Gary D. Bass, Ph.D. testified before the Senate Committee on Governmental Affairs on the Regulatory Right-to-Know Act and Congressional Office of Regulatory Analysis today. Statement of Gary D. Bass, Ph.D. Executive Director OMB Watch Before the Senate Committee on Governmental Affairs On The Regulatory Right-to-Know Act and Congressional Office of Regulatory Analysis April 22, 1999 Thank you for the opportunity to testify today regarding S. 59, the Regulatory Right-to-Know Act (Regulatory Accounting), and the establishment of a Congressional Office of Regulatory Analysis (CORA). My name is Gary Bass, and I am the executive director of OMB Watch, a nonprofit research and advocacy organization. OMB Watch has been deeply involved in monitoring executive branch regulatory matters since its founding in 1983 and has worked to encourage a more open, responsive, and accountable federal government. OMB Watch also chairs a coalition, called Citizens for Sensible Safeguards, that includes more than 300 organizations dedicated to protecting and promoting the interests of consumers, workers, public health, civil rights, and the environment. Speaking for OMB Watch, as well as Citizens for Sensible Safeguards, we strongly oppose the regulatory accounting bill and CORA for similar thematic reasons:

• Both have little practical utility for public policy, yet would carry hefty price tags. As OMB has stated, �Aggregate estimates of the costs and benefits offer little guidance on how to improve the efficiency, effectiveness, or soundness of the existing body of regulations.� Yet with the expanded analytical requirements of S. 59, a substantial resource burden would be placed on OMB and the agencies for cumulative cost-benefit analysis � as well as brand new subanalyses � when regulatory matters are, in fact, handled best on a case-by-case basis. Likewise, CORA adds little to policy-making, as it duplicates work already done by the agencies, OMB, and GAO. This work is readily available to Congress, and as a result, Members have had little difficulty in obtaining cost-benefit information when assessing the merits of agency rules. Undoubtedly, CORA would carry a price tag at least equal to that of the Congressional Budget Office at $25 million, and probably more if it were to truly carry out all its functions, such as cost-benefit analysis of all major rules.
• Both deal in vast analytical uncertainty. OMB has emphasized the uncertainty of regulatory accounting in its first two reports, as have legal and economic experts. Part of the problem here is the masking of value judgements that inevitably occurs in a monetized study of this kind, which actually undermines the public�s �right-to-know.� Moreover, S. 59 marks a significant analytical expansion of previous regulatory accounting requirements, calling for a substantial amount of data that is not now available, such as cost-benefit analysis of paperwork requirements. Similarly, CORA�s data would be unreliable because it would have to conduct its own cost-benefit analysis for each major rule within a 45-day period, and without having been part of the rulemaking process. Such a limited time-frame would likely force CORA to rely heavily on industry estimates.
• Both raise concerns that they could be used as political weapons. Many of the backers of regulatory accounting have also been vocal proponents of other various �reform� measures designed to stem regulatory costs. A regulatory accounting report showing very large costs and small benefits could be a useful tool in advancing this agenda. S. 59, with its slanted analytical requirements, could be seen as an attempt to forcibly bend OMB�s numbers in an ideological direction consistent with the proponents of broad regulatory �reform.� There is also a danger that CORA would be used as a political instrument. It�s not hard to imagine a body like CORA, which would function as an arm of Congress, being influenced by the expectations of individual lawmakers looking to push an ideological agenda. Indeed, under last session�s CORA bill, the House and Senate leadership would control the appointment of CORA�s director, which is especially troubling if data from CORA is to be used as the basis for rejecting agency rules, as its proponents suggest.

