In principle, cost-benefit analysis can inform policymakers and the public about the likely effects of a regulation and help guide policymakers to better decisions, at the expense of a somewhat slower and more resource-intensive rulemaking process.

However, the cost-benefit analysis required by the White House Office of Information and Regulatory Affairs (OIRA) does not provide useful guidance for a large class of rules related to tax and transfer programs because it ignores the distributional and fiscal goals that motivate them. For these rules, referred to as transfer rules, requirements for formal cost-benefit analysis bias the regulatory process, mandate public release of misleading metrics, and delay rules for no gain.

Alternatives to traditional cost-benefit analysis, including those adopted by the Biden administration in 2023, can improve cost-benefit analysis and regulatory policymaking in many domains, but there is no alternative approach that adequately addresses the problems with applying these types of analytical frameworks to transfer rules. 

Notwithstanding the deficiencies of cost-benefit analysis for transfer rules, economic analysis of these regulations can be useful when it helps policymakers judge whether the compliance and other costs imposed by a regulation are worth it. But analysts should do such analysis to inform decisionmaking and help rules stand up to legal challenge. And that cannot be done in a framework that sets aside the transfers that are the reason for these regulations in the first place.

This post, adapted from a recent policy proposal published by the Hamilton Project and the Tax Law Center, argues for exempting transfer rules from White House requirements for cost-benefit analysis and substantive review by OIRA.  

Cost-benefit analysis is inappropriate for tax regulations because revenues and redistribution—the reason progressive taxes exist—generate no net benefits.

Tax regulations have historically been excluded from requirements for formal cost-benefit analysis and OIRA review and there was good analytic reason for that exclusion.

Cost-benefit analysis is not simply a list of the pros and cons of a policy. It is a specific approach to policy analysis that aims to summarize the impacts of a policy in a single number. To produce this number, analysts quantify and then add together the costs and benefits experienced by different people to determine the overall net benefits of the policy.   

A consequence of adding up costs and benefits experienced by different people is that it makes transfers between people (i.e., redistribution) and transfers between people and the government (i.e., the change in net revenues) appear to be irrelevant. For example, transferring $100 from a rich person to a poor person will appear to have no effect because one person’s loss offsets the other person’s gain. Similarly, transferring $100 from taxpayers to the government appears to have no net effect.

However, these types of effects—redistribution and raising revenues—are exactly why the tax system exists. It makes little sense to conduct tax analysis in a framework that sets aside the core reasons for taxation in the first place.

Moreover, while revenues and redistribution do not give rise to benefits in a cost-benefit framework, the costs imposed by taxation are viewed as costs in that  framework. These costs include administrative costs and compliance costs, as well as costs associated with undesirable changes in behavior undertaken to avoid tax.

As a result, cost-benefit analysis will generally find that regulations increasing revenues have costs greater than their benefits.

Cost-benefit analysis is inappropriate not only for tax rules, but also for other types of transfer rules.

While tax regulations were long exempt from White House requirements for cost-benefit analysis, transfer regulations, including regulations relating to Social Security, SNAP, and major health care programs, have been subject to these requirements. Yet the economic rationale for exempting tax regulations from cost-benefit mandates applies with the same force in the transfer rule context: The analysis ignores the redistributive and revenue effects that are fundamental to these programs.  

The primary purpose of these programs is to transfer economic resources between groups of people or between people and the government. Major health care programs shift resources from young to old, wealthy to poor, and healthy to sick. Social Security effectively transfers income within the same person, such as from a person during their working life to their older self in retirement. SNAP shifts resources from higher-income households to those who cannot afford to put food on the table.

While these programs include insurance aspects that people might voluntarily pay for, such as insurance against a future disabling event, they also include redistribution. Moreover, changes in these programs are often motivated by concerns about deficits.

Analysis that sets aside the revenue and distributional effects of regulations implementing nontax transfer programs is not useful. If the USDA issues a regulation that would increase SNAP benefits for certain populations, useful analysis of that regulation must interrogate the merits of that increase. Similarly, if USDA adopts a new program integrity regulation relating to the SNAP program, a useful analysis must interrogate whether and by how much that regulation reduces benefits for ineligible recipients and eligible recipients, as well as its impacts on overall spending.

There are real harms to requiring federal agencies to conduct analysis that is poorly suited to transfer rules. 

While both tax and non-tax transfer rules are ill-suited to cost benefit analysis, they collectively account for a large share of rules subject to OIRA review—roughly half for the period from 2002 to 2023. The Centers for Medicare and Medicaid Services (CMS) accounts for more transfer rules than any other agency because the rules administering Medicare, Medicaid, and the health insurance marketplaces generally fall into this category. CMS transfer rules account for more than 40 percent of all major transfer rules and nearly a quarter of all major rules. During the period from 2018 to 2023, when most tax rules were subject to a requirement for cost-benefit analysis, tax rules accounted for 11 percent of major transfer rules and 6 percent of all major rules.

Requiring formal cost-benefit analysis for transfer regulations biases the rulemaking process against regulations that raise revenue or redistribute income. It can encourage agencies to scale back transfer programs, weaken enforcement of the tax law, and increase inequality. It can force federal agencies to issue misleading analysis that provides critics with an easy means of attacking the agency’s proposed rules. Critics can point to an administration’s own analysis to argue that regulations reduce efficiency or harm the economy without yielding any benefits.

