This is historical material, "frozen in time."
The web site is no longer updated and links to external web sites and some internal pages will not work.
Report to Congress on the Costs and Benefits of Federal Regulations
Estimates of the Total Annual Costs and Benefits of Federal Regulatory Programs
This chapter discusses the total annual costs and benefits of existing Federal regulatory
programs called for by Section 645(a)(1). Before doing so, however, it is important to place the
subject in perspective.
First, when we speak of the costs or benefits of "regulations," we are in, reality, speaking
of the costs and benefits of legislation as well as regulation, for it is usually impossible, and often
not productive, to try to allocate costs and benefits between the authorizing statute and its
implementing or interpreting regulations.
Second, we need to keep in mind the discussion in Chapter I on best practices for
estimating costs and benefits and the limitations on valid and reliable quantitative measures.
Third, it is important to ask: What public policy purposes do aggregate estimates serve if
the ultimate goal is to develop the information necessary to make decisions about specific
regulatory programs or program elements? And, in particular: In what ways can these estimates
help support the recommendations to reform the regulatory system required of the Director by
Section 645 (a)(4)? Clearly, knowing the costs and benefits of individual proposals for
regulatory actions and their alternatives, including the alternative of no action, enables policy
officials to make decisions that improve society's well being. But for reasons discussed below,
knowing the total costs and total benefits of all of the many and diverse regulations that the
Federal government has issued provides little specific guidance for decisions on reforming
For example, four possible outcomes can result from totaling up the costs and benefits of
all existing Federal regulations:
High costs and high benefits.
High costs and low benefits.
Low costs and high benefits.
Low costs and low benefits.
Given the intensity of the debate over regulatory reform, categories (3) and (4) are not
likely outcomes of careful and fair accounting. A priori, it is not clear which of the remaining
two categories is most likely. But does it matter? In each case, the policy guidance would be the
same. Real economic improvement comes from expanding those significant regulatory programs
that provide benefits that are greater than costs and contracting those programs that provide
benefits that are less than costs. The substance is in the details, not in the total.
The implication of this discussion is that an excessive amount of resources should not be
devoted to estimating the total costs and benefits of all Federal regulations. To the extent that the
costs and benefits of specific regulatory programs can easily be combined, some indication of the
importance of regulatory reform can be inferred by the magnitude of these estimates, but
knowing the exact amounts of total costs and benefits, even if that were possible, adds little of
This proposition is important because it is extremely difficult, if not impossible, to
estimate the actual total costs and benefits of all existing Federal regulations with any degree of
precision. There are at least two types of intractable problems that make this so.
The Baseline Problem
In order to estimate the impact of regulations on society and the economy, one has to
determine the counterfactual -- that is, how things would have been if the regulation had not been
issued. In other words, what is the baseline against which costs and benefits should be
measured? With respect to estimating total costs and benefits of all Federal regulations, the
baseline problem has several dimensions.
First, it is impossible to determine the true counterfactual, since it never happened. What
would have happened in the absence of regulation can only be an educated guess. Furthermore,
the greater the hypothesized difference between reality and the counterfactual, the more
problematic the exercise. For example, some estimates of the total cost of regulation include the
cost of compliance with our tax system. But to twist a phrase, one can no more easily imagine a
world without taxes than one can imagine a world without death. It is also difficult to imagine a
world without health, safety, and environmental regulation. Could a civil society even exist
without regulation? In other words, what do we use as the baseline for a world without any
Second, even disregarding the problem of modeling large changes, there are significant
difficulties in determining the counterfactual for individual regulations that one could begin to
aggregate. One can survey firms and other regulated entities on their expected compliance costs
either ex ante, before the regulation is implemented, or ex post, after the regulation has gone into
effect. For both types of studies, the problem of potential bias must be kept in mind. It is often
alleged that strategic behavior may color both regulators' and the regulated's estimates of the
cost of regulation (Hahn and Hird 1991, Hopkins 1991, and Hahn 1996). Agencies are generally
advocates of their programs and businesses generally are not in favor of regulation. In the
ordinary course, therefore, the best studies are ex post studies done by individuals who do not
have vested interests, but do have reputations as objective analysts to uphold.
Often only ex ante cost estimates are available, but even if firms' or agencies' estimates
are unbiased at the time, technological change or "learning-by-doing" may result in those
estimates overstating compliance costs (Hahn and Hird 1991 and Hahn 1996). In fact, there is
much evidence that competition among regulated firms often reduces expected compliance costs
once real time and effort is directed at the problem (Office of Technology Assessment 1995).
While ex post studies are likely to be more accurate than ex ante studies because firms
should by then have had experience with actual regulatory compliance costs, ex post cost
estimates have their own problems. Properly done they are likely to be resource and time
intensive. Firms do not usually keep their cost accounting estimates according to what
regulations are driving them. Thus, when surveyed, firms have to reconstruct causality. A recent
General Accounting Office (GAO) report details the difficulties the GAO had in trying to
determine the total cost of Federal regulation by surveying a sample of firms. The firms reported
great difficulty in estimating their own costs of compliance because they could not easily
separate Federal from State and local regulation and because they did not keep records on
incremental costs of regulation (See GAO 1996, pp. 49-51). Some studies have attempted to
address this problem reasonably successfully by comparing the results of different degrees of
regulation in different localities or time periods.
Moreover, virtually all of the studies of the costs of regulation produced to date are
measuring the expenditures of firms required (ex ante or ex post) by regulation, whereas the cost
to society of regulation should be measured by the change in consumer and producer surplus
associated with the regulation and with any price and/or income changes that may result
(Cropper and Oates 1992). At one extreme, ignoring the consumer surplus loss produced by a
ban understates costs to society because although no compliance expenditures are required,
consumers can no longer buy the product. At the other extreme, calculating compliance
expenditures based on pre-regulation output overstates costs because if the firm raises prices to
cover compliance costs, consumers will shift to other products, which reduces their welfare
losses (Cropper and Oats 1992, p. 722).
