Tracking the value of regulation: can government reporting help bring rationality to regulation?

Hahn, Robert W. ; Layburn, Erin M.

REGULATION IS PLAYING AN INCREASingly important role in the United States and world economies. For example, U.S. regulations aimed at protecting health, safety, and the environment alone are estimated to cost over $200 billion annually, or about two percent of GDP. In 2002, the U.S. government estimated that the total cost of major rules issued between April 1995 and September 2001 was between $50 and $53 billion annually, while the benefits ranged from $48 to $102 billion annually. Yet, the economic impact of federal regulation receives much less scrutiny than the budget.

In an attempt to improve the regulatory process, Congress passed a 1995 law aimed at having the government assess the economic impact of federal regulation. The Office of Management an d Budget, which has primary responsibility for regulatory oversight, was directed to produce the report, which would address such key points as the costs and benefits of federal regulation, the impact of that regulation on local governments, and ways to improve federal regulation.

Critics of the OMB report felt that it would be used simply to highlight the costs of regulation and understate the benefits, because benefits are sometimes harder to quantify than costs. Proponents saw it as a way to achieve greater regulatory transparency and accountability by providing more systematic and timely information on the overall costs and benefits of federal regulation.

In response to regular congressional mandates, the OMB has produced five final reports on the costs and benefits of federal regulation. Have the reports satisfied congressional goals underlying the 1995 legislation? And, how can the government produce better information on the benefits and costs of regulation?

EVALUATION

In order to answer the first question, we need to "score" the five reports that have been produced so far on various dimensions. The advantage of such scoring is that it is reasonably objective. The disadvantage is that scoring cannot easily address some fundamental issues, such as the quality of a particular economic analysis.

In scoring the OMB reports, we evaluated the agency's assessment of the information in the Regulatory Impact Analyses that are performed for each major federal regulation by the issuing agency. Much of the information included in the OMB report comes from different federal agencies. Therefore, the amount and quality of the information on overall costs and benefits that the OMB is able to report is directly related to the quality of the analyses submitted by the agencies.

Throughout our scoring process, we focused on information concerning "non-transfer" rules, which generally include environmental, health, and safety regulations. In contrast, transfer rules are generally designed to shift wealth from one party to another and therefore typically only have small impacts on net economic welfare. President Clinton's Executive Order 12866, which requires agencies to analyze the economic effects of proposed rules that are economically significant, does not exempt transfer rules from the reporting requirement. However, agencies almost never produce regulatory impact analyses for transfer rules. Because agencies do not submit estimates of costs and benefits of transfer rules to OMB, we did not consider OMB's treatment of transfer rules when we considered the quality of the reports on the costs and benefits of regulation.

Costs and benefits In scoring the reports, we noted whether they met certain criteria related to the reporting of costs and benefits. Those results are shown in Table 1.

Executive Order 12866 requires agencies to assess the costs and benefits of rules that are economically significant, and defines a "significant regulatory action" as a rule "that is likely to have an annual effect on the economy of $100 million or more." There are several points worth noting:

* A majority of Regulatory Impact Analyses for new rules examined in the last four OMB reports quantifies some measure of costs and benefits.

* In the past five reports, benefits are monetized less frequently than costs.

* A small number of rules quantify neither costs nor benefits. In 1997, five of 21 rules fell into this category. In 2002, three rules failed to quantify costs or benefits, compared to two rules in 2001. Unfortunately, in 1998 and 2000, the OMB'S reports did not indicate the number of rules failing to quantify costs and benefits.

* The numbers that the OMB does provide cannot easily be compared across the different reports. In the 1998 report, for example, the agency does not clearly state that 16 of the 22 new rules quantify benefits. The reader needs to find and tally seven numbers that the agency does not provide to find that piece of information.

* The fraction of new rules reporting monetized estimates for both some costs and some benefits does not exceed 50 percent in three of the four years for which information is provided.

Standardization of numbers A valuable part of the second, third, and fifth reports is a set of tables in which the OMB standardizes yearly costs and benefits by agency. The agency monetizes the agencies' quantified estimates when it is able to, and converts the estimates to a standardized dollar year so that comparisons can be made across agencies and years. The 2002 report states that in "assembling estimates of benefits and costs ... OIRA has monetized quantitative estimates where the agency has not done so."

