Title Page
 Front Matter
 Table of Contents
 Part I: Mandated standards in a...
 Part II: Microeconomic effects...
 Part III: Aspects of occupational...
 Part IV: Microeconomic effects...
 Part V: Aggregative effects of...
 Part VI: Future outlook and...

Title: Economic effects of government-mandated costs
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00103021/00001
 Material Information
Title: Economic effects of government-mandated costs
Physical Description: Book
Language: English
Creator: Government-Mandated Costs Seminar, (1977
Publisher: University Presses of Florida
Place of Publication: Gainesville
Copyright Date: 1978
 Record Information
Bibliographic ID: UF00103021
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: oclc - 4135740
lccn - 78014399
isbn - 0813006260 :

Table of Contents
        Page i
        Page ii
    Title Page
        Page iii
        Page iv
        Page v
        Page vi
    Front Matter
        Page vii
        Page viii
        Page ix
        Page x
    Table of Contents
        Page xi
        Page xii
    Part I: Mandated standards in a market economy
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
        Page 47
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
    Part II: Microeconomic effects (A)
        Page 53
        Page 54
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
        Page 71
        Page 72
        Page 73
        Page 74
        Page 75
        Page 76
        Page 77
        Page 78
        Page 79
        Page 80
        Page 81
        Page 82
        Page 83
        Page 84
        Page 85
        Page 86
        Page 87
        Page 88
        Page 89
        Page 90
        Page 91
        Page 92
    Part III: Aspects of occupational safety and health in the U.S.
        Page 93
        Page 94
        Page 95
        Page 96
        Page 97
        Page 98
        Page 99
        Page 100
        Page 101
        Page 102
        Page 103
        Page 104
        Page 105
        Page 106
        Page 107
        Page 108
    Part IV: Microeconomic effects (B)
        Page 109
        Page 110
        Page 111
        Page 112
        Page 113
        Page 114
        Page 115
        Page 116
        Page 117
        Page 118
        Page 119
        Page 120
        Page 121
        Page 122
        Page 123
        Page 124
        Page 125
        Page 126
        Page 127
        Page 128
        Page 129
        Page 130
        Page 131
        Page 132
        Page 133
        Page 134
        Page 135
        Page 136
        Page 137
        Page 138
        Page 139
        Page 140
        Page 141
        Page 142
        Page 143
        Page 144
        Page 145
        Page 146
        Page 147
        Page 148
        Page 149
        Page 150
        Page 151
        Page 152
    Part V: Aggregative effects of mandated standards
        Page 153
        Page 154
        Page 155
        Page 156
        Page 157
        Page 158
        Page 159
        Page 160
        Page 161
        Page 162
        Page 163
        Page 164
        Page 165
        Page 166
        Page 167
        Page 168
        Page 169
        Page 170
        Page 171
        Page 172
        Page 173
        Page 174
        Page 175
        Page 176
        Page 177
        Page 178
        Page 179
        Page 180
        Page 181
        Page 182
        Page 183
        Page 184
        Page 185
        Page 186
        Page 187
        Page 188
        Page 189
        Page 190
        Page 191
        Page 192
        Page 193
        Page 194
        Page 195
        Page 196
        Page 197
        Page 198
        Page 199
        Page 200
        Page 201
        Page 202
        Page 203
        Page 204
    Part VI: Future outlook and reform
        Page 205
        Page 206
        Page 207
        Page 208
        Page 209
        Page 210
        Page 211
        Page 212
        Page 213
        Page 214
        Page 215
        Page 216
        Page 217
        Page 218
        Page 219
        Page 220
        Page 221
        Page 222
Full Text

Economic Effects of Government-Mandated Costs

Economic Effects of
Government-Mandated Costs

Edited by
Robert F. Lanzillotti
Director, Public Policy Research Center
University of Florida

University Presses of Florida

Library of Congress Cataloging in Publication Data

Government-Mandated Cost Seminar, University of
Florida, 1977.
Economic effects of government-mandated costs.

1. Industry and state-United States-Congresses.
2. Costs, Industrial-United States-Congresses.
I. Lanzillotti, Robert Franklin, 1921-
II. Florida. University, Gainesville. Public
Policy Research Center. III. Title.
HD3616.U46G65 1977 338.973 78-14399
ISBN 0-8130-0614-7
ISBN 0-8130-0626-0 (cloth)

University Presses of Florida is the scholarly
publishing agency for the State University
System of Florida.





OVER THE PAST DECADE, the character and scope of government regu-
lation of business has become a matter of increasing concern among
economists. There have been repeated calls for reform and selective
deregulation of the sector that has been traditionally directly regu-
lated (for example, airlines). Strangely, at the same time, there has
been an acceleration and intensification of governmental controls
over the private and traditionally nonregulated sector through gov-
ernmental "mandates" covering a wide spectrum of U.S. production.
They include numerous statutes ranging from warnings about the
hazards of cigarette smoking (Public Health Smoking Act, 1970),
standards for "child-resistant" packaging (Poison Prevention Pack-
aging Act, 1970), and safety standards for consumer products (Con-
sumer Product Safety Act, 1972) to safety standards and equipment
for motor vehicles (Traffic Safety Act, 1966), air and water quality
standards (Clean Air Act, 1969-70, and Federal Water Pollution
Control Act, 1972), safety and health standards (Occupational
Safety and Health Act, 1970), and many more.
Professor Murray Weidenbaum has examined the pervasiveness of
these statutory actions in his recent study Government-Mandated
Price Increases (American Enterprise Institute, 1975). His study
suggested the need for further analysis of the micro- and macro-
economic effects of government-imposed standards. This, then, was
the basic reason for the seminar, which consisted of a series of in-
vited papers and discussions based on the prepared papers. The
participants included university economists, economists and other


officials from industry, and officials from government-regulating
The central focus of the papers, commentaries, and general dis-
cussion was to analyze public policy as embodied in various statutes,
the rationale and implicit economic theories upon which they are
premised, and empirical evidence regarding their economic effects.
The papers also provide a kind of agenda for further research on
various economic issues that surround the public policy of mandating
standards in a free-enterprise economy.
We were fortunate to have present in the seminar, in addition to
academicians, economists and others in government and business
who are or have been involved in helping to develop and administer
the public policy of mandated standards. It should be understood,
with respect both to prepared presentations and to comments during
discussions, that the individuals were expressing their personal opin-
ions and do not necessarily reflect the views of governmental agen-
cies or corporations with which they are affiliated.
I would like to express special appreciation for their help in de-
veloping the concept, content, and format of the seminar to James C.
Miller, Milton Z. Kafoglis, Thomas R. Saving, Donald J. Watson,
and Roger D. Blair.
Finally, grants from the General Electric Foundation, the Alumi-
num Company of America Foundation, General Motors Corporation,
Ford Motor Company Fund, Deering-Milliken, Safeway Stores,
George W. Jenkins Foundation, Jim Walter Foundation, and Winn-
Dixie Stores provided the necessary financial support for the con-

Robert F. Lanzillotti
Director, Public Policy Research Center,
and Dean, College of Business
Administration, University of Florida

Government-Mandated Costs Seminar

February 24-25, 1977
Public Policy Research Center
College of Business Administration
University of Florida

Robert F. Lanzillotti
Professor of Economics and Dean,
College of Business Administration,
University of Florida

Nicholas Ashford
Center for Public Policy Analysis
Massachusetts Institute of
Peter Barth
Department of Economics
University of Connecticut
Roger D. Blair
Department of Economics
University of Florida
Morton Corn
Graduate School of Public Health
University of Pittsburgh
Nina Cornell
Council of Economic Advisors
Washington, D.C.
H. Michael Cushinsky
Department of Finance,
Insurance & Real Estate
University of Florida
Louis De Alessi
Law & Economics Center
University of Miami
Howard Dugoff
National Highway Traffic Safety
U.S. Department of Transportation

Henry L. Duncombe, Jr.
General Motors Corp.
New York, N.Y.
James W. Ford
Ford Motor Credit Co.
Dearborn, Michigan
Roy N. Gamse
Economic Analysis Division
Environmental Protection Agency
George A. Hay
Economic Policy Office
Antitrust Division
U.S. Department of Justice
Arnold A. Heggestad
ABT Associates, Inc. and
University of Florida
Thomas D. Hopkins
Executive Office of the President
Council on Wage and Price Stability
Ann R. Horowitz
Department of Economics
University of Florida
Ira Horowitz
Department of Management
University of Florida


Milton Z. Kafoglis
Government Operations and Research
Council on Wage and Price Stability
Marvin Kosters
American Enterprise Institute
Washington, D.C.
Stanley V. Malcuit
Aluminum Company of America
Pittsburgh, Penn.
James C. Miller III
American Enterprise Institute
Washington, D.C.
Thomas G. Moore
Hoover Institution on War,
Revolution and Peace
Stanford University
John F. Morrall III
Executive Office of the President
Council on Wage and Price Stability

Laurence C. Rosenberg
Division of Advanced Productivity,
Research and Technology
National Science Foundation
Thomas R. Saving
Department of Economics
Texas A & M University
Fred G. Secrest
Ford Motor Company
Dearborn, Michigan
Donald J. Watson
Corporate Public Relations
General Electric Co.
Murray L. Weidenbaum
Center for the Study of American
Washington University

Roger G. Noll
Department of Humanities and
Social Sciences
California Institute of Technology

Other Participants

Sanford V. Berg
Department of Economics
University of Florida
Joseph M. Bertotti
Corporate Educational Relations
General Electric Co.
Charles Black
Black, Crow & Eidsness, Inc.
Gainesville, Florida
Eugene F. Brigham
Department of Finance, Insurance
and Real Estate
University of Florida
Joel B. Cohen
Department of Marketing
University of Florida
Leonard Davidson
Department of Management
University of Florida

V. V. DiTommaso
Chrysler Corp.
Detroit, Michigan
Richard A. Elnicki
Department of Management
University of Florida
Henry H. Fishkind
Bureau of Economic and
Business Research
University of Florida
Alex Goldberg
Ford Motor Co.
Dearborn, Michigan
Homer Hooks
Florida Phosphate Council
Lakeland, Florida
James T. Hosey
United States Steel Foundation
Pittsburgh, Penn.


Larry W. Kenny
Department of Economics
University of Florida
Paul E. Koefod
Department of Economics
University of Florida
Edna Loehman
Food and Research Economics
University of Florida
G. S. Maddala
Department of Economics
University of Florida
Robert Q. Marston
University of Florida
Jon L. Mills
Center for Governmental
University of Florida
Michael J. Murray
Rate Department
Tampa Electric Co.
William A. Niskanen
Ford Motor Co.
Dearborn, Michigan
David J. Nye
Department of Finance, Insurance
and Real Estate
University of Florida
Mary B. Peterson
General Motors Corp.
Washington, D.C.

Gerald Reynolds
Jim Walter Corp.
Tampa, Florida
Robert Reynolds
Safeway Stores, Inc.
Oakland, California
R. Blaine Roberts
Department of Economics
University of Florida
Stephen Rubin
College of Law
University of Florida
S. W. Salsburg
Chrysler Corp.
Detroit, Michigan
Michael M. Stump
Black, Crow & Eidsness, Inc.
Gainesville, Florida
J. Taylor Thorington
Jim Walter Corp.
Tampa, Florida
G. Steven Wilkerson
Development and Alumni Affairs
University of Florida
William Wilkie
Department of Marketing
University of Florida
Eric Zolt
Center for Public Policy Analysis
Massachusetts Institute of


Part One: Mandated Standards in a Market Economy
1. Government-Mandated Inflation,
Murray L. Weidenbaum 3
2. The Rationale for Mandated Cost
Increases, Roger G. Noll 23
3. Commentaries by Laurence C. Rosenberg,
Thomas G. Moore, Marvin Kosters, and
Arnold A. Heggestad 38
Part Two: Microeconomic Effects (A)
4. Can Safety Be Mandated?, Nina Cornell 55
5. OSHA and U.S. Industry,
John F. Morrall III 60
6. Commentaries by James W. Ford,
Peter Barth, and H. Michael Cushinsky 79
7. General Discussion 87
Part Three: Aspects of Occupational Safety and Health in
the U.S.
8. The Myth and Reality, Jacqueline
Karnell Corn and Morton Corn 95
Part Four: Microeconomic Effects (B)
9. Mandated Costs: Impact on Small
Business, Milton Z. Kafoglis 111
10. Economic Analysis and Environmental
Regulation, Roy N. Gamse 129
11. Commentaries by Thomas D. Hopkins,
Roger D. Blair, and Ann R. Horowitz 139
12. General Discussion 145

xii Contents

Part Five: Aggregative Effects of Mandated Standards
13. Introductory Comments,
Louis De Alessi 155
14. Welfare Aspects of Mandated Quality,
Thomas R. Saving 157
15. Commentaries by Ira Horowitz and
George A. Hay 186
16. General Discussion 192
Part Six: Future Outlook and Reform
17. Commentaries by Howard Dugoff,
Henry L. Duncombe, Jr., Donald J.
Watson, and James C. Miller III 207

Part One
Mandated Standards in a Market Economy


Government-Mandated Inflation

Murray L. Weidenbaum

OF THE MANY WAYS in which government can affect the rate of infla-
tion, perhaps the least understood method is to require actions in the
private sector that increase the cost of production and hence the
prices of products and services sold to the public. Attention needs
to be focused on these regulatory policy instruments because their
use is becoming more widespread and neither the public nor govern-
ment decision-makers realize their full inflationary effects.'
In theory, the Federal Reserve System could offset the inflationary
effects of regulation by maintaining a lower rate of growth of the
money supply than it otherwise would. In practice, however, public
policy-makers, insofar as they see the options clearly, tend to prefer
the higher rate of inflation to the additional monetary restraint and
the resulting decreases in employment and output. Also, to the extent
that regulation results in real resources being devoted to low-payoff
activities, economic welfare is reduced.
At first blush, government imposition of socially desirable require-
ments on business through the regulatory process appears to be an
inexpensive way of achieving national objectives. Such practices ap-
parently cost the government little and represent no significant direct
burden on the taxpayer. But the public does not escape paying the
cost. For example, every time the Environmental Protection Agency
imposes a more costly (albeit less polluting) method of production
on any firm, the cost of the firm's product to the consumer will tend
to rise. Similar effects flow from the other regulatory efforts, includ-
ing those involving product safety, job health, and hiring and promo-
tion policies.