A third problem relates to the economy and the appropriateness of the baseline for the
purpose for which it is expected to be used. If the objective is to reduce the burden of existing
regulation, even ex post evaluation surveys may be inadequate for they would reflect the cost of
gearing up to comply, not the cost saving of no longer having to comply with a given regulatory
program. While the former is relevant for deciding whether to regulate, the latter would be the
relevant concept if one is considering reducing regulation. There is also the dynamic nature of
the economy, whereby technological advances over time are likely to reduce the start-up cost of
compliance the firm originally faced. In addition, sunk costs, such as specialized capital costs
and the cost of changing procedures already in place, make the cost savings from eliminating
regulation less than the cost of complying with those regulations. Very few studies exist,
especially for health, safety and environmental regulation, that attempt to determine the cost
savings that would result from reducing or eliminating existing regulation.
It is important to note that this dynamic nature of the economy may affect the estimation
of benefits as well as costs. Technological improvements could reduce predicted benefits. For
example, medical progress can reduce the future benefits estimated for health, safety and
environmental regulations, just as productivity improvements in manufacturing reduces the costs
of compliance of some regulations. New drugs or medical procedures can reduce the benefits of
regulations aimed at reducing exposure to certain harmful agents such as an infectious disease or
even sunlight. Regulations aimed at increasing the energy efficiency of consumer products or
buildings may see their expected benefits reduced by new technology that reduces the cost of
producing energy. Furthermore, productivity improvements lead directly to higher incomes,
which lead people to demand better health and more safety. Business responds to these demands
by providing safer products and workplaces, even in the absence of regulation. Individuals with
rising incomes may also purchase or donate land to nature conservancies to provide ecological
benefits. Yet as on the cost side, the baseline that is used is almost always the status quo, not
what is likely to be true in the future.
Fourth, the construction of a baseline may be complicated where, as frequently occurs,
there are several causes of the change in behavior attributed to a Federal regulation. State and
local regulations may also require some level of compliance. The tort system, voluntary
standards organizations, and public pressure also cause firms to provide a certain degree of
public protection in the absence of Federal regulation. As GAO points out, determining how
much of the costs and benefits of these activities to attribute solely to Federal regulation is a
difficult undertaking (GAO 1996). Adding to the complexity, the degree to which these other
factors cause firms and other regulated entities to provide safe and healthful products and
workplaces and engage in environmentally sound practices changes over time, generally
increasing with increasing per capita incomes and knowledge about cause and effect.
Thus, although the National Highway Traffic Safety Administration has significantly
increased the safety of automobiles, it is not likely that if the agency's regulations were
eliminated the automobile companies would discontinue the safety features that had been
mandated. Consumers demand safer cars than they used to and automobile companies are
concerned about product liability. This same phenomenon exists with the environment, although
probably to a lesser extent. Environmentally responsible behavior has become good for the
bottom line. One paper company interviewed by GAO said that it would have incurred a
substantial amount of its compliance costs even if there were no regulations, simply as good
business practices (GAO 1996, p. 51). Over time, this "rising baseline" phenomenon reduces the
true costs and benefits of health, safety, and environmental regulations. Estimates of the
aggregate costs and benefits of regulation that include unadjusted estimates from aging studies
are thus likely to be over estimates of the current costs and benefits of those regulations.
The Apples and Oranges Problem
The studies that have attempted to tote up the total costs and benefits of Federal
regulations have basically added together a diverse set of individual studies. Unfortunately,
these individual studies vary in quality, methodology, and type of regulatory costs included.
Thus we have an apples and oranges problem, or, more aptly, an apples, oranges, kiwis,
grapefruit, etc., problem.
Part of the problem arises because of the nature of regulation itself. There are over
130,000 pages of regulations in the Code of Federal Regulation, with about 60 federal agencies
issuing regulations at the rate of about 4,000 per year. For our purposes, a "regulation" or "rule"
means an agency statement of general applicability and future effect, which the agency intends to
have the force and effect of law, that is designed to implement, interpret, or prescribe law or
policy or to describe the procedure or practice of an agency. Clearly, "regulation" encompasses
a lot of territory. The Hopkins series of studies (1991,1992, 1995,1996), which are the latest
attempts to aggregate the costs of all regulations for which estimates are available and which we
discuss in detail later, include five major categories of regulation:
Environmental. As the EPA points out, the true social cost of regulations aimed at
improving the quality of the environment are represented by the total value that society places on
the goods and services foregone as a result of resources being diverted to environmental
protection. (Cost of a Clean Environment, pp. 1-2 to 1-3.) These costs include the direct
compliance costs of the capital equipment and labor needed to meet the standard, as well as the
more indirect consumer and producer surplus losses that result from lost or delayed consumption
and production opportunities resulting from the higher prices and reduced output needed to pay
for the direct compliance costs. In the case of a product ban or prohibitive compliance costs,
almost all of the costs represent consumer and producer surplus losses. Most of the cost
estimates used in this report do not include consumer and producer surplus losses because it is
difficult to estimate the demand and supply curves needed to do this type of analysis.
Further indirect effects on productivity and efficiency result from these price and output
changes as they filter through other sectors of the economy. According to EPA in the Cost of
Clean report, recent research indicates that compliance cost estimates may understate
substantially the true long-term costs of pollution control (p. 1-3). The estimates used in this
report do not include these indirect and general equilibrium effects.