We do not feel the estimates reported by the OMB are either standardized or comprehensive enough to be used in reliably assessing net benefits. While the agency does monetize some benefits that the agencies have only quantified, the tables take agency numbers as given when an agency has provided a monetized estimate, even though different agencies may use different assumptions to monetize costs and benefits. The OMB notes that "to the extent that agencies have adopted different monetized values for effects--for example, different values for a statistical life or different discounting methods--these differences remain embedded in the tables." Finally, the tables omit any valuation of benefits that the agencies did not quantify. Any cross-year or cross-agency comparison that the tables do allow one to make is still incomplete.

Collection costs of regulation All of the reports address the issue of information collection costs, such as those associated with additional paperwork. In the past three reports, the OMB has also provided a table summarizing information collection costs that result from federal regulation. The table breaks down costs imposed by executive and independent agencies. The costs are significant in some cases, but they do not include many critical costs of regulation, such as impacts on firm's production processes and effects on consumers unrelated to paperwork.

Net benefits Table 2 reviews the OMB's reporting practices on the net benefits of new rules. Each year, of the new non-transfer rules considered, less than half would unambiguously pass a cost-benefit test based on quantified estimates of benefits and costs. At the same time, only a small fraction would unambiguously fail a cost-benefit test. By inference, most rules either do not provide enough information to compare costs and benefits, or there is a large enough range of uncertainty in the agencies' estimates to put the regulations in a gray area where they neither unambiguously pass nor fail. For example, in OMB'S 2002 report, we calculated that net benefits were not obviously positive or negative for 23 of 34 new rules.

While it may be sensible to use ranges when estimating costs and benefits to reflect uncertainty, it is difficult to interpret ranges. First, the agencies rarely provide information about their degree of confidence that the actual value of benefits or costs falls within the range. Second, when estimated ranges of costs and benefits produce a net benefit estimate that ranges from negative to positive, it is unclear whether a regulation is likely to pass or fail a cost-benefit test.

The number of rules passing or failing a cost-benefit test could indicate the effectiveness of the OMB oversight process, although there clearly are other important factors. The flow of rules is not random and is determined by forces outside of the OMB's control--most notably Congress. The number of rules unambiguously passing a cost-benefit test does not show any obvious pattern for new rules over the past five reports.

Aggregate net benefits The reports also consider the issue of aggregate net benefits. Aggregate net benefits can provide useful information on whether a particular set of regulations or programs is enhancing economic welfare at a particular point in time or over a time period. However, as several authors and the OMB have pointed out, there can be problems with adding up the benefits and costs of regulation across different studies because of differences in assumptions and baselines. The 2002 report is the first not to include any information on aggregate net benefits of new rules.

In the past two years, the OMB has discontinued the practice of forecasting aggregate net benefits over time. The 1998 and 2000 reports include tables that forecast the costs and benefits of federal regulations in the years 2005, 2010, and 2015. In both its 1998 and 2000 reports, the OMB forecasted and summed future costs for individual regulations in a summary table.

Sometimes, aggregate benefits can be disaggregated to provide more informative estimates. For example, it may not make sense to combine the net benefits of airline deregulation with the net benefits of safety regulation, but it might be reasonable to determine whether the Environmental Protection Agency's Superfund program's benefits are likely to exceed its costs. Generally, the OMB has not provided much useful information on the benefits and costs of specific programs. In the 2002 report, however, it did begin to provide some data on benefits and costs by agency, information that could be useful for comparing net benefits of regulations across agencies.

OMB's recommendations for reform One of the initial aims of Congress was to have the OMB suggest ways to "reform or eliminate any federal regulatory program or program element that is inefficient, ineffective, or is not a sound use of the nation's resources." The agency has generally been slow to suggest specific reforms. So far, it has not identified any regulatory programs for elimination or improvement.

In the 2001 report, however, the OMB endorsed specific suggestions from the public about reforming, and in some cases eliminating, individual regulations. The 2002 report included information on the status of efforts to improve the regulations that were identified as high priority reform opportunities in the 2001 report. The OMB has taken further measures to improve federal regulation by encouraging agencies to examine the impact of new, economically significant regulations. The 2002 report states that "in response to our request for regulatory reform proposals, we received suggestions on 316 unique regulations and guidance documents covering 26 federal agencies." The report further notes that the OMB has made the decision to change the way it evaluates reform suggestions from an OMB-initiated process to an agency-initiated process. In September of 2001, the agency introduced the "prompt letter," which encourages agencies to issue specific regulations whose benefits exceed costs.

