Statement of
Douglas Holtz-Eakin
Director

CBO's Activities Under the
Unfunded Mandates Reform Act

before the
Subcommittee on Technology and
the House Committee on Rules
U.S. House of Representatives

July 16, 2003

Mr. Chairman and Members of the Subcommittee, I am pleased to be here today to discuss the Unfunded Mandates Reform Act of 1995 (UMRA) and the role of the Congressional Budget Office (CBO) in implementing that legislation.

Through laws and subsequent regulations, the federal government sometimes requires state, local, and tribal governments and various private parties to expend resources to achieve certain goals. Except for amounts collected for certain fees and taxes, those resources are not counted in the federal budget. UMRA was enacted to focus more attention on the costs of such federal mandates. UMRA's supporters had many goals for the legislation, including ensuring that the Congress had information about the costs of mandates before it decided whether to impose them and encouraging the federal government to provide funding to cover the costs of intergovernmental mandates.

To accomplish those goals, title I of UMRA established requirements for reporting on federal mandates and set up various procedural requirements. For example, the House and Senate may not consider a bill unless the committee reporting it has published a CBO statement about whether the bill contains any intergovernmental or private-sector mandates and, if so, what they are estimated to cost. In addition, Members of Congress may raise a point of order against legislation that would create an intergovernmental mandate over the cost threshold specified in UMRA unless the legislation provides funding to cover those costs. That point of order would need to be waived by a majority vote for floor action on the legislation to continue. Such procedural requirements do not stop the Congress from passing bills it wants to, but they can introduce additional hurdles in deliberating unfunded mandates.
 

Trends in Federal Mandates Under UMRA

Title I of UMRA requires CBO to prepare mandate statements for bills approved by authorizing committees. In those statements, CBO must address whether a bill contains federal mandates and, if so, whether the direct costs of those mandates would be greater than the thresholds established in the law. Those thresholds, which are stated in 1996 dollars and are adjusted annually for inflation, are $50 million or more per year for the public sector (state, local, or tribal governments) and $100 million or more per year for the private sector. (In 2003, those thresholds are $59 million for intergovernmental mandates and $117 million for private-sector mandates.)

In the seven years since UMRA took effect, both the amount of information about mandate costs and interest in that information have increased. Since 1996, CBO has provided mandate cost statements for nearly all of the bills reported by authorizing committees. The agency has also provided mandate statements for many proposed floor amendments and some conference reports. Moreover, before proposed legislation is marked up, committee staff and individual Members have on occasion requested CBO's opinion about whether the legislation would create any new federal mandates and, if so, whether their costs would exceed the thresholds set by UMRA.

Over the past seven years, CBO has identified several patterns about federal mandates and their costs.

• Most of the legislation that the Congress considered between 1996 and 2002 did not contain federal mandates as UMRA defines them. Of the roughly 4,000 bills and other legislative proposals that CBO reviewed during that period, 465, or 11 percent, contained intergovernmental mandates, and 561, or 14 percent, contained private-sector mandates (see Table 1). Those percentages have varied only slightly from year to year. Over the seven-year period, the share of bills containing intergovernmental mandates has ranged between 9 percent and 14 percent, and the share containing private-sector mandates, between 11 percent and 19 percent. Generally over the period, legislation considered by the Congress contained more private-sector mandates than intergovernmental mandates.
  
  
• Most mandates examined by CBO would not have imposed costs greater than the thresholds set by UMRA. Only 42 bills, or 1 percent of the bills reviewed, had intergovernmental mandates with annual costs of $50 million or more, by CBO's estimate. Nearly 140 bills, or 3 percent of the bills reviewed, had private-sector mandates of more than $100 million a year. Few of the bills with either kind of mandate, however, contained federal funding to offset the costs.
  
  
• Over half of the intergovernmental mandates that CBO identified were explicit preemptions of state or local authority. For example, a bill in this Congress would preempt a number of state laws that regulate health coverage. Such preemptions of state regulatory laws often do not result in additional costs to state, local, or tribal governments because they do not require those governments to take any action. However, because they limit the exercise of state authority and preclude the application of state laws, they are intergovernmental mandates under UMRA.
  
  
• Few mandates with costs over the UMRA thresholds were enacted into law during the past seven years. Only two intergovernmental mandates with annual costs of at least $50 million became law--an increase in the minimum wage (in 1996) and a reduction in federal funding to administer the Food Stamp program (in 1997). Those enacted mandates represented fewer than 1 percent of the intergovernmental mandates that CBO reviewed from the time UMRA took effect through 2002.
  