The benefits of environmental protection are represented by the value that society places
on improved health, recreational opportunities, quality of life, visibility, preservation of
ecosystems, biodiversity, and other attributes of protecting or enhancing our environment. As
discussed in chapter I, the value is best measured by society's willingness-to-pay for these
attributes. Because most types of improvements in environmental quality are not traded in
markets, benefits must be estimated by indirect means using sophisticated statistical techniques
that generally make benefit estimation more problematic than cost estimation.
Although the EPA issues the great majority of environmental regulations, DOI, DOT, and
the DOE, among others, also issue rules aimed at improving the environment.
Other Social. This category of regulation includes rules designed to advance the health
and safety of consumers and workers, as well as regulations aimed at promoting social goals such
as equal opportunity, equal access to facilities, and protection from fraud and deception. They
are often lumped together with environmental regulation in the category of "Social Regulation."
Social regulation is mainly concerned with controlling the harmful or unintended consequences
of market transactions, such as air pollution, occupationally induced illness, or automobile
accidents. These consequences are commonly called "negative externalities" and regulation
designed to deal with them attempts to "internalize" the externalities. This can be done by
regulating the amount of the externality, e.g., banning a pollutant or limiting it to a "safe" level,
or by regulating how a product is produced or used. Social regulation may also require the
disclosure of information about a product, service or manufacturing process where inadequate or
asymmetric information may place consumers, citizens or workers at a disadvantage. The
techniques and methodological concerns involved in the estimation of the social costs and
benefits generated by these rules is similar to those involved in the estimation of costs and
benefits of environmental regulation discussed above.
Economic. Economic regulation is so-called because it directly restricts firms' primary
economic activities, e.g., its pricing and output decisions. It may also limit the entry or exit of
firms into or out of certain specific types of businesses. The regulations are usually applied on
an industry basis such as agriculture, trucking, or communications. In the United States, this
type of regulation at the Federal level has often been administered by what are referred to as
"independent" commissions, e.g., the FCC, ICC or FERC, whose members are appointed but not
removable without good cause by the President. The economic loss caused by this type of
regulation results from the higher prices and inefficient operations that often result when
competition is prevented from developing.
The costs of such regulation are usually measured by modeling or comparing specific
regulated sectors with less regulated sectors, estimating the consumer and producer surplus
losses that result from higher prices and lack of service, and estimating the excess costs that may
result from the lack of competition. In contrast to social regulatory cost estimates, these
estimates are mainly indirect costs.
Economic regulation may produce social benefits when natural monopolies are regulated
to simulate competition. Although Hahn and Hird (1991) argue that the dollar amount of such
efficiency benefits are small in a dynamic and technologically vibrant economy, this judgement
is an educated guess based on a reading of recent history, rather than the result of an empirical
study. It appears to be based largely on the widely accepted view that the U.S. economy has
become more competitive over time, leaving little scope for benefits from eliminating monopoly,
and on the observation that much of the motivation for economic regulation is to enhance one
group at the expense of another. As noted above, such transfers are not generally considered
social costs or benefits. But even though monopoly power may not be long lasting in a dynamic
U.S. economy, it is not nonexistent at a given point in time.
Moreover, while Hahn and Hird (1991) define economic regulation as including only
regulation of entry, output, and prices, in practice they appear to lump all Federal regulation of
banking and other financial institutions, as well as consumer protection regulation through
mandated disclosure requirements, into the "economic regulation" category in their cost
estimates. In our view, chartering , branching, interest rate, and activity regulation are the only
major categories of banking regulation that conform to the definition of economic regulation
used here. The other categories are "safety-and- soundness" regulation and "consumer
information and protection" regulation, both of which fit more logically into the "other social
regulations" category used in this study (White 1991, pp. 32-33). This definitional issue is
important because the type and magnitude of benefits associated with the different categories of
banking regulation differ greatly. In particular, while costs may exceed benefits for some types
of economic (entry, output, and prices) regulation, safety-and-soundness regulation is essential to
a well-functioning financial system and thus fully justifies the cost (White 1991), and the
consumer protection regulation applicable to banking is similar to consumer protection
information for other industries where there is general agreement that the benefits exceed the
Transfer. As discussed in chapter I, transfers are payments from one group in society to
another and therefore are not real costs to society as a whole. One person's loss is another
person's gain. Examples of transfers include payments to Social Security recipients from
taxpayers and the higher profits that farmers receive as a result of the higher prices consumers
must pay for farm products limited by production quotas. Nevertheless, Hopkins (1991) includes
transfer costs in the total cost of regulations. He does place them in a separate category and
points out that they are different from the real social costs that result from economic efficiency
losses. Also as discussed in Chapter I, OMB's guidance states that transfers should not be added
to the cost and benefit totals included in regulatory assessments but should be discussed and
noted for policy makers.
Process. Process costs, according to Hopkins, are the administrative or paperwork costs
of filling out government forms such as income tax, immigration, social security, procurement
etc. Although there are benefits to the services that these government programs provide and
some minimum amount of process cost is necessary to deliver these services, it makes little sense
to try to place a separate value on administration. Rather, process costs should be viewed as a
"cost of doing business" that should be minimized for a given level or quality of service.
