Last month, the Congressional Budget Office published An Analysis of Certain Proposals in the President’s 2022 Budget. Since then, CBO has completed its analysis of another proposal in the President’s budget, an increase in spending for the Internal Revenue Service’s (IRS’s) enforcement activities. CBO estimates that portions of the Administration’s proposal to increase funding for the IRS by $80 billion over the 2022–2031 period would increase revenues by approximately $200 billion over those 10 years. That estimate does not include changes in revenues resulting from portions of the proposal that involve new information-reporting requirements and other changes to the tax code; those changes are estimated by the staff of the Joint Committee on Taxation (JCT).
The Administration proposes funding for the IRS that is $80 billion greater over 10 years than the amounts in CBO’s July 2021 baseline projections (which reflect the assumption that current laws generally do not change). Two types of funding would be provided: discretionary appropriations, which would mainly be used for enforcement activities; and mandatory funding, which would be used for a variety of activities (not only enforcement but also operations support, business-systems modernization, and taxpayer services).
Spending would increase in each year between 2021 and 2031, though the highest growth would occur in the first few years. By 2031, CBO projects, the proposal would make the IRS’s budget more than 90 percent larger than it is in CBO’s July 2021 baseline projections and would more than double the IRS’s staffing. Of the $80 billion, CBO estimates, about $60 billion would be for enforcement and related operations support.
The Administration also proposes that financial institutions increase their reporting about account inflows and outflows. Part of the increased funding would support the implementation of a new information-reporting system to be used by those institutions. The resulting effects on revenues are estimated by JCT and are not included in CBO’s estimate of an approximately $200 billion increase.
How CBO Estimates the Effect on Revenues of Increased IRS Funding
CBO’s estimate of revenues is based on the IRS’s projected returns on investment (ROIs) for spending on new enforcement initiatives. The IRS estimates those ROIs by calculating the expected revenues that would be raised from taxes, interest, and penalties as a result of the new initiatives and dividing them by their additional cost. (The agency has provided ROIs over the past five years as part of its budget justification.) The IRS’s ROIs ramp up over three years as staff become trained and fully productive, arrive at the peak level, and then stay there. In recent years, peak ROIs have ranged from 5 to 9. That is, a $1 increase in spending on the IRS’s enforcement activities results in $5 to $9 of increased revenues.
CBO adjusts the ROIs so that they better reflect the marginal return on additional spending. First, CBO expects the IRS to prioritize the enforcement activities that it thinks will have the highest average return; additional enforcement spending would therefore have lower returns than previous spending. Second, CBO expects taxpayers to adapt to the IRS’s enforcement activities and adopt new ways of evading detection, so an enforcement activity may have a lower return in later years. Finally, the productivity of the IRS’s enforcement activities will also depend on the IRS’s other capabilities. For example, modernized information technology that stored all of a taxpayer’s information in digital form could increase the productivity of examiners (the employees who detect taxpayers’ noncompliance).
CBO’s estimate of revenues also accounts for the timing of collections resulting from enforcement activity by new hires. Taxes are assessed at the end of an audit; if taxpayers disagree with the assessment, they can appeal and continue to litigate. The length of each step depends on the complexity of the case. CBO estimates that an audit of medium complexity would take 24 months to complete. That time, combined with the expected training time for an experienced new hire, suggests that the IRS would begin to collect revenues 30 months after the new hire joined the agency. (The timing would be longer when cases were more complex or when the taxpayer did not agree to the assessment and appealed.)
What Is Incorporated Into CBO’s Estimate. CBO’s estimate of the change in revenues is relative to the amount of revenues collected under current law (which is reflected in CBO’s baseline budget projections). Under guidelines agreed to by the legislative and executive branches, this change in revenues typically would not be included in a cost estimate for legislation that brought about the change, but it would be reflected in CBO’s baseline budget projections once the legislation was enacted.
CBO’s estimate reflects the assumption that the proposed increase in funding would follow the proposed expansion of information reporting. Expanded information reporting might allow the IRS to better target potentially noncompliant taxpayers; it might also prompt taxpayers to file more accurate tax returns. It might have a positive effect on revenues collected, but it might also reduce the ROIs from enforcement activities, because if returns are more accurate, there will be less noncompliance to audit. In CBO’s and JCT’s judgment, those effects roughly offset each other, on net, resulting in a small positive effect on ROIs.
