Login
Home > Blog

Susan Dudley, Part 7
Dudley's Radical Ideas

7 Nov 2006 in ,

Last Monday we began a multi-part series on Susan Dudley, the Bush administration's nominee to head OMB's Office of Information and Regulatory Affairs. Her confirmation hearing is scheduled for November 13. A quick review:

In today's post we analyze a section in the report prepared by Public Citizen and OMB Watch titled "Dudley's Radical Ideas."

This section of the Opposition Report repeats a couple issues we've already addressed, most notably whether lives or life-years ought to be used as the metric for quantifying the lifesaving benefits of health and safety regulation. We've pointed out that when children are the principal beneficiaries, using life-years produces a larger measure of quantified benefits. Using life-years cannot "devalue the elderly" without simultaneously placing a premium on protecting children.

Dudley's other "radical ideas" discussed in this section of the Opposition Report include her support for:

We analyze each of these items below.

Guidance on Regulatory Analysis, Peer Review, Issuing Guidance in Lieu of Regulation, and Risk Assessment

The Opposition Report is critical of each of these initiatives. For example, the authors say that Circular A-4 "tilt[s] the rulemaking process in favor of corporate special interests," and that OMB's peer review bulletin "delay[s] the release and circulation of important scientific information." OMB's proposed guidance on the issuance of guidance in lieu of regulation "will result in the public being left in the dark about important agency matters," and its proposed bulletin on risk assessment "will continue the Bush administration’s agenda of tainting science with special interest politics."

The Report does not offer a factual basis to support any of these allegations. Nor do the authors list references to scholars or other analysts whose work could be reviewed to ascertain the merits of the authors' conclusions. The authors' fail to establish their conclusions except insofar as one is inclined to accept their presumptions. (In logic, this is called petitio principii, or "begging the question.")

Nevertheless, we can work backward from the authors' conclusions and inquire about whether the premises underlying them are factual. In the case of Circular A-4, the premise is undeniably false. OMB has had regulatory impact analysis guidance in place since at least 1990. Circular A-4 is the latest edition, and differences across editions are rather subtle. Finally, neither Public Citizen nor OMB Watch is on record opposing the edition issued during the Clinton administration.

If OMB's peer review bulletin was issued to delay agencies from releasing scientific information, it should be judged an abject failure. No evidence has been brought forward showing that delay has occurred.

OMB's proposed guidance on the issuance of guidance has not been finalized, so there aren't any data we can observe to evaluate its effects. The Opposition Report predicts that if finalized it will leave the public "in the dark about important agency matters," but this prediction is at least counterintuitive. One major provision of the proposal would require much greater public disclosure. Proposed Sec. III ("Public Access and Feedback") begins with the following requirement:

  1. Internet Access:
    1. Each agency shall maintain on its Web site a current list of its significant guidance documents. The agency shall provide a link from the current list to each significant guidance document that has been made public. New significant guidance documents and their Web site links shall be added promptly to this list, no later than 30 days from issuance.
    2. Each agency shall annually post on its Web site a comprehensive list of its significant guidance documents. The comprehensive list shall identify significant guidance documents that have been added to the list, revised, or withdrawn from the list since the previous comprehensive list.
    3. The guidance document lists shall include the name of each guidance document, any document identification number, and issuance and revision dates.

How greater public access translates into greater secrecy is not easily grasped.

If it is stipulated that the Bush administration actively seeks to "taint[] science with special interest politics," OMB's proposed bulletin on risk assessment appears to be badly designed to achieve that purpose. If finalized unchanged, the bulletin would prohibit agencies from embedding within scientific risk assessment documents anyone's policy judgment about how risk ought to be managed, including the judgments of special interests.

In sum, the conclusions in the Opposition Report about recent OMB initiatives are based on the common logical fallacy of building into the conclusion presumptions that are necessary for the conclusion to be true. When these presumptions are examined, they are either counterfactual or require highly counterintuitive reasoning from the facts available. This is sufficient evidence that Public Citizen and OMB Watch oppose OMB's prior initiatives for unstated instrumental reasons.

Regulatory "Sunsets"

The Opposition Report Criticizes Dudley for proposing that regulations automatically sunset after a specified length of time. Is this a radical idea? Let's begin with the management problem that Dudley will inherit as OIRA Administrator.

