The Regulatory Tax

Over 60 agencies have a hand in federal regulatory policy, ranging from the Environmental Protection Agency (EPA) to the Securities and Exchange Commission (SEC). Together, they enforce over 144,000 pages of rules, with purposes and impacts as varied as the agencies themselves. Some rules are meant to protect health and safety, some to protect (or suppress) economic competition, and some to protect the environment.

Certainly, not all of these regulations are unjustified--many, in fact, are quite beneficial. For instance, most would agree on the need for security rules to protect citizens against terrorism, although the extent and scope of those rules may be subject to debate. Moreover, regulations are not necessarily inconsistent with free-market principles. Some--such as anti-fraud rules and, arguably, the new do-not-call rules for telemarketers--actually reinforce individual and property rights.2

Nevertheless, all rules come at a cost: a "regulatory tax" imposed on all Americans. Of course, Americans do not file regulatory tax forms on April 15, and there is no bottom line indicating how much they pay for these regulations. Yet hidden regulatory costs are staggering by almost any measure. According to the Office of Information and Regulatory Affairs (OIRA), the White House office responsible for reviewing and tracking federal rules, regulations adopted in the past 10 years cost Americans $34 billion to $38 billion annually.3 All federal regulations, OIRA states, could be costing Americans 10 times this amount: some $380 billion.

However, these numbers are low compared to estimates prepared by economists Mark Crain and Thomas Hopkins for the Small Business Administration.4 In 2000, Crain and Hopkins concluded that regulations cost Americans $843 billion (over $8,000 per household).5 This is almost half of the amount collected in federal taxes and close to the $1 trillion paid in personal income taxes that year.6 Put another way, the total is almost a tenth of America's gross domestic product and more than half of the manufacturing sector's output.7

Even these numbers may underestimate the negative effects of regulation. For instance, the Crain-Hopkins study does not include indirect costs. A regulation that increases energy costs would also affect other industries that require energy to produce their products.

Perhaps more important, the magnitudes of some burdens are, by their natures, unknowable. For many economic regulations, the major cost may not be any direct burden placed on consumers or businesses, but constraints on innovation. Assessing such losses is impossible because inventions that never existed cannot be measured. In today's 21st century economy, these unmeasurable costs are perhaps more harmful than the direct, measurable burdens.

In any case, regulatory burdens cause substantial economic harm by reducing economic growth, slowing job growth, and reducing Americans' income. The actual effects vary tremendously depending on the type of regulation,8 but the effect is clear. A recent World Bank study of regulation around the world underlined the connection between economic growth and regulation, finding that "[h]eavier regulation is generally associated with...more unemployed people, corruption, less productivity and investment."9 Interesting, the authors did not find a correlation with better quality of private or public goods.

However, the costs are not just economic. Regulations can--and often do--reduce Americans' health and safety as well. Delays in new drug approvals by the Food and Drug Administration have led to thousands of unnecessary deaths.10 By encouraging the purchase of smaller cars, automobile fuel efficiency standards have contributed to thousands of deaths in car accidents.11 Rules banning health claims on wine bottles have denied Americans information about the beneficial effects of wine on heart health.12

Nixon to Clinton: Record of Past Administrations

Overregulation is nothing new--regulation has been growing in size and scope for decades--and neither are efforts to control regulatory growth. Since the 1970s, Presidents have used a variety of reform measures to control regulatory growth, and with decidedly mixed success.

The first systematic effort to control regulatory growth dates back to 1971, when President Richard Nixon required regulatory agencies to perform rudimentary analyses of each proposed new regulation. President Gerald Ford expanded on this process, making control of regulatory growth part of his war on inflation.13 Ford also set up a Cabinet-level group to focus on other initiatives to control the cost of regulation.

President Jimmy Carter continued regulatory reform efforts by maintaining the practice of conducting economic analyses of proposed regulations and setting up a Cabinet-level Regulatory Analysis Review Group to review proposed new rules. Legislatively, the Carter Administration spurred more regulatory reform than any other Administration before or since. With active support from the Carter Administration, Congress enacted bills deregulating the airline, trucking, and railroad industries, saving American consumers billions of dollars, as well as the Paperwork Reduction Act, which created OIRA. Yet, during the Carter Administration, there was also a substantial increase in the number of new costly regulations imposed by agencies.