Overall picture The picture that emerges from the reports is intriguing. The OMB is focusing less on aggregate estimates of the impacts of regulatory activity and more on improving the process and particular regulations. It has not used its own in-house expertise to render judgments on the quality of regulatory analysis given to it by the agencies. It is, however, taking greater advantage of suggestions from interested parties to identify regulations that are in need of reform or elimination. It is also actively searching to help identify new regulations whose net benefits are likely to be substantial, and using prompt letters to alert agencies to new regulatory opportunities. Finally, changes in the agency's use of the return letter suggest that the OMB may be assessing and making changes to its oversight role. The OMB's use of the return letter is authorized in Executive Order 12866, which states that when the agency denies approval of a regulation that an agency has submitted, it must also send a letter explaining why the regulation was not approved. The OMB may send a return letter to explain a problem it has with a draft rule or with a draft analysis of the rule's impact. The agency issued a higher rate of return letters in the year preceding its most recent report than it had since 1984, and also instituted the practice of making return letters publicly available on its website.

The OMB is also making attempts to improve the quality of regulatory analysis and to make the regulatory process more transparent. In its 2000 report, the agency published guidance to agencies on how to conduct Regulatory Impact Analyses. The OMB announced in its 2002 draft report that it was soliciting comments on what should be considered in updating the guidance that was published in the 2000 report. This year's final report lists five topics that the OMB intends to consider in developing updated guidance to agencies.

Since the most recent report to Congress, the OMB has involved all of the executive agencies in significant actions to improve the quality of information disseminated by federal agencies. A 2001 appropriations bill directed the agency to issue government-wide guidelines that would improve the quality, objectivity, utility, and integrity of federal agencies' information. The OMB issued final information quality guidelines to all agencies in September 2001. In the guidelines, the agency directed all agencies to issue their own agency-specific guidelines that described how the agency would comply with the OMB'S information quality standards, ensure the correction of information that did not comply with the guidelines, and report to the OMB director on complaints received about information quality. In an October 2002 memo, the Office of Information and Regulatory Analysis administrator reported that OMB had completed a review of the different agency-specific guidelines, and that OMB's "implementation of the Information Quality Law represents the first time that the Executive Branch has developed a government-wide set of information quality guidelines, including agency-specific guidelines tailored to each agency's unique programs and information."

The OMB has recognized that the Internet is a valuable way of making regulation transparent. It has made more information available to the public by publishing prompt letters, return letters, memos, and press releases on its own website. There is some indication that the efforts to facilitate public involvement in the regulatory process are working: The OMB received 1,700 comments on 267 regulations and 49 guidance documents in response to its most recent request for reform suggestions. It has also directed agencies to use the Internet to make more information available to the public: "Agencies should use their websites to keep the public informed about information on a timely basis. Specifically, each agency or office should establish an information quality site on its website." In addition, the OMB's willingness to work with agencies to make Regulatory Impact Analyses more available on the Internet is highly commendable. The online collaboration will make the regulatory process more transparent and hold lawmakers more accountable.

RECOMMENDATIONS FOR REFORM

Earlier AEI-Brookings Joint Center analyses have strongly recommended that the OMB make greater use of its in-house expertise in judging the quality of regulatory analyses. After reviewing five annual reports, we have concluded that this recommendation is not likely to be implemented. It is simply too costly politically for the agency to criticize, either implicitly or explicitly, an agency's analysis of an economically significant regulation. The primary venue the OMB uses to offer such criticisms is its return letters, which are now posted on the OMB website. The letters are a useful first step in providing feedback on regulatory analyses that are problematic, but they are only a first step.

To improve the quality of regulatory analysis and promote greater regulatory transparency, we offer the following three recommendations:

The OMB should ask the agencies to provide a standard, objective scorecard for each major regulation submitted to OMB for review.

The agency should make all of the scorecards readily available and provide an analysis of them. By requiring agencies to submit a scorecard that evaluates their own analyses of proposed regulations, the OMB can encourage agencies to pay more attention to fundamental criteria. For example, if an agency needs to report whether it has quantified and monetized pollution benefits, whether it used the same baseline to estimate costs and benefits, whether it identified a best estimate for the regulation's expected net benefits, and whether it identified the dollar year in which it has stated its estimates, the agency will be less inclined to submit an analysis that neglects those features.

We do not believe that this requirement would impose a significant burden on the agencies, as they will be familiar with the contents of the analyses they have performed. Answering a standardized set of straightforward questions about the analyses should require a minimal amount of extra time and resources.