Since 1981, presidents have sought to manage the large and growing regulatory state. Executive Order 12291 called for a "look back" at this body requirements already codified in the Code of Federal Regulations. Responsibility for implementing this look-back was given to the newly established Presidential Task Force on Regulatory Relief and successor Council on Competitiveness, both of which were headed by the sitting vice president. White House oversight generated numerous controversies but few concerned existing regulations subject to validation, revision or sunsetting. Almost all students of regulatory process likely would agree that the look-back provisions in EO 12291 did not succeed.

In 1993, EO 12291 was replaced with Executive Order 12866. Sec. 5 directs federal agencies to reevaluate their body of existing regulations. The stated purposes of this directive could have been lifted from EO 12291. They were:

to reduce the regulatory burden on the American people, their families, their communities, their State, local, and tribal governments, and their industries; to determine whether regulations promulgated by the executive branch of the Federal Government have become unjustified or unnecessary as a result of changed circumstances; to confirm that regulations are both compatible with each other and not duplicative or inappropriately burdensome in the aggregate; to ensure that all regulations are consistent with the President's priorities and the principles set forth in this Executive order, within applicable law; and to otherwise improve the effectiveness of existing regulations:

EO 12866 thus replaced one look-back procedure with a different one. Yet there is little evidence that the new procedure was any more successful. The Clinton administration made another indirect look-back through its National Performance Review. We haven't researched the literature to see if there are any independent evaluations of the permanent successes of that program, but it tended to focus more on federal procurement than federal regulation.

Presidents are not alone in their resort to hope. Several regulatory reform bills debated (but not enacted) during the 1990s included look-back provisions of one type or another. There seemed to have been a consensus in Congress that something needed to be done, but agreement was hampered by partisans on both sides who complained that the proposed look-back provisions were either too weak (Heritage Foundation) or too strong (OMB Watch).

Proposals for regulatory "sunsets" seem to reflect the exasperation of regulatory reformers (such as Dudley) about the persistent failure of prior look-back efforts. Agencies tend to be focused on writing their next rounds of regulations with little regard for whether an existing regulation might have been the cause of the problem they are now addressing or whether there are "changed circumstances" make an existing regulation superfluous. EO 12866 contains language directing agencies to consider both of these phenomena, but there is little evidence that this actually happens.

Agencies' lack of enthusiasm for program evaluation is not limited to regulatory matters. Indeed, the problem was noticed first on the budget side. Public administration libraries are littered with failed management programs; does anyone still remember the Planning, Programming, and Budgeting System or Zero Based Budgeting? Each system was aimed at reforming or weeding out ineffective spending programs. The most recent such initiative is OMB's Program Assessment Rating Tool (PART). (We have no opinion about whether PART has been a success. We admit to some satisfaction that TSA's passenger screening technology is rated as "not performing" but surprise that PART reveals only 4% of federal programs are judged "ineffective".)

The idea of a regulatory "sunset" is very similar to Zero Based Budgeting, which was the management solution of choice for the Carter administration. No one seriously thought that ZBB would cause all federal programs to be "zeroed out," nor was there any significant concern that the task of defending existing programs from scratch would "would be an enormously wasteful enterprise that would leave [agencies] little or no time to look forward" (p. 31). The point of ZBB was to examine programs from the ground up to ensure that they were accomplishing their mission, and if not, consider whether they ought to be reformed or terminated. Regulatory sunsets would almost certainly operate similarly.

It's understandable that interest groups opposed to the systematic review of federal budgetary programs (such as OMB Watch) also would oppose the systematic review of federal regulatory programs. Their opposition may reflect an informed calculation that programs they favor would not perform well. Or they could be sufficiently uncertain about how well their favored programs would perform that organizational risk-aversion leads them to oppose program evaluation in general. Alternatively, program evaluation could be rationally viewed as a threat because they think there's a high likelihood that evaluation will lead to demands for reform, and those reforms will undermine their policy objectives. In short, the expected costs to these interest groups of program evaluation outweighs the expected benefits.

But presidents of both parties have sought (and largely failed) to devise effective schemes for identifying and removing regulatory deadwood. They are caught between interest groups within their own party, some of which will support look-backs and others which will not, and permanent Executive branch bureaucracies that have essentially no interest in evaluating their own problems. Look-back schemes tend to fail because they don't change bureaucratic incentives, and no one has yet figured out how to solve that problem. It's possible that regulatory sunsets could be the Holy Grail that management reformers have been looking for. It's more likely that regulatory sunsets will be the latest triumph of hope over experience.

