*This post is part of a symposium on Modernizing Regulatory Review. For other posts in the series, click here.

The Treasury Department and the White House Office of Management and Budget on June 9 entered into a Memorandum of Agreement (2023 MOA) with the effect that tax and certain other Treasury regulations will not be subject to OIRA review under section 6 of EO 12866. This reverses the course set by the 2018 Memorandum of Agreement (2018 MOA), which had sent many more tax regulations to OIRA for review. 

To implement the 2018 MOA, OIRA adopted a framework for analyzing tax regulations that did not treat tax revenues as a benefit, despite revenue being the tax system’s primary goal. The incomplete and biased analyses that resulted did not improve tax regulation. But they caused a bottleneck in implementing tax statutes and responding to the large unmet demand that taxpayers have for regulations to clarify their legal obligations.

The 2023 MOA will improve regulatory review processes in tax. It also relieves the broader regulatory modernization project of having to try to fit the square tax peg into the round Circular A-4 hole. That project faces a difficult and unresolved question of how Circular A-4 should coherently address the fiscal impacts of regulations in other areas of law where budgetary impacts are less central than in tax. The 2023 MOA will not itself resolve all problems in tax regulation. But the administration has already made progress with initiatives in the tax rulemaking process that are intended to improve analysis, stakeholder input, and interagency coordination, and there are further tools it can use to modernize tax regulatory review.

Problems with tax regulatory review under the 2018 MOA in theory and practice

Prior to 2018, virtually all tax regulations were exempt from requirements under EO 12866 for cost-benefit analysis of significant regulatory actions and OIRA review of those analyses. The 2018 MOA greatly increased the number of tax regulations subject to OIRA review: there were, on average, some 13 times as many EO 12866 reviews of tax regulations each year following the 2018 MOA as there had been on average from 1981 to 2017.

OIRA did not explicitly set out its analytical approach to tax regulations, but adopted an inappropriate framework for analyzing tax regulations that did not count revenue – the primary goal of the tax system – as a benefit. Compliance burdens, however, counted as a cost (with filers who would otherwise pay the most tax tending to have the largest compliance costs). Under this framework, a regulation that blessed a form of corporate tax avoidance would be treated as generating net benefits: corporate taxpayers would face a reduced cost of avoiding taxes, but the lost revenue would not count as a cost. This was an analytical thumb on the scale of regulations that lost revenue but saved compliance costs.

Given these analytical problems, it’s unsurprising that in practice, cost-benefit analysis of tax regulations under the 2018 MOA delivered poor results. Tax experts criticized many regulations implementing the 2017 Tax Cuts and Jobs Act (TCJA) for giving unnecessary windfalls to certain groups, but the cost-benefit analyses for those regulations oftenfailed to identify or analyze those windfalls. The process also failed to critique regulations so generous to taxpayers that they arguably fell outside the bounds of IRS and Treasury statutory authority. 

Analysis of tax regulations under the 2018 MOA also created a regulatory bottleneck. The dramatic increase in OIRA reviews of tax regulations and the resources dedicated to that process led to substantial delays in issuing tax regulations. There is (and was, even prior to the 2018 MOA) overwhelming unmet demand from taxpayers and other stakeholders for the IRS to issue regulations addressing how the IRS will administer particular statutory provisions. Such regulations provide filers certainty and the ability to rely on the agency’s guidance. Filers especially want the IRS to speak to new law, so there is little practical choice about whether the IRS will issue regulations on significant new statutes like the TCJA or, now, the Inflation Reduction Act (IRA), even when the statute does not explicitly require it.

With clear downsides and benefits that were murky at best, it was past time to end the failed experiment with OIRA tax regulatory review. 

The 2023 MOA is the right route to an important improvement

There was not a clear route to coherent analysis of tax regulations within the Circular A-4 and EO 12866 framework, so the 2023 MOA took the correct approach. 

Under Circular A-4, the impact of regulatory choices on the budget has generally been ignored in agency cost-benefit analyses, treated as a “transfer” that is neither a cost nor a benefit. One possible justification for this approach is that fiscal impacts are not a primary policy goal of regulatory policy areas outside of tax, and the tax system can always adjust for any fiscal impacts (and may be able to do so more efficiently than other tools). Even by that logic, the one place where fiscal impacts can’t be ignored is in dealing with tax regulations that sit at the core of the fiscal system.

But, as noted above, under the 2018 MOA, OIRA did ignore fiscal impacts of tax regulation, by effectively adopting a mechanical cut-and-paste application of the Circular A-4 framework in a way that is unsuitable for tax. As Leiserson and Looney wrote:

While the treatment of revenues as a transfer makes sense for the analysis of mandates intended to correct market failures, it is poorly suited to tax analysis. Indeed, taking the A4 framework at face value, any tax system would appear to fail a cost-benefit test: taxation consumes real resources, like taxpayers’ time and IRS enforcement costs, and discourages productive activity, such as work and investment, solely to transfer resources from the public to the government.