Mandated Standards in a Market Economy

These higher prices, we need to come to recognize, represent the
"hidden tax" of regulation which is shifted from the taxpayer to the
consumer. The regulatory "tax" would not be shifted in this manner
if the mandated effort-for example, environmental cleanup-were
conducted or at least financed by the government itself. Moreover,
to the extent that government-mandated requirements impose similar
costs on all price categories of a given product (such as passenger
automobiles), this hidden tax tends to be more regressive than the
federal income tax or state sales taxes. That is, the costs may be a
relatively higher burden on lower income groups than on higher in-
come groups.
It is not inevitable that every regulatory activity increase infla-
tionary pressures. In those instances where regulation generates social
benefits (such as a healthier and thus more productive work force)
in excess of its social costs, inflationary pressures should be reduced.
What is at issue is not the worth of the objectives of these govern-
ment agencies. Rather, the point is that these forms of governmental
intervention in the private sector involve heavy costs. With some
care and effort, the regulatory process might be revised so as to de-
rive at lower costs many of the same benefits as are now achieved-
if decision-makers begin to think in terms of least-cost solutions.

The Impacts of Government Regulation
Before we take on the task of investigating alternatives for reforming
the regulatory process, it may be helpful to examine more closely
the various economic effects of regulation, many of which impact
on price formation importantly but indirectly.

Regulation and Inflation
At times the impact of regulation on the prices that consumers pay
is direct and visible. For example, in the case of the passenger auto-
mobile the federal government has required the producers to incor-
porate a wide array of specified safety and environmental features.
The Bureau of Labor Statistics each year costs out the effect on the
price of the average car. Through 1974 the cumulative cost increase
per vehicle of these mandated features came to $320 (or $3 billion
for all the vehicles sold in that year).2
Government regulation increases the overhead cost of producing
goods and services by imposing a rising burden on paperwork. As of
November 30, 1976, there were 4,418 different types of approved

Government-Mandated Inflation

federal forms, excluding tax and banking forms. Individuals and
business firms spend over 143 million man-hours a year filling them
out, according to the U.S. Office of Management and Budget. A few
examples may convey a sharper impression of reality than the aggre-
gate figures. A small, 5,000-watt radio station in New Hampshire
spent over $26 just to mail to the Federal Communications Commis-
sion its application for renewal of its license-and that was prior to
the last postal rate increase. An Oregon company operating three
small television stations reported that its license renewal application
weighed 45 pounds.
At the other end of the size spectrum, Exxon Company, U.S.A.,
is required to file more than 400 reports each year to 45 federal
agencies. The Standard Oil Company of Indiana maintains 636 miles
of computer tape to store the data that it must supply to the Federal
Energy Administration.
The paperwork requirements of federal agencies inevitably produce
a "regulatory lag," a delay that can run into years and can be a
costly drain on the time and budgets of private managers as well as
public officials. The Federal Trade Commission averages nearly five
years to complete a restraint-of-trade case. It took the Federal Power
Commission eleven years to determine how to regulate the price of
natural gas all the way back to the wellhead. The regulatory lag ap-
pears to be lengthening. Ten years ago the director of planning of the
Irvine Company obtained in 90 days what was then called zoning
for a typical residential development. In 1975, a decade later, the
company received what is now called entitlement to build for one of
its developments, following two years of intensive work by a special-
ized group within the company's planning department aided by the
public affairs staff.3 The preparation of environmental impact state-
ments has become a major source of paperwork. The report for one
off-shore oil field in the Santa Barbara Channel, for example, required
nearly 1,300 pages and took two years to prepare.4
Other aspects of governmental regulatory activities also can be
costly. Several research efforts examining building regulations have
documented repeated instances of increases in the price of housing
as a result of local building codes. A study at Rutgers University re-
ported that overly stringent or outdated codes increase housing costs
by somewhere between 5 and 10 percent of total unit costs.5
Government inspectors are increasingly frequent, albeit unwel-
comed, visitors to business premises. Milk plants also experience an
extraordinary variety of inspections. More than 20,000 state, county,

Mandated Standards in a Market Economy

local, and municipal milk jurisdictions exist in the U.S. A USDA
study reveals that milk plants are inspected about twenty-four times
annually, even though the Public Health Service recommends only
two inspections a year. In one state, each milk plant averaged ninety-
five inspections during a year. One milk plant, licensed by 250 local
governments, three states, and twenty other agencies reported that it
was inspected forty-seven times in one month in 1964.

Regulation and Innovation
One hidden cost of government regulation is a reduced rate of intro-
duction of new products. The longer it takes for a new product to be
approved by a government agency-or the more costly the approval
process-the less likely that the new product will be created. In any
event, innovation will be delayed.
Professor Sam Peltzman of the University of Chicago has esti-
mated, for example, that the 1962 amendments to the Food and
Drug Act are delaying the introduction of effective drugs by about
four years as well as leading to higher prices for pharmaceutical
products.6 As a result in large part of the more stringent drug regu-
lations, the United States was the thirtieth country to approve the
antiasthma drug metaproterenol, the thirty-second country to approve
the anticancer drug adriamycin, the fifty-first to approve the anti-
tuberculosis drug rifampin, the sixty-fourth to approve the anti-aller-
genic drug cromolyn, and the one-hundred-sixth to approve the anti-
bacterial drug co-trimoxazole.7
According to Thomas G. Moore, regulation by the Interstate
Commerce Commission delayed the introduction of unit trains by at
least five years and delayed full use by the Southern Railroad of the
"Big John" cars used to carry grain.8 Ann Friedlaender has estimated
the loss due to retarded innovation at between $12 million and $41
million a year."

Regulation and Capital Formation
Federal regulation also affects the prospects for economic growth
and productivity by levying a claim on a rising share of new capital
formation. This effect of regulation is most evident in the environ-
mental and safety areas. According to the U.S. Council on Environ-
mental Quality, private outlays for pollution control in 1975 were
$3.8 billion higher than would have been the case in the absence of
federal environmental requirements.10 Similarly, the McGraw-Hill

Government-Mandated Inflation

Department of Economics estimates the cost to American industry
of meeting the occupational health and safety regulations at about $3
billion a year. Thus these two programs alone account for 6 percent
of total capital spending in the private sector of the American econ-
omy ($113 billion in 1975).
Capital formation may also be adversely affected by the uncertainty
about the future of regulations governing the introduction of new
processes and products. An example is furnished in the report of a
task force of the U.S. Energy Resources Council dealing with the
possibility of developing a new synthetic fuel industry. In evaluating
the impact of the Federal Water Pollution Control Act Amendments
of 1972, the task force reported, "It would be next to impossible at
this time to predict the impact of these requirements on synthetic
fuels production.""1
In considering the National Environmental Policy Act of 1969,
the task force stated that the major uncertainty was not whether a
project would be allowed to proceed, but rather the length of time
that it would be delayed pending the issuance of an environmental
impact statement that would stand up in court. In assessing the
overall impact of government regulatory activity on the establishment
of a new energy industry, the task force concluded, "In summary,
some of these requirements could easily hold up or permanently
postpone any attempt to build and operate a synthetic fuels plant."12

Regulation and Employment
Government regulation, albeit unintentionally, can have strongly ad-
verse effects on employment. The minimum wage law, for example,
has priced many teenagers out of labor markets. One recent study
has shown that the 1966 increase in the statutory minimum wage
reduced teenage employment in the United States by 320,000 below
what it otherwise would have been in 1972. Thus, as a result of that
one change in government regulation, the youth unemployment rate
in 1972 was 3.8 percentage points higher than it would otherwise
have been.13
In construction labor-where unemployment rates are substan-
tially above the national average-government regulation also acts
to price some segments of the work force out of competitive labor
markets. Under the Davis-Bacon legislation, the Secretary of Labor
promulgates "prevailing" wages to be paid on federal and federally
supported construction projects. A variety of studies has shown that

Mandated Standards in a Market Economy

these federally mandated wage rates are often above those that actu-
ally prevail in the labor market where the work is to be done.14

Regulation and Entrepreneurial Functions
One of the unmeasurable effects of government regulation is what it
does to the basic entrepreneurial nature of the private enterprise sys-
tem. To the extent that management's attention is diverted from
traditional product development, production, and marketing concerns
to meeting governmentally imposed social requirements, a significant
bureaucratization of corporate activity results.
In employee pension fund management, for example, the recently
enacted pension regulation has shifted much of the concern of fund
managers from maximizing the return on the contributions to a more
cautious approach of minimizing the likelihood that the managers
will be criticized for their investment decisions. It thus becomes safer
-although not necessarily more desirable for the employees cov-
ered-for the pension managers to keep more detailed records of
their deliberations, to hire more outside experts (so that the respon-
sibility can be diluted), and to avoid innovative investments.15
In the occupational safety and health area, professional safety
staffs are often diverted from their basic function of training workers
in safer operating procedures to filling out forms, posting notices,
and meeting other essentially bureaucratic requirements. The empha-
sis shifts to such trivia as a discussion of the following types of ques-
tions: How big is a hole? When is a roof a floor? How frequently
must spittoons be cleaned? Of greater concern no doubt is the detail
of the regulations. OSHA directives, for example, contain very spe-
cific requirements for virtually every piece of equipment used in the
production of steel. These requirements range from such major items
as coke ovens all the way down to such minutiae as the ladders used
in plants and the mandatory 42-inch height from the floor for por-
table fire extinguishers.
The results measured by any improvement in safety are almost
invariably disappointing. Two major studies of the occupational
safety and health (OSHA) program to date have yielded negative
findings. Nicholas A. Ashford concluded that "The OSHA Act has
failed thus far to live up to its potential for reducing job ,injury and
disease. . OSHA has had little measurable impact in reducing
injuries and deaths."16

Government-Mandated Inflation

In a more detailed statistical analysis, Robert S. Smith reported
similar findings, ". . the estimated effects [of OSHA] on injuries
are so small that they cannot be distinguished from zero.""7 Appar-
ently, the original concern of the public and the Congress to reduce
accidents has been converted to obeying rules and regulations. The
disappointing results lead to a predictable reaction: redouble the
existing effort-more rules, more forms, and more inspection; thus
there will be higher costs to the taxpayer and higher prices to the
More recent statistics on occupational injuries and illnesses are
hardly reassuring. The reported overall accident and illness rate did
decline in the most recent period for which data are available, from
10.4 per 100 workers in 1974 to 9.1 in 1975. However, the number
of workdays lost to injuries and illnesses per 100 workers actually
rose to 54.4 in 1975 from 53.1 in 1974. Apparently, on the average
the affected workers took more time off than in the previous year.
This could indicate that the injuries and illnesses that did occur in
1975 were typically more severe. Perhaps once again OSHA showed
its instinct for the trivial since it was successful in reducing the num-
ber of minor accidents and illnesses.

