In December 2017, President Donald Trump signed into law the Tax Cuts and Jobs Act, the most significant tax legislation in 30 years. The rushed passage of the law meant that—even more so than usual—the Department of the Treasury and the IRS would need to fill in the details of the statutory ambiguities through regulation. Then, in April 2018, the Treasury Department and the White House Office of Management and Budget agreed to substantially expand the set of tax regulations reviewed by the Office of Information and Regulatory Affairs, part of the Office of Management and Budget. This agreement also expanded the set of regulations for which a cost-benefit analysis is required.1
More than 2 years later, it is clear this experiment in cost-benefit analysis of tax regulations has failed. In a period of incredible regulatory activity from the Treasury Department and the IRS, the cost-benefit analyses released alongside proposed and final regulations provide little information relevant to assessing the merits of those regulations. Moreover, while tax experts criticize many of the TCJA regulations for providing unmerited windfalls to favored groups, the cost-benefit analyses for those regulations often fail to identify these windfalls or provide critical analysis of them.2 Regulatory giveaways in the interpretation of the new deduction for income from pass-through businesses, the tax on global intangible low-taxed income, and the base erosion and anti-abuse tax, for example, were subject to little or no critical analysis.
The weaknesses of these analyses are rooted in the framework that the Office of Information and Regulatory Affairs mandates for the cost-benefit analysis of federal regulations.3 This framework is fundamentally ill-suited to the evaluation of tax regulations. First and foremost, this framework treats revenue impacts as neither a cost nor a benefit even though raising revenues is the primary purpose of taxation. Lax regulatory interpretations that generate windfall gains for recipients often have few or no costs. Indeed, a regulation that simply gives up on preventing some form of corporate tax avoidance could generate net benefits in this framework because corporations would face a reduced cost of avoiding taxes and the revenue loss itself would not be treated as a cost. In contrast, more stringent regulatory interpretations that result in higher revenues may have only incidental benefits because the revenues themselves are not counted as a benefit.Download File
Cost-benefit analysis of U.S. tax regulations has failed. What should come next?
The framework also relegates changes in the distribution of the tax burden to second-tier status. Though it may seem neutral to instruct the Treasury Department and the IRS to ignore changes in the distribution of the tax burden in assessing the costs and benefits of tax regulations, it is not. High-wealth taxpayers are generally better able to avoid tax than low-wealth taxpayers. This framework thus puts a thumb on the scale for reallocating taxes from the wealthy to everybody else. The reduction in tax avoidance is deemed a benefit while changes in the distribution of the tax burden are deemed neither a cost nor a benefit. This framework for cost-benefit analysis is often described as disregarding redistributive impacts, but it would be more accurate to say that it adopts a specific normative view on how to judge redistributive impacts.
Evaluating the merits of a tax regulation requires assessing the impacts of that regulation on tax revenues and the tax burden. The burden imposed by a tax increase is the reduction in living standards for individuals and families that results from that tax increase, ignoring the benefits from increased spending or future spending cuts avoided. Similarly, the reduction in burden resulting from a tax cut is the increase in living standards that results from the tax cut.4 Do the revenue losses prevented justify the burden imposed by a more stringent interpretation of the law, taking into account who would bear that burden? Or does the burden reduction, taking into account who would benefit, justify the revenue loss? The cost-benefit analysis that the Office of Information and Regulatory Affairs instructs agencies to produce cannot answer these questions.
Indeed, regulatory giveaways in the implementation of the Tax Cuts and Jobs Act probably avoided critical scrutiny in part because the analytic frame does not conceive of the revenue loss from those giveaways as a cost and thus does not focus attention on them. If decision-makers within the Department of the Treasury and the IRS used these analyses to guide their regulatory choices, then they would likely be making less-informed choices than they would if they simply ignored the analyses.
Recognizing that the existing approach to cost-benefit analysis is inappropriate for tax regulations, the Office of Management and Budget—the Office of Information and Regulatory Affairs’ parent agency—should have issued new guidance on conducting economic analysis specific to the case of tax regulations.5 Yet more than 2 years after the expanded review of tax regulations began, it has issued no public guidance on how more informative cost-benefit analysis of tax regulations could be conducted. If anything, the movement is in the wrong direction. In December 2019, the Office of Management and Budget solicited comments on a potential technical change to the cost-benefit analyses of regulations relating to spending programs that, if adopted, would create parallel problems in that context.6
In light of this performance, the next administration should eliminate the requirement for cost-benefit analysis of tax regulations and OIRA’s authority to review that analysis. Instead, the Treasury Department should provide a qualitative and, when feasible, quantitative evaluation of tax regulations grounded in the fundamental considerations of tax policy; specifically, the impacts on:
- The level and distribution of the tax burden
- Compliance costs7
The traditional modes of tax analysis developed for the legislative context—revenue estimates and distribution analyses—are exactly the modes of analysis required to evaluate the effects of tax regulations on revenues and on the level and distribution of the tax burden. In the regulatory context, these are appropriately supplemented with additional analysis of compliance costs. These tools are the analog of the cost-benefit framework, taking into account the distinct purpose of tax regulations.
The Department of the Treasury’s analysis of the impacts on revenues, burden, and compliance costs should focus on the decision points where the department and the IRS have discretion to regulate differently, as this is the analysis that would most directly inform regulatory decision-making, and the analysis should be conducted only when the relative effects of different interpretations of the law are substantial.
Historically, the Treasury Department and the IRS rarely evaluated the economic effects of tax regulations under the view that the agencies were rarely exercising discretion in interpreting the tax laws. To the extent that the Treasury Department and the IRS were not evaluating the economic effects of tax regulation frequently enough, the appropriate reform is not to apply an inappropriate analytic framework or require after-the-fact analysis of enacted statutes through a prejudicial lens. Rather, it is to require that when the Department of the Treasury and the IRS are using their discretion in interpreting the tax laws, they apply the tools of tax analysis to better understand such regulations’ economic effects.
