Gotham Gazette

The Place for New York Policy and politics

Gotham Gazette image by Ya-Hsuan Huang

The contretemps over Judd Gregg's aborted nomination to be commerce secretary reveal once again the fault lines over the 2010 census between Republicans and Democrats. Gregg's difference with the administration over economic policy attracted much of the attention, but the census also reportedly played a major role in the whole affair.

The Census Bureau is in the Commerce Department, and the nomination of Gregg, a Republican senator from New Hampshire, prompted an outcry from black and Latino groups who feared a census under Gregg's direction -- in the Senate, he had voted against fully funding the 2000 census --would result in an undercount of minorities and immigrants. In response, the Obama administration guaranteed that it would watch over the census from the White House. That in turn sparked complaints from the GOP and right-wing bloggers about political interference by the Obama administration in census 2010.

Gregg has returned to the Senate, but the controversy over next year's national headcount remains. None of this is surprising or new, since how the 2010 census is conducted will greatly affect the division of power and the allocation of money in New York State and the city, as well as elsewhere in the United States. Framed as neutral counts of the "whole population," the U.S. census, as all censuses, concerns the allocation of political power and money. The 2010 count will:

  • Lead to no change or a loss of one or two congressional seats in New York State.

  • Affect the distribution of the State Senate and Assembly seats.

  • Affect allocations of federal and state funds.

Who gets counted depends on how the counting is done. That has been the source of controversy for decades.

How Many New Yorkers?

The city has claimed an undercount in every census since 1950. Things turned particularly ugly in 1980. A suspicious fire in the census office responsible for counting Bedford-Stuyvesant occurred the weekend before Washington was to send a team to investigate a massive undercount in the area. This and other problems led to a lawsuit demanding an adjustment for the 1980 count.

To resolve the issue, the Census Bureau agreed to add a survey, revisiting some households in person, to make it possible to adjust the 1990 census. Adjusted counts were prepared, and the Census Bureau staff and director recommended using them, but Commerce Secretary Robert Mossbacher, who went on to run George H.W. Bush's reelection campaign in 1992, rejected the plan.

Information also existed to adjust the 2000 Census, but thanks to the work of Kenneth Prewitt, then director of the Census Bureau, and the funding that Judd Gregg and others tried to block, the 2000 census stood without adjustments, a decision with which most professional demographers agreed. The release of the 2000 census made it plain that the 1990 Reagan-Bush census was seriously flawed. especially with respect to populous states with big cities and large minority and immigrant populations.

The Issues in 2010

The census count, however, continues to be mired in controversies. These include:

  • The very definition of what constitutes a household. Census enumerators simply overlooked many households in New York City in 1990, leading to the severe undercount. Now census staff suggests that, if there is any question about how many units are in a given structure, they will consult with the property owner. While that may sound reasonable, landlords who have illegally divided their property may not answer truthfully. This would lead to a lower count for New York City, especially among poor people and minorities. The bureau makes many such rules and they have serious ramifications.

  • The use of statistical sampling to increase the accuracy of the count for those unwilling to come forward. This was resolved in 1999 by the Supreme Court at least in regards to congressional apportionment. The court ruled the sampling could not be used in determining congressional seats.

  • Methods used by the census to fill in results for people who enumerators know live at a specific address but who either cannot be reached or choose not to answer all of the questions. For decades, the census has used the characteristics of nearby households to fill in results -- methods known as imputation and allocation. Following the 2000 census, Utah wanted to stop the bureau from using this technique in an effort to get itself an extra congressional seat at the expense of North Carolina, but the courts agreed census could continue the practice.

  • The counting of military abroad (which is done) and counting other Americans abroad, including Mormon missionaries (which is not), another issue brought forth by Utah but not changed by the court.

  • The counting of prisoners at prison instead of at their place of residence when arrested. Here the National Academy of Sciences panel on the census and the Second Circuit Court of Appeals has strongly questioned the bureau's current practice of counting prisoners as living in the prison. This is a major issue in New York State, where most of the prisoners come from the five boroughs but, for the purpose of the census, are counted as residents of the upstate counties where they are incarcerated. This boosts the amount of aid the upstate communities receive and also increases their official populations, giving them additional legislative seats. If the census changed or modified its method of counting prisoners, it could have a big effect on the allocation of New York State Senate and Assembly seats.

  • The counting of immigrants. Many people do not want immigrants or at least illegal immigrants included in the census. The courts years ago said that such residents are part of the population. Nonetheless the issue is continually raised, especially in the context of redistricting.

  • Counting of students in their dorms, which also has been resolved in favor of the students, but like the counting of immigrants is always raised by affected politicians, who may not be comfortable with students in their districts.

Simply put, a professionally run Census Bureau with a commitment to count the "whole population" using modern methods and techniques will be a boon to New York City (see the results from 2000). On the other hand if there is little or no commitment to a complete count (see the recent blog posting of Bruce Chapman, bureau director during the 1980 controversies and now president of the pro-Creationist Discovery Institute), New York City will be undercounted again. It is his original blog post that stirred up the right-wing blogs and politicians to accuse the administration of interfering with the census. If this attack upon the census is successful, New York City might be miscounted and appear to lose population.

That this would happen during the Obama administration seem unlikely. Nonetheless, the census is at very high risk according to the Government Accounting Office, and with the count only 13 months away, it is well past time to confirm a census director. The choice of many is Kenneth Prewitt, who ran the successful 2000 Census and appears to be willing to step into that role again. New Yorkers should hope that he does.

Andrew A. Beveridge has taught sociology at Queens College since 1981, done demographic analyses for the New York Times since 1993, and been in charge of Gotham Gazette's demographics topic page since 2000. The opinions expressed are his alone.