Approaches to Regulatory Reform
A new way of looking at the microeconomic effects of regulatory
programs is needed. A parallel can be drawn to macroeconomic
policy-making, where important and at times conflicting objectives
are recognized and attempts at reconciliation or trade-off are made
(for example, as among economic growth, employment, income dis-
tribution, and price stability). At the microeconomic level, it is like-
wise necessary to reconcile the goals of specific government programs
with national objectives.
Healthy working conditions, for example, are an important na-
tional objective but not the only important national objective. More-
over, society has no stake in selecting the most costly and disruptive
methods of achieving a higher degree of job safety. Similarly, we
should relate environmental protection, product safety, and other
regulatory efforts to costs to the consumer, availability of new prod-
ucts, and the employment of the work force. In part, this reconcilia-
tion needs to be made at the initial stages of the governmental

Mandated Standards in a Market Economy

process, when the President proposes and the Congress enacts a new
regulatory program. Governmental officials also should realize that
each increment of regulatory power reduces the extent of individual
freedom and of private-sector discretion.
Many of the proposals to reform government regulation involve
the "sunset" mechanism-the compulsory periodic review of each
major regulatory program to determine whether it is worthwhile to
continue it in the light of changing circumstances. A benefit/cost
analysis would provide a quantitative mechanism to aid in making
such value judgments.
At the operating level, the Congress, together with the administra-
tors of the statute, need to understand that there may be more than
one way of achieving a desired objective-and that the search for
more efficient solutions is not synonymous with a "green eye shade"
approach to social goals. Contrary to the allegations of many of the
traditional supporters of regulatory programs, analytical approaches
should not dilute the effectiveness of these programs. Rather, given
the limited resources available to government, analytical approaches
should in the long run enhance the accomplishments of regulatory
activities and thus strengthen their public support.

Benefit-Cost Analysis
One device for broadening the horizons of government policy-makers
and administrators is the economic impact statement. Policy-makers
should be required to consider the costs (and other adverse effects)
of their actions as well as the benefits.
This is not an entirely novel idea. In November 1974, President
Gerald Ford instructed the federal agencies under his jurisdiction to
examine the effects of the major regulatory actions on costs, produc-
tivity, employment, and other economic factors. This was a useful
first step but is subject to severe shortcomings. First of all, many of
the key regulatory agencies-ranging from the Consumer Product
Safety Commission to the Federal Trade Commission-are so-called
independent agencies, which are beyond the President's jurisdiction
in these matters.
Second, even in the case of the regulatory activities that come
within presidential jurisdiction, the existing policy is limited to the
regulations that, in the issuing agency's own estimation, are "major."
Third, the agencies covered by the Executive Order are only required
to examine the economic aspects of their actions; the weight they

Government-Mandated Inflation

give to economic factors remains in their discretion-to the extent
that Congressional statutes permit them to give any consideration to
economic influences at all.
Within these constraints, the Council on Wage and Price Stability
has intervened in many cases of proposed regulation to offer its
analyses of the benefits and the costs of the proposed action. The
agencies have rarely welcomed this advice, but the publicity given
some of the Council's analyses may have at times provided a useful
deterrent to the more traditionally minded personnel of regulatory
agencies, as well as serving a broader public educational purpose.
A broader approach is warranted, one with a strong legislative
mandate. In the fashion of the environmental impact statements (but
hopefully without as much of the trivia), it would be helpful if the
Congress enacted a statute requiring each regulatory agency to assess
the impact of its proposed actions on the society as a whole, and
particularly on the economy. Certainly the nation has no interest in
choosing the most expensive or least efficient ways of achieving its
environmental, safety, or other social goals, and should not select
Much would depend on the "teeth" put into any required economic
impact statement. Merely legislating the performance of some eco-
nomic analysis by an unsympathetic regulator would serve little
purpose beyond delaying the regulatory process and making it more
costly. However, limiting government regulation to those instances
where the total benefits to society exceed the costs would be a major
and desirable departure from current practice. Those who question
the ability to make benefit/cost comparisons of regulatory activities
should be required to justify taking these powerful government ac-
tions in the absence of adequate knowledge of their effects.
An interesting and recent case in point is furnished by some of the
perverse effects of the federal laws intended to eliminate discrimina-
tion in the extension of credit. The Equal Credit Opportunity Act of
1974 makes it illegal for creditors to discriminate on the basis of sex
or marital status. Two research studies performed by the staff of the
Federal Reserve System, however, have shown that women tend to
be better credit risks than men, other things being equal; but the
effect of the regulations designed to implement this law make it
illegal for creditors to use this information.
Therefore, the researchers conclude that women may wind up get-
ting less rather than more credit than they did before the regulation

Mandated Standards in a Market Economy

was effective-in any event that could be the effect of the specific
provision at issue. Thus, it is not feasible to assess the benefits of
the Equal Credit Opportunity Act. In contrast, the cost to creditors
of adapting to it are estimated at approximately $300 million the first
year and at an annual rate of $118 million in subsequent years. These
costs-for attorney fees, new forms, computer time, additional record
storage, etc.-will probably be passed on to consumers in one form
or another.'1
To an eclectic economist, government regulation should be car-
ried to the point where the incremental costs equal the incremental
benefits, and no further. Indeed, this is the basic criterion that is
generally used to screen government investments in physical re-
sources. Overregulation-which can be defined as regulation for
which the costs exceed the benefits-would be avoided under this

Budgeting as a Management Tool
Attention should be given to the role of the budget process in man-
aging regulation. In those cases where an agency's regulations gener-
ate more costs than benefits, the agency's budget for the coming year
should be reduced.
Budget reviewers, be they examiners in the executive branch or
committee staffs in the legislature, face the perennial question of
how to measure the effectiveness of an agency that does not provide
marketable outputs. The traditional response is to concentrate on
the inputs utilized (as, for example, workload statistics). Benefit/
cost analysis, cost effectiveness analysis (which is in effect the search
for least-cost solutions), or other quantitative forms of program
evaluation may provide useful alternatives in such cases.
Because the requested appropriations for the regulatory agencies
are relatively small portions of the government's budget, limited
attention has been given to these activities in the budget process. In
view of the large costs that they often impose on the society as a
whole (those "hidden taxes" shifted to the private sector), greater
attention than now given is warranted to the reviews of the appro-
priation requests for regulatory programs.
The wide dissemination of data on the economic impacts of gov-
ernment regulation also may serve to alter the balance of forces now
exerted by interest groups on the decision-making process. At pres-
ent, interest groups are most often well aware of the benefits they

Government-Mandated Inflation

would receive from a proposed regulation, and thus they mobilize
their forces to promote that regulation. However, information on the
adverse consequences of the regulation, if widely distributed, might
generate countervailing pressures from other groups.19

Changing Attitudes toward Regulation
Basically, however, it is attitudes that need to be changed. Experience
with the job safety program provides a cogent example. Although
the government's safety rules have resulted in billions of dollars in
public and private outlays, the basic goal of a safer work environ-
ment has not been achieved.
A more satisfying answer to improving the effectiveness of govern-
ment regulation of private activities requires a basic change in the
approach to regulation, and one not limited to the job safety pro-
gram. Indeed, that program is used here merely as an illustration.
If the objective of public policy is to reduce accidents, then public
policy should focus directly on the reduction of accidents. Exces-
sively detailed regulations are often merely a substitute-the normal
bureaucratic substitute-for hard policy decisions.
Rather than emphasis being placed on issuing citations to em-
ployers who fail to fill forms out correctly or who do not post the
required notices, it should be placed on the regulation of those em-
ployers with high and rising accident rates. Perhaps fines should be
levied on those establishments with the worst safety records. As
the accident rates decline toward some sensible average standard,
the fines could be reduced or eliminated.
The government should not be much concerned with the way a
specific organization achieves a safer working environment. Some
companies may find it more efficient to change work rules, others to
buy new equipment, and still others to retrain workers. The making
of this choice is precisely the kind of operational business decision-
making that the government should avoid but which now dominates
many regulatory programs. In the short run, however, we need to
acknowledge the obvious-that the pace of government regulation
of private-sector activities is following an upward sloping trend line.
This is clearly evident in any analysis of recent legislation.
In the Ninety-fourth Congress, for example, the following were
some of the many extensions of federal activity which were enacted:
The Resource Conservation and Recovery Act of 1976 (providing
federal regulation of hazardous wastes), the National Forest Man-

Mandated Standards in a Market Economy

agement Act of 1976 (restricting clear-cutting of timber on national
lands), the Tax Reform Act of 1976 (denying tax benefits for costs
of demolishing or replacing historic buildings), the Coastal Zone
Management Act Amendments (giving the states greater review
power over federal permits for off-shore developments), the Solid
Waste Management Act, the Toxic Substances Control Act (setting
up a new system of controls over wide categories of products), the
"Redlining" Disclosure Act of 1975, the Securities Act Amendments
of 1975 (expanding the SEC's powers to include firms that record
and publish stock quotations and transactions), and the Energy Pol-
icy Conservation Act of 1975.
We need to acknowledge of course that many of these programs
were enacted to deal with serious environmental or social problems.
The point being made here is that, regardless of their desirability,
these new laws all represent extensions of federal authority over the
private sector and therefore may entail significant costs as well as

Alternatives to Regulation
The effects of a continuation of the current wave of expansion in
government regulation of business need to be viewed in a long-term
context. When we examine the sector of industry that already is most
subject to government supervision-defense production-the results
are disconcerting. It is precisely the companies that are most heavily
dependent on military contracts that report the largest cost overruns
and the greatest delays. The society does not get the benefit of effi-
ciency and innovation expected from private industry. The ultimate
consequences of governmental assumption of basic entrepreneurial
and managerial functions are surely a topic worthy of considerable
attention and study.
The promulgation by government of rules and regulations restrict-
ing or prescribing private activity of course is not the only means of
accomplishing public objectives. Codes of behavior adhered to on a
voluntary basis may often be effective.2" Trade associations on occa-
sion have served such a socially useful function in upgrading the
level of business performance.21
Government itself has available to it various powers other than the
regulatory mechanism. Through its taxing authority the government
can provide strong signals to the market. Rather than promulgating
detailed regulations governing allowable discharges into the nation's
waterways, the government could levy substantial taxes on those

Government-Mandated Inflation

discharges. Such sumptuary taxation could be "progressive," to the
extent that the tax rates would rise faster than the amount of pollu-
tion emitted by an individual polluter. Thus, there would be an in-
centive for firms to concentrate on removing or at least reducing
the more serious instances of pollution.
The use of taxation would be meant neither to punish polluters
nor to give them a "license" to pollute. Rather it would be using the
price system to encourage producers and consumers to shift to less
polluting ways of producing and consuming goods and services. The
cost of removal of pollution for each organization, compared to the
size of the tax, would determine the level of environmental cleanup
that it pursues. Those that can control pollution more cheaply will
clean up more (and thus pay less tax). Those with higher control
costs will clean up less (and pay more pollution taxes). As pointed
out by Marc Roberts and Richard Stewart, this approach attempts
to achieve a given level of environmental quality with minimum
resource use by equalizing the marginal cost of pollution control.22
In the case of the traditional one-industry type of government
regulation (as of airlines, trucking, and railroads) a greater role
should be given to the competitive process and to market forces.
Unlike the newer forms of regulation on which this paper concen-
trates, the older forms of regulation are often mainly barriers to
entry into a given industry, protecting existing firms from competi-
tion by potential new entrants. It is in this limited sense that deregu-
lation is a viable option. The elimination of regulation in the safety,
ecology, and related areas does not appear to be a realistic alterna-
tive in view of the nation's long-term social concerns.
Indeed, any realistic appraisal of government regulation must ac-
knowledge that important and positive benefits have resulted from
many of these activities-less pollution, fewer product hazards, re-
ducing job discrimination, and other socially desirable goals of our
society. However, the "externalities" generated by federal regulation
cannot justify the government's attempt to regulate every facet of
private behavior. As Henry Owen and Charles Schultze have pointed
out, a reasonable approach to this problem requires great discrimina-
tion in sorting out the hazards that are important to regulate from
the kinds of lesser hazards that can best be dealt with by "the normal
prudence of consumers, workers, and business firms."23
Perhaps after all, and surprisingly to some, Professor F. A. Hayek
said it best in his Constitution of Liberty and provided the final note
for this paper: "a free market system does not exclude on principle

Mandated Standards in a Market Economy

. . all regulations governing the techniques of production. . They
will normally raise the cost of production, or what amounts to the
same thing, reduce overall productivity. But if this effect on cost is
fully taken into account and it is still thought worthwhile to incur
the cost to achieve a given end, there is little more to be said about
it. The appropriateness of such measures must be judged by compar-
ing the over-all costs with the gain; it cannot be conclusively deter-
mined by appeal to a general principle."24