In this report, I first critically evaluate the published cost-benefit analyses for regulations implementing the Tax Cuts and Jobs Act. I then discuss how the weaknesses of these analyses are rooted in the framework that the Office of Information and Regulatory Affairs requires for cost-benefit analyses, which is fundamentally inappropriate for tax regulations. I conclude by recommending a better approach: The next administration should eliminate the requirement for cost-benefit analysis as applied to tax regulations and instead require a Treasury Department-led economic analysis of tax regulations, focusing on the impacts on revenues, on the level and distribution of the tax burden, and on compliance costs.
Cost-benefit analyses of TCJA regulations provide little insight on the merits of the regulations
The passage of the Tax Cuts and Jobs Act in December 2017 initiated an intensive period of rulemaking at the Treasury Department and the IRS. The new requirement for cost-benefit analysis of tax regulations raised the possibility that decision-making on these regulations would be guided by more rigorous analysis. Unfortunately, the cost-benefit analyses of the regulations implementing the new tax law provide little insight on the merits of the regulations. There are four main reasons for this weakness. The analyses:
- Treat revenue impacts as neither a benefit nor a cost and direct little or no attention to those impacts even though the primary purpose of taxation is to raise revenue
- Ignore impacts on the level and distribution of the tax burden
- Define benefits in terms of efficiency gains in a manner that risks misleading decision-makers and the public about the effects of the regulations
- Adopt a baseline that obscures the key question of how the Department of the Treasury and the IRS use their discretion when issuing regulations
I now consider each of these weaknesses in turn.
Revenue impacts are treated as neither a benefit nor a cost and receive little or no attention even though the primary purpose of taxation is to raise revenue
The primary purpose of taxation is raising revenues. Different regulatory choices can result in higher or lower revenues. Thus, the impact of regulatory choices on revenues must play a central role in any economic analysis of tax regulations. It is only by weighing impacts on burden and compliance costs against revenues that decision-makers can determine whether a regulation imposes an appropriate burden in order to prevent tax avoidance and the associated revenue loss, or whether the burden and compliance costs are excessive for the revenue loss avoided.
Similarly, determining whether a desired reduction in burden or compliance costs comes at a reasonable cost in terms of foregone revenues or an excessive cost requires an estimate of the revenue loss. However, consistent with OIRA’s required framework for cost-benefit analysis, revenues are relegated to second-tier status—neither a cost nor a benefit but a transfer—and are largely ignored in the analyses of TCJA regulations.
For example, in 2019, the Department of the Treasury and the IRS finalized a regulation effectively prohibiting the use of charitable contributions as a means of circumventing the limitation on the deductibility of state and local taxes enacted by the Tax Cuts and Jobs Act.8 This regulation provides a striking example of how the inappropriate treatment of revenues in cost-benefit analyses of tax regulations plays out in practice.
The Tax Cuts and Jobs Act limited the federal income tax deduction for state and local taxes to $10,000. But charitable contributions remained deductible without limit. Thus, one relatively easy way to avoid the limit on state and local taxes would be for states to provide a credit against state and local taxes for contributions made to charitable entities established by states or localities that provide similar public services as those governments. In response, the Treasury Department and the IRS issued a regulation stating that the state or local tax credits would be treated as a benefit received by the taxpayer in exchange for the taxpayer’s contribution, which would have the effect of eliminating any charitable deduction for those contributions for federal income tax purposes.
In the preamble to the final regulation, the Department of the Treasury and the IRS state multiple times that a primary purpose of the regulation is to prevent revenue losses attributable to erosion of the tax base contrary to the purpose of the statute. For example, in responding to comments on the proposed regulation, the Treasury Department and the IRS state, “[t]he final regulations are also supported by important tax policy considerations, including the need to prevent revenue loss from the erosion of the limitation [on the deduction for state and local taxes].”9
In the regulatory impact analysis itself, the Treasury Department and the IRS further identify revenues as an important reason the regulation is needed: “[d]isregarding the value of state and local tax credits received or expected to be received in return for charitable contributions would precipitate revenue losses that would undermine the limitation on the deduction for state and local taxes adopted by Congress under the Act.”10
Notwithstanding the central role of revenues in the case for the regulation, there is no discussion of revenues whatsoever in the cost-benefit analysis, consistent with the Office of Information and Regulatory Affairs’ mandated framework for such analyses. Instead, the discussion focuses on benefits such as a reduction in “inefficient choices motivated by the potential tax benefits” of the workarounds and a more neutral treatment of charitable contributions to different organizations and on costs such as compliance costs for taxpayers who receive tax credits in exchange for charitable contributions and for organizations that issue tax credits.11
In short, the cost-benefit analysis provides no insight on the merits of the rule because it ignores the core purpose of the rule: preventing revenue losses inconsistent with congressional intent.
Despite the limitations of the cost-benefit analysis of this regulation, it is something of a high-water mark for the acknowledgement of the role of revenues in the case for regulation. Even if not part of the discussion of costs and benefits, revenues are acknowledged in discussing the need for regulation. In contrast, some of the highest-profile TCJA regulations have little or no discussion of revenues at all.
The recently issued final regulations regarding the limitation on the deductibility of business interest provide an illustrative example of an analysis that simply ignores revenues.12 The Tax Cuts and Jobs Act imposed a limit on the amount of interest businesses can deduct on their tax returns. The Treasury Department and the IRS issued regulations providing guidance on the determination of this limit.
