic undercount. All that Article says is that there shall be an actual enumeration every subsequent 10 years in such manner as Congress, by law, shall direct. All it does is impose on Congress the responsibility to provide for the taking of a decennial census. It does not say that Congress and Congress alone has the responsibility to decide the meaning of, and implement, Article 1, Section 2, Clause 3. It does not speak to the guarantee of equal representation for equal numbers of people, which is the crux of this case.
The same argument was made in Wesberry with respect to Article 1, Paragraph 4. It was argued that Article 1, Paragraph 4 had given Congress exclusive authority to protect the rights of citizens to vote for Congressmen, just as it is here argued that Article 1, § 2, Clause 3 give Congress the exclusive authority to decide whether an adjustment should be made for the black and Hispanic undercount differential. The Court rejected this contention with words that are equally applicable here (376 U.S. page 6 7, 84 S.Ct. page 529):
"We made it clear in Baker that nothing in the language of that article gives support to a construction that would immunize state congressional apportionment laws, which debase a citizen's right to vote, from the power of the courts to protect the constitutional rights of individuals from legislative destruction, a power recognized at least since our decision in Marbury v. Madison [1 Cranch 137, 2 L.Ed. 60] . . . The right to vote is too important in our free society to be stripped of judicial protection by such an interpretation of Article 1."
It should be noted that the Court very carefully referred to Article 1 as a whole, not merely to Article 1, Clause 4. In Powell v. McCormack, 395 U.S. 486, 89 S.Ct. 1944, 23 L.Ed.2d 491 (1969), the court held that Article 1, Paragraph 5, which provides that each House shall be the judge of the election returns and qualifications of its own members, did not constitute a textually demonstrable constitutional commitment to the Congress of the question of whether Congress could refuse to seat a member who possessed the qualifications of age, citizenship and residence set forth in the Constitution. If Article 1, Paragraph 4, and Article 1, Paragraph 5 do not constitute textually demonstrable constitutional commitments of those questions to the Congress, then surely Article 1, Section 2, Clause 3 does not constitute a textually demonstrable constitutional commitment to the Congress of the question of whether the Constitution requires an adjustment be made in the official population count for the black and Hispanic undercount differential. That question is not for the Congress, but for the judiciary, whose constitutional function it has been ever since Marbury v. Madison to "say what the law is." This court has authority to decide the issue before the court.
IIIThe decennial census is a constitutionally mandated attempt to enumerate the whole number of people in the United States. It is an undertaking of vast proportions. Vincent Barabba, Director of the Census, testified that in the most recent census 275,000 people were hired, and more than 80,000,000 questionnaires were mailed out, and 410 district offices were maintained throughout the country.
To facilitate the taking of the census, the United States is divided into a number of regions, which, in turn, are divided into enumeration districts. Each of the 410 district offices has approximately 750 enumeration districts, and cities are divided into census tracts, which are areas small enough to enable the Bureau of Census to monitor change within the cities.
It was conceded at trial that every census has necessarily involved an undercount. The phenomenon irresistibly flows from the magnitude and the manner of the undertaking. Defendant Barabba stated, at page 3 of his affidavit of May 6, 1980,
"Despite the elaborate procedures designed to insure that all persons are counted, it is generally recognized that it is impossible to count every one. Entire housing units may be missed by the Census Bureau. The number of persons living in enumerated housing units may not be reported completely by the householder. Both situations lead to an undercount of the population. Given the magnitude of the undertaking, and the nature and causes of the undercount, the Census Bureau is unable to determine the location of all missed living quarters, or the identity of all householders who report incomplete information . . ."
"Following the 1950, 1960 and 1970 censuses, estimates of undercount were prepared and announced publicly by the Bureau. According to our most recent calculations, the estimates of the per cent of the population undercounted in 1950, 1960 and 1970 are 3.3, 2.7 and 2.5 respectively."
Most significant for the purposes of this lawsuit is the fact that it is equally well accepted that there is a differential undercount. Otherwise stated, it is agreed by all that the undercount impacts upon blacks and whites in a differential manner, resulting in a more severe underenumeration of black people. Exhibit 4, a Bureau of Census publication entitled "Estimates of Coverage of Population by Sex, Race, and Age: Demographic Analysis," February 1974, states at page 7:
"The estimates of the coverage of the population in the 1970 census that have been derived indicate that about two-thirds of the persons missed in that census were males, and about two-thirds were white. According to the preferred set of estimates, . . . the net omission rate for males in 1970 (3.3 per cent) was nearly twice that for females (1.8 per cent), and the net omission rate for Negroes (7.7 per cent) was about four times as great as for the white population (1.9 per cent)."
If anything became clear from the testimony at trial, it was that black people are repeatedly differentially undercounted at a rate approaching 4 to 1. In other words, black persons are four times more likely to be undercounted than whites. And the testimony also established that almost as large a percentage of Hispanics have been undercounted.