Reducing regulatory burdens was one of four "pillars" of President Ronald Reagan's 1981 economic recovery agenda.14 Upon taking office, Reagan established the Task Force on Regulatory Relief, chaired by Vice President Bush, to oversee regulatory reform efforts.15 In addition, he issued an executive order directing that no regulation be promulgated unless the potential benefits outweighed the potential costs and placing the newly created OIRA in charge of reviewing new regulations.16

President George H. W. Bush similarly accorded high priority to regulatory policy, establishing the Council on Competitiveness, a new Cabinet-level effort headed by Vice President Dan Quayle, to spearhead regulatory efforts. The Administration's most focused reform effort was a 90-day freeze in 1992 on new regulations during which agencies were directed to review existing rules and eliminate those found unnecessary. The freeze and review was later extended another 120 days.

However, these deregulatory efforts were overshadowed by new regulations under the first Bush Administration. These included several huge new regulatory programs, including the 1990 amendments to the Clean Air Act and the Americans with Disabilities Act, which Congress passed with President Bush's support.17

Under President Bill Clinton, the basic regulatory review structures were kept in place, but limiting regulatory burdens was--for the first time in two decades--not made a priority. From 1993 to 1999, OIRA returned an average of only two rules per year to agencies for reworking, in contrast with the average of over 31 per year returned in the previous decade. No proposed rules submitted to OIRA were turned down in the final three years of the Clinton Administration.18

The Bush Record

The present Bush Administration's record on regulatory reform is mixed. To the Administration's credit, OIRA has been reinvigorated and is more active and more effective than ever before.19 During the first year under Bush OIRA Administrator John Graham, the office returned more proposed rules to agencies for reconsideration than during the entire Clinton Administration. It has also established strict guidelines for agencies' cost-benefit analyses of their rules and most recently proposed guidelines for peer review of those analyses. The President also issued an executive order strengthening the role of the Small Business Administration in regulatory review.

At the same time, the Bush Administration has not focused much public attention on reduction of regulatory burdens. It has been the topic of few presidential addresses and rarely makes it onto "key priority" lists. The word "deregulation" has apparently been excluded from the presidential lexicon. An on-line search of presidential statements shows no use of the word by President Bush since 2001.20 Bush did call for reducing regulatory burdens on small businesses in his 2004 State of the Union speech and in his acceptance speech at the 2004 Republican Convention, but these were only brief references.21

Of course, an absence of talk does not necessarily indicate a lack of interest in reform. Many consider "deregulation" an inaccurate term for reform, and the word does not have much meaning to the average person.22 Furthermore, as reform efforts have matured, the debates over regulatory policy have become less cohesive. Instead of policy debates over "regulation" in a generic sense, debates tend to take place over specific issues that regulation affects: e.g., telecommunications, environmental policy, and corporate governance.

On many of these specific issues, the Bush Administration has successfully blocked new regulation. Most notably, Bush announced early in his term that he would not support the Kyoto Protocol on climate change, an international agreement that would have required substantial new controls, and he declined to propose new ergonomic rules. In some other areas, rules have been made less burdensome. For instance, the EPA rewrote emissions rules that had discouraged power plants from upgrading their facilities. A few rules have been eliminated almost entirely, such as the Department of Transportation's rules on the operations of airline computer reservation systems.

However, in other cases, the Bush Administration has missed opportunities for regulatory reforms. For instance, in 2002 and 2003, the Administration stayed on the sidelines as the Federal Communications Commission (FCC) was reviewing regulations on telephone companies. A strong pro-deregulatory stance could have changed the balance at the commission.23 Instead, the FCC adopted rules that were more extensive than many, including the FCC chairman, believed were necessary.24

In 2002, OIRA requested public comments on potential regulatory changes and received 1,700 suggestions, resulting in 161 discrete recommendations for change at executive branch agencies. In fall 2003--18 months after the process started--OIRA announced that the relevant regulatory agencies had agreed to look into 45 of the 161 recommendations.25 Little has been done since.

Measuring Regulatory Trends

The ultimate test of an Administration's performance is results: How has the burden of regulation changed? How does that change compare to changes under previous Administrations?