The final regulation departs from the proposed regulations in two ways that will generally reduce taxes. First, the final regulation narrows the definition of what counts as interest, increasing the number of deductible payments that are not subject to the limit. Second, the final regulation modifies the computation of the limit itself in a way that generally relaxes it. In assessing these changes, however, the Treasury Department and the IRS do not discuss the probable loss of revenues that would result from their decision to relax the limit in the final regulation, relative to the proposed regulation. Absent a discussion of the revenue impacts, there is little basis on which decision-makers could decide whether the reduction in compliance costs and burden resulting from these implicit tax cuts is worth it.
Similarly, the final regulations implementing the high-tax exclusion from the tax on global intangible low-taxed income fail to interrogate revenue impacts in a meaningful way.13 The Tax Cuts and Jobs Act identified a certain type of business income—termed global intangible low-taxed income, or GILTI—and imposed a tax on this income to reduce tax avoidance. The so-called high-tax exclusion exempts certain income from this tax.
In assessing the benefits and costs of the regulations governing the high-tax exclusion, the Department of the Treasury and the IRS focus on the certainty and clarity provided by more detailed guidance in interpreting the statute, the benefits of aligning the tax treatment on economically similar transactions, the potential impact on wasteful tax avoidance, and the potential reductions in compliance costs. At no point does the discussion directly evaluate the revenue impacts of the different options.
The inattention to revenues in the analysis of the high-tax exclusion is particularly striking because the exclusion is largely an administratively implemented tax cut. The Treasury Department and the IRS finalized regulations implementing the basic contours of the GILTI provisions with a much narrower exclusion at the same time that they proposed the high-tax exclusion.14
Indeed, the final regulations implementing the GILTI provisions explicitly state that an expanded high-tax exclusion is not available until the proposed regulations implementing such an exclusion are effective. The analysis of the high-tax exclusion is thus an analysis of a tax cut in which the revenue loss is not treated as a cost, but the reduction in tax avoidance behavior resulting from the tax cut—that companies would not have to maneuver around the narrower exception to minimize taxes—is treated as a benefit. Such an approach to the analysis is precisely what the inappropriate framework set forth by the Office of Information and Regulatory Affairs requires and what concerned critics of that framework.
These are only three examples of a consistent pattern. The cost-benefit analyses of regulations implementing the Tax Cuts and Jobs Act largely ignore revenues and, as a result, provide little useful guidance for decision-makers. Moreover, decisions guided by the analysis that was produced would be skewed by the inappropriate lack of emphasis on revenues.
The level and distribution of the tax burden are ignored
In addition to downplaying revenue impacts, the cost-benefit analyses of the TCJA regulations almost entirely ignore the level and distribution of the tax burden. Who received a tax cut as a result of the preferential treatment for real estate brokers, S corporation banks, and foreign banks provided by the regulations implementing the deduction for income from pass-through businesses?15 What was the reduction in burden for those taxpayers? Who received a tax cut as a result of the generous interpretation of rules governing Opportunity Zones, and what was the reduction in burden for them?16 The cost-benefit analyses of the TCJA regulations are largely silent on these issues, even as distribution analyses are central in the broader tax policy debate.
Attention to the distribution of the tax burden is at the very core of the economic analysis of tax policy. In the simplest academic treatment of the economics of income taxation, the income tax exists only because policymakers have an interest in the distribution of the tax burden.17 Policymakers have adopted income as an indicator of the ability to pay tax, and allocate the tax burden to families based on their income, among other factors.
The reasons for attention to the distribution of the tax burden go well beyond interest in the distribution of the tax burden to lower- and higher-income families. One function of tax regulation is to prevent tax evasion, the nonpayment of taxes that a person or business is legally required to pay. Through regulation or other forms of guidance, for example, the Treasury Department and the IRS have the authority to identify tax shelter transactions that must be reported to the IRS. Such guidance does not change the amount of tax owed but can change the amount of tax collected.
Disregarding the distribution of the tax burden in determining the costs and benefits of regulations amounts to the assumption that policymakers should be indifferent as to whether a dollar is raised from someone cheating on their taxes. This injects an unsupported, normative view on the appropriateness of tax evasion into the analysis—treating it as entirely neutral—when the motivation for certain tax regulations is that people should pay the tax they are legally obligated to pay.
Absent an analysis of how a tax regulation increases or decreases the tax burden, and how the change in the burden varies across taxpayers, decision-makers have little basis on which to judge the merits of a regulation. The silence of the TCJA regulatory analyses on this front further emphasizes the extent to which they provide little useful guidance to policymakers.
Benefits consist largely of claimed efficiency gains that risk misleading decision-makers and the public
In lieu of revenue gains or burden reductions, the primary benefits claimed for TCJA regulations are increases in efficiency. In describing the benefits of the regulation prohibiting workarounds for the cap on the federal income tax deduction for state and local taxes, for example, the Treasury Department and the IRS state:
This regulation likely reduces economically inefficient choices motivated by the potential tax benefits available if this regulation were not promulgated. Under the prior law and baseline scenarios, state and local governments have an incentive to fund governmental activities through entities that are eligible to receive deductible contributions and to establish tax credits … The final rule with Notice 2019–12 substantially diminishes this incentive to engage in economically inefficient tax-avoidance behavior.18
Similarly, in the regulations finalizing the high-tax exclusion for the tax on global intangible low-taxed income, the Treasury Department and the IRS argue that if they adopt a more limited exclusion:
taxpayers with high-taxed gross tested income would have an incentive to structure their foreign operations in order to [convert income that is not eligible for the exclusion into income that is eligible] … Because businesses are largely not currently structured in this way, such an organization would entail restructuring, which would potentially be costly and only available to certain taxpayers yet would not provide any general economic benefit … This outcome may lead to higher compliance costs and less efficient patterns of business activity relative to a regulatory approach that provides a broader GILTI high-tax exclusion.19
Reliance on efficiency as the metric for evaluating tax regulations is inappropriate and risks misleading policymakers. To make an informed decision about tax policy choices, policymakers need to be able to compare revenues and burden, taking into account the distribution of that burden. Efficiency is simply the wrong metric. Indeed, the inattention to revenues and the distribution of burden discussed in previous sections is a consequence of this focus on efficiency.20
One way of understanding the problem with using efficiency for this purpose is to recognize that there is a fundamental link between taxes and spending. An increase in tax revenues from some policy change allows for higher spending or a reduction in other taxes. A reduction in revenues requires lower spending or an increase in other taxes. Of course, there is no requirement that these adjustments occur contemporaneously or even through explicit policy changes.
