Summary

In past years, the federal government has annually distributed over $300 billion in federal assistance through grant programs using formulas driven in part by census population data. Of the more than $580 billion in additional federal spending, the American Recovery and Reinvestment Act of 2009 will obligate an estimated additional $161 billion to federal grant programs for fiscal year 2009. The U.S. Census Bureau (Bureau) puts forth tremendous effort to conduct an accurate count of the nation's population, yet some error in the form of persons missed or counted more than once is inevitable. Because many federal grant programs rely to some degree on population measures, shifts in population, inaccuracies in census counts, and methodological problems with population estimates can all affect the allocation of funds. This testimony discusses (1) how census data are used in the allocation of federal formula grant funds and (2) how the structure of the formulas and other factors can affect those allocations. This is based primarily on GAO's issued work on various formula grant programs and the allocation of federal funds.

Federal grants use various sources of population counts in their funding formulas. They include the decennial census, which provides population counts once every10 years, and also serves as the baseline for estimates of the population for the years between censuses--known as postcensal estimates. Other sources of population data include the Bureau's American Community Survey and the Current Population Survey conducted by the Bureau for the Bureau of Labor Statistics, which provides monthly data. The degree of reliance on population in funding formulas varies. For example, the Social Services Block Grant formula allocates funding based solely on a state's population relative to the total U.S. population. Other programs use population plus one or more variables to determine funding levels. Medicaid, for example, uses population counts and income to determine its federal reimbursement rate. On the basis of simulations GAO conducted of federal grant allocations by selected federal grant programs--for illustrative purposes only--we found that changes in population counts can affect, albeit modestly, the allocations of federal funds across the states. For example, in 2006 we found that compared to the $159.7 billion total federal Medicaid funding in 2004, 22 states would have shared an additional $208.5 million in Medicaid funding, 17 states would have lost a total of $368 million, and 11 states and the District of Columbia would have had their funding unchanged. In total 0.2 percent of Medicaid funds would have shifted as a result of the simulation. In addition to population data, various other factors related to the design of federal grant programs may mitigate the effect that population changes can have on the distribution of federal funds. For example, in order to prevent funding losses from a formula change, several programs include hold-harmless provisions guaranteeing that each recipient entity will receive a specified proportion of the prior year's amount or share regardless of population changes.