OMB Circular A-4 — Regulatory Analysis & Cost-Benefit Standards

OMB Circular A-4 is the federal government's master playbook for regulatory economics — the document that tells every executive branch agency exactly how to justify a regulation before it can take effect. Issued by the Office of Management and Budget and enforced by OIRA, A-4 specifies the methodology agencies must use when conducting a Regulatory Impact Analysis (RIA): how to measure benefits, how to measure costs, what discount rate to apply to future effects, how to value a human life saved, and how to account for uncertainty. Any federal rule with $100 million or more in annual economic impact must have an RIA that meets A-4's standards to survive OIRA review — and to survive a court challenge.

The circular was first issued in 2003 and operated largely unchanged for two decades. In November 2023, OMB issued the most significant revision to A-4 since its original publication — a methodological overhaul that changed the primary discount rate from 7% real to 2% real, added mandatory guidance on distributional effects and environmental justice, and updated how agencies should handle benefits that cannot be monetized. Those changes have substantial consequences: lowering the discount rate dramatically increases the present value of long-term regulatory benefits (clean air, climate, public health), making it easier to justify rules with costs today and benefits spread over decades. The 2023 revision remains in effect as of 2026, though it has been politically contested.

Legal Authority

• 31 U.S.C. § 1111 — OMB general authority to supervise agency management and regulatory activities; provides statutory grounding for OMB's authority to prescribe analytical standards for agency rulemaking
• 2 U.S.C. § 1532 — Unfunded Mandates Reform Act; requires agencies to assess the costs of significant federal mandates (>$100M) on state, local, and tribal governments and the private sector; A-4 provides the analytical methodology for UMRA assessments
• 5 U.S.C. § 601 — Regulatory Flexibility Act; requires regulatory flexibility analyses for rules with significant economic impact on small entities; A-4 provides the cost measurement framework for RFAs
• Executive Order 12866, § 6(a)(3)(B)(i) — Requires agencies to provide OIRA with "an assessment of the potential costs and benefits of the regulatory action" for significant rules; A-4 defines what that assessment must contain
• OMB Circular A-4 (September 17, 2003; revised November 9, 2023) — Establishes methodology for Regulatory Impact Analysis: benefit measurement, cost measurement, discount rates, Value of Statistical Life (VSL), distributional analysis, and treatment of uncertainty

Key Mechanics

A-4 governs how agencies must construct a Regulatory Impact Analysis (RIA) for any economically significant rule ($100M+ annual impact). Key methodological requirements: (1) Alternatives analysis — agencies must evaluate a range of regulatory alternatives (not just the preferred option); (2) Benefits quantification — benefits must be expressed in monetary terms where feasible; non-monetized benefits must be qualitatively described and clearly identified; (3) Cost quantification — total economic costs (compliance costs, administrative burdens, opportunity costs) must be estimated in present-value terms; (4) Discount rate — under the 2023 revision, the primary discount rate is 2% real (down from 7% real under the 2003 version); this change dramatically increases the present value of long-horizon benefits (30-50 year climate benefits, public health); a secondary 1.7% real rate is also recommended; (5) Value of Statistical Life (VSL) — the dollar value assigned to a statistical life saved by a regulation (approximately $11-12 million in 2026 dollars, updated periodically by OIRA); used to monetize mortality risk reductions; (6) Distributional analysis — the 2023 revision made distributional and equity analysis mandatory for major rules, requiring agencies to assess how benefits and costs are distributed across income groups and communities. Completed RIAs accompany rules submitted to OIRA for review and are posted publicly; courts review whether the agency's cost-benefit analysis was rational, not whether A-4 was followed to the letter.

Overview

What This Circular Requires

A-4 applies whenever an agency is developing an "economically significant" rule — defined under EO 12866 as a rule likely to have an annual economic effect of $100 million or more, or one that creates significant adverse effects on a sector of the economy, productivity, jobs, or the environment. For such rules, the agency must prepare a full Regulatory Impact Analysis (RIA) that satisfies A-4's methodology before submitting the rule for OIRA review.

The core requirement is straightforward: agencies must show that a regulation's benefits justify its costs. But A-4 specifies in considerable detail how that comparison must be made. Agencies must identify the market failure or other problem that justifies government intervention (externalities, information asymmetries, market power, coordination failures). They must define a realistic "baseline" — what would happen in the absence of the rule, accounting for existing regulations and market trends. They must identify and analyze a reasonable range of regulatory alternatives, not just their preferred approach. And they must estimate costs and benefits in a common unit (dollars) so they can be compared.

