Graham Pushes Regulatory Priorities in President's Budget
by Guest Blogger, 3/4/2002
As administrator of the Office of Management and Budget’s (OMB) Office of Information and Regulatory Affairs (OIRA) -- which has authority to review and possibly reject or amend agency regulatory proposals -- John Graham is pressing agencies to adopt particular analytical methods to assess regulatory costs and benefits that would rig the result and undoubtedly lead to less protective health, safety, and environmental standards. Graham points to his preferences in the president’s recent budget submission to Congress, noting that new formal guidance to agencies on how to conduct monetization is forthcoming: “OMB has committed to update periodically its guidelines for regulatory analysis, which are used when OMB reviews agency rulemakings.” In particular, Graham stresses the importance of “league tables” for setting regulatory priorities. These tables are intended to compare the costs and benefits of one type of regulation, such as auto safety, to another, such as environmental protection. As an example, Graham presents his own league table of rules already on the books, which includes four rules from the Environmental Protection Agency (EPA), four from the Department of Transportation, and two from OSHA. From this table, Graham notes “the tendency for safety rules to be more cost-effective than health rules.” For instance, DOT’s rule on head impact protection scores well while EPA’s NOx SIP Call rule on air quality and OSHA’s rule on methylene chloride (to prevent against cancer) do not. Ideally, according to Graham, league tables should be used to compare programs across agencies at the beginning of the regulatory process for proposed rules (rather than rules that have already been adopted). This “is more useful for synoptic purposes or for decision making by governmental entities with inter-agency responsibility (e.g., appropriations committees and OMB),” Graham writes. Based on Graham’s presentation, this seems to suggest a trimming of agency budgets that deal primarily with health or the environment, such as EPA, and perhaps redistributing the savings to safety agencies, such as DOT, that deliver more “cost-effective” rules. As Graham points out, “The table suggests that we need to do a better job at both refining estimates of the cost-effectiveness of regulatory proposals and setting priorities for the use of the nation’s limited resources to protect citizens from health, safety, and environmental risks.” Yet this policy position is a direct result of Graham’s analytical choices, which are open to question. In formulating the league table, Graham employs his favored method to monetize the costs and benefits of each rule. As explained in this OMB Watch analysis, the mere process of monetization, regardless of method, inevitably fails to capture crucial benefits. In the case of Graham’s league table, the benefit estimates are derived almost exclusively from avoided fatalities. They exclude or devalue other impacts, such as morbidity, effects on ecosystems, and equity considerations. Moreover, even for measures of avoided fatalities, which Graham has included, benefits are greatly understated as a result of Graham’s analytical preferences. Specifically, Graham monetizes benefits by focusing on life-years saved (as opposed to the number of individual lives saved, as commonly practiced), assuming no benefit until the first life-year is saved. From the beginning, then, this method of computing benefits will inevitably bias the system against regulations such as cancer prevention -- which has a long latency period -- that primarily benefit the elderly (who have fewer “life-years” remaining) and people in the future. Yet on top of this, Graham then discounts the value of life-years saved in the future by 5 percent for EPA rules (7 percent for other rules) from the point that the first life-year is expected to be saved. To use an example from Georgetown Law Professor Lisa Heinzerling, a regulation that, on average, prevents fatality at the age of 35 would save 42 life-years assuming a life expectancy of 77 years. Discounting 5 percent from each life-year starting at age 36, the present value of the 42nd life-year saved would be approximately 1/8 of a year. Not surprisingly, no such discounting is advised on the cost side of the equation to account for inevitable economic growth -- which produces more resources to be spent on regulation over time -- or well-documented adaptive effects, such as technological advances or “learning by doing.” As a result, cost estimates frequently prove overblown in the real world. For instance, EPA estimated in 1990 that acid rain controls would cost electrical utilities about $750 per ton of sulfur dioxide emissions; yet the actual cost today is less than $100 per ton, billions of dollars less than what was initially anticipated. With overblown cost estimates, combined with Graham’s approach to discounting and use of life-years, it’s hard to see how the government would ever again produce a regulation that primarily is directed against diseases of old age. From these value-laden analytical choices, Graham suggests the emphasis should be on safety regulation, which can save young people today, while we should curtail health and environmental regulation, which do not fare well under his “cost-effectiveness” test. Yet, the need to pose such a tradeoff remains unclear. Certainly, safety regulations are important (although so far, Graham seems weak even here, as evidenced by his recent rejection of the Department of Transportation’s (DOT) tire safety standard). But should this type of regulation preclude other regulations to protect public health and the environment? Graham has previously accused the government of “statistical murder” for failing to pose such tradeoffs, and his promotion of league tables is clearly an attempt to move in that direction. This formulation assumes a fixed national budget for risk reduction, where a dollar spent on Risk A means a dollar less to spend on Risk B. Yet the United States has a $9 trillion economy, of which only a tiny fraction is devoted to risk reduction. Of course, agencies and decision-makers must prioritize activity within fiscal constraints. But frequently the hard tradeoffs posed by Graham -- pitting one life-saving measure against another -- are unnecessary and lead down a path to inaction. In the end, the effort to monetize benefits and the debate over methodology is just a distraction that masks the true policy choices that must be made. Because of the many assumptions and analytical games that go on to arrive at a dollarized figure -- which only the practitioner can sort out and truly understand -- it would be more useful to decision-makers and more transparent to the public if non-monetary benefits were simply described (quantitatively to the extent feasible) by stating, for example, the expected number of lives saved over a given period of time. As it stands now, numbers are thrown around like rhetorical grenades and no one really knows what they mean.