CBO’s estimate includes “direct revenues” and “protected revenues.” Direct revenues are generated from the IRS’s auditing and collection efforts. Protected revenues result when the IRS prevents a taxpayer from recouping previously assessed and paid taxes—for example, when the IRS prevents fraudulent refunds or disallows claims in taxpayers’ amended returns.
The estimate reflects CBO’s expectation that the increased enforcement activities would change the voluntary compliance rate—that is, the share of taxes owed that are paid voluntarily and on time—only modestly. The magnitude of that effect is highly uncertain, however, and the empirical evidence about the effects of audits on taxpayers’ behavior is inconclusive. Research about such deterrence finds varying responses, depending on the type of taxpayer. People generally increase their reported income in the years following an audit, but people with higher income generally do not, and neither do corporations. (For more discussion, see Box 1 in CBO’s July 2020 report Trends in the Internal Revenue Service’s Funding and Enforcement.)
How the Current Analysis Differs From Previous Analyses. In that July 2020 report, CBO estimated that a $40 billion increase in enforcement funding would raise $103 billion (for a net effect of $63 billion). The methods used for this estimate differ in several ways from the methods used for that one.
First, CBO used updated ROIs that incorporated the IRS’s most recent estimates of the return on enforcement activities. CBO then adjusted the ROIs to reflect both direct revenues and protected revenues, increasing the peak ROI from 6.4 to 7.1.
Second, CBO’s current methods allow for positive interaction between enforcement spending and other IRS funding. That is, CBO accounts for ways in which increased capabilities, such as more digitization of taxpayers’ information and greater visibility of income flows, can increase the productivity of enforcement activities.
Third, this analysis reflects a longer time frame for receiving enforcement revenues because of the complexity of audits associated with high-wealth individuals, large corporations, and partnerships. Taxpayers with greater resources may be more likely to appeal assessments or to litigate their disputes in the U.S. Tax Court, delaying the receipt of assessed taxes. As a result, revenues from some audits will not be received until later than CBO estimated in its July 2020 analysis.
Sources of Uncertainty. The change in revenues resulting from an increase in the IRS’s funding could be different from CBO’s estimate. It depends on the IRS’s ability to hire experienced candidates, changes in voluntary compliance, and the interaction of enforcement funding with the IRS’s other capabilities.
The IRS intends to hire mid- and senior-level people with private-sector experience who will not require a great deal of training to become productive. But it might not be able to hire its desired mix of candidates. If it hired less experienced candidates, it would have to spend more resources training them. Not only would they take longer to become productive, but current staff members would have to devote more time to training them. A related source of uncertainty in CBO’s estimate is attrition: If it proved higher than expected, personnel would have fewer years at full productivity.
An increase in the IRS’s funding could signal that the agency was more capable of detecting noncompliance, thus increasing voluntary compliance and revenues. However, if there were fewer noncompliant taxpayers to audit, the ROIs from the IRS’s enforcement activities would drop, and the direct revenues from increased enforcement would be lower than CBO estimated.
Finally, it is unclear how much the greater information reporting or the increased IRS spending in areas other than enforcement (such as technology) could improve examiners’ productivity. Greater nonenforcement spending might increase overall revenues but decrease ROIs—for example, if improved services for taxpayers enabled those taxpayers to more accurately determine their tax liability, reducing the pool of noncompliant taxpayers to audit.
Effects on Taxpayers
The proposed increase in spending on the IRS’s enforcement activities would result in higher audit rates than those underlying CBO’s baseline budget projections. Between 2010 and 2018, the audit rate for higher-income taxpayers fell, while the audit rate for lower-income taxpayers remained fairly stable. In CBO’s baseline projections, the overall audit rate declines, resulting in lower audit rates for both higher-income and lower-income taxpayers. The proposal, by contrast, would return audit rates to the levels of about 10 years ago; the rate would rise for all taxpayers, but higher-income taxpayers would face the largest increase. In addition, the Administration’s policies would focus additional IRS resources on enforcement activity aimed at high-wealth taxpayers, large corporations, and partnerships. CBO estimates that if the proposals were enacted, tax compliance would be improved, and more households would meet their obligation under the law.
Higher audit rates would probably also result in some audits of taxpayers who would later be determined not to owe additional taxes. However, the Administration’s proposal for more information reporting, as well as additional spending on IRS technology, might reduce the burden on compliant taxpayers by allowing the IRS to better target noncompliant ones and to reduce the number of audits that resulted in no change in tax assessment.
Phillip L. Swagel is CBO’s Director.