By assuming that revenues are a mere transfer and that a dollar has the same value regardless of who receives it, efficiency-based analysis of tax changes sidesteps this constraint. However, efficiency-based analysis does this by, in effect, evaluating not the impacts of the specific tax change proposed, but rather a hypothetical alternative policy change with no net budgetary impact.
As a result, there is a fundamental indeterminacy in the specification of the proposed policy. Is the evaluation an evaluation of the proposed policy, combined with the imposition of a head tax or grant to every person that would offset its budget effects? Of the proposed policy, combined with a change in spending that happens to be valued exactly dollar for dollar? Importantly, this means there is a sense in which it is simply not possible to conduct a distribution analysis of the tax proposal for which the efficiency impact is being evaluated because it is ambiguous what that proposal even is.
While this critique applies in principle to efficiency-based analyses of any type of regulation, the consequences are far more severe in the tax context. For many regulations, raising revenues is not the underlying motivation for the policy. In the tax context, it is. Moreover, Congress has adopted a progressive income tax precisely to control the distribution of the tax burden. Conducting analysis not of the regulation under consideration but of another hypothetical policy, the revenue impacts of which have been set to zero and the distributive impacts of which are indeterminate—as an efficiency-based analysis implicitly does—serves no useful purpose.
If decision-makers already have an analysis of the impacts of a regulation on revenues and burden, then efficiency analysis of tax regulations is at best superfluous and at worst prejudicial. The impacts on revenues and burden summarize the impacts of a tax regulation. Evaluating tax regulations in terms of revenues and burden simply stops short of taking a normative view on the distribution of the tax burden imposed or adopting assumptions about how budgetary shortfalls will be closed in order to quantify the net benefits.
An analysis of tax regulations in terms of revenues and burden appropriately leaves these normative judgments to regulatory decision-makers. Evaluating tax regulations through the lens of efficiency imposes specific normative judgments under the guise of technocratic analysis.
Finally, efficiency-based analyses risk misleading decision-makers and the public not just because readers may be unaware of the implicit assumptions in them, but also because government documents and reports refer to regulations that increase efficiency as regulations with net benefits and refer to regulations that reduce efficiency as regulations with net costs. The terms “costs” and “benefits” appear to refer to familiar and intuitive concepts, but in this case, those familiar concepts have been replaced with unfamiliar technical concepts that entirely ignore or improperly identify relevant costs and benefits. A system in which beneficial tax regulations end up being described as having net costs due to the use of an unjustified technical framework is a system biased against the issuance of those regulations.
The baseline used in TCJA cost-benefit analyses means analyses largely ignore the key question of how the Treasury Department and the IRS use their discretion
The baseline adopted by the Department of the Treasury and the IRS for conducting cost-benefit analyses of TCJA regulations reflects the enactment of the Tax Cuts and Jobs Act but assumes that the agencies do not issue any regulations. Using this baseline means the analyses provide information primarily on the decision of whether to regulate. Analysis of this question serves little practical use. There is no serious debate as to whether the Treasury Department and the IRS should issue regulations interpreting a major tax overhaul or whether issuing regulations would be beneficial. The more relevant question is how the Treasury Department and the IRS should interpret the law when they have discretion in how they do so. It is in these areas where economic analysis can usefully inform regulatory decision-making by clarifying the impacts of different choices. It is also in these areas where economic analyses can clarify the impacts on people affected by the agencies’ regulatory choices.
Notably, while the treatment of revenues and burden described and criticized above is consistent with the Office of Information and Regulatory Affairs’ framework for cost-benefit analysis, the baseline adopted by the Treasury Department and the IRS is not. Yet contrary to the criticism made here, the Office of Information and Regulatory Affairs instructs agencies to use a pre-statutory baseline that would direct attention neither to the agencies’ use of discretion in issuing regulations nor to the impact of issuing regulations at all—but instead to the effects of the statute itself. As discussed further below, such an approach would not only fail to clarify the impacts of different regulatory choices, but also—in the context of a law as complex as the Tax Cuts and Jobs Act—is impossible to implement in a useful way.
Cost-benefit analyses of TCJA regulations fail to identify regulatory giveaways
A striking manifestation of the weaknesses of the cost-benefit analyses of TCJA regulations is the extent to which they are either silent on controversial decisions in the regulations or fail to critically evaluate those decisions.
In prior work with Adam Looney, then a senior fellow at The Brookings Institution, I highlighted this weakness in the regulations relating to the new deduction for pass-through businesses, specifically the decisions to extend eligibility for the deduction to real estate brokers and S corporation banks and to enable more businesses to take advantage of the deduction by construing the phrase “reputation or skill” narrowly.21 But these were far from the only regulatory giveaways in implementing the Tax Cuts and Jobs Act.