For benefits that cannot be monetized — hard-to-value environmental improvements, avoided suffering, cultural preservation — the 2023 A-4 revision significantly strengthened agencies' ability to rely on them. Under the prior 2003 guidance, OIRA could return a rule that appeared to fail cost-benefit analysis on quantified values alone. The 2023 revision explicitly directs agencies to describe non-monetized benefits carefully and states that a rule should not be judged to fail cost-benefit analysis solely because some benefits resist quantification. This matters enormously for environmental and public health rules where the science supports a benefit but the economic measurement is contested.

A-4 also requires sensitivity analysis — agencies must show how their conclusions change if key assumptions are varied. A rule that passes cost-benefit analysis only under optimistic assumptions is a weak rule; A-4 requires transparency about that uncertainty so OIRA reviewers and courts can assess the robustness of the agency's case.

The Discount Rate: The 2023 Revision's Biggest Change

Discounting converts future costs and benefits to present value — a benefit received 30 years from now is worth less today than the same benefit received next year, both because society can invest resources productively in the interim and because people generally prefer benefits sooner. The choice of discount rate determines how much future harms and benefits count in a cost-benefit analysis, and it has outsized effects on rules addressing long-horizon problems like climate change, toxic contamination, or infrastructure.

Before 2023: A-4 required agencies to run their analyses at both 7% real (representing the opportunity cost of capital — what private investment could earn) and 3% real (representing the social rate of time preference). The 7% rate served as the "primary" rate for most analyses, which systematically understated the benefits of long-term regulations.

After the 2023 revision: The primary discount rate is 2% real, reflecting updated estimates of the social rate of time preference (how much society values present over future consumption), guided by empirical evidence on real interest rates and long-run growth. Agencies must also run analyses at 7% real as an alternative scenario, ensuring transparency about the sensitivity of results to the discount rate assumption. For rules with effects extending across generations, agencies may use 1.7% real as a long-run rate grounded in empirical estimates of the growth-adjusted Ramsey rate. The change from 7% to 2% as the primary rate can increase measured benefits by a factor of two or three for rules with long-time-horizon effects — a transformation for climate, public health, and infrastructure rules.

The Value of a Statistical Life

When a regulation reduces mortality risk — by cleaning up air pollution, requiring better safety equipment, or imposing stricter food safety standards — agencies must place a dollar value on those avoided deaths to compare them to compliance costs. The Value of a Statistical Life (VSL) is not the value of any identified individual's life; it is the aggregated willingness-to-pay to reduce the risk of death across a population. If 100,000 people each pay $100 to reduce their individual risk of death by 1-in-100,000, the implied VSL is $10 million.

A-4 does not specify a single mandatory VSL; instead, it directs agencies to use estimates grounded in high-quality labor market and stated-preference studies, with guidance on how to update them for real income growth over time. In practice, most agencies use VSL estimates close to the values published by EPA, DOT, and HHS, which as of 2023–2026 cluster around $10–13 million (in 2020 dollars). The 2023 A-4 revision updated the guidance on income-adjusting the VSL — agencies may adjust VSL upward for higher-income beneficiary populations but must be transparent about the adjustment and should also conduct analysis using a uniform VSL.

The VSL is one of the most politically sensitive inputs in regulatory analysis. Rules that produce large estimated mortality reductions — such as EPA air quality standards or OSHA workplace safety rules — rest on VSL calculations that can make or break the cost-benefit case. Industry critics argue VSL estimates are inflated; public health advocates argue they should be higher. A-4 tries to anchor agencies in empirical estimates while allowing methodological transparency.

Key Provisions

• Section III.a — Baseline: Agencies must define the "baseline" scenario (the world without the rule) clearly, accounting for other regulations already in effect, market trends, and voluntary industry actions
• Section III.b — Alternatives: Agencies must identify and analyze a range of regulatory alternatives, including more and less stringent versions of the rule and a "do nothing" option
• Section III.c — Benefits: Agencies must quantify and monetize benefits where feasible; for non-monetized benefits, describe them qualitatively and explain why monetization is not feasible
• Section III.d — Costs: Agencies must quantify direct compliance costs, indirect costs (price increases, output changes), and opportunity costs; avoid double-counting
• Section III.e — Distributional analysis (2023): Agencies must analyze who bears costs and who receives benefits, disaggregated by income, geography, race, age, and other relevant characteristics; identify effects on low-income and disadvantaged communities
• Section III.f — Uncertainty: Agencies must disclose key uncertain parameters and conduct sensitivity analysis showing how conclusions change with different assumptions
• Section IV — Discount rates: Primary rate 2% real; alternative 7% real; intergenerational analyses may use 1.7% real (all as of the 2023 revision)
• Section V — Distributional weights (2023): Agencies may, but are not required to, apply distributional weights that give greater weight to benefits accruing to lower-income populations
• Section VI — Transfer payments: Cost-benefit analysis should focus on real resource costs and benefits; transfer payments (e.g., taxes, fines) are not costs to society but are distributional effects that should be described