Extensive expert testimony as to the underlying causes of the differential undercount was taken. Although the precise weight to be attributed to the various factors involved is not known, all of the experts agreed that some or all of the cluster characteristics associated with race are major factors. In this regard, Dr. Philip M. Hauser, a witness for the plaintiffs, stated that the following characteristics should be considered important:
1. lower income leading to diverse and unique housing patterns and living arrangements which frequently involve the lawful or unlawful conversion of housing units to accommodate increased numbers of people;
2. greater antagonism or resistance to government;
3. relatively less education; and
4. increased numbers of individuals within the harder-to-enumerate categories, such as young males.
Thus, the decennial census undercount results in dilution of plaintiff's vote relative to the votes of whites within the State of Michigan. The undercount gives rise to a constitutional violation of the one-person one-vote principle because blacks simply are not counted as much as whites, and this is particularly true in the inner cities of this country. See exhibits 14 and 15, and the uncontroverted testimony of plaintiffs' witness, Dr. Carl E. Taeuber, which established that blacks are overwhelmingly concentrated in the urban areas of the United States.
Dr. Taeuber, a sociologist at the University of Wisconsin, has developed a segregation index which numerically reflects the degree of segregation by race within various geographical areas. On a scale of 0 to 100, 100 reflects complete segregation. Dr. Taeuber indicated that in 1940, 1950, 1960 and 1970 every major United States city, including Detroit, reflected a high segregation index. He further stated that, in his opinion, the pattern would not change in 1980. He testified, and the Court accepts his testimony, that, of 22.6 million black people in the country, 16.8 million are concentrated in 34 metropolitan areas, each of which had more than 100,000 black people. The segregation index for blacks in Detroit in 1970 was 91, and the same pattern was replicated in smaller cities in Michigan, the segregation indices for Flint, Grand Rapids, Lansing and Saginaw being 87, 90, 93 and 88, respectively.
Defendants did not controvert any of the foregoing facts because, in large measure, Dr. Taeuber's research was based on data from the Census Bureau. These and other data establish as a matter of fact that the differential undercount results in a distortion of constitutional magnitude diluting plaintiff's vote in relationship to white voters in the nonurban areas of the State of Michigan. Such distortion clearly violates the principle of one-person one-vote.
It was also established that Hispanics, as a whole, reside in segregated urban areas or in non-urban areas in the southwest in segregated groups, and possess cluster characteristics similar, if not identical, to the cluster characteristics of black people. A differential undercount exists as between the general white population and Hispanics, and that undercount is compounded by the number of undocumented Hispanics living in the United States.
All of the experts agree that in deriving an estimate of the undercount for Hispanics, novel problems are presented arising from the fact that accurate birth, death, immigration and emigration statistics are not available. Thus, a demographic analysis, comparable to the demographic analysis for white people and black people, which can adjust the actual count (see infra) cannot be completed for Hispanics.
With reference to defendants' claim that the decennial census has only been a simple headcount, and that is what the Constitution mandates, plaintiffs' expert, Dr. Philip Hauser, a former director of the Census, testified that the decennial census was nothing more than an estimate of the population. He defined the census as an estimate derived from a headcount and numerous other devices which could involve a series of judgments for error. All of the plaintiffs' experts pointed to the 1970 decennial census as a graphic illustration of devices employed by the Bureau of Census to impute population which had not been enumerated in the actual headcount and whose utilization in 1970 rebuts the notion that the census is only a simple headcount.
Exhibit 8, "Effect of Special Procedures to Improve Coverage in the 1970 Census", Bureau of Census, 1974, is a census publication describing the various methods employed by the Bureau to adjust the official count in 1970. In speaking of the imputation program which added 4.9 million people to the census, Exhibit 8 states as follows:
"The second type included a group of projects intended to locate households or individuals that were missed in the initial field activities and add them to the census counts. In some of the most important of these projects, sampling was used, and, as a result, only estimates of missed persons and housing units could be added to the counts . . ."
As to other methods of imputation and adding persons to the actual headcount, namely the nation vacancy check, the post-enumeration Post Office check, and the supplemental forms operations, the Bureau stated in Exhibit 8, page 2:
"The remaining three-fourths of the added persons, and one-half of the added housing units came from the second category of projects, which consists of operations in which the post-census investigations were used to detect deficiencies in census coverage. The population and housing counts for each small area of the United States were adjusted to reflect the results of these checks. The characteristics of persons and housing units added in this manner were those of persons and housing units previously enumerated."
Thus the Bureau of the Census admits that the foregoing operations added people to the 1970 census, and that these people were added by imputation when the Bureau did not have hard evidence, such as a census questionnaire, that they even existed in the quantity imputed.
The procedures adopted by the Bureau in 1970 prove that when the Bureau deemed it necessary to adjust census figures to preserve their statistical integrity, it has added people. In 1970 10.2 million people were missed, and the Bureau imputed approximately 4.9 million back into the count, leaving 5.3 million as the undercount.