1. This paper draws heavily from the author's previous work, especially
Government-Mandated Price Increases (Washington: American Enterprise In-
stitute, 1975); "Reducing Inflationary Pressures by Reforming Government
Regulation," in Contemporary Economic Problems, ed. William Fellner (Wash-
ington: American Enterprise Institute, 1976); and Business, Government, and
the Public (Englewood Cliffs, N.J.: Prentice-Hall, 1977).
2. Weidenbaum, Government-Mandated Price Increases, p. 26.
3. Richard M. Geiler, "Development Regulations Must Be Reasonable,"
Urban Land (October 1976): 3.
4. Don Dedera, "What on Earth Is an EEE-EYE-ESS?," Exxon USA (1st
Quarter 1976): 14.
5. George Sternlieb and David Listokin, "Building Codes, State of the Art,
Strategies for the Future" (Report submitted to the HUD Housing Review
Task Force, June 1973).
6. Sam Peltzman, "An Evaluation of Consumer Protection Legislation: The
1972 Drug Amendments," Journal of Political Economy (September/October
1973): 1090.
7. Testimony by Dr. William Wardell, University of Rochester School of
Medicine and Dentistry, before the Senate Committee on Labor and Public
Welfare, Subcommittee on Health, Washington, September 27, 1974.
8. Thomas G. Moore, Statement before the Senate Subcommittee on Trans-
portation and Aeronautics, 92d Congress, Serial #92-79, p. 1337.
9. Ann Friedlaender, "The Social Costs of Regulating the Railroads," Amer-
ican Economic Review (May 1971): 226.
10. U.S. Council on Environmental Quality, Fifth Annual Report (Wash-
ington: Government Printing Office, 1975), p. 175.
11. Synfuels Interagency Task Force, Recommendations for a Synthetic
Fuels Commercialization Program, Report submitted to the President's Energy
Resources Council (Washington: Government Printing Office, 1975), vol. 1,
p. C-22.
12. Ibid., p. 134.
13. James F. Ragan, Jr., "Minimum Wage Legislation and the Youth Labor
Market," Working Paper No. 8 (St. Louis: Center for the Study of American
Business, Washington University, 1976), p. 29.
14. John P. Gould, Davis-Bacon Act (Washington: American Enterprise
Institute, 1971); Armand J. Thieblot, Jr., The Davis-Bacon Act (Philadelphia:
University of Pennsylvania, The Wharton School, 1975).

Government-Mandated Inflation

15. Shoya Zichy, "How Small Funds Are Coping with the New Pension
Law," Institutional Investor, September 1975, pp. 19-20.
16. Nicholas A. Ashford, Crisis in the Workplace: Occupational Disease
and Injury (Cambridge: MIT Press, 1976), p. 17.
17. Robert S. Smith, The Occupational Safety and Health Act (Washing-
ton: American Enterprise Institute, 1976), p. 70.
18. James F. Smith, "Effects of Regulation on the Price and Availability of
Consumer Credit" (Summary of remarks before the National Economists Club,
Washington: November 23, 1976), p. 1.
19. Roland McKean, "Property Rights within Government, and Devices to
Increase Efficiency in Government," Southern Economic Journal (October
1972): 177-86.
20. See Roland McKean, "Economics of Ethical and Behavioral Codes."
Working Paper No. 11 (St. Louis: Center for the Study of American Business,
Washington University, 1976).
21. See Michael S. Hunt, "Trade Associations and Self-Regulation: Major
Home Appliances," in Regulating the Product: Quality and Variety, ed. Rich-
ard E. Caves and Marc J. Roberts (Ballinger, 1975), pp. 39-55.
22. Marc J. Roberts and Richard B. Stewart. "Energy and the Environ-
ment," in Setting National Priorities: The Next Ten Years, ed. Henry Owen
and Charles L. Schultze (Washington: Brookings Institution, 1976), p. 424.
23. Owen and Schultze, "Introduction," ibid., p. 10.
24. F. A. Hayek, Constitution of Liberty (Chicago: University of Chicago
Press, 1960), pp. 224-25.

Appendix-The Rising Cost of Government Regulation

The direct cost of federal government regulation of business hit a
high of $2.9 billion in the fiscal year 1976. On the basis of appropri-
ations already enacted, these costs are estimated to reach $3.5 billion
in the current fiscal year, a 21 percent increase. The budget for fiscal
1978 recently transmitted by the outgoing Ford administration would
bring the annual total federal regulatory cost to approximately $3.8
billion next year, an 85 percent rise over the 1974 level.
As shown in Table 1.1, the budgets of the regulatory agencies
continue on a steadily upward growth trend. Any additions that will
be proposed by the Carter administration-such as a new consumer
advocacy agency or a strip mining control law-would push these
numbers higher still, if not offset by reductions in other areas.
The most expensive sectors of federal regulatory activity are not
the traditional economic regulating commissions which typically focus
on an individual industry. Rather, the largest budgets have been
assigned to a broad category of social activities including environ-

18 Mandated Standards in a Market Economy

ment, energy, and consumer safety and health; the major agencies
in this latter group include the Environmental Protection Agency,
the National Highway Traffic Safety Administration, the Food and
Drug Administration, and the Animal and Plant Health Inspection
Service. The jurisdiction of these agencies generally extends to the
great bulk of the private sector, including manufacturing and service
industries which are not generally thought of as "regulated." Tables
1.2 through 1.6 contain detail on each of the major sectors of federal

N 'l N fir
00 00 t- 0N 00

vl t e % \0t

VI N0 Nt-
.1- t~ \ r


S0 D 0 S C4 00

3s 00> It 00 N (
Z 'p-N ( -o

- 'O O0 N -OO MC =
N 0 c )0 O- 0. &
s- eU M C

0 U
3 I- II VIm )'; I
0 a 00 e 00 alOO 30
-- c". iBrc-

e 08 '

In en z u
C wa
in CO V)

0\ N VI \t f
- N o N en

p- i-* g g

O 0' 00 Cfm C1 -\ 0 i S c
o\ t~ m *< w < ^ o. ,
e. m "' Hy 0
N w *^ 1-"z Ui-B~-

en r4n BI

*^ 0 .1 '.0 y\0
oa o N, u, C
0\ -< 'O fl (MM
0 m m ti
-" N

0 0 to*
G 1. m

C >0 L5 0 0
0~o ~




















W .0


w .4



Mandated Standards in a Market Economy

(Fiscal years, millions of dollars)


Department of Agriculture
Animal and Plant Health
Inspection Service
Packers and Stockyards
Department of Health, Education,
and Welfare
Food and Drug Admin-
Department of Housing and
Urban Development
Office for Consumer Affairs
and Regulatory Functions
Department of Justice
Antitrust Division
Drug Enforcement Admin-
Department of Transportation
National Highway Traffic
Safety Administration
Federal Railroad Admin-
Department of the Treasury
Bureau of Alcohol, Tobacco,
and Firearms
Customs Service*
Consumer Product Safety
National Transportation
Safety Board


1974 1975 1976 1977 1978

314 345 377 429 437

4 5 5 6 6
318 350 382 435 443

165 201 218 240 277

1 2 **

14 18 21

98 132 146 176 184
112 150 167 203 213

157 150 151 182 224

7 9 15 19 21
164 159 166 201 245

79 95 103 122 135
228 299 334 358 383
307 394 437 480 518

38 46

11 14

1,094 1,299 1,419 1,621 1,755

*Activities extend beyond business regulation (breakdown not available).
**Less than $1 million.

Government-Mandated Inflation

(Fiscal years, millions of dollars)

1974 1975 1976 1977 1978

Department of the Interior
Mining Enforcement and
Safety Administration
Department of Labor
Employment Standards
Labor-Management Services
Occupational Safety and
Health Administration
Equal Employment Opportunity
National Labor Relations
Occupational Safety and
Health Review Commission
Federal Metal and Nonmetallic
Mine Safety Board of Review


59 68 84 102 105

56 72 84

24 27 37

99 106

49 54

69 90 109 129 132
149 189 230 277 292

56 59

61 67

68 72

82 88

5 5 6 7 7

310 379 446 536 564
310 379 446 536 564

*Less than $1 million.

(Fiscal years, millions of dollars)

Agency 1974 1975 1976 1977 1978

Environmental Protection
Agency 232 317 363 471 487
Federal Energy Administration 33 121 136 237 269

Total 265 438 499 708 756


Mandated Standards in a Market Economy

(Fiscal years, millions of dollars)

Agency 1974 1975 1976 1977 1978

Cost Accounting Standards
Board 1 1 1 2 2
Council on Wage and Price
Stability 1 2 2
Securities and Exchange
Commission 35 44 51 58 58

Total 36 45 53 62 62

*Less than $1 million.

(Fiscal years, millions of dollars)

Agency 1974 1975 1976 1977 1978

Civil Aeronautics Board
Commodity Futures Trading
Federal Communications

89 81 91 99 93

3* 3* 11 14 13

Commission 38 48 53 57 60
Federal Maritime Commission 6 7 8 9 9
Federal Power Commission 27 34 36 43 43
Federal Trade Commission 32 39 44 54 58
International Trade Commission 7 8 10 12 12
Interstate Commerce Commission 38 44 47 61 62
Nuclear Regulatory Commission 80 86 180 236 271
Renegotiation Board 5 5 6 6 6


325 355 486 591 627

*Expenditures for Commodity Exchange Authority.


The Rationale for Mandated Cost Increases

Roger G. Nol

THE PURPOSE OF THIS paper' is to explore the reasons why standards
regulation is imposed upon private enterprise. The unfelicitous phrase
"mandated cost increase" is taken to mean a government legislative
or administrative action that increases the amount of resources that is
required to produce a unit of output. The purpose of the action is
taken to be to improve the quality of a product in its primary use or
to improve such secondary consequences of the product as the en-
vironmental, health, and safety effects of its production and use. The
form taken by these interventions is the setting of performance or de-
sign standards for products, production processes, and places of work.
Alternatives to standard-setting regulation include (1) no govern-
mental intervention, which implies reliance on markets and an ex-
tensive body of civil law that defines and enforces various rights;
(2) reliance on informational strategies, including publicly supported
creation and dissemination of information and mandatory require-
ments that business provide information about products and work-
places; and (3) the use of economic incentives-taxes and bounties-
to achieve in more decentralized fashion the same objectives of the
standard-setting process.
A rationale for standards regulation has two components. The first
consists of an explanation of why government intervention into pri-
vate decisions might be the result of a rational decision-making proc-
ess, rather than a result of electoral, legislative, and bureaucratic
stupidity. The second consists of a demonstration that in some em-
pirically plausible circumstances, standards may be at least as efficient
as the best alternative public strategy. The paper does not deal with

Mandated Standards in a Market Economy

an equally important issue, the problem of a failure of policy owing
to the presence of perverse incentives operating on government of-
ficials.2 Certainly any practical evaluation of government intervention
must take account of problems of regulatory failure.
This paper offers no defense for the long list of inappropriate, even
silly, standards that have been promulgated by regulatory agencies.
Government probably does have a tendency systematically to under-
value informational strategies in relation to standards. The principal
point of the paper is that, all things considered, a reasonable theoreti-
cal argument can be made for considering standards in certain cir-
cumstances. Of course, such a conclusion can only be based upon a
second-best argument: standards make sense in large measure be-
cause the alternatives can be worse, rather than because they have
great intrinsic merit.

Reasons for Intervention
The core of the argument for standards lies in the explication of the
sources of a political demand for government intervention into market
transactions to deal with issues such as product quality, environmental
degradation, or the safety and healthfulness of products and places of
work. A discussion of the rationales for intervention naturally begins
with an examination of the sources of failure in the combined market
and civil liability system. The perspective taken in the following dis-
cussion of these issues is distinctly that of the neoclassical economist.
Individual preferences are taken to be normatively meaningful and
stable, and behavior is assumed to be motivated by the desire to at-
tain the most preferred feasible bundle of public and private goods.
The paper does not deal with rationales for standards that are based
upon the manipulability of individual preferences or that are pater-
nalistic (for example, no matter how much information is available
about the consequences of individual actions, some people must be
protected against their wills from systematically making decisions
that cause them damage).
Oi (1973) has provided a relatively complete statement of the
mechanism by which markets and liability law interact to deal with
the kinds of qualitative issues with which standards are designed to
deal. While his analysis deals directly with the issue of the quality of
consumer products, the results may be easily generalized to cover the
other issues as well. The basic microtheoretic result is that if all in-
stitutions (markets and courts) operate costlessly, if all of the con-

The Rationale for Mandated Cost Increases

sequences of an economic transaction bear directly upon the parties
to the transaction, and if the parties are fully informed about these
consequences, the resulting allocation of resources will be economi-
cally efficient.
That some of the consequences of a transaction may be possibili-
ties but not certainties creates no special problems for the theory as
long as insurance can be purchased to cover risks. Responsibility for
the consequences of contingent events is divided between the parties
by the civil liability system. Economic agents pay insurance premiums
equal to the expected values of their shares of these consequences,
which are included in their calculations of the net price of the transac-
tion. Alternative divisions of liability affect the ultimate allocation of
resources only insofar as they affect the distribution of income and,
thereby, the pattern of product demand and factor supply; all divi-
sions of liability between participants in a market, under the preceding
assumptions, produce an economically efficient result.