How It Affects You

If you work at a federal agency: Every economically significant rule your agency issues requires an RIA that satisfies A-4. Assign an economist to lead the analysis from the earliest stages of rulemaking — retrofitting an RIA after a rule has been drafted is a recipe for OIRA return letters and court losses. The 2023 revisions add mandatory distributional analysis: you must analyze who bears costs and who receives benefits, not just the aggregate totals. Build internal capacity to handle this, or contract it to regulatory economists. Your OIRA desk officer should see drafts of the RIA before formal submission.

If you are a federal contractor, consultant, or regulatory economist: A-4 analysis is a specialized field with a growing market. Federal agencies often contract out RIA work for major rules. The 2023 revision's distributional analysis requirements have created demand for economists with equity analysis and environmental justice expertise. Familiarity with the 2023 discount rate guidance and the VSL literature is now prerequisite for serious regulatory economics work. Resources: the full 2023 A-4 circular is at whitehouse.gov/omb/information-for-agencies/circulars/.

If you represent a regulated industry or advocate for regulatory reform: An inadequate RIA is one of the most effective grounds for challenging a final rule under the APA's arbitrary-and-capricious standard. Courts look at whether the agency "examined the relevant data and articulated a satisfactory explanation" — a weak RIA that ignores major costs, uses implausible assumptions, or fails to analyze alternatives provides concrete grounds for reversal. OIRA also returns rules with inadequate RIAs. Filing detailed public comments on proposed rules that identify methodological weaknesses in the agency's draft RIA creates a record for both OIRA review and future litigation.

If you are a researcher, journalist, or policy analyst: RIAs for all economically significant rules are public — available in the rulemaking docket at regulations.gov and summarized in OIRA's annual regulatory review reports. The 2023 revision to A-4 is a landmark document in regulatory economics and reflects a deliberate policy choice to value long-term and distributional effects more heavily than the prior methodology. Reading A-4 alongside an agency's actual RIA reveals the analytical choices — and gaps — that determine whether a regulation survives or is struck down.

The 2023 Revision: Why It Was Controversial

The September 2023 A-4 revision drew sharp criticism from business groups, Republican members of Congress, and some economists who argued that:

1. Lowering the discount rate from 7% to 2% effectively tilts cost-benefit analysis in favor of more regulation by inflating the present value of future benefits — without a commensurate adjustment to the cost side
2. Distributional analysis introduces subjective and politically contested value judgments about which populations matter more into what is supposed to be a technical economic analysis
3. The comment process for the revision was rushed and the final rule departed from the proposed guidance in ways that reduced stakeholder input on key changes

Defenders of the revision argued that the 7% discount rate reflected 1970s estimates of private capital returns that are no longer empirically valid, that modern welfare economics supports lower social discount rates, and that ignoring distributional effects was itself a normative choice masquerading as neutral analysis.

The Trump administration that took office in January 2025 signaled intent to revisit the 2023 A-4 revision, and agencies were directed to conduct sensitivity analysis using higher discount rates in addition to the 2% primary rate — a de facto return toward dual-rate analysis. As of 2026, the 2023 revision technically remains in effect but is subject to ongoing regulatory and political pressure.

State and Local Government Implications

A-4 directly governs only federal rulemaking; state agencies are not bound by it. However, A-4's methodology has substantially influenced state-level regulatory analysis requirements. Many states that have enacted their own regulatory review laws modeled the analytical requirements on A-4, and federal grant conditions sometimes require state regulatory processes that are consistent with A-4 standards. States receiving major federal environmental grants (EPA, Transportation) may be expected to demonstrate that their state regulatory decisions are analytically sound under frameworks similar to A-4.

Recent Developments

• November 9, 2023 — OMB issued the first major revision to A-4 since 2003 (effective for proposed rules submitted to OMB after February 29, 2024): primary discount rate reduced from 7% to 2% real; distributional analysis required; VSL guidance updated; strengthened treatment of non-monetized benefits
• Early 2025 — Trump administration directed agencies to conduct sensitivity analysis using higher discount rates alongside the 2% primary rate; signaled intent to revise A-4 methodology
• Ongoing — Several litigation challenges to major EPA and OSHA rules cite the 2023 A-4 discount rate changes as rendering the RIA methodologically improper; outcomes pending in the D.C. Circuit
• Companion revision — OMB Circular A-94 (discount rates for government investment programs) was also revised in 2023 to align with A-4's updated social rate of time preference framework