S. 59, the Regulatory Right-to-Know Act Mr. Chairman, when introducing S. 59, which requires OMB to perform a yearly cumulative cost-benefit analysis, you indicated a desire to build on previous regulatory accounting riders and OMB�s two subsequent reports released in September of 1997 and February of this year. But before moving forward with this legislation � which marks a significant analytical expansion of the previous riders � it is first important to consider some of OMB�s conclusions. Problems with Existing Regulatory Accounting OMB makes a special effort in both reports to point out that rulemaking decisions are made on a case-by-case basis, as they must be, and that throwing all of the government�s diverse regulations, from environmental standards to economic controls, into the same pot has little practical utility for public policy. �[W]e still believe that the limitations of these estimates for use in making recommendations about reforming or eliminating regulatory programs are severe,� OMB states in its second report. �Aggregate estimates of the costs and benefits offer little guidance on how to improve the efficiency, effectiveness, or soundness of the existing body of regulations.� Further calling into question S. 59's applicability to policy-making is the inherent uncertainty involved in cumulative cost-benefit analysis. OMB discusses a litany of factors that, in its words, make it �difficult, if not impossible, to estimate the actual total costs and benefits of all existing Federal regulations with any degree of precision.� These include:

• The �apples and oranges� problem. The studies OMB bases its report on, and indeed OMB�s report itself, have simply added together a diverse set of individual studies that vary in quality, methodology, and type of regulatory costs examined. To produce its estimates for costs and benefits for regulation prior to 1988, OMB relied heavily on a 1991 study by Robert Hahn and John Hird. The Hahn-Hird study does not include benefit estimates for all regulations (e.g., consumer product safety was not counted), but still showed costs and benefits to be about the same. Even more interesting was that the Hahn-Hird data was not new; it was actually based on an earlier 1982 study. As a result, the Hahn-Hird study does not reflect the benefits of key environmental regulation that occurred under the Clean Air Act during the 1980s, such as the reduction of airborne lead and fine particles in the air. Taking this into account, OMB supplemented the Hahn-Hird work with two EPA studies � �Cost of a Clean Environment� (1990) and �The Benefits and Costs of the Clean Air Act, 1970 to 1990" (1997). EPA�s 1997 report was not included as part of the first report and, as a result, the second report contains substantially higher aggregate benefit estimates. �In addition to using different assumptions about baselines and time periods, the studies use different discount rates, different valuations for the same attribute, and different concepts of costs and approaches to dealing with uncertainty, to mention a few,� OMB writes. In the end, a regulatory accounting effort will always involve adding apples and oranges, with results more akin to rotten tomatoes, in which the final numbers, far from creating transparency, are virtually impenetrable.
• Dated studies and analysis. The older the study, the less reliable it is. That is because business learns to adapt to regulation and reduce costs over time through technological advancements, �learning by doing,� and other factors. The studies used by OMB were essentially static estimates that did not try to predict future adaptive effects. Moreover, because there are no studies comparable to Hahn-Hird that cover regulations after 1988, OMB relies on Regulatory Impact Analyses (RIAs) � which are conducted by agencies during major rulemakings � for rules since 1988. The RIAs used by OMB are especially unreliable because they were conducted before any adaptive effects could take hold (whereas the other studies were retrospective), and as a result are likely to overstate costs dramatically. 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.
• Setting a baseline. To estimate the impact of regulations on society and the economy, you must first determine how things would have been in the absence of regulation � in other words, set a baseline against which to measure costs. But because it is impossible to know what would have happened without regulation, this can only be an educated guess. This problem is accentuated the larger the regulatory changes. �If we use as a baseline a world with no regulation, one can reasonably argue that the benefits of regulation must clearly swamp any likely cost,� OMB writes.
• No accounting of equity. None of the analyses used by OMB�s two reports provide quantitative information on the distribution of benefits or costs by income category, geographic region, or any other equity-related factor.