Here’s just one case in point: The regulations implementing the base erosion and anti-abuse tax, a provision of the Tax Cuts and Jobs Act intended to prevent foreign-owned companies from shifting income from the United States to their foreign parents, provided an exclusion for interest paid by certain large financial institutions.22 Though controversial and potentially involving large sums of revenue, this special treatment received no explicit attention in the cost-benefit analysis of either the proposed or final regulations.23
In addition to giveaways on which the cost-benefit analyses are entirely silent, there are also giveaways for which the regulations simply provide little useful guidance. The Opportunity Zone regulations illustrate this scenario. The Opportunity Zone regulations provided an exceptionally favorable interpretation of the laws, such as by providing a generous interpretation of what it means for substantially all property to be used within a zone.24 The cost-benefit analysis of the Opportunity Zone regulations acknowledges several of these generous interpretations, but the failure to interrogate the revenue and burden impacts of regulatory choices means that there is little basis on which to assess them. How much revenue was lost to the generous interpretation of the provisions? To what extent did that implicit tax cut merely generate windfalls to Opportunity Zone investors as opposed to deliver any material gains for the broader public?
The regulatory analysis is silent on these issues.
An even more severe version of this issue arises with the high-tax exclusion from the tax on global intangible low-taxed income. As noted above, the high-tax exclusion effectively received its own cost-benefit analysis because it was adopted after the initial round of GILTI regulations. Thus, in principle, policymakers could have been informed by a detailed analysis of the benefits and costs of this exception. The published cost-benefit analysis, however, advances a case for the exclusion based on certainty and clarity, and provides no information about the estimated revenue cost of the high-tax exclusion or who the anticipated beneficiaries are.
The failings of the TCJA cost-benefit analyses are rooted in the OIRA framework
The framework for cost-benefit analysis of federal regulations that the Office of Information and Regulatory Affairs instructs agencies to use is inappropriate for tax regulations. The fundamental flaws of the cost-benefit analysis of TCJA regulations are a result of a reliance on this inappropriate framework.
Circular A-4, issued by the Office of Management and Budget, provides guidance to agencies on cost-benefit analysis. It instructs agencies that “the revenue collected through a … tax is a transfer payment” and “[y]ou should not include transfers in the estimates of the benefits and costs of a regulation. Instead, address them in a separate discussion of the regulation’s distributional effects.” As I have noted in prior work, this treatment obscures the relevant trade-offs in developing tax regulations because it implies that certain effects of a tax regulation, such as an increase in compliance costs, should be enumerated as social costs, while the core purpose of raising revenues should be treated as offering no net benefits—even though Congress enacted the tax system for that very purpose. The treatment of revenues as neither a benefit nor a cost is thus firmly rooted in the instruction that the Office of Information and Regulatory Affairs gives to agencies for conducting cost-benefit analysis.
In lieu of attention to revenues and the distribution of the tax burden, the cost-benefit analyses of TCJA regulations focus on efficiency. The focus on efficiency is likewise consistent with OIRA instructions. Circular A-4 states:
Benefit-cost analysis is a primary tool used for regulatory analysis. Where all benefits and costs can be quantified and expressed in monetary units, benefit-cost analysis provides decision makers with a clear indication of the most efficient alternative, that is, the alternative that generates the largest net benefits to society (ignoring distributional effects). This is useful information for decision makers and the public to receive, even when economic efficiency is not the only or the overriding public policy objective.25
While the overall approach to costs and benefits adopted by the Department of the Treasury and the IRS is consistent with OIRA instructions, as noted above, there is one important difference between the framework set forth in Circular A-4 and the agencies’ approach to cost-benefit analysis for TCJA regulations. Circular A-4 instructs agencies to use a pre-statutory baseline in the case of regulations that largely restate statutory requirements. In other words, under this view, the Treasury Department and the IRS should examine not the impact of the specific choices made in issuing TCJA regulations, but rather the combined effect of the legislation passed by Congress and the choices made by the Treasury Department and the IRS.
As discussed above, a useful economic analysis of tax regulations should focus on the agencies’ use of discretion in interpreting the tax law. Using a pre-statutory baseline would instead direct attention to the effects of the legislation itself, obscuring the impacts of the agencies’ discretionary decisions.
Even more fundamentally, however, the use of a pre-statutory baseline is simply untenable in the case of TCJA regulations. The Tax Cuts and Jobs Act was a wide-ranging reform of the U.S. tax system. There were numerous interconnected parts, and the effects of each provision depend on the presence or absence of other provisions. Thus, were the Treasury Department and the IRS to attempt an analysis of each regulation relative to a pre-statutory baseline, the analysis would tell readers more about the meaningless counterfactuals the Treasury Department and the IRS adopted for evaluating each piece of the legislation and corresponding regulations than anything useful about the legislation or regulation itself.
Alternatively, the Department of the Treasury and the IRS could wait to produce a regulatory analysis until after they had issued all of the regulations implementing the Tax Cuts and Jobs Act. But at that point, it would be too late to inform any regulatory choices and would primarily serve not as an evaluation of those regulatory choices but as an after-the-fact critique of the major choices made in developing the legislation through an inappropriate lens.
Notably, even though the Treasury Department and the IRS chose not to use a pre-statutory baseline for TCJA cost-benefit analyses, the Office of Information and Regulatory Affairs’ draft report to Congress on regulations adopted in 2017, 2018, and 2019 includes what appear to be intended as estimates of revenue impacts of TCJA regulations relative to a pre-statutory baseline.26 The use of the pre-statutory baseline in this report illustrates one of the many ways reliance on such a baseline can generate misleading results.
The centerpiece policy of the Tax Cuts and Jobs Act was a reduction in the corporate tax rate from 35 percent to 21 percent, which the staff of the congressional Joint Committee on Taxation estimated would cost $1.4 trillion over the subsequent decade.27 However, the Treasury Department and the IRS have not issued regulations on the rate reduction—a highly consequential but relatively simple feature of the Tax Cuts and Jobs Act—so it does not appear in the draft report. In contrast, the Treasury Department and the IRS have issued regulations on many of the law’s complex revenue-raising provisions. Thus, the Office of Information and Regulatory Affairs reports to Congress an increase in revenues that is more informative about which provisions require the Treasury Department and the IRS to issue regulations than how the department and the IRS have interpreted the law or even anything about the law itself.