It is therefore clear that the defendants' contention that the Constitution mandates that the decennial census be nothing but a simple, straightforward headcount rings hollow. Their contention that the adjustments in 1970 were somehow different, because the Bureau had some indication as to where to place the imputed persons, simply is not persuasive. In fact, such an assertion is astonishing in light of the "sampling" techniques utilized to impute persons and housing units that the Bureau estimated existed.
One of the chief issues at trial was the reliability of a method of distributing the undercount rate from the national level down to the state and sub-state levels known as the synthetic method. The simple synthetic method is an arithmetical procedure which can be used to distribute the total undercount of the United States by age, race or sex to local areas. It is a procedure employed to apportion undercount figures among the states and sub-state areas. New population figures derived by the use of the synthetic method are referred to as the adjusted population count.
In its simplest application, the national rate of undercount for any one of the variables (age, sex or race) is used to expand the unadjusted census figures for each of the local areas for which census data exists. For example, in any area with more than a de minimus number of blacks, the national rate of undercount of blacks, multiplied by the unadjusted census figures, would yield the net adjustment for black people in that locality. Said adjustment would in turn be added to the census figure to derive the adjusted census count for that locality. Computations can be carried on for sub-groupings within each of the variables (age, race or sex).
The reliability of this method was seriously challenged by Jacob Siegel of the Census Bureau, but Dr. Barbara Bailar, Associate Director of the Census, conceded that the synthetic method, or a variation thereof, was utilized as a part of the 1970 census in the imputation operations previously discussed. Most particularly this method was employed in the vacancy operation and in the computer method used to impute persons and characteristics, and even housing units. Both plaintiffs' and defendants' experts characterized this computer method as the "hot deck" method.
It is therefore clear that the decennial census is not, and has not been, for at least the past decade, a simple, straightforward headcount. Rather, the decennial census is a relatively accurate estimate of the population developed through the use of self-enumeration by questionnaire, statistical techniques, and computer-control devices.
Plaintiffs contend that procedures already known to and well understood by the Bureau, and already employed by the Bureau in one form or another, can be used to adjust for the differential undercount. Their experts testified that through the use of the demographic method, an estimate for the national undercount by age, race and sex could be easily generated. This contention remained uncontroverted at trial. All the experts, both plaintiff and defendant, conceded that the estimate of the national population derived through the use of the demographic method is more accurate than the actual census count. The demographic method, according to the affidavit of defendant Barabba, is a comparison of the expected number of persons estimated independently with the actual number enumerated. The expected number can be described as the number born since a particular year, minus the number of such persons who have died, and adjusted for the number who have moved into or out of the country. A modification of this technique was used in the 1950, 1960 and 1970 censuses to provide national undercount estimates of the total population for race, age and sex.
At trial, defendant Barabba testified that as of April 1, 1980, the total number of people in the country, including the undercount, was 227,224,000 and the actual census count will show approximately 222,000,000. It was the demographic method which allowed him to arrive at these figures.
It is significant that only a few months before trial, defendant Barabba stated in his affidavit as follows:
"All the procedural and promotional efforts, including those given as examples in paragraph 7, are expected to yield rates of coverage completeness, particularly for certain population segments, which are better than previous censuses. Nevertheless, I am unable to predict today how successful we will be; nor can I predict what the undercount rates will be for the Nation, for the various geographic areas, and for the various population subgroups. This will not be known by anyone until the census results are tabulated and then compared with expected estimates. For this reason the Census Bureau is continuing to study the undercount problem and the issue of adjustment."
The inconsistency between the affidavit and Mr. Barabba's testimony leads the Court to believe that the Census Bureau presently has a good idea of the undercount for the 1980 census. Whether this was obtained through sampling or some other method is not known. Nevertheless, it is abundantly clear that the development of an estimate of the national rate of undercount is not dependent upon the tabulation of all the actual census results, nor is an estimation by age, race and sex. The demographic method can be employed for statistically defensible estimates of the national undercount for blacks and whites by age and sex. However, it cannot be used to develop estimates for Hispanics.
It is acknowledged that the demographic method cannot be used to develop undercount estimates below the national level. This is primarily attributable to the fact that migration between states is very heavy, and there are no records comparable to national immigration and emigration records.
Thus the heart of the controversy as to method of adjustment pertains to how or whether the national rate of undercount can be apportioned to the state and sub-state areas in a fashion that results in adjusted census figures which are closer to the truth than the unadjusted census count. To obtain distribution of the national undercount to state and sub-state areas, plaintiffs contend that two methods are available to defendants. The first is the distribution of the national undercount through the use of the "simple synthetic" method in a manner which will result in adjusted census figures which are, it claims, closer to the truth than the raw, unadjusted census data. All of plaintiffs' experts so testified.
The reliability of the synthetic method in arriving at the undercount in urban areas is attributed to the fact that blacks reside overwhelmingly in the urban areas of the United States, that they are homogeneous populations, and in the main are subject to similar income, similar education and similar segregated housing.