In order to meet the requirements of the regulatory accounting report, OMB has, not surprisingly, found it necessary to put cumulative costs and benefits in terms of dollars and cents. And indeed, S. 59 puts a premium on monetization, asking OMB to show �net benefits or net costs.� In its last report, OMB demonstrates the method for showing �net benefits� through a benefits minus the cost calculation. Yet agencies often evaluate benefits using qualitative factors, such as the reduction in health or safety risks to children, whereas costs are more easily stated in monetary terms. This analytical discrepancy is only accentuated when you attempt to monetarily add up all federal regulation at once and can produce numbers that are greatly misleading. When seemingly qualitative factors are converted to monetized figures � as OMB has begun to do with agency RIAs to fulfill its regulatory accounting obligations � value judgements become hidden behind a mask of technical expertise. For instance, OMB's most recent report incorporated the estimated benefits of reducing lead in gasoline, including the prevention of IQ loss in children. Although it�s hard to imagine a parent who would regard their child�s drop in IQ as adequately captured by an estimated loss of future earning capacity, this is actually one of the many value judgements buried in OMB�s numbers. Problems with S. 59 Despite all of the uncertainty described above, and in the face of warnings from OMB, S. 59 would make cumulative cost-benefit analysis even more problematic. Specifically: 1. It seeks to dramatically expand analytical requirements contained in the previous appropriations riders. S. 59 � which has removed language from the previous appropriation rider requiring analysis only �to the extent feasible� � calls for OMB to estimate the annual costs and benefits of rules and paperwork (a) in the aggregate, (b) by agency, agency program, and program element, and (c) by major rule. In addition, OMB would have to assess the direct and indirect impacts of federal rules on federal, state, local and tribal governments, the private sector, small business, wages, and economic growth. The inclusion of these new subanalyses, all aimed at elevating cost considerations, make it even more likely that �net benefits� will be understated. Notably, the bill calls for no such specificity in evaluating benefits, although there are certainly subcategories here worth considering � including effects on vulnerable populations, such as children, the elderly and the disabled. But the biggest problem with these new requirements is that much of the information called for is not currently generated during agency rulemakings. When the first appropriations rider was passed, a colloquy between Sen. Stevens and Sen. Levin made clear that the intent was not to generate new data or studies, but rather to pull together existing information. �I expect a rule of reason will prevail: Where the agencies can produce detail that will be informative to the Congress and the public, they should do so,� Sen. Stevens said at the time. �Where it is extremely burdensome to provide such detail, broader estimates should suffice.� S. 59 represents a departure from this logic and takes an all-things-are-possible approach. For instance, under the Paperwork Reduction Act, agencies are not currently required to conduct cost-benefit analyses for paperwork (either on the whole or specifically for the sub-categories listed in S. 59); rather, the agency is to assess �practical utility� and burdens imposed. Nor do agencies currently conduct analysis by �program element,� meaning a cluster of related rules. And still another problem is that S. 59 applies to all regulations, including minor rules for which an RIA is not currently done and no data is available for OMB to apply. This might mean that agencies would need to spend resources and time on cost-benefit analysis, even for small, regularly renewed rules. Testifying against similar legislation in the House, former OMB Deputy Director Ed DeSeve recently explained, �... agencies may have to be called upon to compile detailed data that they do not now have, and undertake analyses that they do not now conduct, using scarce staff and contract resources, regardless of any practical analytic need as part of the rulemaking process.� 2. It requires OMB to issue guidelines on agency cost-benefit analysis and make recommendations on agency policy. The requirement that OMB issue such guidelines is puzzling since OMB only recently issued its �Best Practices� document, containing guidelines for cost-benefit, after extensive interagency discussion. Moreover, the last regulatory accounting rider, introduced by you Mr. Chairman, also contained the same requirement that OMB issue new guidelines for cost-benefit analysis; it makes little sense to require that OMB repeat this task only a short time later through S. 59. In addition, it is important to keep in mind that there will always be differences in the way cost-benefit analysis is conducted between federal agencies because of the many different functions they perform. But S. 59 seems to assume that there must be uniform approaches to the cost-benefit calculation, granting OMB, in consultation with the Council of Economic Advisors, the power to �standardize� across government the �most plausible measures of costs and benefits� and to review agency submissions �to ensure consistency with the guidelines.