The benefits of this exercise could he significant. It could help regulators and decision-makers determine the strengths and weaknesses of the underlying analysis. It would also facilitate a straightforward assessment of the degree to which regulatory analyses are meeting several important criteria.

Summary statistics on the analysis could be compiled easily by looking at a set of scorecards completed by federal agencies. The quality of federal regulatory analysis could then be assessed at the level of an individual regulation, a specific program or statute, a specific department, or agency. The OMB or the public would be more able to assess which regulations were producing the highest net benefits to society, or which agencies were consistently failing to estimate overall impacts of their regulations.

The OMB should calculate summary statistics based on the scorecards, observe common strengths or deficiencies in agencies' analyses, and draw conclusions about ways in which agencies' analyses could be improved. An examination of past reports shows that the OMB's overall assessment has important omissions. For example, the reports consistently fail to note how many rules provide a point estimate of net benefits, provide a range of net benefits, or consider alternatives.

Independent agencies should be required to do regulatory analyses for economically significant regulations.

Independent agencies' analyses should be subject to the same requirements as executive agencies' analyses, and should also be included in OMB's annual report. Currently, Executive Order 12866 does not apply to independent agencies. Those agencies--federal agencies that have been established by Congress to have a certain amount of independence from the president--account for a significant amount of regulation, but when it comes to analysis, they get a virtual free ride.

Independent agencies are frequently engaged in big, and sometimes massive, regulatory proceedings. Think of the Securities and Exchange Commission in the wake of corporate accounting scandals and the Federal Communications Commission decisions over the sharing of telephone lines and spectrum frequencies. Analysis of the costs and benefits of such regulations is becoming more critical. Regulations from the independent agencies should receive the same level of scrutiny that is applied to regulations from executive agencies. Like the executive agencies, independent agencies should have to work with the OMB in estimating the effects of new regulations and should be required to get the OMB's approval of all new regulations, to the extent permitted by law.

To implement this change, a new executive order would be needed that requires the OMB to review regulations from both independent and executive agencies. If the OMB is not allowed to review regulations from independent agencies, then Congress should develop an alternative mechanism for review that is similar to the OMB oversight process.

Congress should create a Congressional Office of Regulatory Analysis or a separate agency outside of the executive branch to assess the economic impact of important regulatory activity occurring at all federal regulatory agencies.

We believe that a separate regulatory oversight agency outside the executive branch is sound for three reasons: first, it can provide an independent check on the analysis done in the executive branch by the OMB and the agencies; second, it can help to make the regulatory process more transparent; and third, Congress can use the independent analysis to help improve regulation and the regulatory process.

The agency's Office of Information and Regulatory Affairs faces inherent limits in the scope of its review of individual regulatory proposals. The OIRA administrator is nominated by the president, who also nominates the heads of the various regulatory agencies. Therefore, there is likely to be some implicit understanding that the head of OIRA is not to press the agencies excessively hard because he or she is part of the same administration as the agency heads.

The constraints on OIRA are manifested in its annual report, in which it has, so far, simply accepted the benefit and cost estimates compiled by the agencies instead of providing any of its own assessments. A new office of regulatory analysis outside of the executive branch would not have that conflict of interest and could more easily criticize the analysis done by federal regulatory agencies. Competition between agencies has the potential to enhance the analysis produced by OIRA and its independent competitor, much like competition between the Congressional Budget Office and the OMB has done in the budget arena.

Secondly, the new office could help make the regulatory process more transparent by providing a more independent and public voice than OIRA. We recommend that this office be encouraged to submit comments during the public comment period of proposed regulations. The comments will give the agencies early warning of what the non-executive branch office is likely to say in its report to Congress after the rules are issued as well as make the regulatory process more transparent for the public.

Finally, we believe that this additional independent analysis could help improve regulation and the regulatory process in general. The new office could help provide a more complete picture of the regulatory process, especially in areas that the OMB has not examined carefully. For example, we only have a very incomplete understanding of the benefits and costs of regulatory activities at independent agencies.

Furthermore, a new regulatory analysis office could help Congress periodically assess the need to modify its own regulatory statutes as well as major regulations. As it is now, if and when Congress chooses to do so, it will have to rely on the agency's own estimates of the impacts of a rule and on any other data that interested parties may or may not have submitted in the rulemaking record. Significantly, Congress now has no credible, independent source of information upon which to base such decisions. That is analogous to the pre-CBO Congress, which had to make budget and appropriations decisions based solely on the information developed by the executive branch.