Adding these various categories together, as Hopkins and others have done, does two
things. It produces large numbers and it creates confusion. It produces large numbers by
including "costs" that are not normally considered as part of the regulatory reform debate. For
example, costs such as the burden of filling out income tax forms or doing the paperwork needed
to get visas, passports, small business loans, and veterans benefits are not what one usually
thinks about when worrying about the cost of regulation. Nor do we usually think that the
income gained by farmers from price support programs or the increased sales by domestic
businesses as a result of trade protection are costs of regulation. Congress did not seek oversight
of these types of costs when, in the last Congress, it debated legislative proposals for
comprehensive regulatory reform, such as S. 343 and H.R. 9, or when it passed the Unfunded
Mandate Reform Act of 1995 or the Small Business Regulatory Enforcement Fairness Act of
Adding these categories of regulation together with health, safety and environmental
regulation also creates confusion if the ultimate goal is to make recommendations for regulatory
reform because the appropriate policies to reduce any adverse effects from these programs are
very different. To reduce price supports, modify international trade protectionism, and minimize
non-cost-effective health, safety, and environmental regulation would take very different paths.
It may make it appreciably more difficult. Lumping them together does not enlighten the search
for appropriate reforms.
In sum, adding up the costs and benefits of the various regulatory programs may give us a
rough estimate of the magnitude of the impact of regulatory activities on the economy and make
it clear that regulation plays an important role in our economy. We do not disagree that this is
important information to have. Indeed, we can use the total cost figures to begin to track the
extent of this activity relative to other aggregate data. For example, our calculations indicate that
regulatory costs are about 4 percent (3.7%) of GDP in 1997. We have also looked at 1988, and
found that regulatory costs were then roughly the same percentage. From this comparison, we
can say that there has been no material growth in the cost of regulation relative to the size of the
economy in the last decade.
However, these data provide little useful information about what to do next. If what is
intended is to make regulation more efficient, one needs to estimate the incremental costs and
benefits of individual regulations, or specific provisions of individual regulations, on a case-by-case basis. If what is intended is to reduce the burden of existing, health, safety and
environmental regulation, one needs to estimate how firms would react to the removal of
requirements, not how they acted when the requirements were originally imposed. If what is
intended is to improve the cost-effectiveness of new regulations, one needs to know what factors
are preventing future regulations from being more cost-effective. But none of this information is
found in the aggregate estimates of the costs and benefits of regulation done to date.
2. Our Estimates of the Costs and Benefits of Existing Regulations
To meet the requirements of Section 645(a)(1), we surveyed the existing literature on the
total costs and benefits of regulation, supplementing it with information we have obtained from
reviewing regulatory impact analyses over the last ten years under Executive Orders 12291 and
12866. Our review of the literature revealed only one comprehensive study that attempted to
estimate the total costs and benefits of all Federal regulations (Hahn and Hird 1991). Hahn and
Hird's estimates were peer reviewed and published in one of the top economics/ legal journals
specializing in regulatory issues, the Yale Journal on Regulation. In addition, EPA issued a
report to Congress at about the same time known as the Cost of Clean report (EPA 1990). The
Cost of Clean report is recognized as the most thorough and careful attempt to estimate the
compliance cost of environmental regulation published to date.
The Hahn and Hird study compiled cost and benefit estimates from over 25 studies
published mostly by academics in peer reviewed journals, e.g.,. Hufbauer (1986) for international
trade, Wenders (1987) for telecommunications, Gardner (1987) for agricultural price supports,
Morrison and Winston (1986 and 1989) for airlines, Crandall (1986) for highway safety, and
Crandall (1988), Denison, (1979), and Viscusi (1983) for Occupational Safety and Health. It
should be noted that although all of these studies are generally recognized as the best available,
they are not without shortcomings. For example, the Crandall (1988) and Denison (1979) studies
relied upon for the cost of OSHA regulations used survey data that included expenditures that
firms would have made on safety in the absence of OSHA regulation.
The Cost of Clean report's estimates of costs are based on annual survey data from the
Department of Commerce's "Pollution Abatement and Control Expenditures" (PACE) reports,
regulatory impact analyses of major EPA regulations, and special analyses by EPA program
offices or contractors. The PACE report surveys, which were conducted through 1994, but
discontinued thereafter, cannot be used without careful adjustments because they contain
pollution control expenditures that are not Federally mandated. EPA is continuing its efforts to
review the costs and benefits of certain of its regulatory programs. It has completed reports on
drinking water (EPA 1993) and surface water (EPA 1995) and is presently working on a report
required by the Clean Air Act Amendments of 1990 on the costs and benefits of the Clean Air
Act, which it plans to submit to Congress in October of 1997. A draft of this report presents
estimates that appear to be at variance with some of the numbers we report below (EPA 1997).
In addition, we used information about the costs and benefits of major regulations
reviewed by OMB under Executive Orders 12291 and 12866. (We include the rules published in
1987 and 1988 to allow for a lag between publication of the rule and the expenditure of funds for
compliance and realization of benefits.) The rules included are generally all final rules with
annual costs of $100 million or more issued by Executive Branch agencies, which we believe
capture at least 90 percent of the costs added by all rules. The cost estimates themselves are
agency estimates that have gone through OMB review and the Administrative Procedure Act
requirements for notice and comment by the public.
Using the estimates for Federally mandated regulatory costs from the Cost of Clean
report (1990, Table 8-9D) for environmental regulation and Hahn and Hird's estimates for other
social regulation for a 1988 base, we added the cost of all major regulations reviewed by OMB
under Executive Orders 12291 and 12866 and issued by the agencies between 1987 and 1996.
Table 1 shows our calculations for the costs of social regulations.