� This expansion of authority would put OMB in the position of prescribing value-laden analytical judgements to agencies that each face their own unique methodological obstacles in assessing costs and benefits. In many cases this would require agencies to actually conduct two separate assessments, one to meet the demands of an underlying statute, the other to meet the demands of OMB and its regulatory accounting report. S. 59 goes even further in expanding OMB power by requiring �recommendations to reform inefficient or ineffective regulatory programs or program elements.� OMB is a body that reviews agency analysis and coordinates regulatory plans. It should not be within its jurisdiction to set policy at other federal agencies, especially on the basis of something with such dubious reliability. And indeed, in both reports required by the appropriations riders, OMB expresses great reluctance in making recommendations based on its findings. 3. It requires OMB to subject its findings and guidelines to �peer review.� S. 59 requires OMB to subject its findings and guidelines to peer review provided by �a nationally recognized public policy research organization with expertise in regulatory analysis and regulatory accounting.� There are only a handful of groups who would qualify under this language, and virtually all are more concerned with the cost side of the regulatory equation. Given that the bill instructs that OMB �shall use the peer review comments� � not simply consider the comments � in preparing its report, this could allow a single, privileged organization to greatly bias results and achieve a disproportionate amount of influence over the future of agency cost-benefit analysis. The exclusive format of the peer review process actually undercuts the bill�s all-inclusive public notice and comment process. For the first two reports, which were not subject to peer review, the comment period meant that everyone enjoyed the same fair shot at influencing OMB�s final product. That would no longer be the case with the addition of the �peer review� section. 4. It would move in the direction of a regulatory budget and appears to be constructed as a political weapon. In light of all the uncertainty involved in cumulative cost-benefit analysis, it seems fair to question the motives of those who say regulatory accounting is about �right-to-know.� While some, no doubt, honestly believe this to be the case, the idea of regulatory accounting actually originated during the Reagan Administration as part of a proposal to create a regulatory budget, which later resurfaced again in the Contract with America. Under the Contract with America proposal, federal agencies would have to cap regulatory costs at a certain percentage of our GDP; if costs exceed that cap, agency rules would have to be eliminated and no new regulations could be issued. In fact, this proposal actually required cuts in regulatory costs by reducing the cap by a set percentage each year. But to institute such an approach or any type of regulatory budget, you must first have a system that aggregates regulatory expenditures on an ongoing basis, and S. 59 would put proponents of regulatory budgeting halfway to their final goal. In addition, many of the backers of regulatory accounting have also been vocal proponents of other various �reform� measures designed to stem regulatory costs. A regulatory accounting report showing very large costs and small benefits could be a useful tool in advancing this agenda. In this respect, however, OMB�s first two reports were major disappointments. (The first found $298 billion in benefits and $208 billion in costs for social � i.e., health and safety � and environmental regulation, and the second found $170 billion to $224 billion in annual costs and $258 billion to $3.55 trillion in annual benefits, expressed in ranges �to reflect the substantial uncertainty in the estimates.�) S. 59, with its slanted analytical requirements, could be seen as an attempt to forcibly bend the numbers in an ideological direction consistent with the proponents of broad regulatory �reform.� In the absence of higher cost estimates from OMB, another regulatory accounting study, by Thomas D. Hopkins, that yielded very high estimates of costs is often cited despite a methodology that was refuted by OMB. Among its many problems, the Hopkins study includes process costs that are not normally considered a part of the regulatory reform debate, such as the burden of filling out income tax forms or doing the necessary paperwork to obtain visas, passports, small business loans, and veterans benefits. This does two things, according to OMB: �It produces large numbers and it creates confusion.