Incomplete Information among Producers
Oi introduces informational problems into his model asymmetrically
by investigating the consequences of two sources of incomplete in-
formation that prevent producers from knowing the value of the dam-
age that their products and workplaces inflict upon clients and workers.
His first case retains the assumption regarding full information in
only a technical sense. For example, the probability that a given de-
sign feature of a product will injure a customer is known by both
producer and consumer. But individual differences in the monetary
value attached to a particular hazard are known only to the person
accepting the risk. If a producer knew the economic value of a hazard
to a particular customer or worker, the producer could set individ-
ualized prices and wages that would effectively insure against the pro-
ducer's liability for the contingent event. However, if individual dif-
ferences cannot be measured by the producer, or can be measured
only at a cost, assigning liability to the producer produces an ineffi-
ciency because prices cannot be set separately for each person to in-
clude the optimal insurance policy against the contingent event. By
contrast, in a caveat emptor economy, consumers and workers incor-
porate their personal preferences about hazards and risks into de-
cisions about consumption, occupation, and insurance.
Oi's second relaxation of the full information assumption is to in-
corporate the possibility that consumers (or workers) can affect the
probability that they will be harmed by a product in ways that do

Mandated Standards in a Market Economy

not affect the compensation they will receive from a liable producer
should the contingent event take place. In this circumstance, Arrow's
(1963) "moral hazard" problem arises. In what is essentially an ap-
plication of Calabresi's (1970) "least cost avoider" principle, the pres-
ence of moral hazard leads to the conclusion that producers should
not be liable; but the conclusion here is not well founded. Presumably,
insurers will face an informational problem similar to the one faced
by producers so that moral hazard will create inefficiencies in the
market for insurance against occupational and product risks. Never-
theless, insurers could have more opportunity to assign a person to
an approximately correct risk pool because they can gain claim ex-
perience for each person across a wide variety of economic activities-
in contrast to producers' experiences, which are limited to events in-
volving their own products and places of work. If so, caveat emptor
may be preferred to full producer liability, but as a second-best rather
than an optimal solution.

Inefficiencies of Civil Law
The inefficiencies of the civil liability system provide an additional
set of arguments against reliance on full producer liability. Most ob-
viously, when civil law suits are so expensive, even if they are settled
prior to trial, litigation to recover damages involves a net loss to the
plaintiff. To the extent that both litigation costs and damages are paid
by a negligent party, excessive incentive is created to improve the
quality and safety of products and places of work; to the extent that
litigation costs are paid by the damaged party, incentives to recover
damages that are less than the litigation costs are eliminated. In prac-
tice, the litigation costs of the damaged party are divided between the
litigants so that both inefficiencies are possible and at least one is
Another difficulty with the civil liability system arises if the cause
of a contingent event can only be determined probabilistically, even
after it occurs. Although the doctrine of contributory negligence mud-
dies the waters to some extent, in general the civil liability system is
designed to assign responsibility for a contingent event either to a
particular economic agent or simply to nature. This can be interpreted
as assigning liability to a particular agent if the probability that the
agent caused the event (or could reasonably have prevented the event)
exceeds some minimum. This probability is the standard of proof;
the burden of proof can be interpreted as assigning to one of the liti-

The Rationale for Mandated Cost Increases

gants the responsibility for showing that the standard of proof is or is
not satisfied. In civil law, the standard of proof is less than the com-
plete certainty of the "no reasonable doubt" standard of criminal law.
In other words, the civil standard is that a "preponderance of evi-
dence" supports the proposition that a particular party was respon-
sible for the event in that reasonable actions to avoid the event were
available but were not taken. A quantitative interpretation of this
standard is impossible; however, the minimum probability of causal-
ity that is sufficient to establish liability is probably greater than 50
percent but less than unity.
The burden of proof in civil cases usually rests with the party
claiming to be damaged. Should the plaintiff satisfy the standard of
proof, the compensation that is awarded is intended to be the full
damages. Thus, defendants do not pay the expected value of the dam-
ages they create; instead, defendants who contribute to the likelihood
of a damaging event by more than the standard pay all damages,
while those who contribute less than the standard pay nothing.
As a practical matter, a system of complete producer liability is
impossible within this conceptual structure of civil law. In particular,
civil law provides too great an incentive to reduce hazards that, ac-
cording to the civil standards of proof, can be blamed on producers,
and no incentive to reduce hazards which have numerous possible
causes, one of which is producer negligence.
For example, machinery-related injuries in the workplace would
tend to fall in the first category, while long-term health effects of the
working environment would tend to fall into the second. Proving that
a worker contracted cancer because of a hazardous substance in the
working environment is essentially an impossible task because the
illness is likely to have numerous other potential causes and because
a relatively small proportion of workers are likely to contract cancer
in a particular working environment.
A third inefficiency of the civil liability system arises from its im-
plicit prohibition of certain transactions. One might expect that if the
costs of preventing a particular hazard exceeded the expected dam-
age, a liable producer would negotiate an agreement with the party
at risk to accept the hazard and agree not to hold the producer liable
for the contingent event. In most instances, civil law does not recog-
nize such an agreement as enforceable. This doctrine not only stands
in the way of those who wish to become indentured servants-a con-
stitutional issue-but also renders meaningless the prose on parking

Mandated Standards in a Market Economy

receipts that alleges to limit the liability of the parking lot should an
automobile be damaged while in the care of its agents. Recently this
issue has become important in the debate about medical malpractice
litigation. One proposed solution to the problem is for doctors and
hospitals to charge lower fees for patients who agree to waive the
right to sue for malpractice; however, these waivers usually could not
be enforced should the patient later decide that malpractice had tran-
The final difficulty of the civil liability system to be examined here
arises from the problems associated with awarding compensation.
First, compensation is limited to, at most, the net worth of the negli-
gent legal entity, including insurance coverage. Should damages ex-
ceed the limit, the defendant is driven to bankruptcy. Second, the re-
lationship between damages and compensation is tenuous at best. In
part, this arises from the problems raised by Oi of individual dif-
ferences in valuing hazards, although a court is in a better position to
compel honesty than is a party to a transaction. Nevertheless, litiga-
tion does not constitute a forum in which the incentives are likely to
be present for true revelation of economic values. Courts naturally
focus on the monetizable aspects of damages: medical costs, loss of
income, trial costs, and so on. However, as Bergstrom (1976) has
shown in the context of evaluating human life, these establish neither
a minimum nor a maximum bound on the true cost of injury to an
injured person.
The analysis thus far supports the conclusion that a system of com-
plete producer liability is both inefficient and, within the structure of
the operating rules of civil law, impossible. No case has yet been
made for government intervention. What has been shown is that a
civil liability system with producers fully liable will not produce an
efficient result if, for some reason, a caveat emptor economy fails to
achieve efficiency by means of the incentives provided through market

Costly Information
One source of problems in the caveat emptor world is that informa-
tion is not fully available at zero cost. The presence of costly infor-
mation complicates models of the market economy, especially if the
method of acquiring information is by sequential search. Economic
theory is not very well developed in this area, but thus far, as Wilde
and Schwartz (1977), for example, have indicated, strong assump-

The Rationale for Mandated Cost Increases

tions are required for search models to have normatively interesting
solutions. But even neglecting the search problem, costly information
creates an inefficiency in a caveat emptor world because information
is a public good. Information about the quality of a product and the
hazards of a place of work is useful to all of its customers and work-
ers. For each consumer or worker to assemble the same information
constitutes duplication of effort and, all other things being equal, vio-
lates the principle of assigning responsibility to the least-cost avoider.
Facing information cost asymmetries, individuals can demand in-
formation as a condition of purchase, but this does not solve the
problem. In the absence of a contractual agreement to the contrary,
the information that is provided need not be complete, in which case
persons must learn by experience. If so, the problem is back on the
uncertain turf of search models. With a contract, liability is shifted
back to the producer, in which case all the problems of the civil jus-
tice system reenter the picture. Moreover, writing contracts raises
transactions costs even if contracts are enforceable through civil liti-
gation. Williamson's (1975) theory of vertical integration is based
largely on the inherent impossibility of iron-clad contracts. He con-
cludes that vertical integration may be motivated by the impossibility
of an efficient market due to contracting problems. Exactly the same
problems, but without Williamson's solution, can impede product and
factor markets as well.
Still another possibility is that an information industry will emerge,
providing independent assessments of products and occupations. Of
course, such an industry does exist, but the nature of its product as-
sures that it will be inefficiently small. The industry must sell infor-
mation at a price that covers the costs of dissemination and a portion
of the costs of generating the information. The latter is a public good,
and, consequently, does not affect the marginal cost of disseminating
the information. Including it in the price causes too little use of the
information. Moreover, potential users of the information have an
opportunity to avoid payment by engaging in sharing arrangements
with other users.
If technical information about hazards is supplied, an economic
agent can convert the technical information into an assessment of the
value of engaging in an economic transaction. This kind of informa-
tion processing also is costly. Moreover, if one economic agent can
find another agent with similar preferences, information-processing
costs can be cut in half if the first simply delegates to the second the

Mandated Standards in a Market Economy

job of making decisions for both. If preferences are not the same for
both agents, a second-best calculation determines the desirability of
delegation: assigning decision-making responsibility to another agent
is best if the divergence of the bundle of goods that the agent acquires
from that which is optimal is less of an economic loss than the costs
of individualized information processing.
Individuals are likely to find that collective decisions about con-
sumption are more attractive when (1) individual preferences are
more homogeneous and (2) the relevant information is more com-
plex and, therefore, more costly to assemble, process, and interpret.
Collective decisions, by reducing individual information-processing
costs, increase the amount of information that is optimal as a basis
for decisions, and the optimal amount of information increases as the
size of the group that is sharing the cost increases. By the same token,
a group may have to become more heterogeneous in order to grow
larger, in which case collective decisions become more inefficient in
terms of the mean divergence of the group decision from that which
is best for each member. Of course, to solve problems of cost-sharing
and potential free-riding, collective action to conserve on decision
costs requires coercive powers, which implies the use of government
as the means of carrying out collective processes.

External Costs
Since the concept of an external cost is not only relatively well de-
veloped theoretically but has even seeped into the popular lexicon,
elaborate discussion of externalities is unnecessary. Briefly, an ex-
ternal cost arises if the production or use of an economic good re-
duces the welfare of members of society other than those who were
directly engaged in its production and consumption. Unless a mech-
anism exists that causes external costs to be considered by those
making decisions about production and consumption activities, ra-
tional economic behavior will lead to excessive use of goods that gen-
erate external costs.
Normally, external costs are regarded rather technically as taking
the form of involuntary inputs to some other production or consump-
tion activities. An example is the undesirable effect of automotive
transportation on air pollution and highway congestion.
In some instances, the institutional arrangements surrounding a
particular economic activity may be a source of external costs. This
arises when part of the direct cost of providing a private good is not

The Rationale for Mandated Cost Increases

paid by the person who receives the good. For example, private health
insurance reduces the financial risks of illness by spreading part of
the cost of medical services among the insured group. As long as in-
surance coverage and reimbursement are not strictly tied to the risks
to health that are accepted by an insured individual, insurance creates
the familiar problem of "moral hazard" in which individuals have too
little incentive to take actions that reduce their chances of injury or
illness. By accepting greater risks, an individual imposes external costs
on other insured parties because their premiums will be raised to
cover part of the higher expected medical costs that are the conse-
quence of more risky behavior. Of course, moral hazard is a con-
sequence of information failures, as discussed above.
As Coase (1960) has shown, the presence of an external cost does
not necessarily lead to the conclusion that government intervention
is necessary to achieve economic efficiency. In fact, reflection on why
external costs lead to collective action reveals that externality, per se,
is not the central cause of a demand for government intervention.
Suppose that the rights of economic agents are well defined, and a
particular external cost is created solely by one economic agent and
borne exclusively by another. In this case, negotiations between the
two agents and, as a last recourse, civil litigation offer alternatives to
governmental intervention through legislative and administrative ac-
tion. The reason that either the receptor or the source would have for
selecting the last alternative would not be due to the special nature of
external costs, but because some agent perceived that a better private
result could be obtained through interventionist strategies rather than
through the other alternatives. The reasons for such a perception are,
in general, the same as the reasons that parties to any private transac-
tion might seek intervention to increase their bargaining strength.
When more than one economic agent indivisibly suffers a particu-
lar external cost, a caveat emptor market mechanism alone cannot
produce an efficient allocation of resources. In this instance, the ex-
ternal cost cannot be reduced for one receptor without reducing it for
all. In the absence of collective action among receptors to enter the
market as a group, individual receptors will respond to the private
costs and benefits of alternative actions to reduce the external cost.
Although all receptors benefit from an action that reduces the external
cost, each individual in responding to only the personal benefits will
lack sufficient incentive to cause an economically efficient degree of
reduction in the external cost suffered by all.

