The Fiscal Impact of the Census Undercount on Cities  A 34-City Survey
January 1999


It has been long-standing policy of The U.S. Conference of Mayors that an accurate census count is vitally important to every city. Beyond the importance of the count to fair representation of citizens in federal and state legislatures, the census figures are the basis for the distribution of funds for a variety of programs – housing, community and economic development, transportation, job training and low income home energy assistance, among them. If a city’s population is undercounted, that city stands to lose both federal and state funds. The U.S. Bureau of the Census estimates that in the 1990 census, eight million people were undercounted, and another four million were counted twice.

Census Bureau officials assert that trying to count every individual using the traditional methods would be extremely expensive and would provide no assurance of improved accuracy. Indeed, three separate panels convened by the National Academy of Sciences have recommended that the Bureau of the Census use sampling in the 2000 census to save money and improve accuracy.

The U.S. Conference of Mayors supports the use of statistical sampling and the specific efforts by the Bureau of the Census to employ them in the 2000 census. It has opposed Congressional efforts which have attempted to deny funds to the Bureau for its use in planning and preparing to employ statistical sampling in the 2000 census. It also has sided with the Census Bureau in two law suits – one brought by Congress, the other by private parties – which seek to prohibit the Bureau from using statistical sampling in the 2000 census. The Conference of Mayors and a number of cities filed an amicus brief in these cases. Oral arguments were heard by the Supreme Court in those cases on November 30, 1998 and the High Court’s decision, for the plaintiffs, was announced January 25,1999. The Conference believes that regardless of this decision, the Bureau of the Census has the authority to use statistical sampling to adjust the 2000 census for non-apportionment purposes, such as the distribution of federal funds. This position was argued by the Solicitor General in one of his briefs to the Supreme Court and by Brain S. Currey of O’Melveny & Myers, LLP, the attorney for the Conference and the cities who prepared the amicus brief, in a January 21, 1999 letter to U.S. Commerce Secretary William M. Daley.

Currently, no national data exist on the fiscal impact of the undercount on local governments. In an effort to estimate the impact of the 1990 undercount on cities, and to estimate the likely impact of similar inaccuracies in the 2000 census, the Conference of Mayors analyzed information obtained initially from 34 cities. (Additional analyses of data from additional cities are planned.) Specifically, the cities were asked to report the federal and state funds which they lost over the last decade due to the 1990 census undercount – both total dollars lost and the amount lost per person. They also were asked what that loss – in total dollars and per-person dollars – would be for the next decade if the 2000 census were to maintain the same level of inaccuracy as that of the 1990 census.

This report provides initial findings based on the information provided by the 34 cities. All data included in the report were provided by the cities, with the exception of the population statistics in the table on page five which are either 1990 census figures or a subsequent update provided by the city, as shown in the July 1998 edition of The Mayors of America’s Principal Cities, published by The U.S. Conference of Mayors.