� Citing this one obviously skewed report over and over again, as some proponents of regulatory accounting have done, contradicts their stated desire to provide better information, and instead seems to be an attempt to mislead the public. Inevitably, this leads one to question whether S. 59 is really about the public�s �right-to-know,� and is in fact more about building a political weapon. In summary, by allowing crucial value judgments to be masked by monetized figures, we believe a report of this kind implies a sort of detached objectivity that simply doesn�t exist, and in doing so creates less transparency, not more, as proponents suggest. Moreover, the slanted analysis required by S. 59 appears to be intended as a political weapon to undermine critical health, safety, and environmental standards. Certainly such a regulatory accounting has no real utility for public policy, as OMB has pointed out. And yet, as constructed by this legislation, it could prove extremely burdensome for already cash-strapped federal agencies. Congressional Office of Regulatory Analysis S. 1675 of the 105th Congress � the Congressional Office of Regulatory Analysis Act � would have set up a congressional office to review agency rulemakings and conduct its own cost-benefit analysis for every major rule (and non-major rule upon the request of Members). It has been suggested that through the creation of CORA, Congress would be better informed on agency rules, and more likely to use the recently-enacted Congressional Review Act (CRA). As Sen. Shelby pointed out on Feb. 25, 1998, when introducing S. 1675, neither the Senate or the House has moved a resolution of disapproval through the expedited track provided for under the law and no rule has been struck down. Yet if you look at the recent case examples, there appears to be little confusion among Members. Is there anyone here on this Committee who in the last Congress didn�t develop an opinion on OSHA's methylene chloride rule or EPA's recent clean air standards? Most, if not all, know exactly how they feel. And if they don�t, there is a wealth of information already made available to Congress to help Members make prudent decisions. The CRA requires that agencies submit all proposed rules to the parliamentarian and leadership in each chamber. In addition, the General Accounting Office must prepare a report on each agency rule and submit it to the appropriate congressional committees in both the House and Senate. (In fact, this information can be viewed by anyone who has access to the world wide web � www.gao.gov) Thus, the intimate details of each agency rulemaking (e.g., the cost-benefit analysis, risk assessment, and small business panel recommendations) are right there at the finger tips of each Member and readily available to the relevant oversight committees. And if after reviewing all this information Congress still has questions, congressional leaders can hold hearings. There have been many hearings on the CRA and specific regulations such as OSHA�s methylene chloride rule and EPA's clean air standards. The true reason that these sort of rules � which have been vocally opposed by some Members � have not been considered under the disapproval process is political. There is a fear among those who might vote to strike down rules that they would be branded anti-environment or anti-worker as a result. More research will not address these political considerations, and it is not likely to lead to more resolutions of disapproval as Sen. Shelby hopes. But apart from whether a Congressional Office of Regulatory Analysis would accomplish its stated purpose, the proposal has many other problems. Specifically: 1. It would create a costly new government apparatus that would duplicate functions already performed by OIRA and the individual agencies. Under Executive Order 12866, OMB's Office of Information and Regulatory Affairs (OIRA) must review all major rules (rules with an annual economic impact of $100 million or more, or rules OMB so designates) and other nonmajor rules that OIRA believes warrant consideration. For 1998, this amounted to the review of 486 agency rules; the content of these reviews is readily available to Congress. CORA would duplicate all the work done by OIRA, including an annual report estimating the total cost of federal regulations on the U.S. economy. Although these responsibilities are time-consuming and expensive � OIRA operates on an annual budget of $5 million � previous CORA proposals have sought to go further than simply creating a second OIRA. CORA would also engage in activities currently handled by individual agencies, performing an additional "Regulatory Impact Analysis" for each major rule, and some tasks currently required of the Congressional Budget Office under the Unfunded Mandates Reform Act of 1995. In explaining the necessity for this duplication, the bill states that �in order for the legislative branch to fulfill its responsibilities ... it must have accurate and reliable information on which to base its decisions.� This is true, but it assumes that information from CORA would be more reliable than that coming from OIRA, the agencies, and GAO. Considering the nature of rulemaking, and all of its components, along side CORA�s 45-day review period, it�s hard to see how this could be the case. Predictably, this new and redundant regulatory review apparatus would cost taxpayers millions, carrying with it few or no benefits. Shelby�s version authorizes whatever appropriations are necessary to fulfill the office�s requirements, the sky�s the limit. To conduct a cost-benefit analysis for a major rule, it costs an average of $570,000, according to the Congressional Budget Office. It�s unclear how many Regulatory Impact Analyses CORA would conduct each year. But taking CBO�s estimate into consideration, if CORA were to do an RIA for each of the 75 economically significant rules in 1998, the office would have cost about $43 million for the year. If it were to do an RIA for all of the 486 major rules reviewed by OMB last year, the office would have cost about $277 million. Now we realize Congress is not going to appropriate that much, but certainly CORA would have to be given