Mandated Standards in a Market Economy

Economists tend to think of government legislative action as the
only alternative to the market mechanism for solving the problem of
shared external costs, but in reality the civil liability system also deals
with this problem. A liability system that is costless and that accords
persons the right to be completely free of externally imposed costs
effectively solves the problem of even shared external costs whenever
the external cost is attributable to a specific source. In this case, the
receptors each sue the source for the damages they experience, and
each source faces the appropriate incentive for abating the external
cost. Collective action among those suffering externally imposed costs
is not required for the costless liability system to produce an efficient
Of course, the preceding paragraph is full of counterfactual assump-
tions about the efficacy of a system of producer liability; however,
the invalidity of these assumptions undermines the effectiveness of the
liability system for dealing with any problem of civil law, not just
problems of external costs. Externality per se does not stand as an
additional source of failure of the civil liability system. Thus, the
principal consequence of shared externality is that the receptors can-
not effectively resort to use of the market without organizing for

Market Failure and Liability Law
It is tempting to write off the law as merely a mechanism for dealing
with equity issues. Surely the legal literature provides ample ammuni-
tion for this point of view by its ready references to concepts of fair-
ness and by its common failure to distinguish economic efficiency from
the distribution of income. However, as the preceding discussion im-
plies, liability law can also be interpreted as an attempt to solve the
problems of incomplete, costly information and external costs in a
decentralized, individualistic manner. Producer liability is a natural
response to a world in which contracts that covered all conceivable
eventualities would be too costly to write to be worthwhile; informa-
tion is costly and available at different costs and with different quality
to different people, and economic activities affect people who are not
directly involved in them. In principle, producer liability can remove
the need to engage in collective action in order to capture informa-
tional or contractual economies of scale, or to solve the coordination
problem in acting to reduce an indivisible external cost.
Unfortunately, the civil liability system does not work very well
when individual damages are small or difficult to measure or when

The Rationale for Mandated Cost Increases

liability is difficult to ascertain. Thus, informational problems of the
kind that make producer liability seem attractive on the basis of the
principle of least-cost avoider also are likely to strain the effectiveness
of civil litigation.

Taxes, Subsidies, and Standards
In this section, the comparative advantages of alternative interven-
tionist strategies are examined. Most of the discussion centers on the
comparison between taxes and standards, and the relationship of the
efficacy of taxes to the nature of the problem that the policy inter-
vention is designed to solve. Before examining these issues, a few
words are in order, for completeness, about "clean-up" subsidies.

As a purely theoretical matter, taxes and subsidies are normally re-
garded as identical as a means of inducing better but more costly
production methods. The basis for this conclusion is the Coase (1960)
argument that the assignment of property rights does not affect the
ability of a market system to achieve an efficient allocation of re-
sources. A tax implicitly makes producers liable for whatever unde-
sirable consequence is traced to a production process, whereas a
subsidy implicitly gives producers a right to continue imposing that
Regardless of the merits of the equity issues involved in assigning
property rights, subsidies pose two problems. The first is how to deal
with entrants into the industry. If entrants, too, are subsidized to
avoid polluting, the effect is to induce overproduction in the industry.
This is because entrants are given property rights in resources do-
nated to abatement, which amounts to a redistribution of income in
favor of entrants. Both sources of revenue, sales from production and
income redistribution, will be considered by firms contemplating en-
try; yet only the former is relevant to the efficient expansion of the
industry. If entrants are not subsidized nor subjected to pollution
standards, they have implicitly been given property rights to impose
uncompensated pollution costs on others. If so, the entrant will face
costs and charge prices that, in relation to other products with fully
internalized costs, are too low.
The second difficulty with subsidies arises from their distributive
character as governmental programs. Specific facilities would make
investments that qualified them for payments. The difficulty with all

Mandated Standards in a Market Economy

such specific, targeted expenditure programs is that Congress faces a
great temptation to convert them into pork barrel programs. Each
legislator will desire a fair share of subsidies for the home district,
and, as Weingast (1978) has shown, each legislator has an incentive
in this case to act in a manner that causes excessive expenditures for
the program as a whole. The problem is exacerbated if measures of
the social benefits of the program are elusive and ripe for manipula-
tion. For the problems at issue in the paper, this is sure to be the
case because informational problems give rise to the political demand
for these interventions.
By comparison, both taxes and standards fully internalize the costs
of achieving the reduction in hazards that is the objective of the in-
tervention, and do so without introducing other, extraneous expendi-
ture incentives in the legislature. They result in price structure, re-
gardless of the state of entry, that incorporates either the costs of
abatement to the standard or the pollution tax.

The conceptual basis for taxes, as opposed to standards, is that the
government need not know as much to implement a tax scheme as a
standards scheme. All the government need do is estimate the bene-
fits to be derived from an improvement in performance, as measured,
say, by injuries for safety regulation or by emissions for external costs
of pollution, and then set a tax per unit on the undesirable perfor-
mance characteristics that is based on the estimate of the value of
improvement. The government need not know anything about the
costs of improvements; firms that face costs of improvement that fall
short of the tax will improve, while others will simply pay the tax.
The case for taxes is, of course, weakened if the benefits of im-
provement are much more difficult to measure than the costs of im-
provement. The preceding section emphasized that one source of
demand for government intervention was the absence of sufficient in-
formation about individual valuations of improvements in technical
hazards. This measurement problem lies at the heart of Oi's initial
argument against producer liability and of Arrow's discussion of
moral hazard. Moreover, if collective action does lead to more in-
formation about hazards but less dissemination and evaluation of it
by individual actors, market artifacts that produce measures of the
value of hazard reductions in the face of new information will not be
present. In short, the very informational problems that give rise to

The Rationale for Mandated Cost Increases

the demand for intervention can make the selection of a tax funda-
mentally arbitrary.
Nevertheless, standards have a potential difficulty not shared by any
tax, however arbitrary. If firms are relatively heterogeneous in their
cost structures, a given uniform standard will not be the least-cost
method of achieving the overall performance objective. Firms with
low costs of improved performance will improve too little compared
to firms with high costs of improved performance.
Conversely, if firms can be separated according to the costs of im-
proved performance that they face, and if costs are relatively homo-
geneous within groups, then standards can be set separately for each
group in such a manner that the marginal cost of improvement is the
same for every firm. From an economic efficiency perspective, such a
standard system is equally effective as the equivalent tax.
From a political perspective, the standard system is probably pref-
erable to industry. First, standards are less costly because an equiva-
lent tax system imposes both an abatement cost and a tax payment to
cover whatever pollution is not abated, whereas a standard system
imposes only the former. Moreover, if the intensity of industry oppo-
sition to a policy influences the objectives of the policy, the standards
system, by imposing less cost on firms, will produce somewhat more
abatement than will the tax system, although the normative implica-
tions of this result are ambiguous.
Another point of controversy over taxes versus standards essen-
tially boils down to whether policy should regulate inputs or outputs.
An output standard requires that the undesirable events-injuries,
emissions, and so on-occur at no more than some admissible rate,
whereas input standards specify the resources and technology to be
used in producing output. Input standards demand that the regulator
know the details of the production processes that are associated with
the particular problem that regulation is expected to ameliorate,
whereas output standards, in order to be effective without shutting
down an industry, only require the knowledge that some technology
exists that makes the standard technically and economically feasible.
Moreover, input standards provide no incentive-indeed, because ob-
taining approval is costly, a disincentive-to find alternative tech-
nologies that achieve the same objectives at lower costs. For these
reasons, output standards are generally regarded as more efficient
than input standards.

Mandated Standards in a Market Economy

Taxes are more amenable to output regulation than are standards
owing to the stochastic nature of outputs. Adverse health conditions,
injuries, and quality vary randomly over time in the presence of con-
stant rates of economic activity. If a firm is prohibited from exceeding
a target output standard, it will have to vary output-indeed, even
shut down-when it experiences unlucky draws. A tax system enables
the firm to balance good with bad luck over a longer period of time
by varying tax payments rather than its scale of operations. Normally,
the latter is more expensive so that with standards a firm faces more
uncertainty than it would under a tax system that was designed to
achieve the same target. Consequently, in order to reduce risks, a
standard that is equivalent in effect to a tax must be based on more
information about the random aspects of the conditions it seeks to
Output standards have the additional undesirable effects that they
discourage innovation to reduce pollution beyond the current target,
whereas taxes create a continuing incentive to improve performance
so that tax payments can be reduced. This is, indeed, the strongest
argument for taxes over standards. Nevertheless, it is contingent on
the practicality of regulating outputs. If outputs are more costly to
measure than inputs, the preceding argument may be undone by its
The typical form of input regulation is to specify a particular tech-
nical method as a design standard. Periodic inspections to ascertain
whether the device is in use and effective are the necessary monitor-
ing activity to enforce the policy. Output regulation for the purpose
of calculating taxes or enforcing standards normally demands more
extensive monitoring. In the case of ambient levels of hazardous sub-
stances or health and safety effects of products and occupations, the
monitoring must include a sufficiently large sample so that output
and therefore taxes over an entire period can accurately be estimated.
Obviously, the greater the cost of monitoring output, the less likely
it is that output standards or taxation will be desirable despite their
more desirable effects on firm behavior.

The choice between taxes and standards in a world of imperfect in-
formation and costly monitoring is not as clear as the theoretical su-
periority of taxation implies. Although taxes clearly begin the debate
with a theoretical advantage, in the end the selection of interven-

The Rationale for Mandated Cost Increases

tionist techniques is more an empirical than a theoretical issue. The
case for taxes is strongest when outputs are relatively easy to moni-
tor, the economic benefits of better performance are relatively easy to
ascertain, and the individual economic agents whose behavior the
policy intervention is supposed to change face heterogeneous costs of
improved performance. At the other extreme, input standards are
strongest if benefits and outputs are especially difficult to measure,
economic agents face relatively homogeneous costs of improved per-
formance, and the particular technical requirements of improved
performance are well known and unlikely to change. Output standards
are strongest when technology is changing rapidly, benefits but not
outputs are difficult to measure, an equivalent tax policy would gen-
erate substantial revenues, and firms can be categorized into groups
that face similar economic and technical situations.

1. Part of the costs of preparing this paper were provided by the National
Science Foundation, grant No. APR 75-16566. The author thanks Bruce Owen
for his numerous useful comments on an earlier draft.
2. A general theory of the incentive structure operating on government of-
ficials is presented in Fiorina and Noll (1978), and a specific discussion of the
legislative signals given to safety regulatory agencies is provided in Cornell,
Noll, and Weingast (1976).

Arrow, K. "Uncertainty and the Welfare Economics of Medical Care." Amer-
ican Economic Review 53 (1963), 941-73.
Bergstrom, T. "Preference and Choice in Matters of Life and Death." Depart-
ment of Economics, Washington University, 1976.
Calabresi, G. The Costs of Accidents (Yale, 1970).
Coase, R. "The Problem of Social Cost," Journal of Law and Economics 3
(1960), 1-44.
Cornell, N. W.; Noll, R. G.; and Weingast, B. "Safety Regulation." In Setting
National Priorities: The Next Ten Years, edited by H. Owens and C. L.
Schultze. Washington: The Brookings Institution, 1976.
Fiorina, M. P., and Noll, R. G. "Voters, Bureaucrats and Legislators: A Ra-
tional Choice Perspective on the Growth of Bureaucracy." Journal of Public
Economics (forthcoming 1978).
Oi, W. Y. "The Economics of Product Safety." Bell Journal of Economics and
Management Science 4 (Spring 1973), 3-28.
Weingast, B. "A Rational Choice Perspective on Distributive Policy Making."
Working Paper No. 27, Center for the Study of American Business, Wash-
ington University, 1978.
Wilde, L., and Schwartz, A. "Equilibrium Comparison Shopping." Social Sci-
ence Working Paper No. 184, California Institute of Technology, 1977.
Williamson, O. Markets and Hierarchies (Free Press, 1975).