Finally, there is an interesting irony worth noting. The management problem that program evaluation seeks to solve is a governmental failure entirely analogous to the externality-based market failures that animate many demands for regulation. Let's recount the similarities. The beneficiaries of the regulatory status quo have a concentrated interest in its preservation, whereas the beneficiaries of reform are diffuse taxpayers and voters. They face insuperable transactions costs trying to organize to represent their interests effectively. If they do, they face the added hurdle that their opponent across the negotiating table holds all the keys to power.

In the case of an externality-based market failure, the usual prescription is regulatory intervention that changes property rights so that the party which can remedy the externality at lowest cost is assigned the duty to do so. Because industry is usually deemed to have caused the market failure, industry gets assigned this duty. The same prescription would logically apply in the case of an externality-based governmental failure. The party that can remedy the externality at lowest cost should be assigned the duty to do so. In the case of governmental failure, the party able to remedy the externality at lowest cost is the regulatory agency.

Of the many previous attempts to gain management control over existing regulation, none have reassigned property rights this way. If regulatory sunsetting could be designed to accomplished that reassignment, it might actually work.

Regulatory Budgeting

The Opposition Report correctly says Dudley has supported assigning budgets to regulatory agencies that would place limits on what costs they could impose. Because this would mimic the procedures used to allocate appropriations, it is called "regulatory budgeting." (The authors of the Opposition Report call it regulatory rationing, a term that is usually applied to actions by regulatory agencies that limit consumer access to such things as health care, housing, and electricity. OMB Watch appears to be alone in using this term to describe regulatory budgeting.)

In theory, agencies would be "appropriated" an amount of cost they could impose during a fixed time period, and they would have to stay within that budget. As long as the budget is immovable, advocates hope that agencies would use the discipline of the budgeting process to maximize the social benefits that could be obtained within their legislatively authorized programs.

In practice, regulatory budgeting would almost certainly fail to achieve what its advocates hope. In testimony before the House Subcommittee on Energy Policy, Natural Resources ands Regulatory Affairs, I offered the following best-case scenario in response to former OIRA Administrator and OMB Director James Miller's endorsement of regulatory budgeting:

In his testimony last year Dr. Miller said, “OMB should be given a stronger role in policing this bias by replacing agency reports of benefits and costs with more objective estimates…” Whereas Dr. Miller would implement this through a regulatory budget, I am less sanguine about the likely effectiveness of such an approach. Nothing in the concept of regulatory budgeting overcomes the perverse incentives agencies have to understate costs.

In my judgment, a regulatory budget would exacerbate these perverse incentives. As it stands now, an agency’s incentive to understate costs is largely driven by the fact that high costs (irrespective of the magnitude of benefits) generate bad public and Congressional relations. But an enforced regulatory budget would limit what regulations an agency could issue. In principle, once an agency’s budget is reached it would be done for the fiscal year just as if it had spent its budget appropriations. Excess fiscal spending is controlled by the Anti-Deficiency Act, and additional budget dollars cannot simply be conjured up. It is difficult to imagine how to craft, much less enforce, an Anti-Deficiency Act for regulatory costs.

Agencies would respond to a regulatory budget much like they do to the Information Collection Budget—by reducing their estimates as necessary to make them fit under the allowable ceiling, not by reducing the paperwork burdens they impose (pp. 12-13).

The authors of the Opposition Report have a palpable antipathy for regulatory budgeting not because it would fail, but because they fear it might succeed. In their view, the benefits obtained through regulation are entitlements on which it is immoral to set any price:

Regulatory protections of the public health, safety, civil rights, environment, and other public interests are not a species of fiscal activity, meaningful only in terms of the costs imposed on corporate special interests when the federal government finally forces them to do the right thing as corporate citizens. They are, instead, entitlements in the truest sense of the word... In the face of harmful pollution, unsafe products released into the national marketplace, and other hazards that corporate special interests expose us to without otherwise being forced to internalize the attendant public costs, we are entitled to regulatory safeguards. Our government owes us nothing less.(pp. 33-34, emphasis in original).