While our estimates do not include the costs of regulations with costs below $100 million
nor regulations from independent agencies such as the CSPC or NRC, and there is a possibility
that agencies understate the costs of proposed rules (Hopkins, 1992, p. 13), we believe that, if
anything, the estimates overstate actual direct costs because of the rising baseline phenomenon
discussed above. For example, as a sensitivity analysis, it does not seem implausible that, for
environmental and other social regulations over ten years old, no more than half of compliance
costs would likely be saved if these Federal regulations magically disappeared over night. The
automobile companies are not likely to make their cars less safe or less fuel efficient. Similarly,
the great majority of firms are not likely to stop controlling asbestos and cotton dust fibers or
lead dust and benzene emissions in the workplace if these regulations were abolished. Nor
would the judicial tort system likely tolerate increased levels of harmful pollution or harmful
products. If this scenario is correct, then the cost of social regulation in 1997 would fall to $130
billion (136/2 + 62 = 130), or $93 billion for environmental regulations and $37 billion for other
To the cost estimates for environmental and other social regulation, we must add the costs
of the other types of regulation, i.e., economic and process regulation. We use the Hahn and
Hird estimate for the efficiency cost of economic regulation for 1988. Because the great majority
of these regulations are issued by independent regulatory agencies (e .g., the FCC, the ITC, and
the FDIC) that were not required under Executive Orders 12291 or 12866 to submit information
on benefits and costs of regulations to OMB, we did not have our own data to update the 1988
baseline. Instead, we relied on a study by Hopkins (1992) who derived an estimate of $81 billion
for the efficiency costs of economic regulation for 1997.
Hopkins (1992) made several additions to Hahn and Hird to update economic regulation
costs to 1997: $10 billion for surface transportation costs, $5 billion for the Jones Act, and $5
billion for banking regulations (p. 27). However, since Hopkins made his estimates in 1992,
there have been several developments resulting in further deregulation, especially in banking --
with the passage of the Interstate Banking and Branching Efficiency Act of 1994 and the recent
actions taken by the four Federal bank regulatory agencies to reduce regulatory burdens -- and in
telecommunications -- where the FCC has also taken steps to increase competition and
implement the Telecommunications Act of 1996. To take into account the recent deregulation in
financial services and telecommunications, we reduced the Hopkins' estimate by $10 billion.
We do not include Hopkins' estimate of the transfer costs of economic regulation,
because, as noted above, we do not believe that transfers are costs that should be included in total
cost of regulation estimates. However, if one insists on adding transfer costs using the Hopkins'
methodology, the total would increase by about $140 billion because he assumes transfer costs
are about twice efficiency costs. However, by definition, this transfer amount must also be added
to total benefits .
We also do not include most of the process or paperwork cost estimated by Hopkins and
others (Hopkins 1991 and 1992 and Weidenbaum and DeFina 1978) because these costs are for
the most part already included in cost estimates supplied by the agencies and reviewed by OMB
or are tax paperwork costs which we point out above do not logically fit with traditional
regulation. However, the cost of paperwork imposed by the independent agencies should be
included because those agencies do not submit the cots of the underlying regulations to OMB
and thus they would not be included in our Executive Order 12291 and 12866 estimates.
According to OMB's latest estimates from its information collection budget (OMB 1997), the
burden hours of paperwork imposed by the independent agencies was about 390 million hours
(or about $10 billion in costs using a $26.50 per hour estimate to take into account the fact that
these agencies' paperwork often require some professional expertise to fill them out (Hopkins
1991)). Most of these costs are the costs of disclosing information to consumers or investors to
prevent fraud and deception (FTC, SEC and FDIC represent about 75 percent of the total). We
include these costs in Table 2.
Although for the reasons stated above we do not believe it is productive to add the cost of
tax paperwork to the total cost of regulation, these are real costs and should be considered in their
own context. Indeed IRS paperwork is by far the largest component of OMB's information
collection budget. According to our latest estimates, the total burden hours for Treasury for FY
1997 is about 5.3 billion (OMB 1997), of which 99 percent is for the IRS. Using the $26.50
estimate as the opportunity cost of an hour's time in filling out tax forms produces a total cost of
about $140 billion. This estimate by itself may be informative but it should not be added to our
total cost estimate. If it were added, the benefits of our tax system (revenue received was about
$1.5 trillion for FY 1997) should also be added to our benefit totals.
Therefore, our best estimate of the total cost of regulation for 1997 is shown in Table 2.
Aggregating benefits from individual regulations poses special problems even beyond
those discussed above for aggregating costs. There are several important limits to such an
exercise. First among these is uncertainty. Because so much of the uncertainty in possible
benefit estimation is unknown, and so little is known about the relationships among benefit
estimates of different regulations, analysts have virtually no basis for aggregating benefits in a
manner that might preserve information about the likely distribution of aggregate benefits.
Second, as noted above, benefits, like costs, may be presented as monetized, quantified,
or in narrative forms. For a variety of reasons, many of them understandable, if not legitimate,
agencies often do not express beneficial effects in monetizable terms that can easily be
aggregated. What is being described may not be readily amenable to quantification or
monetization (e.g., the value of greater national security or of increased individual privacy), or
the agency may have chosen not to develop monetized estimates because of resource or time
constraints. Moreover, while some of the effects are present as quantified estimates, these cannot
be summed if they are not expressed in common units. Of course, when effects are not expressed
in quantitative terms, this aggregation problem is even more acute. We can only conclude that
estimates of the total benefits of regulation will be understated by an unknown amount until all
significant benefits are monetized.
Because of the difficulty of estimating benefits, there are very few studies that attempt to
estimate the total benefits as well as costs of regulation. Indeed the only study that has attempted
to estimate the total benefits of all regulations is the study by Hahn and Hird that we relied upon
for the 1988 cost baseline. Hahn and Hird present a broad range of estimates of the
annual benefits of regulation in billions as of 1988, shown in Table 3, which we have converted to 1996 dollars
using the CPI.