The Department of the Treasury and the IRS have largely followed the Office of Information and Regulatory Affairs’ direction on the framework for cost-benefit analysis, and thus the resulting analysis provides little useful guidance for regulatory decision-makers, but they have departed from that framework in one important way in an effort to provide more meaningful analysis of regulatory choices. Yet following OIRA guidance in that one area would only lead to less useful analysis.
The path forward
The cost-benefit analyses of TCJA regulations provide little insight on the merits of those regulations. Coverage of regulatory giveaways in the popular press must rely on academic commentary, nonpublic analyses, and industry reaction to regulatory announcements because the analyses do not answer basic questions about the revenue losses and who benefits.28 And while the Congressional Budget Office has revised down their estimate of corporate revenues related to the Tax Cuts and Jobs Act by $110 billion, its public statements are ambiguous as to what portion of that change is attributable to lax regulatory interpretations or other factors.29
More fundamentally, however, the problem is not just about inadequate disclosure but a biased perspective, too. The focus on efficiency, and corresponding inattention to revenues and burden, injects bizarre and inappropriate normative assumptions into the purportedly technocratic analysis of tax policy.
The shortcomings of the cost-benefit analyses of TCJA regulations are rooted in the shortcomings of the framework for cost-benefit analysis mandated by the Office of Information and Regulatory Affairs. In de-emphasizing revenues and burden, and focusing on efficiency, the Treasury Department and the IRS are following the Office of Information and Regulatory Affairs’ instructions as to how agencies should conduct cost-benefit analysis. OIRA guidance, however, is well-suited only to a limited set of economic questions, primarily those related to externalities and decision-making biases, and when distribution is not a primary concern.
Quantitative economic analysis of tax regulations is possible, but it must be structured to provide useful guidance in making regulatory decisions. The foundations for this analysis are the traditional tools of tax analysis:
- Revenue estimates
- Distribution analyses
- Estimates of compliance costs
These tools are the analog of the cost-benefit framework, specialized for the case of tax regulations.30
The next administration should eliminate the requirement for cost-benefit analysis of tax regulations. Instead, the Department of the Treasury should provide a qualitative and, when feasible, quantitative evaluation of tax regulations grounded in fundamental principles of tax policy, centering the impacts of the regulation on revenues, burden, and compliance costs. Moreover, this analysis should focus on the decision points where the Treasury Department and the IRS have discretion to regulate differently. This analysis would most directly inform regulatory decision-making. It should not evaluate the legislation Congress has already enacted.
One criticism of this proposed approach is that, as the Treasury Department has resisted estimating revenue impacts of TCJA regulations, it would also resist producing the revenue estimates required for this approach.31 Importantly, this is not an argument against the recommended approach in principle, only in practice. In essence, the argument suggests the only alternative to biased analysis is no analysis, and then argues that biased analysis is better than no analysis. As this report establishes, however, the biases implicit in the current approach to cost-benefit analysis of tax regulations are severe, and the approach should be rejected.
Moreover, that argument fails to interrogate how the inappropriate nature of the current approach facilitates the Treasury Department’s resistance to producing revenue estimates. The OIRA framework for cost-benefit analysis de-emphasizes revenues and thus weakens its ability to make the case that the Treasury Department and the IRS should estimate revenue impacts of regulations. Indeed, while OIRA staff emphasized to the Government Accountability Office that its guidance requires agencies to conduct distributional analysis of regulations when relevant—which they indicate would include estimates of revenue impacts in the case of tax regulations—cost-benefit analyses without distributional analysis are widespread.32
It is clear that the Office of Information and Regulatory Affairs does not prioritize distributional analyses. Adopting a framework that centers the importance of revenues would substantially strengthen the case for producing estimates of regulations’ revenue impacts.33
A related defense of traditional cost-benefit analyses of tax regulations is that a cost-benefit analysis supplemented by a revenue estimate and a distribution analysis would include all of the information in that revenue estimate and distribution analysis, and thus provide policymakers with the relevant information to make a decision. This is true, but it is no defense of the current framework for cost-benefit analysis. It amounts to the claim that if the Treasury Department and the IRS conduct the analysis according to the current framework, and then also conduct a revenue and burden analysis, decision-makers consequently would have the information provided by the latter useful analysis. The analysis required by the Office of Information and Regulatory Affairs not only consumes resources without adding value, but also, as discussed above, risks confusing decision-makers and members of the public who do not understand the assumptions implicit in it or how to interpret the results. The requirement that it be done should be discarded.
A third criticism of the Department of the Treasury-led approach is that the Office of Information and Regulatory Affairs operates as a “dispassionate and analytical ‘second opinion,’” operating as a check on an agency’s internal biases at either the career or political level.34 But the failure of the cost-benefit analyses of TCJA regulations to identify or highlight the giveaways in the TCJA regulations undermines this defense of the Office of Information and Regulatory Affairs’ role.
Indeed, the failure to identify these giveaways is probably tied to the OIRA framework for cost-benefit analysis and its organizational culture. William West, a professor at the Bush School of Government and Public Service at Texas A&M University, writes that “a common denominator among [OIRA] personnel is the belief that government intervention has the potential to produce more harm than good” and that “[m]ost officials feel that economic efficiency should be an important (although, for many, not the only) criterion for evaluating administrative policies.”35
Consistent with this view, as noted above, Circular A-4 not only asserts that efficiency deserves a central role in the regulatory process, even when it “is not the only or the overriding public policy objective,” it also defines the net benefits of a regulation as the impact on efficiency. This approach appears to have been extended to tax regulations, even though it amounts to evaluating hypothetical, indeterminate alternative tax regulations in lieu of the regulations under consideration and injects normative assumptions on the appropriate distribution of the tax burden into the analysis.