The validity of the synthetic method for obtaining counts in urban areas is premised upon homogeniety of population. It is not necessarily true that the simple synthetic method would be accurate in suburban areas, or in areas where the black population is very small.
It is conceded by plaintiffs that the simple synthetic method necessarily involves the assumption that factors attributing to the national rate of undercount are, relatively speaking, uniformly duplicated in each local area for which it is used to make an adjustment. As pointed out above, such assumption cannot be valid universally. But plaintiffs contend that the assumption is sufficiently valid in urban areas in dealing with the undercount of the vast majority of black persons, because of their inordinate urban concentration and similar cluster characteristics. Robert B. Hill, the Director of Research for the National Urban League, an expert for the plaintiffs, testified as to the urban concentration of black and Hispanic people, and stated that the simple synthetic method would be particularly appropriate in large urban areas because of the homogeniety of the population.
The second method is the current population survey match, which is a system of comparing the lists of persons in one or more survey or sets of records with the list of persons in the census to determine who was missed in the census. The method involves matching on a person-to-person basis. In theory, the method can be used to distribute among the states and other large areas the national undercount estimate derived by demographic analysis. This method can also be used to determine the Hispanic undercount.
Finally, all of the experts agreed that some areas might not merit an adjustment at all. Such areas are those particularly easy to enumerate because of stable living patterns and generally higher education. The suburbs are small enough and possess limited numbers of Black persons so that the total population is generally counted.
Defendants conceded at trial that, assuming the current population survey match and appropriate corrections were carried out, the Bureau would be capable of distributing the national undercount for blacks to state and sub-state areas. In addition, they conceded that such a match would enable them to develop, in a statistically defensible manner, an estimate of the national undercount for Hispanics, and distribute that estimate among the states and sub-state areas.
It is therefore clear that there are proper and adequate methods of adjustment of the raw census data that can be used by the Census Bureau to give reasonably accurate and statistically defensible counts that are more accurate than the raw enumeration of the Census Bureau.
The Court having determined that there are proper and adequate methods for adjusting the raw census data to give a reasonably accurate and statistically defensible adjusted count at the national, state and sub-state levels, the question remains whether such an adjustment is mandated.
Plaintiffs' case is bottomed on the dual constitutional argument that Article 1, Section 2, Clause 3 of the Constitution requires the Bureau of Census to make an adjustment for the number of black and Hispanic people that will not be included in the decennial census, and that this Article imposes an affirmative duty upon the Secretary of Commerce and the Bureau of the Census to arrive at an enumeration of the whole number of persons within the United States, which is as close to the truth as is practicable. They claim that a failure to adjust the data from the raw population enumeration of the decennial census for a known underenumeration of black and Hispanic people violates Article 1, Section 2, Clause 3, and the 14th Amendment, which guarantee that all people will have their vote counted equally with all other persons.
Plaintiffs, in their post-trial brief, cogently stated the issues before the Court. Does the Constitution require that an adjustment be made in the official population count of the decennial census for the anticipated differential undercount of Black and Hispanics? If the Constitution does require that such an adjustment be made, how is the Constitutional obligation to be implemented with respect to the 1980 decennial census?
We have seen that in both the 1970 and the 1960 censuses, blacks were undercounted at a rate four times that of whites. We have also seen that the racial undercount differential for the 1980 census is likely to be at least as large as in past censuses.
The Census Bureau testified that if its coverage improvement methods were uniformly successful, the overall undercount rate could be reduced from its 1970 level of 2.5 per cent to 1.4 per cent, but that the 4 to 1 ratio differential of completeness would remain. The Bureau further stated that in the unlikely event that only coverage improvement methods directed at blacks were successful, the overall undercount rate will be reduced to 2.3 per cent, and the differential of completeness would be reduced to 2.7 to 1. It further estimated that the Hispanic undercount rate in the 1970 census was intermediate between black and white, closer to black, and that in 1980 it will be closer to the black undercount than to the white.
Defendants contend that Article 1, Section 2, Clause 3 of the Constitution prohibits statistical adjustment of the decennial census count for the purpose of apportionment of representatives among the states.
They claim that interpretation of the phrase "actual enumeration" in Article 1, Section 2, Clause 3 must begin with the words themselves, and that the terms "census" and "enumeration" mean nothing more or less than a headcount. They say that the use of the modifier "actual" with the word "enumeration" can only reinforce the conclusion that the framers of the Constitution intended a headcount, and nothing but a headcount. They further rely upon the fact that, with the exception of the 1970 census when imputations were performed which added approximately 4.9 million people, the census has been, since 1790, an actual headcount and nothing more.
