March 21, 1996
Supreme Court Upholds 1990 Census as Valid, Despite Undercount
Text of the Supreme Court Opinion Supreme Court Syllabus
By LINDA GREENHOUSE
ASHINGTON -- The Supreme Court on Wednesday upheld the validity of the 1990 census, ruling unanimously that the federal government had no constitutional obligation to adjust the results to correct an acknowledged undercount in big cities and among minorities.
While the decision, written by Chief Justice William Rehnquist, ends a long-running lawsuit, it almost certainly will not resolve a continuing policy debate over the best way to count the nation's population.
New York City and a coalition of other big cities had brought the lawsuit in 1988 to challenge the Bush administration's refusal to use statistical sampling methods to adjust the 1990 census figures.
In the Supreme Court, the Clinton administration supported its predecessor's policy, telling the court that the decision whether to make a statistical adjustment was within the discretion of the secretary of commerce.
The court accepted that argument Wednesday, with Rehnquist concluding that "the secretary's decision not to adjust need bear only a reasonable relationship to the accomplishment of an actual enumeration of the population."
Last month the Census Bureau, a Commerce Department agency, announced that as a way of saving money on the census for the year 2000, it would directly count only 90 percent of the population and rely on statistical sampling methods to calculate the remainder.
At a hearing earlier this month in the House of Representatives, some members of the Government Reform and Oversight Committee, which oversees the Census Bureau, raised questions about the bureau's plans.
Interest in this issue is hardly abstract. Various methods of counting can influence the allocation of congressional seats among the states, legislative power within states, and the flow of money from Washington for federal programs that operate according to population-based formulas.
Partisan considerations have never been far below the surface in this debate, because those who are missed by conventional census-taking methods are disproportionately poor, urban-dwelling members of minority groups -- and traditionally Democratic.
Had the Census Bureau used the sampling technique it considered for the 1990 census, representation in the House would have changed for two states: California would have gained a seat at the expense of Wisconsin.
New York City would have gained an estimated 230,000 people, not enough for an additional House seat but enough to have an effect on redistricting for the state Legislature, which is also based on the census.
Wisconsin Attorney General James E. Doyle hailed Wednesday's decision as a "great victory." But the U.S. Conference of Mayors said that "mayors across this nation are very disappointed" by the ruling.
In New York City, Mayor Rudolph Giuliani said that if the undercount had been adjusted, the city would have received as much as $500 million more in federal aid. "Certainly we want changes for the next census," he said at a news conference after the ruling was announced.
At the core of the legal challenge to the 1990 census was the racially disparate undercount, the existence of which no one disputed. The census missed about 2 percent of the population as a whole, some 4 million people. But it missed 4.8 percent of the black population and 5.2 percent of the Hispanic population.
New York and the other plaintiffs, which included Los Angeles, Chicago, Houston, Philadelphia, Washington and other big cities, as well as the National League of Cities and the National Association for the Advancement of Colored People, argued that the refusal to adjust the undercount was an infringement of the fundamental right to have one's vote counted equally.
Wisconsin and Oklahoma intervened in the lawsuit on the Commerce Department's side, to defend their own interests against a statistical adjustment.
The plaintiffs lost after a 13-day trial in U.S. District Court in Brooklyn in 1993. The 2nd Cicruit U.S. Court of Appeals, in Manhattan, reinstated the lawsuit in 1994.
Applying the analysis of the Supreme Court's one-person, one-vote decisions, the appellate court held that because the constitutional principle of equal protection was at stake, the government had to defend its policy by showing that the refusal to adjust was not only reasonable but also essential to some legitimate governmental purpose.
At that point, Wisconsin and Oklahoma appealed to the Supreme Court, followed two months later, after a heated internal debate, by the Clinton administration. The administration was in an awkward position because Commerce Secretary Ronald Brown had been a leading critic of the Bush administration's refusal to adjust the census when he served in his previous job as the Democratic national chairman.
But the administration decided it needed to defend the government's prerogatives in this area. The federal courts should not "take sides in a statistical dispute between statisticians and demographers," Solicitor General Drew S. Days III told the court when the case, Wisconsin vs. New York, No. 94-1614, was argued Jan. 10.
That proved to be a winning argument. The appellate court "failed to recognize that the secretary's decision was made pursuant to Congress' direct delegation of its broad authority over the census," Rehnquist said, noting that the Constitution gives Congress authority to conduct the census "in such Manner as they shall by Law direct."
Under a federal law, the Census Act, Congress has delegated that authority to the Commerce Department.
In declining to adjust the census, Robert Mosbacher, who was secretary of commerce under President George Bush, said that the "distributive accuracy" of the census -- the relative distribution of population among the states -- was more important than strict numerical accuracy to accomplish the constitutional purpose of the census, apportioning congressional districts among the states.
While a statistical adjustment could improve the numerical accuracy of the census, he said, there was no proof that it would improve the distributive accuracy.
This was "a reasonable choice in an area where technical experts disagree," Rehnquist said.
In his opinion, the chief justice noted that no census had achieved perfect accuracy and that there had been evidence of a racially disparate undercount since at least 1940. Thomas Jefferson complained that the first census, in 1790, contained a significant undercount, the chief justice said.
The opinion contained one hint that the federal courts may still have more to say about the census-taking debate. The Constitution refers to an "actual enumeration," and some people who oppose any form of statistical adjustment have argued that this phrase precludes anything except a conventional head count.
"We do not decide whether the Constitution might prohibit Congress from conducting the type of statistical adjustment considered here," the chief justice said in a footnote.
Doyle told the justices at the argument that his state would invoke the "actual enumeration" language to challenge any statistical adjustment that hurt Wisconsin.
Copyright 1996 The New York Times Company