The Opposition Report raises other interesting objections to regulatory budgeting. For example:

[N]ot all costs have the same moral or ethical value. Some regulatory costs represent the cost to industry of what it should have done as a good corporate citizen in the absence of regulation (p. 34).

Costs borne by firms are less important morally than costs borne by households. To the extent that firms have enjoyed "illicit" profits, large regulatory costs are morally desirable. "[T]hey knew of the harms they were creating but failed to act" (p. 35). The more it costs to comply with a regulation, the better it must be.

It's not clear how widely the public shares this "cost theory of benefit." It has a certain socialist-populist appeal that most recently seems to have been visible during last Spring's run-up in gasoline prices. Proposals for anti-"price gouging laws" and "windfall profits taxes" might fit the "cost theory of benefit" mindset.

But it's impossible to find any reference to such a notion anywhere in the economics literature, and by this indicator we infer that it is the Opposition Report authors' views that are appropriately described as "radical." The authors might reply that economists, as a group, are highly unrepresentative because they are notably conservative, Republican-leaning, or beholden to corporate special interests. There is ample empirical data to prove this is false.

William McEachern has studied the political contributions of members of the American Economics Association, the professional society to which 55% of economists belong. His sample of 2,000 AEA members yielded a 5.1 to 1 ratio of Democrat to Republican contributors during the 2003-04 election cycle. When he examined leaders of the AEA, he found a 9 to 0 ratio among editors; a 19 to 2 ratio among authors published in the AEA's flagship journal, the American Economic Review, and a 32 to 1 ratio among authors of its Papers and Proceedings issue. McEachern found similarly asymmetric partisan contributions for other AEA journals. For the Journal of Economic Literature, the ratios were 5 to 0 for editors, 16 to 0 for referees, and 24 to 0 for authors. Among 247 book reviewers with US affiliations, there was not a single Republican contributor. For the Journal of Economic Perspectives, the ratios were 2 to 0 for editors, 12 to 0 for advisory board members, and 20 to 1 for authors. "[W]hen all the qualifications are in, " wrote McEachern, "we have the following plain facts: the AEA is a predominately Democratic organization. Those responsible for the journals are especially Democratic, and they run the journals in a manner that tends to reflect that particular ideology" (p. 173).

Amending Laws to Permit Benefit-cost Decision-making

The Opposition Report criticizes Dudley for advising Congress to change statutes that currently prohibit Executive branch decision-makers to take account of cost. They term this "The End of 'Safety First' Laws" (p. 35). How radical is that?

As a substantive matter, it certainly could lead to significant changes in the decisions that officials make. For example, many scientists believe that there is no level of exposure to fine particulate matter that is safe. Yet the law directs EPA to set National Ambient Air Quality Standard at a level that protects public health with an "ample margin of safety." If these scientists are correct, then the only legally defensible standard for particulate matter is zero. It's unlikely that Congress intended EPA to set such a radical air pollution standard. For one reason. At the time it wrote the law no one seriously believed that particulate matter posed a health risk at every positive exposure level. For another reason, trying to achieve a zero standard would lead to national bankruptcy.

Procedurally, asking Congress to reconsider its intentions is an entirely reasonable path to take -- especially for an Executive branch official who lacks the legal authority to exercise any meaningful discretion. Congress could decide, after an appropriately informed and vigorous debate, to follow Dudley's suggestion and give EPA the authority to take cost into account. Alternatively, it could reiterate its originally stated desire for perfectly clean air by directly legislating a standard of zero for particulate matter. Nothing would eliminate "paralysis by analysis" faster than that.

Of course, it's possible that Congress would prefer to duck the question and make EPA continue to struggle trying to reconcile the inflexible demands of the law with the complexities of science. Given the many complaints raised in the Opposition Report about the "paralysis of analysis" that its authors believe Dudley will impose on regulators, this is an instance in which Dudley's approach (asking Congress for clarity, hoping for benefit-cost balancing as the outcome) could help break the current impasse and relieve EPA of its obligation to do the impossible.

A truly radical idea would be for EPA to set the particulate matter standard at zero and force Congress to deal with it.

OIRA Report Cards of Agency Performance

The Opposition Report mentions critically Dudley's recommendation that the OIRA professional staff grade agencies' performance with respect to the conduct of regulatory analysis. This "would consume vast amounts of taxpayer dollars on navel-gazing analyses that would increase the reported estimates of regulatory costs while doing little to inform the public about the lifesaving benefits of sensible safeguards appears to be a concern" (p. 37).