The Court finds nothing in the Constitution that prohibits adjustment techniques. When the Constitution speaks of actual enumeration, it speaks of that as opposed to estimates. The framers wanted, as much as possible, an accurate count because the count was to be used for the determination of Congressional apportionment. It is clear from Wesberry (supra) that the Supreme Court interpreted Article 1 to require a reapportionment among the states, and within the states, based upon actual population count. The face of Article 1, Section 2, Clause 3 makes no mention of a headcount.
It seems clear that the framers were most concerned that representation be achieved in the House on the basis of numbers, and to that end delegated to Congress the task of determining the method by which an "actual enumeration" would take place. In fact, the delegation of the responsibility as to the method of enumeration negates defendants' contention that a headcount is the sole procedure authorized by the Constitution. If the defendants' claim were valid, there would simply be no need for language as to the manner of taking the census.
In Federation for American Immigration Reform v. Klutznick, 486 F. Supp. 564 (D.D.C.) (1980), the Court found that the Constitution required the counting of whole numbers of persons, although it did recognize that for apportionment purposes the straightforward headcount had been used since the first census in 1790. The Court, on page 576, quoted Representative Celler of New York, who said:
"The Constitution says that all persons shall be counted. I cannot quarrel with the founding fathers. They said that all should be counted. We count the convicts who are just as dangerous and just as bad as the Communists or the Nazis . . . The only way we can exclude them would be to pass a constitutional amendment."
As pointed out in the brief Amicus of the U.S. Conference of Mayors, one certainly has to question why the Census Bureau is required to count convicts and Communists and Nazis, but is not required to count, or account for, 2.5 to 3 per cent of the population. The fact is that the Census Bureau is required to count them. Article 1, Section 2, Clause 3 requires it. The 14th Amendment requires it, because census figures must be used for apportionment of representatives among the states and within the states. See Wesberry and Kirkpatrick v. Preisler (supra).
The Courts that have considered this matter are unanimous in holding that the method, as opposed to the fact, of enumeration under the census is a procedural matter, and as such is within the discretion of the Bureau, subject to the Constitution and the disapproval of Congress, the Secretary of Commerce, or the Courts themselves. See Lampkin v. Connor, 360 F.2d 505 (D.D.C.) (1966); Newark v. Blumenthal, 457 F. Supp. 30 (D.D.C.) (1978).
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. Even the Census Bureau itself does not believe this, for in 1970 it imputed several million people to the actual headcount of that year.
The words "actual enumeration" in Article 1, Section 2, Clause 3 do not prohibit an accurate statistical adjustment of the decennial census to obtain a more accurate count. Quite to the contrary, Article 1 of the Constitution, which mandates equal apportionment of representatives, mandates an adjustment to obtain a figure that more accurately reflects the actual population of the United States and state and sub-state areas than the unadjusted headcount. So does the 14th Amendment.
Moreover, as witness Bailar pointed out at trial, the Census Bureau does not conduct a "headcount" which they claim they are required to do by the Constitution, but rather a "household count". The widespread use of imputations in 1970, many of which will continue in 1980, were based, in fact, on tangible evidence of the existence of a housing unit, and, in many instances, the persons occupying it.
Further, in the 1970 census, the Census Bureau imputed persons even where it could not find a particular housing unit in which to put them. Where the individual census reports and the "were you counted" forms could not be geographically identified to an enumeration district, let alone to a particular house within an enumeration district, the Census Bureau located these "created households" in a large enumeration district and put them "into space" within that district.
Plaintiffs do not contend that all of the imputations done in 1970 and planned for 1980 are unreasonable. They point out that the Census Bureau's unbroken historical practice really has been to use modern knowledge and scientific techniques to get further and further away from simple headcounting. Instead of headcounting people, it uses the mail-out form and the mail-out/mail-back format to enumerate most persons today. It created 4.9 million persons by imputation in 1970, and will create many persons by imputation in 1980. Except for the crucial black and Hispanic undercount differential, it uses modern knowledge and scientific techniques to create persons so as to make the official population count more complete and more accurate. Plaintiffs agree this is exactly what the Census Bureau should do.
The decennial census is not a "house count" or a "headcount", or a "headcount plus a house count". It is an estimation by the best means currently available of the population of the United States, and of the states and sub-state areas of the country. The term "actual enumeration" means an official population count, and in order to implement the "equal representation for equal numbers of people" principle of the Constitution, that official population count must be as complete and accurate a count as it is reasonably possible to make.
With the modern knowledge and the scientific techniques available to adjust for the known undercount differentials, and to bring about official population counts that are closer to the truth than those derived from the raw, unadjusted headcount, surely the framers of the Constitution would have intended that such an adjustment be made.
Defendants further contend that even if the Constitution does not prohibit an adjustment for apportionment of Representatives, Congress has by statute prohibited such an adjustment. For their authority, they cite 13 U.S.C. ¶ 195, which provides:
"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' in carrying out the provisions of this title."
In 1976, the Congress, by Public Law 94-521, amended Title 13 U.S.C. § 195 as quoted above. Prior to the amendment, § 195 read:
"Except for the determination of population for apportionment purposes, the Secretary may, where he deems it appropriate, authorize the use of the statistical method known as `sampling' in carrying out the provisions of this Title."