Criticism Despite those important reasons for creating a new regulatory analysis office, critics have voiced many questions and concerns. First, there has been some debate about the appropriate scope of activities for a regulatory analysis office outside of the executive branch. We do not believe that this office should go so far as to replicate everything that the agencies have already done. Conducting its own Regulatory Impact Analysis for every "major" rule would not only be ineffective (the time frame under which the office would have to operate would make it impossible to conduct a comprehensive analysis) but also costly. The point of this independent agency would be to review the Regulatory Impact Analyses and rules--both as they are proposed and when they are issued--for their methodological and factual integrity and for whether they reflect a consideration of reasonable alternatives and whether they are consistent with the authorizing statute. Up to this point, the OMB has shown that its political constraints are too great to allow enough flexibility to critique agencies' analyses.

Finally, critics have worried about the cost of instituting yet another government office. Costs can be kept low by only reviewing major rules, with a focus on those that potentially will have large economic impacts. We have already established that regulation has a large economic impact on the economy. Therefore, even small gains that this new office could make toward reducing unnecessary or inefficient regulation would yield large benefits in the long term.

CONCLUSION

Regulation is becoming increasingly important in many aspects of our economy. It has an important effect on our quality of life and the costs of goods and services; it also affects the ability of firms to compete in an increasingly global economy. The OMB's annual reports attempt to provide an objective characterization of the costs and benefits of federal regulation as well as identify opportunities for reform. If successful, that could have important implications for reviewing the impact of regulations across the world. Countries and regions, such as the European Union, are showing greater interest in introducing new approaches that will allow them to improve their assessment of regulatory activity.

Five U.S. government reports on the costs and benefits of regulation have been completed. On the positive side, each report provides useful information on the costs and benefits of regulation and the regulatory process. In addition, the reports show how the regulatory oversight function in the United States has changed in a relatively short period of time. In particular, recent changes at the OMB suggest a heightened concern with regulatory transparency, a greater focus on identifying regulations that are likely to do more good than harm, and a greater focus on targeting regulations for reform or elimination.

On the negative side, there are clear limits to how far an agency like the OMB can go in providing an objective critique of sister agencies. One of the telling findings of our analysis is that the OMB has taken its sister agencies' analyses of regulations as the basic point of departure for providing information in all of its reports, although OMB has also used its annual report as a place to discuss shortcomings in agencies' analyses. With well-known defects in many of the analyses, we think the degree to which the OMB treats agency numbers as reliable is problematic. In addition, the agency has neither directly criticized regulations nor offered reform suggestions that it produced. Instead, it used suggestions from outsiders as a way of partially insulating itself from the political "heat" it would take for making such suggestions. We think that is an innovative approach to a difficult political problem, but we also believe there are real limits to what the OMB can do in this area.

Even if the agency numbers on the impacts of regulation are taken as given, the OMB can still provide a valuable service by presenting an objective characterization of regulatory analyses that are done by the agencies. That includes information on the number of regulations that are likely to pass or fail a cost-benefit test and the number of Regulatory Impact Analyses that consider alternatives to the regulation.

The basic problem that an agency like the OMB faces in reporting on regulatory policy and developing more effective regulatory policy is that it operates in an intensely political environment. Some of the recent innovations by the OMB, such as prompt letters, represent useful ways of addressing the politics and economics of improving regulation. In that regard, we think that the OMB's move to organize agencies to work on information quality guidelines was a good one. A key insight for those interested in political economy is that strategic use of the administrative process can result in policies that would generally be viewed as enhancing efficiency and transparency.

By and large, we think that the OMB's annual reports represent a significant step forward in providing insights into the regulatory process as well as the costs and benefits of regulation. But a review of the reports also illustrates some of the problems with having a government agency conduct such an analysis. We have recommended that the OMB require agencies to issue a scorecard evaluating each agency regulation, that it summarize the strengths and weaknesses of regulations using this scorecard, that independent agencies be included in the OMB's analysis, and that Congress create an agency or office outside of the executive branch to perform a regulatory evaluation function similar to that of the OMB.

While we are highly critical of some aspects of the reports, we are guardedly optimistic about their potential to help improve regulation and the regulatory process. They demonstrate that it is possible for a federal agency to provide a reasonable characterization of the costs and benefits of regulation and offer concrete suggestions for improving specific regulations. And that is no small feat.

TABLE 1
Five Sets of Findings
Information on regulations and their costs and benefits.

