Skip Navigation

I am pleased to present to the Committee on the Budget the views of the Congressional Budget Office on H.R. 3697, the State and Local Government Cost Estimate Act of 1979. As introduced by Representative Holtzman, H.R. 3697 requires CBO to estimate the costs that would be incurred by state and local governments in complying with the provisions of each reported bill or resolution. These estimates are to be included in the regular estimate of the five-year costs for the federal government of reported bills or resolutions required by Section 403 of the Congressional Budget Act.

Without doubt, it would be valuable for Congressional decisionmakers to know the costs that a bill or resolution would impose on state and local governments. It would also be useful to know the costs and benefits of proposed federal legislation on private industry, individuals, or other sectors of the economy. Currently, the only such information available on a regular basis is the estimated costs to the federal government, as required by Section 403.

While the objectives of H.R. 3697 are important, CBO's ability to achieve them would limited for several reasons.

In the first place, the task could be significant. Last year, CBO prepared 995 cost estimates for bills reported from committees or under consideration by committees. Of course, not all legislation affecting federal spending imposes costs on state and local governments. We estimated last year that 10 percent of all reported bills or resolutions have some impact on state and local governments, and of these probably not more than half are likely to have a significant impact. Furthermore, in the past, CBO has considered state and local budgetary impacts in some of its cost estimates of the federal budgetary impact. For example, our cost estimates of the various welfare reform proposals considered in the last Congress included estimates of the fiscal relief that would have been accorded the state governments.

Second, the lack of data and of appropriate estimation techniques would limit our ability to estimate the impact on state and local governments. The impact of some types of legislation would be relatively easy to estimate. The increased costs to state and local governments of a Social Security tax hike or an increase in the minimum wage would fall into this category. It might be very difficult, however, to estimate the impacts of other types of legislation. Let me offer two examples.

Consider, for example, the difficulty of estimating the costs to state and local governments where the impact varies from jurisdiction to jurisdiction and the data on that variance is not easily available. This difficulty can be illustrated by noting the complexities that would have been involved in estimating the cost impacts on state and local governments of Section 504 of the Rehabilitation Act of 1973. That section states:

No otherwise qualified handicapped individual in the United States . . . shall, solely by reason of his handicap, be excluded from the participation in, be denied the benefits of, or be subjected to discrimination under any program or activity receiving Federal financial assistance.

This sweeping requirement covers not only the rehabilitation services provided by that act, but also all other activities receiving federal assistance. This included, to name just two major areas affected, the construction of any educational facilities--on the elementary, secondary, or postsecondary level--funded in part by federal grants, and the construction and purchase of equipment for subway systems, funded by grants from the Urban Mass Transit Administration. Estimating the costs of just one aspect--the modification of existing state university and community college facilities to permit barrier-free access for the handicapped--would require detailed data on the characteristics of the educational plants of institutions receiving higher education grants. Lacking such data, estimates would have to be developed for a sample of universities with campuses of various sizes, terrains, layouts, and building ages. Then, the total cost for institutions would have to be estimated from profiles of the value of university buildings and changes in enrollments over time, based on an assumption that the cost for an individual university is proportional to the value and age of its structures.

Or consider the difficulty of estimating the costs to state and local governments of legislation that is written very generally but will have implementing regulations, which will not be written until after the legislation has been enacted into law. Again, Section 504 of the Rehabilitation Act can serve as an illustration. The estimate of costs to educational institutions of complying with the regulations on removal of physical barriers could not realistically be done until the Department of Health, Education, and Welfare (HEW) had promulgated its standards. The standard finally adopted for existing buildings required that federal grant recipients must--through the elimination of physical obstacles or through other methods--operate their programs so that a program viewed in its entirety is readily accessible to handicapped persons. In order to estimate the costs of this bill before the regulations were issued, CBO analysts would have been required to make policy judgments, such as whether or not methods other than removal of physical obstacles would be permissible.

In both of these cases, where data are lacking and where the impact depends on regulations written after a bill's enactment, the preparation of the cost estimates would be marked by uncertainty and would take a long time--too long perhaps to be timely for the filing of a bill's report after final committee approval.

A third problem is that these estimates would have to be for the nation as a whole. In many cases, it would be impossible to break down the costs on either a state-by-state basis or a state-versus-local-government basis. This limitation may affect the utility of CBO's estimates for some users. It might not be possible to determine whether the burden would fall unfairly on certain jurisdictions--for example, on rural western counties versus eastern cities.

In view of the limitations I have just listed, a more limited approach under which CBO would be required to prepare estimates of the impacts on state and local governments for only those bills likely to have a major impact would seem to be more feasible than such a requirement for all bills. CBO's experience with inflationary impact analysis might serve as a model. This year, at the direction of the Appropriations and Budget Committees, CBO developed the capability to provide inflation impact estimates for selected bills. Additional funds and staff were provided in the Legislative Branch appropriations bill for fiscal year 1979. A ten-person unit was created to prepare these estimates for selected bills identified by the Budget Committees as being important in terms of their impact on inflation.

With a limited increase in staff and funds for computer support and special surveys, a similar unit could be established to do a select number of state and local impact estimates. After some period of time during which CBO could build its capacity and develop appropriate data bases and methodologies, the Congress could reassess the usefulness of state and local impact statements and decide whether the effort should be expanded to cover additional bills.