Defendants argue that Congress was aware of the longstanding view of the Census Bureau that an adjustment for apportionment of representatives was prohibited by the Census Act, and the Congress' 1976 Amendment of ¶ 195, without changing the prohibitory language concerning apportionment of representatives, was a ratification of that interpretation. This is not the fact.
In 1976, when Congress amended the statute, it must be assumed that it was well aware that, in the 1970 census, several devices to impute people that had not been counted were used by the Census Bureau. The amendment to ¶ 195 does nothing more than tighten the language of the prior statute and substitute the word "shall" for the word "may". It also more clearly refers to congressional apportionment rather than simply referring to apportionment. In net effect, the amendment serves to strengthen the Census Bureau's power to use sampling techniques where congressional apportionment is not involved.
Judge Guy, in his opinion, Young v. Klutznick (supra), dealt with this issue. He said:
"With respect to defendant's argument that 13 U.S.C. § 195 precludes the use of statistical techniques to determine the number of persons in the United States for the purposes of congressional apportionment, the Court is unable to interpret the statutory language so expansively. Rather, the Court finds that ¶ 195 indicates that for less constitutionally significant purposes, the agencies charged with the responsibility of gathering data of this kind may decide to use statistical tools to reach certain conclusions. Plaintiffs readily admit that it would be impermissible to allow congressional apportionment to be based on population figures derived solely by statistical techniques, and the Court is inclined to agree that this would be neither permissible nor prudent. That is not to say, however, that either ¶ 195 or the Constitution itself prohibits the use of statistics in addition to the more traditional measuring tools to arrive at a more accurate population count. Consequently, the Court does not find that plaintiffs' complaint should be dismissed because the course of conduct they request is prohibited by federal statute."
"If there were any doubt on the question, the silence of Congress in the face of the long continued practice of the Department of the Interior in construing statutes [in a particular manner] . . . must be considered `equivalent to consent to continue the practice until the power was revoked by some subsequent action by Congress.'"
This Court agrees with Judge Guy's reading of ¶ 195. There is nothing contained in 13 U.S.C. § 195, as amended, which would suggest that the Congress was interested in terminating the Census Bureau's practice, manifested in the 1970 census, of adjusting the census returns to account for people who were not enumerated. All that § 195 does is prohibit the use of figures derived solely by statistical techniques. It does not prohibit the use of statistics in addition to the more traditional measuring tools to arrive at a more accurate population count.
This interpretation is consistent with a finding that the statute is constitutional. A court should avoid construction of a statute that would render it ineffective and unconstitutional. Were the Court to read the statute as contended by defendants, it would have to find the statute unconstitutional, because the Constitution mandates the most accurate count that can be obtained, and this necessarily includes more than an unadjusted headcount.
Defendants next contend that, even if the Constitution and the Census Act permit an adjustment for apportionment of representatives, the Census Bureau's decision whether to perform such an adjustment is committed to agency discretion, and is not subject to judicial review.
They argue that Section 10 of the Administrative Procedures Act, 5 U.S.C. § 701, withholds from judicial scrutiny those cases where statutes preclude judicial review, or the action is committed to agency discretion by law. They state that where it appears from the terms of the statute, and from its legislative history, that Congress intended the ultimate decision-making authority, in a particular area, to be within the administrative agency rather than the courts, judicial review is foreclosed.
Relying upon Panama Canal Company v. Grace Line, 356 U.S. 309, 78 S.Ct. 752, 2 L.Ed.2d 788 (1958), the defendants state that the Census Bureau's decision whether or not to adjust the official population counts requires the balancing of highly technical factors, which is necessarily accomplished by informed calls of judgment. These "calls of judgment" and the exercise of discretion by the Census Bureau, defendants contend, have been committed to agency discretion and are not subject to judicial review. They argue that the types of decisions to be made by the Director, in exercising his broad discretionary authority, involve specialized managerial and scientific knowledge of a kind at which a specialized agency, and not a United States District Court, is most expert.
In short, defendants argue that the Census Bureau is supreme in determining the method of counting, and that even though constitutional questions are involved, as they clearly are here, this Court may not interfere with the decisions of the Census Bureau because of the specialized managerial and scientific knowledge of the Bureau.
The Court totally disagrees. The contentions of defendants would completely change the law. A court is not precluded from reviewing the actions of an administrative agency just because the administrative agency acts in a technical field. This is especially true when Constitutional issues of the dimensions involved in this case are at stake. To say that a United States District Court cannot look into a matter such as is before this court, where the one-person one-vote concept and the distribution of billions of dollars of federal funds are at issue, is to say that a United States District Court cannot take action to vindicate rights where major Constitutional and statutory rights are involved, merely because the determination of such rights is highly technical. This is not the law. Nothing in the law insulates the Census Bureau from proper judicial scrutiny. To hold otherwise would be to unconstitutionally limit the powers and make administrative agencies supreme over the Federal Judiciary. This is not only bad public policy, it is bad law.