 1997 1998 2000

Date range of economically (4/96-3/97) (4/95-3/96) (4/98-3/99)
significant final rules that and
were not included in a (4/97-3/98)
previous report

Does OMB take the agency's Yes Yes Yes
numbers as given?

Number of new non-transfer 21 22 22
rules discussed in the report

Did the report state how many new social regulations
(or "non-transfer") rules ...

 ... quantified some costs? No 19 of 22 16 of 22
 ... monetized some costs? 16 of 21 19 of 22 16 of 22
 ... quantified some 14 of 21 16 of 22 16 of 22
 benefits?
 ... monetized some 8 of 21 13 of 22 12 of 22
 benefits?
 ... monetized at least No 13 of 22 10 of 22
 some costs and some
 benefits in the
 agency's analysis?
 ... quantified neither 5 of 21 No No
 costs nor benefits?

Are any quantitative cost or Yes Yes Yes
benefit estimates provided
for independent agencies'
regulations?

Did OMB monetize any costs or No Yes Yes
benefits that the agencies'
RIAs did not?

Did OMB monetize lives saved No Yes Yes
when an agency did not?

Did OMB monetize any No Yes Yes
pollution benefits.

 2001 2002

Date range of economically (4/99-3/00) (4/99-9/01)
significant final rules that
were not included in a
previous report

Does OMB take the agency's Yes Yes
numbers as given?

Number of new non-transfer 12 34
rules discussed in the report

Did the report state how many new social regulations
(or "non-transfer") rules ...

 ... quantified some costs? 8 of 12 27 of 34
 ... monetized some costs? 8 of 12 27 of 34
 ... quantified some 9 of 12 20 of 34
 benefits?
 ... monetized some 7 of 72 19 of 34
 benefits?
 ... monetized at least 6 of 12 12 of 34
 some costs and some
 benefits in the
 agency's analysis?
 ... quantified neither 2 of 12 3 of 34
 costs nor benefits?

Are any quantitative cost or Yes Yes
benefit estimates provided
for independent agencies'
regulations?

Did OMB monetize any costs or No Yes
benefits that the agencies'
RIAs did not?

Did OMB monetize lives saved No Yes
when an agency did not?

Did OMB monetize any Yes Yes
pollution benefits.

TABLE 2
Five Sets of Findings
OMB report information on net benefits of individual regulations

 1997 1998 2000

Did the report state how many
new social regulations ...

 ... had positive net benefits 7 of 21 6 of 22 8 of 22
 in the agency's analysis?
 ... had negative net benefits No 4 of 22 No
 in the agency's analysis?
 ... reported a point estimate No No No
 for net benefits?
 ... reported a range for net No No No
 benefits?

Did the report state how net No No No
benefits were calculated?

 2001 2002

Did the report state how many
new social regulations ...

 ... had positive net benefits 3 of 12 9 of 34
 in the agency's analysis?
 ... had negative net benefits Unclear 2 of 34
 in the agency's analysis?
 ... reported a point estimate No No
 for net benefits?
 ... reported a range for net No No
 benefits?

Did the report state how net No No
benefits were calculated?

READINGS

* Benefit-Cost Analysis in Environmental, Health, and Safety Regulation: A Statement of Principles, by Kenneth J. Arrow, Maureen L. Cropper, George C. Eads, Robert W. Hahn, Lester B. Lave, Roger G. Noll, Paul R. Portney, Milton Russell, Richard Schmalensee, V. Kerry Smith, and Robert N. Stavins. Washington, D.C.: AEI Press. 1996.

* Improving Regulatory Accountability, by Robert W. Hahn and Robert E. Litan. Washington, D.C.: AEI Press and Brookings Press, 1997.

* Reforming, Federal Regulation, by Robert E. Litan and William D. Nordhaus. New Haven, Conn.: Yale University Press, 1983.

* Risk and Reason, by Cass R. Sunstein. Cambridge, UK: Cambridge University Press. 2002.

* "Stimulating Smarter Regulation: 2002 Report to Congress on the Costs and Benefits of Federal Regulations and Unfunded Mandates on State, Local, and Tribal Entities," produced by the Office of Management and Budget. Washington, D.C.: Government Printing Office, December 2002

Robert W. Hahn is the executive director of the AEI-Brookings Joint Center for Regulatory Studies and a resident scholar at the American Enterprise Institute.

Erin M. Layburn is a graduate student at Stanford Business School and was the program manager at the AEI-Brookings Joint Center for Regulatory Studies.