However, the answer is not easy to determine. There is simply no explicit, complete, and accurate way to track changes in regulatory costs from year to year.26 However, a number of measures can illuminate the picture, albeit imperfectly. Together, they may provide a fair picture of what is happening in the regulatory world.27 Among these are:

Number of Federal Register Pages. Before any new federal rule can be proposed or finalized, the agency involved must publish it in the Federal Register. Over the years, the size of this daily publication has often been used as a rough measure of federal regulatory activity.

In 2003, the total number of pages in the Federal Register topped 75,798. While less than the astounding 80,332 in 2002, the number is still high compared to other recent Administrations.28 During the Clinton Administration, page totals were generally less than 70,000. During the Reagan and first Bush Administrations, totals were even lower, typically in the 40,000-50,000 range.

However, Federal Register page counts are an extremely crude measure of regulatory activity. For one thing, they are easily manipulated. On occasion, font sizes have even been changed to keep the page count low.

Moreover, the Federal Register contains more than regulations. In addition to the texts of rules, it includes discussions of rules, determinations under rules, requests for public information, proposed rules, and even consent decrees in court cases. In fact, according to OIRA, the increase in Federal Register pages from 2001 to 2002 was due entirely to publication of the Microsoft antitrust settlement.29

Finally, since agencies must publish all rule changes in the Federal Register, it includes actions to eliminate or reduce regulatory burdens, as well as actions to increase them. Thus, while an increase in size may generally signal an increase in agency activity, it does not indicate whether that activity lessens or increases burdens.

Length of the Code of Federal Regulations. Another measure of regulation is the Code of Federal Regulations (CFR). Although not cited as frequently as the Federal Register, the CFR provides a more direct gauge of regulation. Unlike the Federal Register--which is in effect a posting board for all sorts of agency actions--the CFR is the regulatory equivalent of a statute book, including only the text of existing regulations.

In number of pages, the CFR makes the Federal Register look rather small, with the 2003 edition weighing in at 144,177 pages, divided into 50 titles or subject areas.30 While a little less than the record 145,099 pages in 2002, the 2003 total is about 2 percent more than when Bush took office in 2001. This rate of increase is actually greater than the rate during the Clinton presidency, when the page count increased about 7.5 percent over eight years, but about half the size of the increases during the first Bush Administration and--perhaps surprisingly--the Reagan Administration. The CFR exploded during the Carter Administration, increasing by 26 percent over four years.

Within the titles, there are great variations. While the CFR increased by a total of 41 percent from 1980 to 2003, the environment title grew by 262 percent, and the wildlife and fisheries rules increased by 375 percent.31

Yet the CFR also has significant drawbacks as a measure of the regulatory burden. Foremost among these is that the number of pages in a regulation does not necessarily indicate a heavier burden. For instance, a 500-page regulation that outlines numerous exceptions and conditions could actually impose a much lesser burden than a one-line prohibition of a certain activity.

Number of Major Rulemaking Proceedings. Another gauge of regulatory trends is the number of federal rulemaking proceedings. The total number of final rules promulgated each year is high--4,148 were published in the Federal Register in 2003.32 Many of these rules, however, are quite limited in impact. The number of major rules (those with an annual impact of $100 million or more) promulgated is much smaller. The U.S. General Accounting Office (GAO) reported 49 such rules in 2003.33

However, many of the major rules are not regulatory in the commonly understood sense of the word because they do not directly impose limits or mandates on private activities. Many rules are budgetary in nature, such as those that establish rules and conditions for federal spending programs.34 Others are annual determinations, such as the number of birds that can be hunted in certain areas. Eliminating these "non-regulatory" rules leaves a much smaller number of what might be called "major regulatory" rulemakings. There were 21 such rulemakings reported by the GAO in 2003.

Significantly, not all of the major regulatory rulemakings increase the burdens on the private sector. Federal agencies must publish a rule whenever they make any changes in their regulations, even if those changes decrease burdens. Even completely repealing a regulation would require publication, which raises the questions of how many of the new rules promulgated by federal agencies have increased private-sector burdens and how many have decreased those burdens.

Based on the author's review of each major rule reported by the GAO from 1997 through 2003,35 a strong majority of each year's rules have been pro-regulatory, although the portion has been less since Bush took office.

Of the 153 rules reviewed, 37 (about 25 percent) decreased regulatory burdens. Of the 121 regulations from the Clinton Administration, 27 (22 percent) were deregulatory.36 The Bush Administration was slightly more deregulatory, with 10 of 32 rules (31 percent) decreasing regulatory burdens.