  Twenty-one private-sector mandates with costs over the $100 million threshold were enacted into law. Of those, eight involved taxes, four concerned health insurance (requiring portability of insurance coverage, a minimum time for maternity stays, changes in Medicare coverage that imposed new requirements on private health insurance providers, and parity in insurance coverage providing mental health benefits and other medical benefits), four dealt with regulation of industries (telecommunications reform, changes in milk pricing, country-of-origin labels for certain foods, and a new safety requirement for automobiles), two affected workers' take-home pay (increases in the minimum wage and in federal employees' contributions for retirement), one imposed new requirements on sponsors of immigrants, one changed procedures for the collection and use of campaign contributions, and one imposed fees on airline travel to fund aviation security.
  
  
• Only a small fraction of the almost 1,600 public laws enacted since 1996 contained federal mandates that were not reviewed by CBO. In a few cases, mandates were enacted in appropriation bills, which CBO does not review under UMRA. In other cases, bills containing mandates were not considered by an authorizing committee, and sometimes, mandates were added to the legislation after CBO's review.

The Narrow Scope of UMRA

The numbers I am presenting today should be viewed in light of the fact that UMRA defines federal mandates and their costs narrowly. UMRA's "success" is tempered in some observers' view for three main reasons: because conditions for obtaining federal grants are generally not considered to be mandates; because indirect, or secondary, costs are not accounted for in UMRA; and because some bills are specifically excluded from UMRA's requirements. Many of the proposals to amend UMRA over the past seven years have focused on addressing those perceived limitations.

According to UMRA, the conditions attached to most forms of federal assistance (including most grant programs) are not mandates. Yet complying with such conditions of aid can sometimes be costly, and states often think of new conditions on existing grant programs as duties not unlike mandates. Two examples of such conditions are the requirements for receiving federal funding under the No Child Left Behind Act and the Individuals with Disabilities Education Act, which call for, respectively, designing and implementing statewide achievement tests and preparing individualized education plans for disabled children. All such requirements, while potentially costly for state and local governments, are clearly conditions for receiving federal assistance and thus are not considered mandates under UMRA.

Between 1996 and 2002, CBO identified more than 600 bills that would impose those types of "nonmandate" costs on state, local, or tribal governments. In most cases, however, CBO estimated that such costs would not be significant or would be covered by federal funding authorized in the bills.

UMRA requires CBO to estimate the direct costs of compliance for entities affected by federal mandates; it is those direct costs that are used to determine whether mandate costs in a bill exceed the relevant thresholds. But federal mandates also often have secondary effects, including the effects on prices and wages when the costs of a mandate imposed on one party are passed along to other parties, such as customers or employees. Those effects of federal legislation on other levels of government or the private sector are not subject to the requirements of UMRA. When such indirect effects are significant, however, CBO includes that information in its cost statements.

When sufficient time and data are available, CBO also provides quantitative estimates of the size of those effects. For example, CBO analyzed the indirect effects of proposed requirements for parity in insurance coverage of mental health conditions and other medical conditions, including possible impacts on workers' take-home pay, health insurance coverage, and fringe benefits. Similarly, CBO's analyses of proposed increases in the minimum wage routinely include the possible impact on the employment of low-wage workers. But such analyses can be complicated and are not always doable in the short time often available for cost estimates.

The scope of UMRA is further narrowed by the fact that the law does not apply to legislative provisions that deal with constitutional rights, discrimination, emergency aid, accounting and auditing procedures for grants, national security, treaty ratification, and title II of Social Security (Old-Age, Survivors, and Disability Insurance). Fewer than 2 percent of the bills that CBO reviewed in the past seven years contained provisions that fit within those exclusions. Many of the excluded bills dealt with national security or Social Security and generally did not contain costly mandates. Yet some of the excluded bills (for example, the election reform bills enacted in 2001, which dealt with the constitutional right of citizens to vote) contained costly requirements for state and local entities; however, because of the scope of UMRA, CBO did not estimate those costs as part of its review, and the requirements were not subject to the point of order established by that law.
 

Challenges to CBO in Implementing UMRA

For the most part, UMRA's provisions have been straightforward to carry out. In some cases, however, determining clearly whether a bill would impose a mandate as defined in the law or whether the cost would exceed the statutory thresholds has not been possible.

Determining what constitutes a mandate under UMRA can be complicated. For example, the law defines a mandate as "an enforceable duty except . . . a duty arising from participation in a voluntary federal program." Although an activity (such as sponsoring an immigrant's entry into the United States) may be voluntary, the federal program governing that activity (immigration laws) is not. In that case, a bill imposing new requirements on the sponsors of immigrants would constitute a mandate under UMRA. In contrast, other federal programs that are truly voluntary in nature may impose requirements on their participants that, by UMRA's definition, are not mandates. For example, a requirement on firms that bid on federal contracts to comply with the Davis-Bacon Act (which requires that employees be paid the prevailing wage rates) is not a mandate. Firms are entering into a contractual arrangement, and any requirements are part of that voluntary contract. Those distinctions between what is voluntary and what is mandatory are not always clear.

