On Balance: What’s the Score? The Congressional Budget Office and its Role in the Policy Process

December 8, 2017

By Stuart Guterman

The Congressional Budget Office (CBO) plays an important role in the federal legislative process. CBO’s score on a given bill—that is, its estimate of how it would affect the federal budget deficit—can determine whether Congress decides to go forward with the bill, modify it to get a more favorable estimate, or simply drop it. Given their importance, debates over CBO’s scores and the methods they use to produce them can be as controversial as the bills that are being considered. While this controversy can be politically motivated (with advocates on either side of an issue arguing for a score that is more favorable to their position), it also stems from limited understanding of CBO’s intended role in the process—and reflects the difficulty of conducting analyses of benefits and costs in the context of policy decisions.

Because Congress wanted an objective scorekeeper to develop accurate budget estimates, it created CBO in the Congressional Budget and Impoundment Control Act of 1974. Among other functions, CBO is required by law to produce a formal cost estimate for nearly every bill that is approved by a full committee of either the House or the Senate. It also produces informal cost estimates for a much larger number of legislative proposals. This scorekeeping function has become more important over time. CBO’s Update to the Budget and Economic Outlook estimates the current federal budget deficit at almost $700 billion and projects it will more than double over the next decade, and Congress is acutely conscious of the budgetary implications of potential legislation.

Informed decision-making, however, involves consideration of multiple factors, which are not limited to the budgetary implications of the decision (in fact, although CBO is required to estimate the impact of unfunded mandates on state governments and the private sector, its primary focus is on the federal budget). Moreover, determining the broader impact of potential legislation is a much more complicated endeavor, involving diverse and often conflicting preferences and values, as well as expectations about the future that may be determined by events and responses that cannot be foreseen. These other factors—although they may be at least as important—may not easily be monetized.

A case in point is CBO’s analysis with the Joint Committee on Taxation (JCT) of the tax reform bill recently passed by the Senate Finance Committee; one provision in that bill would repeal the individual mandate, a requirement enacted in the Affordable Care Act of 2010 that all individuals have health insurance coverage or pay a penalty. The CBO/JCT estimates that eliminating the individual mandate would reduce the federal budget deficit by $338 billion from 2018 through 2027, mainly by reducing the amount the federal government would spend on health insurance subsidies to individuals and Medicaid subsidies to states; that reduction in spending would offset some of the deficit-increasing impact of the bill, which nonetheless is projected to increase the budget deficit by $1.4 trillion over ten years.

However, CBO/JCT also estimates that, by 2027, the number of people with health insurance would decrease by 13 million; moreover, they project that average premiums in the non-group health insurance market (through which people who don’t have health insurance directly through an employer or the government may obtain coverage) would be about 10 percent higher than under current law. How are these costs and benefits to be compared? Although its estimated impact on the federal budget is relatively straightforward (although still controversial), the broader implications of the provision are less easily represented in a comparable metric.

Moreover, although the distributional impact of prospective legislation—particularly in the case of a major tax bill—can be critically important, it generally is not reflected in the bill’s score. While the JCT estimates that total federal taxes for all income groups would be $240.1 billion lower in 2019 under the bill passed by the Senate Finance Committee than under current law, more than three-quarters of that reduction ($181.9 billion) would accrue to households with incomes of $100,000 or more, and almost half ($117.8 billion) to households with incomes of $200,000 or more. In 2027, federal taxes would be $27.3 billion (2.9 percent) higher than under current law for households with incomes under $75,000, while federal taxes for households with incomes of $100,000 or more would be $15.9 billion (0.5 percent) lower.

It is important to understand what these estimates mean, and the circumstances under which they are most—and least—useful. Although its methodology always can be improved, by far the most egregious error regarding CBO’s work is the tendency of policy makers to overemphasize seemingly precise cost estimates relative to the many other (but certainly harder to define and quantify) factors that should be considered in policy decisions.

In evaluating the impact of potential legislation, knowing the score often is not enough.


Stuart Guterman, currently an independent consultant, has spent more than 30 years as a health policy researcher and policymaker, including time as deputy director of the Medicare Payment Advisory Commission, director of the Office of Research, Development, and Information at the Centers for Medicare and Medicaid Services, and Vice President at the Commonwealth Fund. He also has spent time at the Health Care Financing Administration, the Prospective Payment Assessment Commission, the Urban Institute, and the Congressional Budget Office.