Over one-third of all categorized rules (57 of the 153 rules) were promulgated by independent agencies and thus were outside the OMB regulatory review process. The overwhelming majority of these were attributable to two agencies: the FCC and the SEC.

Interestingly, far larger portions (about 50 percent) of independent agency regulations were deregulatory. In fact, the FCC had more deregulatory actions than regulatory actions, a distinction shared only with the Nuclear Regulatory Commission, another independent agency.37

The reason for the higher percentage of deregulatory actions at these independent agencies is unclear. One factor may be that both the FCC and SEC administer 1930s-era economic regulations that have been undergoing significant change.38 The FCC's deregulatory record, in fact, was due largely to proceedings liberalizing radio spectrum rules.39 Moreover, it should be noted that regardless of the deregulatory actions of the independent agencies, they are still a major source of new regulation, accounting for about 25 percent of all rules that increased burdens.

When independent agency rules are excluded, the difference between the Clinton and Bush Administrations' regulatory records becomes much more stark. Less than 7 percent of rules by executive branch agencies during the Clinton years were deregulatory. By contrast, over 28 percent of such rules during the Bush years have reduced burdens. This is a significant difference. Yet increases in regulations have still outnumbered decreases by more than two to one.

The sharpest contrast between the two Administrations is in the number of pro-regulatory actions, with the Bush Administration adopting just over seven major rules per year that increased burdens versus over 20 per year under Clinton.40 This difference is particularly noticeable at a few individual agencies. For instance, the Bush EPA has increased burdens twice, as opposed to the Clinton EPA's 26 times. The Department of Labor increased burdens nine times under the Clinton Administration compared to none under Bush. According to these numbers, President Bush has not reversed the growth of federal regulation, but he has slowed it substantially.

Of course, counting regulatory and deregulatory rulemaking does not show the full regulatory picture. Many key questions involve not whether to regulate or to deregulate, but rather how and how much. For instance, the FCC's 2002 decision modifying its rules on telephone competition was virtually ordered by a court. Relaxing the existing rules was a given. The real battle was on how much to reform. The final, controversial decision was technically deregulatory but was largely a victory for the pro-regulatory side because it kept key provisions in place.

Second, counting only the number of actions hides the actual impact of each decision: A rule costing $100 billion is weighed the same as one costing $1 billion. As a result, many important rulemakings--such as the virtual repeal in 2003 of regulations on airline computer reservation systems--are not reflected in these figures.

Cost Estimates. Ideally, the best way to measure regulatory burdens would be to look at the actual costs of each new rule. Over the past few years, the ability of OIRA and agencies to do this has progressed substantially. As outlined above, the federal government has engaged in economic analyses of new regulations for some 30 years, starting with basic "inflation" impact statements and evolving into the fairly complex cost-benefit analyses used today for all major new regulations. Moreover, OMB is required by statute to report information on regulatory costs to the Congress, both in the aggregate and by major rule.

According to these cost measurements, the annual cost of new rules in the Bush Administration has been just under $1.5 billion.41 By contrast, the annual costs totaled $5.7 billion under President Clinton and $8.5 billion under the first Bush Administration.

This is further evidence that the Bush Administration has been keeping regulatory costs under control--or at least avoiding excesses. Much of the difference seems to be fewer extremely costly rules. Only one rule costing over $1 billion was promulgated during the first two years of the current Bush Administration. By contrast, there were seven such rules in 1992 during the first Bush Administration.

However, these statistics should be viewed with caution for three reasons:

1. Despite efforts to expand the use of cost analysis, many major rules are still adopted without a quantification of costs. In fiscal 2003, at least 23 major rules were promulgated by federal agencies, but costs were quantified for only 15.
2. The numbers are based on analyses performed by regulatory agencies themselves as part of their justification for their rules. Although the analyses were approved by OIRA as part of the review process, they do not present a truly independent assessment of regulatory costs. Moreover, although OIRA recently took steps to standardize the methodologies and assumptions used in these studies, most are far from uniform, making it difficult to aggregate the numbers meaningfully, much less judge their quality.42
3. The cost-benefit analyses do not gauge decreases in regulatory burdens. Under such analyses, there is no such thing as a negative cost. Thus, at best, if a rule is eliminated, the cost is scored as zero. It therefore is not helpful in determining whether regulatory costs are increasing or decreasing on balance.