Defendants next argue that, assuming judicial review of the agency action is warranted, a review of the Census Bureau's actions indicates that their decision not to adjust the population count is lawful and proper in all respects, and that, therefore, judgment for defendants is required. Defendants state that if the Director of the Census Bureau's actions here are subject to judicial review, the standard of review is extremely narrow. Defendants contend that in reviewing the Agency action in this case, the only relevant determination is whether the decision of the Director of the Census Bureau not to adjust population counts submitted to the President on January 1, 1981 to reflect an estimate of the undercount constitutes arbitrary and capricious action, and abuse of discretion, or is otherwise contrary to law.
The Court agrees that it must accord great deference to the actions of the Secretary of Commerce and the Director of the Census Bureau. But this does not mean that, when statutory and Constitutional rights of the magnitude involved here are present, the Court is tied to the narrow type of review appropriate for the decisions of an administrative agency.
In the first place, this case is not a review on the record of an administrative tribunal's decision. It is true that when a court is reviewing the decisions of an administrative tribunal on the record, clear and convincing error must be shown to reverse the decision, and the findings of fact of the administrative agency are not to be overturned unless clearly erroneous. But defendants misconceive the issue. The law relating to reviews of an administrative tribunal simply have no applicability here. This is an original action brought in a United States District Court for declaratory relief and for injunction, raising significant statutory and Constitutional issues. As such, the issues raised must be judged and decided based upon the record made and the law presented to the Court. Thus, the law cited by the defendants relating to review of an administrative tribunal has no application to this case.
Clearly, the Constitution and laws of the United States require that the plaintiffs receive the relief which they seek. The 1980 census figures must be adjusted to reflect the undercount, and to abolish the differential undercount between blacks and whites, and between Hispanics and whites.
Defendants next contend that even if the Court orders an adjustment of the census to reflect the undercount, there is no possible way that it can be done accurately within the time limits provided by statute, which require a report to the President of the total national census on or before January 1, 1981, and a report to the states and some 39,000 sub-state areas on or before April 1, 1981. 13 U.S.C. § 141(b), 141(c).
Since 1929 the Bureau of the Census has been obligated to deliver its tabulation of total population by states within eight months after the census date. For the 1980 census, the time limits were enlarged from eight months to nine months ( 13 U.S.C. § 141(b)). The granting of the additional month to report to the President reflected congressional judgment that an increased time period would better enable the Bureau of the Census to meet its constitutional mandate and in all likelihood was connected to difficulties the Bureau experienced in 1970. The Senate report accompanying Senate Bill 3688, the predecessor of the Public Law amending 13 U.S.C. § 141(b) in 1976, stated:
". . . Subsection (b) of Section 141 provides nine months instead of the currently authorized eight months to complete the required tabulation of the population by States for the reapportionment of the United States House of Representatives. It is for the purpose of apportioning representatives that the United States Constitution establishes a decennial census of population. The additional month provided the Bureau of the Census is intended to better enable it to meet this constitutional requirement." 1976 U.S. Code Cong. and Admin.News, page 5467.
It should be noted that Titles 2 and 13 of the U.S. Code, which form the backbone of the current legislative scheme, have a common root. The census legislation of 1929, which contained the essence of the provisions now found in Titles 2 and 13, not only required the completion of the tabulation in eight months, but also required that on the first day, or within one week thereafter, of the second regular session of the 71st session of the Congress, the President should transmit to the Congress the statement showing the whole number of persons in each state, and the number of representatives to which each state would be entitled tinder the apportionment of the then existing number of representatives. It is therefore clear that the nine month time period in 13 U.S.C. § 141(b), and the 12 months time period in 13 U.S.C. § 141(c) have significance only because they reflect the judgment of the Congress as to how the task of reapportionment can be administratively handled.
The question remains whether these time limits are mandatory upon the Bureau of the Census and the Secretary of Commerce if they cannot be met without sacrificing a count more accurate than the unadjusted head count of persons in the United States. Are time limits established by Congress mandatory or discretionary?
In Diamond Match Company v. United States, 181 F. Supp. 952, (U.S.Customs Court, 1960), the customs collector failed to mail the plaintiff's notice within the prescribed statutory time limit. The Court stated: (page 958)
"As a general rule, a statute which provides a time for the performance of an official duty will be construed as directory so far as time of performance is concerned, especially where the statute fixes the time simply for convenience or orderly procedure . . . Statutes fixing the time for the performance of acts will ordinarily be held directory where there are no negative words restraining the doing of the act after the time specified, and no penalty is imposed for delay."