Traditional cost-benefit analysis can even help policymakers obscure their reasons for adopting a policy. For example, an administration’s true motivation for weakening regulations that prevent tax avoidance may be that they believe rich people already pay too much in taxes. However, the administration could claim, citing their own cost-benefit analysis, that their primary goal is to reduce administrative and compliance costs for taxpayers.

It is not enough to augment the cost-benefit analysis with alternative metrics that an agency believes justify the rulemaking. The problem is that the cost-benefit analysis itself is misleading. For example, suppose Treasury issues a regulation that has negative net benefits according to formal cost-benefit analysis, but the agency determines the regulation is worthwhile because it raises revenues. That determination is effectively a rejection of the cost-benefit framework. Agencies should stop performing this misleading analysis.

Requiring agencies to do formal cost-benefit analysis and then having OIRA review that cost-benefit analysis also introduces unnecessary delay into the rulemaking process. The OIRA review process can take several months or longer, and it can involve multiple rounds of review, with substantive feedback and requests for changes at each round.

Finally, the resources devoted to preparing these analyses are necessarily drawn from other agency activities. Given the deficiencies of the analyses in practice and the fact that OIRA’s requirements are ill-suited to transfer rules in general, many of those other activities would likely be a better use of resources.

The wastefulness of the process is even more severe when—as is currently the case with Treasury—an agency resists the guidance provided by OIRA and yet continues to go through the motions of the analysis. While this resistance avoids some of the harms of requiring an inappropriate mode of analysis, it generates little in the way of other benefits to justify its costs.

Transfer rules should be exempt from requirements to conduct formal cost-benefit analysis and from substantive OIRA review.

The White House should adopt broad programmatic exemptions under which tax regulations and regulations relating to Social Security, Supplemental Security Income (SSI), Medicare, Medicaid, the ACA marketplaces, and SNAP would not be subject to cost-benefit analysis requirements. Agencies should also be allowed to exempt transfer rules that are not covered by these programmatic exemptions on a rule-by-rule basis. Regulations that are exempt from mandates for cost-benefit analysis should also be exempt from substantive OIRA review.

Under this proposal, agencies would provide transfer rules in draft form to the White House. OIRA would then continue to fulfill its statutory responsibilities under the Congressional Review Act, Paperwork Reduction Act, and other authorities, and it would serve any necessary coordination functions within the White House. If requested by other parts of the White House or the rulemaking agency, OIRA would also coordinate interagency circulation.

Critically, OIRA would have no formal clearance role and would not be responsible for approving the rule for publication. Its role would be limited to these coordination functions.

Conceptually, the question of whether cost benefit analysis is the right mode of analysis and the question of whether OIRA should review whatever analysis is done are distinct. However, in practice, having OIRA review transfer rules under this proposal is unlikely to be beneficial.

The history of requiring inappropriate analysis of transfer rules means that continued OIRA review of these rules is likely to result in significant confusion and requests for inappropriate analysis. Circular A-4, OIRA’s guidance to agencies on cost-benefit analysis, defines cost-benefit analysis as efficiency analysis, and OIRA staff have consistently held that such analysis is well-suited to transfer rules. Achieving productive OIRA review of transfer rules would require a substantial change in organizational culture for these rules. There is no reason to make that kind of challenging, time-consuming organizational transformation a prerequisite to addressing the existing issues around analysis.

Alternative analytic approaches cannot address the fundamental challenges of applying cost-benefit analysis to transfer rules.

Alternatives to traditional cost-benefit analysis can improve cost-benefit analysis and regulatory policymaking, but there is no alternative approach that adequately addresses the problems with applying these types of analytical frameworks to transfer rules. 

Weighted cost-benefit analysis is perhaps the most compelling alternative. In a weighted cost-benefit analysis, the analyst sums the welfare impacts of a regulation on different people using a set of weights that vary across individuals or families. These weights provide a means of giving transfers to lower-income people higher value in the analysis; weights can be based on estimates of the diminishing marginal utility of income or other factors.

While weighted cost-benefit analysis can be useful in many settings, it is insufficient for dealing with transfer rules. First, weighting cannot resolve the challenge of applying cost-benefit analysis to rules where deficit impacts are a primary consideration. Changes in the deficit matter primarily through their impact on future taxes and spending. But when a rule that increases or decreases the deficit is adopted, it cannot be known what effect that change in the deficit will have on future taxes or spending, so it is not clear how to value the change in the deficit.  

Second, there is no widely accepted set of distributional weights that can serve as the basis for weighted cost-benefit analysis. Not only do different administrations have different views on distributional questions, but a single administration will likely hesitate to publicly commit to a single set of distributional weights that must be used across all contexts.

Finally, using weighted net benefits is poorly suited to anything other than limited types of vertical equity. For example, Congress has often prioritized families with children as beneficiaries of programs. This prioritization could lead to scenarios where children in families with higher incomes are getting benefits before adults with lower incomes. Congress also targets programs based on geography either directly or based on the history of an area.

Rather than mandating inappropriate analysis or waiting for an alternative that is unlikely to exist, the White House should exempt transfer rules from cost-benefit mandates. The full proposal from which this post is adapted can be found on the Hamilton Project website.

Greg Leiserson is a senior fellow at the Tax Law Center at NYU Law and a former Deputy Assistant Secretary for Tax Analysis at the U.S. Treasury Department. Kathleen Bryant is a policy advisor at the Tax Law Center.