resources at least equal to CBO at $25 million. A House version of this bill (H.R. 1704) was reported out of the Judiciary Committee and the Government Reform and Oversight Committee during the 105th Congress. And precisely because CORA has the potential to be so outrageously expensive, it was amended to limit its annual appropriation to roughly the same level as OIRA�s. But considering that the scope of CORA�s activities would be far greater than OIRA�s, this surely would not be enough. We believe that regardless of funds, the information generated by CORA would be unreliable, for reasons explained below. But without proper funding, this undoubtedly would be the case. If Members are truly concerned about the quality of analysis coming out of the agencies, perhaps Congress should use the funding that some seem ready to apply to CORA and appropriate it to the agencies. Just within the last several years the President has signed into law the Small Business Regulatory Enforcement Fairness Act, the Unfunded Mandates Reform Act, and amendments to the Paperwork Reduction Act � all of which require agencies to perform rigorous new regulatory cost assessments. The obligations under these laws would be more easily fulfilled with greater resources, and the results would likely be better as well. 2. It runs counter to current efforts to streamline the government. Members of this Congress have often raised objections to agencies that perform apparently redundant functions. The administration has responded to such criticism through E.O. 12866 and the Vice President�s "Reinventing Government" initiatives, both of which have attempted to increase government efficiency. CORA, however, would run counter to these efforts by duplicating functions at OIRA and the individual agencies. Such redundancy calls into question whether CORA could stand up to the same sort of rigorous cost-benefit analysis so valued by many Members of this Committee. 3. It contains the unreasonable expectation that CORA conduct its own Regulatory Impact Analysis for all major rules. OIRA does not do this, and for good reason. Cost-benefit analyses are extremely time-consuming, require significant expertise, and are done within the context of each rulemaking. Yet Sen. Shelby�s bill seems to imply that CORA would do the various types of analyses within a 45-day period before reporting to the appropriate committee. Even if CORA gets a head start on its requirements � say when the agency publishes a Notice of Proposed Rulemaking�� such analysis would still be unworkable. Without being a part of that rulemaking (e.g., without being involved in the agency�s public comment period, SBREFA panels, etc.), it would be impossible for CORA to make a credible, independent estimate at both cost and benefits. CORA could essentially copy agency findings, but if that�s the case, the bill does not meet its stated purpose. More likely, the limited time-frame would force CORA to rely very heavily on estimates from regulated interests. Under this scenario, however, the proponents would lose the independence they say they want � which would be especially troubling if Congress intends to use information generated by CORA as a basis for rejecting agency rules. Adding to the concern here is CORA�s lack of pubic accountability. When agencies choose a regulatory option that is "arbitrary or capricious," they can be sued. But the public would have no recourse for sloppy work produced by CORA. During the 1994 debate over unfunded mandates, Robert Reischauer, director of CBO at the time, was very skeptical of the legislative branch�s ability to conduct these sorts of highly technical and time-consuming cost estimates, calling it "impossible in any practical sense." Congress heeded Reischauer's warning by narrowing the scope of analysis that CBO is to do under the Unfunded Mandates Reform Act. Yet CORA would move Congress directly into areas that Reischauer warned would be dangerous. 4. It would place CORA in the position of describing �lower cost� regulatory alternatives, raising Constitutional concerns over separation of powers. Sen. Shelby�s bill requires CORA not only to conduct detailed cost-benefit analyses, but also determinations of "potential net benefits" and descriptions of alternative regulatory approaches that could "achieve the same regulatory goal at a lower cost" and cost-benefit analyses of these approaches. These types of assessments are not required of agencies at this time and, most significant, put public protections secondary to finding "lower cost" regulatory approaches. Moreover, it moves CORA in the direction of subordinating the powers granted to the executive branch to execute the laws of the land. Congress has every right to establish laws and revise them, but CORA would place the legislative branch in the role of describing regulatory alternatives for the way the executive branch is to execute. In addition, under Sen. Shelby�s proposal, CORA could utilize executive branch facilities and personnel, upon approval from OMB and the agency head, without reimbursement to carry out work it needs done. This opens a process in which CORA could have a direct impact on agency activities and decisions. 