In addition, in some cases CBO has been unable to make such a determination about intergovernmental mandates, mostly because UMRA is unclear about how to view a bill that might raise the costs of an existing mandate without imposing a new one. In those cases, the bills contain provisions that by themselves would not establish any new enforceable duties and would not directly amend existing mandates. However, those provisions would have indirect effects on existing mandates, making them more expensive to carry out. For example, a proposed joint resolution in the last Congress would have provided Congressional approval of the site at Yucca Mountain, Nevada, for the storage of nuclear waste; while the resolution, by itself, would have established no new enforceable duty on state, local, or tribal governments, shipments of nuclear waste to the Yucca Mountain site would have increased costs to the state of Nevada for complying with other existing federal requirements. UMRA is unclear about whether a bill's effect on the costs of existing mandates should be counted as a new mandate cost when the bill itself contains no new enforceable duty.

When CBO determines that a legislative proposal contains a federal mandate, the agency may face numerous challenges in estimating the costs. CBO has been unable to estimate mandate costs for about 10 percent of the bills containing mandates. Those uncertainties arose for several reasons. In many instances, estimating the costs of a mandate at the legislative stage, before regulations to implement it have been developed, may be impossible. Even the mandated parties may not be able to estimate costs reliably without knowing what the regulations to carry out the mandate will entail. In such cases, even determining how many state and local governments or entities in the private sector would be affected by a mandate may be impossible. In other cases, the entities that would be subject to a particular mandate are diverse and would not be affected uniformly, making it difficult to develop a meaningful estimate of the costs of compliance from a limited sample.

Some mandates occur in bills that would extend, and sometimes expand, an expiring mandate. In such cases, an ambiguity in UMRA's definition of direct costs makes it unclear whether to measure the change in the costs of extending the mandate relative to the current level of mandate costs or to assume that the mandate would expire and measure the costs of its extension as if that were a new mandate.

Finally, the legislative process often does not allow enough time for a comprehensive analysis of all relevant effects. Fortunately, UMRA requires CBO to determine whether the costs of complying with mandates would exceed specific thresholds and to provide cost estimates only for mandates that do. If UMRA required CBO to provide more-detailed estimates for each mandate, the agency's job would require considerably more resources.
 

Proposals to Expand UMRA

The Unfunded Mandates Reform Act has clearly increased both the demand for and the supply of information on the costs of federal mandates. Moreover, that information has played a role in Congressional debate about several issues over the past seven years.

To date, lawmakers have made only one, relatively minor, change to UMRA. The State Flexibility Clarification Act of 1999 (Public Law 106-141) requires authorizing committees and CBO to provide more information in committee reports and mandate statements for legislation that would "place caps upon, or otherwise decrease, the federal government's responsibility to provide funding to state, local, or tribal governments" under certain large entitlement grant programs (such as legislation that would cap the federal contribution to Medicaid). In general, that requirement for additional information applies to few bills, and no legislation reported by authorizing committees since the requirement was enacted has been affected by it.

Since UMRA's enactment, lawmakers and other interested parties have proposed additional ways of expanding or changing title I in several ways. Most proposals seek to increase the types of bills that would be subject to UMRA's cost-estimating and point-of-order provisions. One proposal would build on UMRA's perceived success in focusing Congressional attention on unfunded intergovernmental mandates by expanding the law to allow for a point of order against bills that contain private-sector mandates with costs over the statutory threshold (as the law currently does for intergovernmental mandates). Such an expansion could establish an additional hurdle for private-sector mandates and could increase the demand for additional cost information about such mandates. Another proposal would expand UMRA's definition of a mandate to include certain conditions in large entitlement grants administered by state or local governments. Both of those proposals were included in the Mandates Information Act, which was considered by the Congress in 1998 and 1999 but never enacted.

Other proposals to change or expand UMRA have included broadening the definition of intergovernmental mandate to include new conditions on any existing grant program; narrowing the exclusions discussed above to apply only to the provisions allowing for a point of order and not to the requirement that CBO provide cost information; and eliminating the threshold so that any mandate, regardless of its costs, could trigger a point of order. Such a change would allow a point of order to be raised when the Congress is considering bills that preempt state and local authority. At least half of the intergovernmental mandates that CBO identifies are preemptions, and because the costs of preemptions rarely approach or exceed the threshold, the point of order cannot be used against them.
 

Conclusion

In closing, Mr. Chairman, in the seven years since UMRA took effect, both the amount of information about the cost of federal mandates and Congressional interest in that information have increased considerably. In that respect, title I of UMRA has proved to be effective. Moreover, numerous pieces of legislation that originally contained a significant unfunded mandate were amended to either eliminate the mandate altogether or to lower its costs. CBO has made a determined effort to provide the Congress with the timely information it needs to implement UMRA, and the agency is committed to continuing that effort in the years to come.
 

---

Charts Presented at the Hearing