Note that while Hahn and Hird do not include any benefits from economic regulation (on
the grounds that they are negligible in most cases), they state that the regulation of natural
monopolies can theoretically produce efficiency gains (p. 253). We believe that Hahn and Hird
are being too pessimistic about the benefits of regulating natural monopolies; indeed, our
experience with local telephone markets suggests that they have appreciably under estimated the
benefits of regulating certain markets. More importantly, it appears that although consumer
protection regulation through disclosure requirements more logically fits with social regulation,
the Hahn and Hird study included some disclosure regulation costs in the economic regulation
category, but they did not include any of its benefits. We are not able to remedy this imbalance,
for while we have been able to estimate the costs of some disclosure regulation from independent
agencies using our information collection burden estimates, we do not have estimates of the
benefits of these regulations. We know, however, that there are significant benefits from various
disclosure requirements. For example, the SEC, by promoting investors' confidence in the
integrity of the capital markets and providing price and trade information, reduces the cost of
raising capital, facilitates innovation, and guides resources to their most valued uses. Thus, in
the table below, we use asterisks for the benefits of economic and paperwork/disclosure
regulation to indicate that these benefits may be significant but that valid and reliable estimates
remain to be done.
Since the Hahn and Hird study, the only systematic study of the benefits together with the
costs of major social regulations, of which we are aware, is a study by Hahn, published jointly by
Oxford University Press and the AEI Press in 1996. In that study, Hahn reviewed the regulatory
impact statements required by Executive Orders 12291 and 12866 for major regulations
produced by agencies between 1990 and mid-1995. Hahn accepted the agency estimates of
benefits at face value, used consensus estimates from the academic literature to value the benefits
(e.g., the Viscusi 1992, estimate for a "statistical life"), and used consistent assumptions across
agencies to produce monetized benefit estimates (pp. 214-217). He found that 54 regulations had
produced almost $500 billion in benefits in present value (discounting at 5 percent and using his
middle value consensus estimates) (p. 218). Hahn also calculated that these regulations produced
$220 billion in net costs (gross costs minus any costs savings produced by regulation).
Unfortunately, we do not have enough information to convert Hahn's present value
estimates to annual estimates so that we could compare them to our annual cost estimates
presented above. However, we can use Hahn's benefit/cost ratio ($500b/$220b) or 2.5, assume
that it holds for the full period since 1988, and calculate an aggregate benefit estimate. It should
be noted, however, that Hahn believes his aggregate net benefit estimates " . . . are likely to
substantially overstate actual net benefits" (p. 224). Both our estimates and Hahn's estimates
would most likely include almost the same set of regulations issued between 1990 and 1995
because we both attempted to be exhaustive in our cost collection effort. According to our
sample, about 80 percent of the costs of social regulation issued between 1989 and 1996 were
issued between 1990 and 1995. Assuming that in 1988, social regulation produced net benefits
of $2 billion as Hahn and Hird suggest, and using Hahn's benefit-cost ratios for environmental
(1.4) and other social regulation (5.3), we calculate that the benefits of regulation in 1996 were as
follows, and we present our cost estimates for comparison:
As explained above, these are very rough estimates and viewed alone not very
informative. Moreover, there is particular uncertainty about the benefits of all categories of
regulation. With that very important caveat, the total numbers on costs and benefits indicate that
regulation has produced about as much, if not more, in benefits as in costs, although according to
these estimates environmental and other social regulation has clearly produced benefits
significantly greater than direct compliance costs. Disaggregating the totals a little reveals that
"Other Social" regulation produces very large net benefits, but if one digs into both the Hahn and
Hird, and Hahn studies in greater detail, it becomes clear that most of the benefits of this
category are produced by highway safety regulation. Hahn and Hird state that they found very
little "credible evidence" that as of 1988, OSHA regulations had produce any significant benefits
(275-276), although Hahn's 1996 study found that OSHA regulations had produced over $50
billion (present value) in net benefits by 1995.
Hahn makes clear that even though his study found that the 53 regulations issued between
1990 and 1995 produce very large net benefits, only 23 would "pass" a cost-benefit test. He also
points out that if the rules that had not passed the test had not been issued, net benefits would
have been $115 billion, or about 40 percent greater (p. 221). He also finds that all safety
regulations have benefits greater than costs, and that regulations based on the Clean Air Act and
the Safe Drinking Water Act had positive net benefits (p. 221) (which is corroborated by the
EPA Drinking Water study (1993). An analysis of the costs and benefits of regulations based on
other regulatory programs produced mixed results. The message is clear: the policy content is in
3. Other Estimates of the Total Costs of Regulation
As noted, the estimates of total costs and benefits that we have provided overstates, we
believe, both the benefits and most certainly the costs of regulation. Nonetheless, our cost
estimates are substantially less than other numbers that are often cited and have gained a certain
credibility in the debate. We would note that, apart from the Hahn and Hird study we used, all
other estimates of total costs do not present benefit estimates. We believe that presenting costs
without benefits is not very informative and potentially misleading. In any event, some
explanation of the difference between our numbers and other numbers that have been cited is
According to a 1995 report to Congress by the Small Business Administration's (SBA)
Office of Advocacy, there are estimates of the total cost of regulation generated by the Heritage
Foundation as high as $810 billion to $1.7 trillion for 1992 with benefits reportedly netted out.
We cite this study because it is the largest estimate of the costs of regulation that we are aware
of. Our reference to it should not be construed as any endorsement of it; indeed, it has not been
peer reviewed, it has not been published in a reputable journal, and, most importantly, the basis
for the estimate has not been made publicly available. Our own view is that the numbers are
either wrong or are measuring something other than what we are talking about.