5. It contains no language requiring CORA to operate in the sunshine. During the 1980s, OMB was permitted to operate in secret with little public accountability. Rules would go to OMB, changes could be made, and no one would know exactly why. Similarly at CORA, significant decisions on agency rules affecting everything from small business to the environment to children�s health could be made without ever providing a proper explanation to the public. This is especially significant if Congress is going to use CORA findings as a basis to reject agency rules. As the Freedom of Information Act has been advanced, OMB has opened up slightly (though problems still remain since it is not subject to the same statutory requirements as federal agencies). But Sen. Shelby�s bill doesn�t touch the subject of whether or not FOIA would apply to CORA, nor does it spell out any other mechanisms to bring CORA into the sunshine to ensure greater public accountability. More importantly, CORA raises serious concerns involving the Administrative Procedure Act. Under the APA, agencies are required to take a number of steps (e.g., public notice and comment) to ensure openness. Agencies can also be sued if the agency decision is "arbitrary or capricious," providing important checks and balances. CORA would have to conduct cost-benefit analyses just like federal agencies, but unlike federal agencies would not be bound to the APA. This means important decisions at CORA that could lead to the defeat of health, safety, or environmental protections might be made without any input from the public. In the absence of public accountability, it is possible that CORA could be used as a tool to advance a political agenda rather than a source of objective analysis on agency rules. 6. It would politicize the evaluation of agency rules. It�s not hard to imagine a body like CORA, which would function as an arm of Congress, being influenced by the expectations of individual lawmakers looking to push an ideological agenda. Indeed, under Sen. Shelby�s bill, the House and Senate leadership control the appointment of CORA�s director, which is especially troubling if data from CORA is to be used as the basis for rejecting agency rules. 7. It contains a regulatory accounting provision that could become a congressional regulatory budget. There are many problems with the requirement in Sen. Shelby�s bill that CORA do an annual report on the "total cost of Federal regulations" on the U.S. economy (many of which are discussed in our comments above on S. 59). First, this would require significant work. CORA would not be able to review every rule generated by the executive branch, and therefore would need to establish a process for determining costs for every rule. Currently, OMB does not keep such information either. Second, the regulatory accounting provision does not define what is meant by its requirement to estimate total costs. Does this include indirect costs? In the past, business has used such vague language to create opportunities for showing significant cost (e.g., lost business opportunity) relative to benefit, inflating burdens and justifying a decision not to regulate. Third, there have been many recent attempts to quantify the cumulative costs of federal regulations by independent organizations and other researchers, yet in virtually every case, these studies vary by hundreds of millions of dollars � influenced by the various ideological underpinnings of the researchers. Likewise, it is easy to see how CORA�s study could be influenced by Members of Congress looking to push an ideological agenda. Fourth, the requirement does not instruct CORA to provide an annual estimate of the total benefit of federal regulations, including the economic benefit of regulation. This would create a one-sided figure that could be greatly misused. Finally, as an annual requirement, the regulatory accounting provision raises serious concerns that it could become a backdoor approach to creating a regulatory budget � something strongly opposed by the public interest community but called for in the Contract with America. 8. It is not necessary. Under the Congressional Review Act, GAO must provide an analysis of each agency rule to the appropriate congressional committees. Furthermore, information on OIRA�s regulatory review and the agency�s rulemaking is also delivered to Congress. This gives lawmakers all the tools they need to exercise necessary executive branch oversight. Supporters of CORA have failed to identify a compelling reason to transfer GAO's functions to a new congressional agency. Although CORA purports to enhance congressional knowledge of agency rulemaking, Members have exhibited little confusion in this regard. For instance, most Members from the last Congress were able to form well-developed opinions on OSHA�s rule on methylene chloride and EPA�s new clean air standards without an expensive apparatus like CORA. In summary, the fact that Congress has not used the CRA is a function of political will, not a lack of information, and therefore CORA would not lead to more resolutions of disapproval as its proponents hope. But it would create a costly new government apparatus to perform a myriad of functions already performed by other government entities. This is not a wise use of resources and contradicts recent efforts to streamline government. In addition, an array of problematic side-effects would result from CORA�s creation, such as its license to operate in secret and questions regarding the separation of powers between the executive and legislative branches of government. Furthermore, there are questions about CORA's mandated requirements and why they exceed those imposed on agencies.