Laurence C. Rosenberg
THE EXPRESSION "MANDATED COSTS" naturally evokes a distaste for the
mechanisms that generate them. Who wants the government to spon-
sor cost increases in the economy? The Weidenbaum paper deals with
the question by finding a plethora of ways in which regulation in-
creases costs, either without corresponding benefits or in ways which
are not the least cost to society. Weidenbaum's main point is not ar-
guable: better regulatory processes might achieve benefits at lower
costs. His examples, when taken in that spirit, amply convey the
feeling that there is room for improvement. However, he has gone too
far. The effective thrust of his paper is that regulation is so heinous
that we are almost always better off without it. Indeed, Weidenbaum
suggests that a most severe benefit/cost test be applied (one unlikely
ever to be met) before intervening in markets. Since Weidenbaum
writes his paper as an economist, it is presumed that his preference
for free-market operations is based on a greater love of neoclassical
economic theory than on a theory that legitimizes government inter-
Roger Noll's paper, while clearly displaying a distaste for regula-
tion, explores reasons why government intervention (in standards
regulation) might be a second-best case, which is superior to non-
intervention alternatives. His argument is framed as a neoclassical
apologist's case for regulation. Noll concentrates on problems of
market failure, externalities, and inefficiencies in the marketplace
and gives special emphasis to problems in the civil liability system
and to information and processing costs. He analyzes conditions un-


der which taxes and subsidies might not be better alternatives to
The Weidenbaum paper is concerned mainly with the question of
who pays the cost of regulation. This is the distributive question. Noll
deals with the allocative question in describing why markets might not
perform their economic function well.
Noll in his discussion of institutional imperfections, and Weiden-
baum in his discussion of regulatory evils, both draw arguments from
static neoclassical theory. Neither Noll nor Weidenbaum incorporates
a discussion of how one moves from the current state of affairs to a
state which is more desirable. Failure to deal adequately with this
question makes for a chasm between economists' proscriptions and
policy-makers' needs.
Neoclassical theory would provide a more appealing basis for policy
if it did not abstract from intergenerational utility comparisons. Static
theory presumes instantaneous change. But in the real world change
occurs over time. Future generations are affected (whether they be
generations of business managers or other economic decision-making
populations). If economists are dissatisfied with the allocative con-
sequences of the regulatory regime, should they not deal with the
process of change by concentrating on disaggregative short-run effects
as well as long-run aggregative welfare effects?
Identification of parties affected by change-measurement of who
reaps the benefits and who bears the costs of regulation-is necessary.
Calculation of efficiency losses due to transition friction, institutional
analyses, and theories of legislative and regulatory agency behavior
are needed. The impact of regulatory change in periods of inflation
and low employment, as opposed to periods of stability and relatively
full employment, should be studied.
It is easy to say that the authors should have written different pa-
pers (perhaps they are not ready to be written). But as the papers
now stand they are interesting material for a debate among econ-
omists themselves or for an audience of sympathetic businessmen.
Some comments on the Weidenbaum paper, taken for what it is,
are in order. Weidenbaum says that regulation is inflationary. But he
uses a rather strange definition of inflation. Inflation is usually de-
fined as a general rise in the level of prices. Included in this general
rise are increases in the per unit price of inputs. Most regulation in-
creases the quantity of inputs required to produce the regulated good
or service. These added inputs may be available at the same, lower,

Mandated Standards in a Market Economy

or higher prices, depending upon marginal productivities and levels
of unemployment. Inflation means an increase in the price of most
goods and services, but regulation alters the quality of goods and
services or reallocates resources to the production of different quan-
tities of them. Sometimes regulations divert capital and labor from
the production of private to public goods and sometimes they inter-
nalize costs. If regulations are inflationary, in the usual sense of the
word, a different set of proofs is required than those which Weiden-
baum offers.
Weidenbaum argues that inspections are morally repulsive and in-
efficient in that they are not the low cost technology for a given out-
put. He suggests that the market could do better alone. Noll points
out why the market might not do better. We know too little about
efficient enforcement systems, but considerable evidence can be
amassed to suggest that both industry and government have both good
and bad inspection systems. Compare OSHA enforcement with pri-
vate automobile company quality inspections. Compare the Internal
Revenue Service audit system with the job of the Better Business
Bureau in promoting honesty. It is not at all clear that government
inspections per se are inefficient mechanisms relative to free-market
Finally, while Weidenbaum admits there may be benefits from reg-
ulation, his examples do not deal with benefits to anyone. For ex-
ample, he complains that a radio station actually had to spend $26
in postage as part of the cost of receiving a monopoly; and so it goes.

Thomas G. Moore
The papers presented by Professors Weidenbaum and Noll take diver-
gent approaches to the subject of this conference. Professor Weiden-
baum's paper in large part consists of a recitation of horror stories
and estimates of the costs such programs levy on us. Professor Noll
analyzes the rationale for these regulations. Murray Weidenbaum
documents the cost of regulation in terms of the prices consumers
must pay, the cost of paper work, the cost of new housing, and the
reduction in the rate of innovation; but he does not indicate the bene-
fits. He also shows that regulation reduces the rate of capital forma-
tion with a consequent reduction in the rate of growth of the economy.
Professor Weidenbaum also cites studies that show that the minimum
wage reduces employment and that the Davis-Bacon Act has signifi-


cantly increased construction costs of government-supported construc-
tion projects. He also has claimed that OSHA regulation has in-
creased management costs while at the same time it has had no or
little measurable impact in reducing injuries and deaths.
To deal with these costs Professor Weidenbaum has suggested that
a benefit/cost approach be applied to all mandated controls. It seems
strange that Professor Weidenbaum would suggest a bureaucratic pro-
cedure-an inflation impact statement-to deal with excess regulation.
Perhaps his benefit/cost approach should be applied to his statute
requiring each regulatory agency to assess the impact of its proposed
I must add that I think it is too simplistic and naive to say that
government regulation should be confined to those areas where total
benefits to society exceed the costs. We know that our studies show
that many if not most governmental regulatory schemes do not meet
this criterion. We also believe that government officials and Congress
are not being purposely wasteful or blindly stupid.
Professor Weidenbaum suggests that using the price system might
be more efficient than regulation for achieving such aims as reducing
accidents on the job. I doubt that he would get much argument from
most people in this room. At the same time, however, he reports that
"We need to acknowledge the obvious . the pace of government
regulation of private sector activities is following an upward sloping
trend line." These two points raise the most important issues facing
this conference. Why has government regulation been growing at an
exponential rate? Why has the government failed to adopt the lowest
cost solution to dealing with the real problems that exist? It does little
good for a group like this to bemoan the growth of governmental
regulation or to tell each other that a greater role for competition
should be provided in the airline, trucking, or railroad industries. Or
that effluent taxes and pollution certificates would reduce the cost and
be more efficient in reducing pollution. We know that. At least our
economic calculus produces that conclusion; but it obviously does not
produce that conclusion in congressional hearings or in the halls of
EPA. Is there something wrong with our calculus; or is our message
not getting through, or is the system perverse?
Professor Noll on the other hand has tried to find a rationale for
governmental regulation. His paper consists of an analysis of why
market solutions may fail to provide the optimum allocation. He iden-
tifies three major types of problems that provide a rationale for gov-

Mandated Standards in a Market Economy

ernment-mandated standards: informational costs, court costs, and
externalities. By themselves none of these provide a necessary or suf-
ficient condition for regulation. Traditional economic theory has as-
sumed perfect information by all players. We know that in the real
world information is not perfect and is costly to assemble. A system
that takes costs into account should not be considered a second best
system. Perhaps what should be condemned is standard price theory
for assuming that information is costless rather than the market sys-
tem for taking it into account. Information though does have some
public-good aspects as Professor Noll quite correctly points out, but
whether these aspects should lead to government-mandated standards
is not clear.
Professor Noll is perfectly right in emphasizing the costs involved in
litigation, but such costs are neither necessary nor sufficient for the
failure of markets. If markets are competitive then product failures
or product safety will be determined by consumer demands and sup-
ply conditions and should be optimum barring significant externalities.
Litigation will affect equity but not efficiency.
Externalities have been well treated in the literature. As I already
pointed out, effluent charges and pollution certificates are likely to be
more effective than government regulation. Professor Noll recognizes
the advantages of such a price solution but asserts that "standards are
strongest if benefits and outputs are especially difficult to measure,
economic agents face relatively homogeneous costs of improved per-
formance, and the particular technical requirements of improved
performance are well known." I am not sure what he has in mind; it
may be an empty set. However, if the benefits and outputs are espe-
cially difficult to measure, perhaps the government had better stay out
because there may not be positive benefits.
One of the greatest problems with Professor Noll's paper is his
failure to include the cost of government-mandated standards. He has
given us a reasonable discussion of the costs of the marketplace and
the costs of the courts; but what about the alternative costs of gov-
ermental regulation? Is he assuming that such costs are negligible?
Professor Weidenbaum has showed that they are significant. To pre-
scribe governmental regulation implies that the outcome will be better
than the outcome of the marketplace or some other solution. I think
that Professor Weidenbaum has made a convincing case that govern-
ment intervention is hardly costless. There is nothing in Professor
Noll's paper demonstrating that private market failures are worse
than the horrors that Professor Weidenbaum has presented.


I also want to make the same complaint about Professor Noll's
paper as about Professor Weidenbaum's. Neither presents any theory
or explanation for the most rapidly growing industry of our time-
regulation. Professor Noll's paper might explain why some regulation
would exist but not why it is growing so rapidly. Professor Weiden-
baum explains why it shouldn't exist but not its rapid growth.
Let me offer some speculation on the subject. Karl Brunner and
Bill Meckling recently finished a paper on "The Perception of Man
and the Conception of Government." In it they argue that the socio-
logical perception of man is that his actions are determined by the
environment in which he grows up and operates. Thus if the environ-
ment were changed, the nature of man can be changed. The economic
view on the other hand is of an essentially self-interested being who,
given his environment, will attempt to further his or her own well-
being within the rules of the game.
To the extent that the sociological view of man is gaining ascen-
dancy it provides a motivation and a justification for regulating the
environment to "protect" individuals. Thus paternalism becomes the
motive for much government action. Many years back, I published in
the Journal of Law and Economics a study of government licensing
of occupations. The objective of the paper was to explain the growth
in occupational licensure. Was it simply the greed of the practitioners?
In the end I had to reject such a simple explanation. My final con-
clusion was that licensing was due at least in part to a paternalistic
view of government toward the average citizen.
Coupled with this sociological view of man is a growing distrust
and hostility to markets and market forces. Within certain intellectual
circles any government action aimed at controlling the market would
be welcome. This view seems to be gaining converts on many cam-
puses; and it has been well documented that the better the university
and the better the faculty member the more likely he or she will be
hostile to liberal-that is, free-market-principles.
I will now admit that my "explanation" for the growth in govern-
mental regulation is not an explanation because I have not explained
why these ideas appear to be winning in the marketplace. The hos-
tility to markets may be due simply to an objection to the fact that
marketplace outcomes do not conform with egalitarian principles. Yet
that "explanation" will still not satisfy my economist colleagues, who
look for reasons in terms of self-interest and market forces. Perhaps
intellectuals object to the marketplace because it provides a source of
power to rival their own ambition to have exclusive power.

Mandated Standards in a Market Economy

Roger Noll's paper disturbs me in one other sense. It seems to me
that there are at least three kinds of "failures" that supposedly justify
standards: externalities, safety controls on consumption, and safety
standards for the work place. All three have different considerations.
For environmental problems taxes are clearly more efficient than
controls. I think that Professor Noll has correctly identified a major
reason that such taxes are not often used: taxes affect the intramar-
ginal units while controls do not raise their costs. Nevertheless, we as
economists should continue to push for their adoption.
Professor Noll has not to my satisfaction explained why certifica-
tion would not be preferable to standards for protecting the con-
sumer. Certification can provide all the information that standards
furnish without constraining the consumer to abide by a bureaucrat's
view of a safe product. Such certification can be justified on the
grounds that information is a public good. But standards are unneces-
sarily rigid and raise the problem already mentioned that regulation
can make things worse rather than better.
Safety in the work place can also be met by providing information.
For example, if working with a certain chemical substance can in-
crease the risk of cancer, this information can be required to be given
each worker with his or her pay check and also posted in the work
place. Such a scheme would permit those who would prefer to bear
risk and thereby receive a higher pay to do so rather than to require
all to be safe.
Therefore, I would conclude that Professor Noll has failed to dem-
onstrate that mandated standards are generally desirable, at least to
the extent now in force. Neither Weidenbaum nor Noll has given us
any explanation for the puzzling growth of what can only be de-
scribed as a system which has failed.