On the other hand, there is a series of Hopkins studies of the total cost of regulation
(1991, 1992, 1995, and 1996), which is both well known and better documented. The Hopkins
estimates have also received attention from the Congress. A recent GAO study, Regulatory
Reform: Information on Costs, Cost-Effectiveness, and Mandated Deadlines for Regulation
(1995), was asked to focus on the Hopkins study because of its prominence and the fact that it
was the only game in town.
Hopkins relied on the paper by Hahn and Hird (1991) that provided estimates of the costs
and benefits of economic and social regulation for 1988, on the 1990 study by the EPA, The Cost
of a Clean, and various reports from OMB: The Information Collection Budget (various years) --
that is, the same materials that we used for our 1988 cost baseline. Hopkins also reviewed two
earlier attempts at adding up the total costs of regulation as of 1976-77 by Weidenbaum and
DeFina (1978) and Litan and Nordhaus (1983) to make estimates of the trend in total regulatory
costs over this decade. He also projected cost to the year 2000, based on estimates from the Cost
of Clean, extrapolations of past trends, and some educated guess work about the future costs of
compliance with regulations required by statutes such as the Clean Air Act Amendments of 1990
and the Americans with Disabilities Act of 1990. Because we focus our attention on the state of
regulation as of 1997, we do not directly critique the earlier studies by Weidenbaum and DeFina
or Litan and Nordhaus, nor do we discuss Hopkins' extrapolations beyond 1997.
Hopkins' cost estimate for 1997 (presented by us in 1996 dollars using the CPI), is shown in Table 5.
One important problem with these estimates is that, with the exception of the Process
estimate, they are based on individual studies that were published, for the most part, between
1975 and 1990 and then, as mentioned above, extrapolated to 1997 based on the Cost of Clean
cost projections for future years for environmental regulation and his own ad hoc "guesstimates"
(his words (1991, p. 11)) for other social and economic regulation. Note that although we also
use data from 1988 and earlier, his approach to update the information to 1997 differs
significantly from ours. Rather than extrapolate, as he did, we used timely information supplied
by the agencies over the period 1987 to 1996 that was subject to notice and public comment and
OMB review, as well as more recent experience. Ideally, to get a realistic picture of the total
costs of regulation, one needs to do a comprehensive study of all regulatory costs facing the
economy at a given point in time. But that would be prohibitively expensive and, as pointed out
above, ex post surveys of the costs of existing regulations have their own problems.
A second problem relates to the appropriateness of Hopkins' adjustments. Specifically,
Hopkins' adds to EPA's Cost of Clean report (the 1988 base), $10 billion for the Clean Air Act
Amendments, $8 billion for Superfund/RCRA, and $1 billion for several DOT environmental
regulations. It is not clear, however, how these figures are derived. Similarly, Hopkins' estimate
for "other" social regulation costs starts with Hahn and Hird (as we did), but adds an additional
$1 billion and an assumed rise of 5% percent per year for OSHA regulations, and adds $4 billion
for the new universal accessibility standards, $500 million for food labeling regulations, $200
million for energy conservation standards, and $1.6 billion for clinical lab regulations. These amounts are taken from a combination of agency and industry sources, although again it is not
clear how the specific numbers were derived.
As noted above, we used Hopkins' updates for the changes in Economic: Efficiency
Costs to 1997, reduced by $10 billion to account for the deregulation that has taken place in
banking and telecommunications since Hopkins derived his estimate in 1992. We did not,
however, include Hopkins' estimate of Economic: Transfer Costs. Hopkins acknowledges that
transfers are exchanges of funds from one group to another, but he argues that the existence of
transfers creates real social costs because they give rise to "rent-seeking behavior." ("Rent
seeking behavior" is behavior that attempts to capture or create excess profits usually by
influencing government actions, such as regulations.) He states that the existence of transfers
creates real costs that exhausts the amount of the transfer as interest groups and their lobbyists,
lawyers and experts campaign for those funds (p. 29). We believe that Hopkins has the causality
wrong. Rather than the existence of a transfer program causing rent-seeking behavior, rent-seeking behavior causes the transfer. It is the possibility that rent-seeking behavior may result in
a gain that causes special interests to form and campaign for special treatment. The transfer
program does not have to exist, just the possibility that one could be set up. Thus to the extent
that rent-seeking behavior imposes real costs on society, those costs would be more appropriately
attributable to our democratic political system than to a particular regulation.
We also believe that Hopkins' has overstated the costs of Process regulation, which for
the most part either represents double counting or more appropriately belongs elsewhere. Most
of Hopkins' estimate is based on the burden hour estimates reported in OMB's annual
Information Collection Budgets (various years ) of the time it takes the public to comply with
information requests made or generated by the Federal government. He multiplies burden hours
by $26.50 per hour (in 1996 dollars), an estimate of the public's opportunity cost for filling out
forms and gathering information. While average private nonagricultural hourly earnings was
$11.82 in 1996 (less than 45 percent of the number he used), Hopkins argues that his time cost
estimate is not too high because about 85 percent of the burden hour estimate is from the
Treasury Department, much of which represents the time it takes high priced tax accountants to
fill out income and corporate tax forms.
We believe the paperwork costs of the tax code should not be included in an estimate of
the total cost of regulation that is being developed to enable proposals for regulatory reform.
First, filling out tax forms is not the result of "regulations" but rather of the tax code itself, with
most regulations merely providing interpretations and clarifications of tax law. Second, Hopkins
assumes a zero baseline -- that is, he implicitly assumes that replacing the revenue generated by
the present tax code could be done with no record keeping or reporting costs. The implicit
baseline is a world without taxes. Third, reforming the tax code is an entirely different public
policy area than regulation, and lumping the two together, especially when the tax numbers are
so large relative to social and economic regulatory costs, just confuses the issue.