Identifying revenue giveaways as a key issue for evaluation in tax regulations requires treating revenues as a key object of interest. Moreover, it potentially requires the Office of Information and Regulatory Affairs to adopt a position as an advocate for reduced net benefits—at least as it defines the term—or even outright costs. This stands in tension with its culture and role in the executive branch for 40 years.
Finally, one could argue that rather than have the Treasury Department evaluate regulations according to a distinct framework, the Office of Information and Regulatory Affairs should instead adopt a distinct framework for tax regulations. Indeed, this is the approach for which I advocate in prior work.36 While Circular A-4 instructs agencies to focus on a definition of net benefits that excludes distributive impacts, Executive Order 12866, which governs the process of regulatory review, sets out a broader definition of benefits:
Further, in choosing among alternative regulatory approaches, agencies should select those approaches that maximize net benefits (including potential economic, environmental, public health and safety, and other advantages; distributive impacts; and equity), unless a statute requires another regulatory approach.37
There is an important debate to be had about broader reforms to the Office of Information and Regulatory Affairs in the next administration, but its failure to address the shortcomings of its analytic framework in the case of tax regulations during the TCJA rulemaking process and the resulting weaknesses of the analyses of TCJA regulatory actions strongly suggests the best course of action for tax regulations, at this point, is to eliminate the requirement for cost-benefit analysis of tax regulations, eliminate OIRA’s authority to review the analysis of tax regulations, and instead replace OIRA-led cost-benefit analysis with Treasury Department-led economic analysis of tax regulations using the tools of tax analysis discussed above.38
The April 2018 agreement between the Department of the Treasury and the White House Office of Management and Budget imposed a requirement for formal cost-benefit analysis on many more tax regulations than had been subject to such a requirement in the past. This agreement came at the beginning of a remarkable period of regulatory activity for the Treasury Department and the IRS as they worked to implement the Tax Cuts and Jobs Act enacted in December 2017.
Unfortunately, the cost-benefit analyses produced for regulations implementing the Tax Cuts and Jobs Act provide little insight into the merits of the regulations. They downplay revenues, the fundamental reason the tax code exists, and ignore impacts on changes in the distribution of the tax burden. Generous interpretations of TCJA provisions that provide windfall gains to favored groups receive little critical attention and, in some cases, none.
These weaknesses are rooted in the framework that the Office of Information and Regulatory Affairs mandates for agency cost-benefit analyses. This framework ignores regulations’ effects on revenues and the level and distribution of the tax burden in assessing costs and benefits, and thus biases the analysis in favor of regulatory giveaways and against regulations that protect the tax base.
The next administration should eliminate the requirement for cost-benefit analysis as applied to tax regulations and instead require Treasury Department-led economic analysis of tax regulations that focuses on the factors most important for evaluating such regulations: impacts on revenues, on the level and distribution of the tax burden, and on compliance costs.
1. The Department of the Treasury and the Office of Management and Budget, Memorandum of Agreement, the Department of the Treasury and the Office of Management and Budget, Review of Tax Regulations under Executive Order 12866 (2018), available at https://home.treasury.gov/sites/default/files/2018-04/04-11%20Signed%20Treasury%20OIRA%20MOA.pdf.
2. See, for example, Jesse Drucker and Jim Tankersley, “How Big Companies Won New Tax Breaks From the Trump Administration,” The New York Times, December 30, 2019, available at https://www.nytimes.com/2019/12/30/business/trump-tax-cuts-beat-gilti.html; Rebecca Kysar, “TCJA’s Business Tax Provisions: Design Flaws and Undemocratic Implementation,” Testimony Before the U.S. House of Representatives Committee on Ways and Means, February 11, 2020, available at https://docs.house.gov/meetings/WM/WM00/20200211/110494/HHRG-116-WM00-Wstate-KysarR-20200211.pdf.
3. Office of Management and Budget, Circular A-4 (2003), available at https://www.whitehouse.gov/sites/whitehouse.gov/files/omb/circulars/A4/a-4.pdf.
4. More formally, the burden concept used in this analysis and in most distribution analyses is the impact of a tax change on household utility evaluated on an annual basis without assuming any additional future policies to offset the deficit impacts. For additional exposition and discussion of this concept and its role in formalizing distribution analysis, see Greg Leiserson, “Distribution Analysis as Welfare Analysis” (Washington: Washington Center for Equitable Growth, 2020), available at https://equitablegrowth.org/working-papers/distribution-analysis-as-welfare-analysis/.
5. Greg Leiserson and Adam Looney, “A framework for economic analysis of tax regulations” (Washington: The Brookings Institution and the Washington Center for Equitable Growth, 2018), available at https://equitablegrowth.org/a-framework-for-economic-analysis-of-tax-regulations/.
6. Specifically, the Office of Management and Budget solicited comments on the incorporation of an estimate of the excess burden of taxation into the cost-benefit analysis of spending programs. As in the case of taxation, this would insert inappropriate normative assumptions into the analysis of spending programs. For example, it would require evaluating anti-poverty programs under the assumption that their primary purpose—poverty reduction—is not a benefit while simultaneously counting as costs a much wider array of costs associated with the programs. In addition, as the Office of Management and Budget suggests in its comment solicitation, it would also lead to double counting, as the costs of taxation would be counted on both the spending and tax sides of the ledger. See Office of Management and Budget, Marginal Excess Tax Burden as a Potential Cost Under E.O. 13771 (2019), available at https://www.reginfo.gov/public/pdf/eo13771/EO13771_marginal_excess_tax_burden.pdf.