Rhetoric aside, it's not exactly clear what it is about OIRA-issued report cards that bothers Public Citizen and OMB Watch. There should be no question that performance appraisals would be a spur to improve quality. They might fear agencies would be driven to conformity with the Administration's political agenda, but Circular A-4 provides a model performance standard that should insulate agencies from any such pressure. (OMB could not get away with criticizing an agency for adhering to the principles of Circular A-4.) If Dudley made these report cards public -- and it's hard to imagine how she could avoid doing so in response to a Freedom of Information Act request -- it would be a huge step-change in transparency that Public Citizen and OMB Watch otherwise would be expected to welcome.

Changes in Regulatory Accounting

The Opposition Report criticizes Dudley's view that costs are significantly understated in OMB's annual reports to Congress on the benefits and costs of federal regulation, and that OMB ought to devote more attention to fully characterizing costs (p. 37). Quoting from the Report:

Accordingly, Dudley has counseled that OIRA should present as robust a picture of costs as possible, even presenting a picture of the costs for rules for which benefits have not been similarly quantified. Given the enormous difficulty of quantifying the benefits of regulation, which can include such abstract but quite real benefits as equity, civil rights, and the preservation of an interconnected ecology, Dudley’s proposal would feed industry’s anti-regulatory propaganda by presenting a skewed, cost-heavy picture of regulatory protections (pp. 37-38).

The authors of the Opposition Report seem to agree with the conventional wisdom that costs are easier to estimate than benefits, and therefore benefits tend to be underestimated. As a theoretical matter, this view is almost certainly false. There are two reasons why.

First, when agencies estimate cost they almost always estimate expenditures. But expenditures are not equal to costs, and cost is not defined as expenditure. OMB's Circular A-4 is technically correct on this point: "'Opportunity cost' is the appropriate concept for valuing both benefits and costs" (p. 18). What is "opportunity cost"? It is the value of benefits foregone when resources are sacrificed to obtain a benefit. To be more accurate in our terminology, we would not refer to the principle tool in applied economics as "benefit-cost analysis." We'd call it "benefit-benefit foregone analysis." We don't because it's really awkward, and we expect analysts to understand the opportunity cost principle. (For the record, in her public interest comment to OMB on its proposed Circular A-4, Dudley commended OMB for getting the opportunity cost principle correct. See p. 9.)

Second, those who estimate benefits and costs have recognizable incentives to do so strategically. That goes for Public Citizen and OMB Watch as well as the Mercatus Center. It's also true for federal regulatory agencies. OMB's reports to Congress are biased because it relies exclusively on estimates obtained from these agencies. This problem cannot be solved directly, but it can be managed through competition:

A better approach is ... to create incentives for the preparation of high quality analyses to be produced. As in every other market, competition is the key to improving quality. When it comes to regulatory analysis, each federal agency has monopoly power over what information is finalized and disseminated. As every freshman economics student learns, monopolies do not foster quality. Whether they work for industry or advocacy groups, outside experts can submit public comments to their hearts’ content... (p. 13).

Enabling all interested parties to give it their best shot, either on a complete regulatory analysis or some critical potion thereof, is the best way to get as close as possible to the elusive truth. Each interest group has strong incentives to detect both inadvertent error and chicanery in the work of its competitors. As mistakes are exposed, benefit and cost estimates are likely to converge. Decision-makers would have better, more creative regulatory options from which to choose; they'd be better informed about the likely effects of each alternative; and they'd be able to make timely decisions that are publicly defensible without resort to lawyerly obfuscation..

Now that's a truly radical idea.

name

Add to Technorati Favorites

[add a comment]

Add a Comment

*
*
*
Check to receive notifications of future comments.
Yes
No

Authors

Topics

Issues

Government Agencies

Jurisdiction

Archives

Blog Roll

Megan McArdle
Becker-Posner
Cafe Hayek
The Commons
Economist's View
Environmental Evaluation and Cost-Benefit News
Environmental Economics
Effect Measure
Federalist
Freakonomics
Greg Mankiw
Homeland Security Watch
Marginal Revolution
Roger Pielke, Jr.
Wine Economics