Hopkins has removed the cost of procurement paperwork, such as that imposed by DOD
and GSA, based on an OMB estimate that in 1990 the procurement paperwork burden was about
30 percent of the total non-tax-related paperwork. He correctly points out that those costs are
mostly paid by taxpayers through higher procurement costs, and thus it would be double
counting to include them as private sector regulatory costs. However, most of the remaining
paperwork costs also represent double counting, because the estimates of regulatory costs for
individual social and economic regulations that he uses already include these costs as a cost of
compliance. Specifically, the compliance cost estimates submitted to OMB and included in our
estimate for the cost of social regulation include associated paperwork costs. Although Hopkins
admits the likelihood of double counting, he dismisses it because "the dominance in this category
of tax-related paperwork suggests this is not likely a serious problem" (1991, p. 14). But once
tax-related paperwork is removed, it becomes a serious problem.
Hopkins also adds to his Process costs estimates $10 billion in 1997 as the amount that
State and local government spent to comply with Federal mandates. However, we cannot
determine a clear basis for his estimate. Because our approach of adding the costs of all social
regulations issued since 1987 should capture State and local regulatory costs, there should not be
a special provision for State and local mandates.
The final piece of Hopkins' Process cost estimate is an estimate of how much more
overhead the U.S. multi-payer health care system generates than Canada's single-government-
payer system. His argument here is that because the United States has less regulation, it has
higher regulatory costs. It is certainly true that regulation can improve efficiency, but it seems
disingenuous to argue that because regulations have not mandated a single payer system or
restricted private payment systems, etc., regulatory costs are increased. These increased cost
estimates (Woolhandler and Himmelstein, 1991), if they are true (they are controversial), are
more properly treated as potential benefits from regulation (or of a government program), not as
costs of not regulating. Additionally, as discussed above, including these costs confuses the
regulatory reform debate.
In sum, in our view, Hopkins' total cost estimate is substantially greater than ours
because he includes inappropriate transfers and process costs and less accurate estimates of the
growth of social regulation since 1988.
4. Assessment of the Direct and Indirect Impact of Federal Rules
A proper assessment of the costs and benefits of regulation would have to take into
account both the direct and indirect impact of regulation on the economy. As reported above, our
estimate of the direct effect is that, in the aggregate, the net benefits of regulation issued to date
is positive. The few studies that have attempted to determine the indirect effects of regulation on
productively and welfare have found significant indirect effects, implying that the direct effects
reported above are significant understatements of the full costs of regulation (Hazilla and Kopp
1990 and Jorgenson and Wilcoxen 1990). However, as Hahn and Hird (1991) point out, it is not
clear how to evaluate these studies and others like them, which are based on huge, complex and often proprietary models of the U.S. economy. This makes it almost impossible to validate the
models or to view the assumptions on which they are based.
These studies have another major problem because they only take into account indirect
cost effects and do not include the indirect beneficial effects that may result from better health
and safer lives. Yet it is generally agreed that healthier people tend to work harder and longer
and save and invest more, thereby increasing the growth of the economy. Therefore, without
knowing what the indirect and general equilibrium benefits of regulation are, one should not
draw conclusions by only looking at the indirect costs. Models that take into account the indirect
benefits and general equilibrium effects of longer life spans, higher levels of environmental
quality, and more equal opportunities remain to be developed.
The best survey of what we know about the full range of indirect costs and benefits of
social regulation was recently published in one of the leading economic journals: the Journal of
Economic Literature (Jaffe, Peterson, Portney, and Stavins 1995). Although concentrating on
environmental regulation, their discussion should apply to health and safety regulation as well
because they are similar in their economic effects and the direct costs of health and safety
regulation are only about one third the amount of environmental regulation. The authors
conclude from a survey of the literature that environmental regulation has little impact on
"competitiveness as measured by net exports, overall trade flows, and plant location decisions (p.
157), " modest adverse impacts on productivity" (p. 151) and "significant dynamic impacts . . . in
the form of costs associated with reduced investment" based on computable general equilibrium
models (p. 151). However, they also point out that, for the most part, these estimates do not take
into account the feedback effect from improvements in the environment (p. 153).
Jaffe et al. also examine the contention that environmental and other social regulation
may actually enhance economic growth and competitiveness by stimulating improvements in
productivity as firms compete among themselves to comply with regulations in the least cost
way. We discussed this proposition above as a reason why the actual costs of compliance ex post
often turns out to be less than predicted ex ante. Several authors have extended this proposition
beyond the ad hoc to include the economy as a whole (Porter 1991 and Gardiner 1994). This
line of reasoning claims that the country that leads in environmental protection will gain a lasting
comparative advantage in international trade in the supplier industries because of having been the
"first mover" into an area that other countries must follow.
We are cautious about extending such claims to the economy as a whole. To be sure,
certain sectors benefit and we may even develop a comparative advantage in them, but other
sectors must invariably lose their comparative advantage because resources are drawn from them
and comparative advantage is by definition a relative phenomenon. Jaffe, et al., (p. 157)
"Thus, overall, the literature on the 'Porter hypothesis' remains one with a high ratio of
speculation and anecdote to systematic evidence. While economists have good reason to
be skeptical of arguments based on nonoptimizing behavior where the only support is
anecdotal, it is also important to recognize that if we wish to persuade others of the
validity of our analysis we must go beyond tautological arguments that rest solely on the
postulate of profit-maximization. Systematic empirical analysis in this area is only
beginning, and it is too soon to tell if it will ultimately provide a clear answer."
We agree with this statement and hope that this report stimulates "systematic empirical
analysis" in this area, as well as work on as the broader issue of how to improve the estimation
of the costs and benefits of regulatory programs discussed in this report.