7. For a more detailed proposal, see Leiserson and Looney, “A framework for economic analysis of tax regulations.” For a discussion of the concept of burden and the conduct of distribution analysis, see Leiserson, “Distribution Analysis as Welfare Analysis.”
8. Internal Revenue Service, “Contributions in Exchange for State or Local Tax Credits,” Federal Register 84 (2019), p. 27513.
9. Ibid., p. 27515.
10. Ibid., p. 27523.
11. Ibid., pp. 27527–27528.
12. Internal Revenue Service, “Limitation on Deduction for Business Interest Expense,” Federal Register 85 (2020), p. 56686.
13. Internal Revenue Service, “Guidance Under Sections 951A and 954 Regarding Income Subject to a High Rate of Foreign Tax,” Federal Register 85 (2020), p. 44620.
14. Proposed GILTI regulations were published in the Federal Register on October 10, 2018. On June 14, 2019, the Treasury Department and the IRS finalized the previously proposed GILTI regulations and proposed further regulations, including the high-tax exclusion, both of which were subsequently published in the Federal Register on June 21, 2019.
15. Internal Revenue Service, “Qualified Business Income Deduction,” Federal Register 84 (2019), p. 2952.
16. Internal Revenue Service, “Investing in Qualified Opportunity Funds,” Federal Register 85 (2020) p. 1866.
17. See, for example, James A. Mirrlees, “An Exploration in the Theory of Optimal Income Taxation,” Review of Economic Studies 38 (2) (1971): 175–208; Emmanuel Saez, “Using Elasticities to Derive Optimal Income Tax Rates,” Review of Economic Studies 68 (1) (2001): 205–229.
18. Internal Revenue Service, “Contributions in Exchange for State or Local Tax Credits,” p. 27527.
19. Internal Revenue Service, “Guidance Under Sections 951A and 954 Regarding Income Subject to a High Rate of Foreign Tax,” p. 44633.
20. As the core critique of this report is the inappropriate nature of the cost-benefit framework mandated by the Office of Information and Regulatory Affairs for the analysis of tax regulations, I set aside the question of the quality of the efficiency analyses provided for TCJA regulations.
21. Leiserson and Looney, “A framework for economic analysis of tax regulations.”
22. Kysar, “TCJA’s Business Tax Provisions: Design Flaws and Undemocratic Implementation.”
23. Internal Revenue Service, “Base Erosion and Anti-Abuse Tax,” Federal Register 83 (2018), p. 65956; Internal Revenue Service, “Base Erosion and Anti-Abuse Tax,” Federal Register 84 (2019), p. 66968.
24. Samantha Jacoby, “Final Opportunity Zone Rules Could Raise Tax Break’s Cost” (Washington: Center on Budget and Policy Priorities, 2020), available at https://www.cbpp.org/blog/final-opportunity-zone-rules-could-raise-tax-breaks-cost.
25. Office of Management and Budget, Circular A-4.
26. Office of Information and Regulatory Affairs, Draft 2018-2019-2020 Report to Congress on the Benefits and Costs of Federal Regulations and Agency Compliance with the Unfunded Mandates Reform Act (2019), available at https://www.whitehouse.gov/omb/information-regulatory-affairs/reports/#ORC.
27. Joint Committee on Taxation, Estimated Budget Effects Of The Conference Agreement For H.R.1, The “Tax Cuts And Jobs Act” (2017).
28. Drucker and Tankersley, “How Big Companies Won New Tax Breaks From the Trump Administration.”
29. Phill Swagel, “Recent Changes in CBO’s Projections of Corporate Income Tax Revenues,” Congressional Budget Office blog, February 7, 2020, available at https://www.cbo.gov/publication/56121.
30. For additional discussion of how quantitative economic analysis of tax regulations can be conducted, see Leiserson and Looney, “A framework for economic analysis of tax regulations.”
31. See, for example, the discussion in Government Accountability Office, Tax Cuts and Jobs Act: Considerable Progress Made Implementing Business Provisions, but IRS Faces Administrative and Compliance Challenges (2020).
33. A related critique is that the distribution analysis needed under this proposal may require incidence or other assumptions specific to the regulation proposed. However, the same is true for efficiency analyses, so the complexity of the analysis does not substantiate a case for the latter mode of analysis over the former.
34. The White House, “Presidential Memorandum Regarding Regulatory Review,” Press release, April 23, 2009, available at https://obamawhitehouse.archives.gov/the-press-office/presidential-memorandum-regarding-regulatory-review.
35. William F. West, “The Institutionalization of Regulatory Review: Organizational Stability and Responsive Competence at OIRA,” Presidential Studies Quarterly 35 (1) (2005): 76–93.
36. Leiserson and Looney, “A framework for economic analysis of tax regulations.”
37. Executive Order no. 12866, Federal Register 58 (1993), p. 51735.
38. While a full treatment of other potential reforms to the Office of Information and Regulatory Affairs is beyond the scope of this report, it is worth acknowledging that some of the critiques made here apply beyond tax regulations. Specifically, the use of a pre-statutory baseline for implementing regulations deserves reconsideration across the board. Analysis of agency regulatory actions should focus on agencies’ discretionary actions, which the analysis can usefully inform. In addition, the use of a framework for cost-benefit analysis that severely undervalues distributive considerations should be reconsidered whenever regulations are motivated by distributive considerations, which would include regulations implementing a wide array of spending programs. For one recent perspective on a reform agenda, see Todd N. Tucker and Rajesh D. Nayak, “OIRA 2.0: How Regulatory Review Can Help Respond to Existential Threats” (New York: The Great Democracy Initiative, 2020), available at https://greatdemocracyinitiative.org/wp-content/uploads/2020/04/OIRA_Final.pdf.