In Ralpho v. Bell, 569 F.2d 607, 627 D.C. Cir. (1977), the Court said:
"In determining the effect of the windup provision, we must cleave as nearly as we can to the congressional design. That design, as evidenced by `the mischief to be remedied' as well as `contemporaneous discussion' convinces us that justice to the Micronesians is not to be renounced simply to avoid even a brief additional lease on life for the Commission. Statutes that, for guidance of a governmental official's discharge of duties, propose `to secure order, system, and dispatch in proceedings' are usually construed as directory, whether or not worded in the imperative, especially when the alternative is harshness or absurdity."
In discussing statutory time periods, Judge Wisdom of the 5th Circuit said, in Fort Worth National Corporation v. Federal Savings and Loan Insurance Corporation, 469 F.2d 47, 58 (C.A.5 1972):
"A statutory time period is not mandatory unless it both expressly requires an agency or public official to act within a particular time period and specifies a consequence for failure to comply with the provision . . .
"Section 1730A(e) does not specify any particular consequence for failure to act within the time period. Nor is there anything in the legislative history indicating such an intent. We conclude, therefore, that the provision is directory . . ."
And in Holbrook v. United States, 284 F.2d 747, 752 (C.A.9, 1960) the Court stated:
"In the absence of direct evidence of legislative intent, a significant consideration in determining whether a statutory requirement should be given mandatory or directory effect is a comparison between the results to which each such construction would lead."
Great discretion is given courts in determining the application of statutes where a strict reading would result in a harsh result. The United States Supreme Court pointed out in Sorrells v. United States, 287 U.S. 435, 446, 53 S.Ct. 210, 214, 77 L.Ed. 413 (1932), quoting from U.S. v. Kirby, 7 Wall. 482, 19 L.Ed. 278:
"All laws should receive a sensible construction. General terms should be limited in their application as not to lead to injustice, oppression, or an absurd consequence. It will always, therefore, be presumed that the legislature intended exceptions to its language, which would avoid results of this character. The reason of the law . . . should prevail over its letter."
"When that meaning has led to absurd or futile results, however, this Court has looked beyond the words to the purpose of the act. Frequently, however, even when the plain meaning did not produce absurd results but merely an unreasonable one `plainly at variance with the policy of the legislation as a whole' this Court has followed that purpose rather than the literal words . . . Emphasis should be laid, too, upon the necessity for appraisal of the purposes as a whole of Congress in analyzing the meaning of clauses or sections of general acts."
". . . Resort to the policy of a law may be had to ameliorate its seeming harshness or to qualify its apparent absolutes . . . The process of interpretation also misses its high function if a strict reading of a law results in the emasculation or deletion of a provision which a less literal reading would preserve."
It is clear from a close reading of 13 U.S.C. § 141(b), (e) that the time limits adopted by Congress were directory. Congress intended to set up time limits so that reapportionment could go forth in an orderly manner. It is obvious that Congressional reapportionment does not have to be accomplished before the 1982 primary and general elections. Therefore, there is no absolute need to have accurate census figures for reapportionment until the latter part of 1981, or the early part of 1982, so as to meet early primary elections. It is also apparent that the distribution of federal funds to local units of governments can be delayed a reasonable time after April 1981 to obtain a more accurate census count, if such delay is necessary for that purpose.
Thus it appears that the dates of January 1, 1981, and April 1, 1981, are not mandatory. If it is necessary that the Census Bureau be given more time to arrive at a more accurate count at the state and sub-state levels so that accurate figures can be provided for the 39,000 local units of government, as well as each of the 50 states, a reasonable extension of time can be granted by this Court.
If the Census Bureau satisfies this Court that it is impossible to obtain accurate adjusted figures in time to meet the January 1, 1981, and April 1, 1981, deadlines, this Court has power to give an extension of time. Such extension would be based upon a clear and convincing showing of the necessity for the additional time and the methods to be used to accurately adjust the census.
For the reasons given in this Opinion, a Judgment will issue embodying the following:
1. An injunction against defendants, enjoining them from certifying to the President on January 1, 1981, and to states and sub-state units of government on April 1, 1981, a population count based on the actual unadjusted headcount.
2. A mandatory injunction ordering defendants to adjust population figures for the 1980 census at the national, state, and sub-state level to reflect the undercount, and to adjust the differential undercount to prevent the known undercount of Blacks and Hispanics, as well as whites. The methods by which this adjustment shall be made shall remain within the discretion of the defendants so long as they are statistically defensible.
3. A Declaratory Judgment, declaring it to be the right of every person within the United States of America on April 1, 1980, to be counted in the census, and to direct that such a count be made to the extent it is statistically defensible.
4. An order to the bureau of the Census to report to this Court within thirty days from the date of this opinion the precise nature of its plan to implement the holdings of this Opinion.
If the defendants show to this Court by clear and convincing evidence that the adjustments ordered above cannot be made on or before January 1, 1981, and April 1, 1981, as required by statute, reasonable extensions of time for the making of such adjustments shall be ordered.
This opinion shall constitute the findings of fact and conclusions of law required by Rule 52 F.R.C.P.