And:

Treating revenues as a transfer is the central limitation in conducting cost-benefit analysis of tax regulations according to the guidance of Circular A-4. However, there are also important conceptual ambiguities in Circular A-4 regarding the treatment of deficits, the role of offsetting policies, and the role of income effects. It also embeds implicit assumptions about the appropriate distribution of the tax burden.

The proposed changes to Circular A-4 do not yet clear a path through these conceptual difficulties that, while thorny in other regulatory areas, are fatal in tax. The proposed change would allow agencies to choose to move transfers into the main cost-benefit analysis, instead of the current approach of treating transfers as neither costs nor benefits. But transfers would broadly be treated as netting: a transfer of value from one party to another would be treated as a cost to the transferor but an equal benefit to the transferee. 

Applying this approach to tax regulations would replicate the outcome of OIRA’s inappropriate post-2018 approach to tax regulations: again, all else equal, a regulation that saved a would-be tax avoider money but cost the fisc the same amount of revenue would generate benefits (to the tax avoider) and costs (to the fisc) that precisely offset. Such a regulation could deliver “net benefits” by reducing compliance costs. 

So, even in the new Circular A-4 framework, all else equal a dollar in the hands of a private actor is worth just as much as a dollar in the hands of the fisc. If applied more broadly, this approach again implies that all else equal, the fiscal state generates no net benefits. This might be an acceptable kludge in areas of regulation where budgetary impacts are not the primary policy goal, but it is a critical problem for coherent analysis of tax regulations at the core of the fiscal state. The more radical revisions of Circular A-4 that would be needed to accommodate tax regulations could bog down the project and prevent it from achieving its other important goals. (See Preamble to Circular A-4 revisions, footnote 23 for just some of the difficulties.) 

The analytical mismatch between Circular A-4 and fiscal rules makes the 2023 MOA a wiser path. It is also a reason why it makes sense to remove tax regulations from OIRA review altogether rather than return to the pre-2018 situation where some major “legislative” tax regulations could be subject to OIRA review. (Recent developments have also left the legislative/interpretive distinction, as applied in various contexts in tax, especially muddled.) In any event, before the 2018 MOA almost no tax regulations went to OIRA, so the practical differences between the 2023 MOA and the pre-2018 status quo will be small. 

Further steps forward

Relieved of a process that drove resources and attention to the wrong questions, the Administration should focus on more promising options for improving tax regulations. Indeed, it has implemented several. 

For example, investment in tax analysis can support improved tax decisionmaking. After a decade of severe cuts to IRS funding, the IRA provided funds to restore and improve IRS and Treasury capacity broadly. The IRS Strategic Operating Plan (SOP) for using those funds commits to embedding evidence-based evaluation across the tax system, including for setting and evaluating SOP goals. The Administration should now provide more detailed funding commitments and plans to improve and apply tax research and analysis capacity specifically, including bolstering the capacity of the IRS Statistics of Income Division and the IRS Research, Applied Analytics and Statistics Division, as well as Treasury’s Office of Tax Analysis (OTA). Better analysis can help better direct resources across the tax system – including selecting and shaping tax regulatory projects. Treasury and the IRS have already made progress developing and using new analytical tools to support distribution and equity analysis in tax. 

Treasury and the IRS have also, consistent with EO 14094, implemented new approaches intended to broaden the range of stakeholders who have a say on the tax regulatory agenda and the substance of tax regulations. In implementing the IRA’s climate tax credits, before issuing any NPRMs, the IRS and Treasury issued nine notices in October and November of last year, inviting stakeholders to comment on what guidance Treasury and the IRS should issue and prioritize, and undertook outreach including a series of stakeholder discussions. This responds to research highlighting that the provision of pre-NPRM comments in tax had been especially opaque relative to other agencies, and inaccessible to many stakeholders. Treasury received thousands of comments on these notices.

Treasury also established an Advisory Committee on Racial Equity to bring “a wide set of outside perspectives and lived experiences to the decision-making table.” The Committee has engaged with a series of tax administration and guidance matters. 

The Administration has adopted new executive oversight and interagency coordination  structures for regulatory projects that have both major tax and non-tax components. EO 14082 established, within the Executive Office of the President, the White House Office on Clean Energy Innovation and Implementation, to conduct interagency coordination of the implementation of the Inflation Reduction Act’s energy and infrastructure provisions. The EO details the President’s implementation priorities and requires agencies to prioritize those goals to the extent consistent with the law. The 2023 MOA clearly does not mean that the President cannot continue to exercise oversight over and require inter-agency coordination for important tax rulemakings. 

Other steps to improve tax regulatory analysis, coordination, and participation are either underway or should be explored. It’s a more diffuse approach to improving tax regulation than focusing on OIRA review. But OIRA review under the 2018 MOA was not a silver bullet, and with Circular A-4 continuing to be a poor fit and no panacea available, using a range of tools to improve analysis, coordination, and participation in tax regulation is a sound way to modernize tax regulatory review. Tools that show promise should be refined, while those that don’t, like the 2018 MOA, should be abandoned.

Chye-Ching Huang is the executive director of the Tax Law Center at NYU Law.