The Irreplicable Role of EPA Staffers

Jana Milford’s appraisal of the state of air quality benefits analysis at the US Environmental Protection Agency, “Counting the Benefits of Clean Air Regulations” (Issues, Summer 2026), will ring true to the policymakers, economists, risk assessors, and epidemiologists who have constructed the tools and techniques so widely viewed as best-practice. She rightly notes the role of independent advisory bodies in helping EPA, regardless of its political leadership, protect the public from the full breadth of air quality effects on human health. We believe, however, that she misses two key ingredients that made this possible: the creative and technically adept civil servants dedicated to performing analyses that advanced the state of the science; and a constructive feedback loop that begins with EPA policy decisions, leads to external scientific advisory boards, and then back again to the work of EPA staff.

The scientific advice of outside experts is meaningless unless the right staff are available to act upon it. In our combined 52 years of service at EPA, we worked alongside researchers who performed world-class research, advanced methods for simulating air quality, quantified the adverse effects of air pollution, and estimated the economic value of the resulting premature deaths and illnesses. One measure of their skill is reflected in the large and growing number who have been recruited by research institutions worldwide.

These individuals helped develop methods and tools that have been adopted globally. Examples abound. The environmental Benefits Mapping and Analysis Program (BenMAP) was conceived and developed by EPA staff and is now used worldwide to quantify and monetize the effects of poor air quality and climate change. Reduced-form techniques for estimating the benefits of reducing air pollution without having to model air quality, including the “benefit per ton” value that Milford mentions, were first deployed by EPA staff. Advanced methods for characterizing uncertainty in the relationship between ambient particulate matter and premature mortality using an expert elicitation protocol (a synthesis of opinions where there is uncertainty due to insufficient data) were developed and led by EPA staff. These remind us of the irreplicable role that a competent and apolitical civil service plays in constructing credible analyses.

We echo the author’s point that external scientific bodies, including the National Academy of Sciences and the EPA’s Science Advisory Board, play a key role in this process. These experts critically assess EPA’s methods, providing feedback to staff that allows them to inform policy decisions with strong evidence. Indeed, EPA staff have repeatedly sought external review of its methods. These panels, comprising world leaders in a variety of fields, provide invaluable feedback that very frequently both extends and validates EPA’s methods. As an example, EPA requested its advisory board to evaluate the agency’s reduced-form tools, as noted above; the board found that the methods were methodologically sound and generally fit for purpose.

As we consider the unprecedented value that clean air brings to human health, we should not lose sight of what makes it possible to quantify and monetize these effects: expert staff who are equal measures dedicated and creative. Any plan to restore EPA’s leadership must surely include adequate funding—but just as important will be its ability to recruit and retain the scientists who will again make its work great.

Jana Milford comments on the Environmental Protection Agency’s final rule for emissions from stationary combustion turbines, released in January 2026, which sets out the costs for industry but abandons a full cost-benefit analysis. EPA offers the rationale that uncertainty associated with estimating the benefits of controlling emissions is too great for meaningful cost-benefit analysis. I share Milford’s concerns about the implications of the agency’s position and offer a complementary perspective as an epidemiologist who has done research for decades on the harms of air pollution.

Cost-benefit analysis involves models, both on the costs and benefits sides of the comparison. The models explore counterfactual or “what if” scenarios. For stationary combustion turbines, models for benefits would estimate reductions in emissions and consequent drops in population exposures and then apply epidemiological models of risk to exposure scenarios to calculate how much morbidity and mortality would be avoided with the revised regulation. The EPA has long-standing, well-documented approaches for assigning dollar value to the avoided disease burden. Inherently, the benefits-side calculations bring acknowledged uncertainties. But costs to industry are no different and are estimated with a model, also with uncertainties.

I am struck by the asymmetry of the handling of the implications of uncertainty comparing benefits and costs. For costs to industry, the Economic Impact Analysis described in the final rule offers a superficiality: “It is important to note that the cost estimates presented in this chapter are subject to multiple sources of uncertainty.” For benefits, a far lengthier discussion of uncertainty is provided, prefaced by: “Historically, however, the EPA’s analytical practices often provided the public with a false sense of precision and more confidence regarding the monetized impacts of fine particulate matter (PM_2.5) and ozone than the underlying science could fully support, especially as overall emissions have significantly decreased, and impacts have become more uncertain.” (The technical designation refers to particulate matter less than 2.5 micrometers in diameter.) There is no metric for comparing uncertainties for cost and benefit estimates, and the basis for the asymmetric conclusion, leading to retention of cost but not of benefit estimates, is not clear.

I also share Milford’s concerns about the implications of the EPA’s characterization of the epidemiological evidence on airborne PM_2.5 and ozone for the National Ambient Air Quality Standards set under Sections 108 and 109 of the Clean Air Act. Quoting the Economic Impact Analysis: “In particular, the EPA is interested in evaluating the validity of estimating the benefits of air quality improvements relative to the National Ambient Air Quality Standards (NAAQS) for PM_2.5 and ozone.” Is this a warning that EPA may question the voluminous epidemiological evidence that has been foundational for the particulate matter covered by the NAAQS? Milford lays out the stringent process through which EPA uses evidence to assess the need for revision of a NAAQS, a process implemented while I was chair of EPA’s Clean Air Scientific Advisory Committee. This transparent process has worked, but I am concerned that it could be undermined by dismissing critical findings from epidemiological studies or tarnishing epidemiological evidence as too uncertain for the NAAQS revision process.