What Should Be Done

While no existing regulatory yardstick is perfect, as a whole, such yardsticks shed light on general regulatory trends. Regulatory growth seems to have slowed compared to growth during the Clinton Administration and the first Bush Administration. However, the burden of regulation on Americans is still increasing.

There is no magic bullet that will reverse this trend. Regulatory policy is a multiheaded creature, with hundreds of separate issues. The debates over telecommunications competition rules may have little to do with the debates over air pollutants and drug approvals. Nevertheless, the Administration and Congress can take a number of steps to ensure that the full costs and consequences of regulation are weighed as rules are considered. (To avoid a net increase in federal expenditures, any additional spending required for these reforms should be reallocated from the approximately $30 billion that is now spent annually by federal regulatory agencies.)

Specifically, the Administration and Congress should:

• Continue to strengthen the Office of Information and Regulatory Affairs. OIRA has been reinvigorated during the current Bush Administration, playing an active role in consideration of new rules and ensuring that their full costs and benefits are considered before they are promulgated. However, OIRA is still badly outgunned in regulatory battles, with over 4,300 regulatory agency staffers for every OIRA staffer. OIRA should be provided with additional resources to do its job better.
• Establish a congressional Office of Regulatory Analysis. A congressional office charged with providing Congress with information on the cost and impact of regulation--and any alternatives--would provide another independent source of regulatory analysis. This new office could be modeled on the Congressional Budget Office, which provides Congress with information on spending programs and acts as both a complement to and a check on the Office of Management and Budget.43
• Establish a regulatory review office in each regulatory agency. Consideration of the costs of regulation should not begin when a proposal leaves an agency, but should take place within an agency as well. However, to be effective, this review should be from outside the specific office or bureau developing the policy. Therefore, each agency should have its own regulatory review office that is structurally separate from the units originating the rules and that examines all important agency rules before they are endorsed by the agency.
• Designate "regulatory reform czars" at each agency to identify unneeded regulations. Often, the best way to ensure that an issue is considered is to make a specific individual responsible for it. In 1992, as part of the first Bush Administration's 90-day regulatory review initiative, each agency was required to designate an officer, informally known as a "regulatory czar," to identify and eliminate unnecessary agency regulations. No new staff positions were created because the individuals typically were the general counsels or policy directors of the agencies involved. (In the future, such officers could be heads of agency regulatory review offices.) These officers were asked to spearhead efforts to reduce regulation at their agencies, meeting regularly with the Vice President to report on progress. Certainly, not every one produced a success story, but some did become zealous advocates of reform inside their agencies.
• Require independent agencies to submit cost-benefit analyses to OIRA. Independent agencies--such as the Federal Communications Commission and the Securities and Exchange Commission--produce a substantial share of the major rules finalized each year. The overall impact of these agencies is even greater because they cover some of the economy's most dynamic and vital sectors. Yet their rules are not subject to OIRA review before they are promulgated, and only rarely are their costs and benefits formally analyzed. This problem could be resolved by subjecting independent agency rules to the OIRA review process. If that cannot be done, they should at least be required to prepare cost-benefit analyses of all planned significant rules and to forward the analyses to OIRA for non-binding review.
• Emphasize that reducing the regulatory burden is a key Administration priority. None of the above steps will turn the tide of regulation unless President Bush states clearly that reducing regulatory burdens is a key priority in his agenda. He should emphasize, especially to his own agency heads, that reducing the regulatory burden is a key goal, not just another bit of Washington rhetoric that can be disregarded when politically convenient. He must set the underlying tone for the many regulatory debates within his Administration, making clear that actions to reduce regulation will be viewed favorably. He has begun to do this with strong statements on overregulation in several recent speeches, and he should continue this. Otherwise, reform efforts will surely fail in the face of bureaucratic and political opposition.

Conclusion

Regulation is a hidden tax on Americans, imposing a burden almost as heavy as income taxes. While President Bush has done better than many of his predecessors in limiting new regulations, less has been done to eliminate existing, unneeded rules. The Administration should act to ensure that the growth of the regulatory burden on Americans is not just slowed, but reversed.

James L. Gattuso is Research Fellow in Regulatory Policy in the Thomas A. Roe Institute for Economic Policy Studies at The Heritage Foundation. Margaret Hamlin of the Heritage Center for Data Analysis assisted in preparing this report.