Marvin Kosters

In commenting on Murray Weidenbaum's paper, I am reluctant to
be very critical of it since it draws so heavily on his AEI publication.
One of the important things he has done in his paper is to surface the
idea that governmental regulation imposes costs and to try to create
a public recognition of the kind of costs that are imposed by regula-
tion. He has contributed to a wider recognition of the fact that while
publishing a set of regulations is a very low-cost government activity
-at least low cost to the government-it does impose very high costs


on the economy in the society. He has pointed out ways in which
those costs are felt, including inflation. In spite of the fact that I am
very much a monetarist about that, I agree that regulation will impose
extra costs on a product and production processes and thereby result
in inflation.
It is much more difficult to get a reasonable amount of price stabil-
ity than if one were alternatively reducing excise taxes here and there
to reduce inflation as one went along. It seems likely that the idea
that regulation imposes costs will become part of the conventional
wisdom and appear in newspapers and magazines. I don't know
whether this will influence much the political contest, which is, in part,
behind the growth of regulation.
In this connection, there was a column in Sunday's Washington
Post by Jack Anderson on the issue of regulations and their costs. In
Washington people usually do not like to see their ideas, or particu-
larly their names, mentioned in Jack Anderson's column for obvious
reasons, but here he talks about the massive costs of government
regulation and paperwork. The first reason I mention this fact is it
seems to indicate that the idea that regulations do impose costs is
reaching a wider audience. The second reason I mention it is he pre-
sents some cost figures, which amount to something like $200 billion
per year imposed by government bureaucrats through their forms and
regulations. Jack Anderson is not a very good source of documenta-
tion to economists in their research studies, but his estimate is prob-
ably not much softer than most other cost estimates we have in this
area. The third reason I mention the Jack Anderson column is that
it indicates something in common with papers such as Weidenbaum's
and other works in this area that focus on costs. It is intended much
more to deplore the costs than to illuminate what ought to be done
about the situation.
Murray Weidenbaum does give some suggestions about what might
be done to improve regulatory performance. His observations about
the way this kind of regulation imposes costs are reasonable, although
they do not follow very closely or directly from his analysis. I do not
know precisely what kind of analysis should lead us to specific recom-
mendations for improved performance, but I do have an uneasy feel-
ing we need to do some homework in this area to obtain analysis that
does point more directly toward sources of regulatory improvement.
Roger Noll's paper outlining the elements of the rationale for a
standards approach or a detailed standards approach to regulation
does a good job of pulling together strains of the rationale for that

Mandated Standards in a Market Economy

kind of approach. One could take exception to Noll's arguments on
other grounds, but the main source of uneasiness I have about it is
that most of the rationale is more or less static in its approach. Prob-
lems associated with adjustment to change with innovation, new ap-
proaches to reducing pollution, or safety do not really get a good deal
of attention in the analysis. Even though he emphasizes the informa-
tion problems that underlie it, he does not really say very much about
how alternative approaches might themselves have different implica-
tions for the information that is generated by the process itself.
I will not attempt to develop a case for the superiority of taxes to
the mandated-standards approach, but at least I will try to explain
what I have in mind. For example, in the case of water quality stan-
dards, we are dealing with the problem that arises in part from ex-
ternalities of dealing with production of a public good, presumably
cleaner water. Usually a kind of engineering-like approach is used to
figure out feasible effluent levels that can be graded, with detailed
production processes and a standard to that established for different
kinds of emissions for different kinds of products.
One of the problems here is that there is a great deal of variation
throughout the economy in the kinds of plants. For example, some
plants may be nearly obsolete, but they are worth running a year or
two, perhaps paying a tax for it, but not really worth revamping. It
might not be socially optimal to shut them down immediately. Some
plants are new and can probably meet the standards very easily, and
others are worth restructuring.
This means that differences of that sort need to be dealt with
through a bureaucratic exceptions process. I am somewhat dubious
about that kind of process because I have sat on committees dealing
with these in connection with wage and price controls. Two important
considerations at least are involved in dealing with such exceptions.
One, it is always regarded as important to hold business's feet to the
fire to meet certain standards, a reasonable proposition since if busi-
ness's feet are not held to the fire management does not have the
inclination to engage in costly activity. The other consideration is that
one cannot really afford to shut down a major production activity so
therefore one needs to be reasonable. In dealing with the latter situa-
tion taxes are much better suited to it than a set of standards with a
bureaucratic process for deciding how to apply them.
One might respond that once we get all firms up to a certain
standard the process will have been brought into equilibrium and we


will not have this kind of dynamic problem. However, it seems to be
much more likely that we will have changing standards than a stable
world. First of all, all kinds of things change in the world, the feasibil-
ity of all kinds of approaches and so on. Second, if we are to keep in
the aggregate a stable level of environmental pollution, say water
pollution, then we will really be required to establish tighter and
tighter restrictions per unit of output if units of output grow. There-
fore, we won't really be in an equilibrium in which we have finally
turned to engineering standards in a stable world of standards and
technology. Further, of course, there are lots of problems with taxes,
some of which are only suggested in Noll's paper.
Another way of looking at the question of standards and regula-
tions, of course, is in terms of what they mean for property rights.
One of the characteristics of the way standards are typically devel-
oped (for example, the water quality standards that establish different
effluent levels for canning apricots and for cherries and different levels
for brine cherries and for sweet cherries) is based on the difficulty of
obtaining a given level of effluent. If one were to place a tax on each
unit of effluent, there would be a difference of assignment of property
rights implicit in that kind of standard setting. The political process
contains a notion that fairness or equity means not penalizing dirtier
activities just because they happen to be dirtier but taking into ac-
count what is feasible. Charles Schultze in his Godkin lectures pointed
to a variant of that when he talked about the principle in Congress
of "not doing direct harm." Finally, a question about taxes has to do
with the orders of magnitude we are talking about. For example, if
you take the Council on Environmental Quality's estimates of costs
of meeting pollution requirements, according to their current estimate
we will be spending some $20 to $30 billion per year in operating
costs and about that much in new capital investment-just to meet
pollution requirements. It would be interesting to see what Congress'
reaction would be toward giving EPA the authority to handle taxes
of that magnitude when we quarrel a great deal about a $10- to $15-
billion reduction in income tax schedules.

Arnold A. Heggestad
Professor Weidenbaum delineates many of the costs of regulation. He
is especially concerned with the regulations designed to achieve spe-
cific social goals-the standards regulations that aim for goals such

Mandated Standards in a Market Economy

as worker safety, environmental protection, or equity in employment.
As Professor Weidenbaum points out, but perhaps should stress, the
regulations do offer benefits to society as well as their costs. We should
not lose sight of the benefits.
Many of the regulations cut across a wide swath of different mar-
kets, firms, and even industries. That may well achieve their goals in
most cases. They cannot be analyzed by looking at isolated inci-
dences. Rather, more objective studies are required that investigate
their effects, both costs and benefits, wherever they appear.
There is a tendency to overstate the cost of regulations relative to
their benefits and thus overstate the social ills of regulation. The
benefits are generally spread over a large number of people. A regu-
lation may offer a small benefit to a large segment of society. In some
cases, its benefits may not even be definable in money terms-espe-
cially when there are secondary effects, such as in market failure
cases. Its costs, on the other hand, are generally focused on the few
firms supplying the product. Many of the costs are direct, such as the
cost of compliance, the cost of reporting, or the cost related to chang-
ing the production process. The indirect costs may also be large. The
industry, however, has a greater incentive and is in a better position
to measure indirect costs than the corresponding indirect benefits to
society. In summary, the present system encourages an overemphasis
on the costs relative to the benefits of regulation.
We can illustrate this by analysis of the Equal Credit Opportunity
Act (ECOA) as passed in 1974 and amended in 1976. The Act
states that a creditor shall not discriminate against any applicant on
the basis of sex or marital status. In 1976 the protected classes were
extended to include race, color, religion, national origin, age, and re-
ceipt of public assistance benefits. The law is enforced by prohibiting
use of these variables in the credit scoring models used to determine
if an individual is a good credit risk.
The costs of the regulation have been thoroughly discussed by
James Smith.' The direct costs of compliance, including costs of legal
counsel, printing of forms, training, and programming, are $30.9 mil-
lion. The cost of maintaining separate records for men and women
is $143.9 million. Further, there are recurring costs of reporting and
record keeping. The indirect costs may also be large. The intent of
ECOA is to remove irrational discrimination. Its enforcement will
also prevent rational discrimination if, say, it is sex related to credit
worthiness. One recent study finds, in fact, that women are better


credit risks.2 Preventing firms from using this information will reduce
the ability of firms to make proper credit decisions. Smith estimates
this involves an additional 2 percent loan loss or $293.3 million.
The benefits of ECOA are related to the extent of discrimination
before the law. Weidenbaum argues that there is no empirical evi-
dence of discrimination in credit. Therefore, the ECOA offers no
benefits, only costs. Smith, however, after carefully examining the
record, concluded: "Nevertheless, the evidence offered concerning
discriminatory experiences is of sufficient volume to ensure that at
least some creditors have engaged in 'social discrimination.' "3
Discrimination is very difficult to measure empirically. It is neces-
sary to control for differences in cost and differences in demand. This
has been done in housing and in employment and has been proved
to be substantial in many cases.4 Credit discrimination may be even
more difficult to prove because of market segmentation, usury laws,
and natural self-selection processes. Initial evidence from a survey in
California indicates that there is evidence of credit discrimination by
race. All else constant, blacks have less debt.5 One should not ignore
this evidence when assessing a law such as ECOA.
Professor Weidenbaum also suggests the need for a benefit/cost
analysis of all regulation. I agree that this information is necessary
and clearly in the national interest to pursue. The questions, however,
are highly complex. Let me illustrate by a study currently underway
for Abt Associates under the auspices of the National Science Foun-
dation on the costs and benefits of regulation of consumer financial
There are two major reasons the industry is regulated. First, regu-
lation is designed to prevent failure of depository institutions and the
accompanying distortions of the payments mechanism. Second, regula-
tion is designed to protect consumers against abuses of creditors. A
partial list of the regulations imposed on the industry will indicate
their complexity.

A. Regulation of Firm Behavior
1. Capital requirements
2. Restrictions on asset and liability management
3. Mandatory deposit insurance
4. Restrictions on management conduct via periodic
5. Holding company investment restrictions

Mandated Standards in a Market Economy

B. Restrictions on Competition
1. Restrictions on new charters
2. Branching restrictions
3. Enforced product specialization
4. Interest rate limits for liabilities
C. Direct Consumer Protection Regulations
1. Usury laws
2. Truth in lending laws
3. Equal Credit Opportunity Act
4. Limitation on creditor/debtor rights and remedies

Our study of these regulations, which is being conducted by a ten-
person research team, has demonstrated that the regulations are highly
interdependent in their effects. A regulation will have different effects
depending on the severity of the other regulations. For example, the
extent of discrimination, and therefore the benefit of ECOA, will de-
pend on the severity of usury ceilings by changing the opportunity
cost of discrimination. It will also depend on the degree of compe-
tition in markets since discrimination is unlikely to exist in highly
competitive markets. Similarly, restrictions on liability prices limit
the desire to compete for loans and make discrimination more likely.
Thus, a study of ECOA cannot ignore the incidence of other regula-
This interdependence imposes the need to analyze the effects of all
regulations in an industry simultaneously. To handle this problem we
need to consider the complementarities and overlaps between regula-
tions. This may be achieved by mathematical modeling of firm or
industry behavior to allow simulation of the effects of changes in
regulations. Another solution is to use social experiments where regu-
lations are changed on a small scale. A recent example of such ex-
perimentation in finance is seen in the NOW accounts in New England,
which allow interest on demand deposits and competition between
thrift institutions and commercial banks.
In summary, Professor Weidenbaum is correct in our need for
analysis of costs and benefits of regulations. We should pursue this,
but we should not forget two facts: first, there is a natural tendency
to overstate costs relative to benefits, and second, benefit/cost analysis
of a set of regulations is extremely complex. Great care should be
taken in accepting partial analyses of the separate incidences of reg-

Commentaries 51

ulation without considering the cumulative effect of the entire set of

1. James F. Smith, "The Equal Credit Opportunity Act of 1974: A Cost-
Benefit Analysis," Journal of Finance (May 1977).
2. Gary A. Chandler and David C. Ewert, "Discrimination on the Basis of
Sex under the Equal Credit Opportunity Act," Working Paper No. 8 (West
Lafayette, Indiana: Credit Research Center, Purdue University, 1976).
3. Smith, "The Equal Credit Opportunity Act."
4. See, for example, Gary S. Becker, The Economics of Discrimination
(Chicago: The University of Chicago Press, 1971).
5. Robert Shay and John Marshall, "Discrimination in Consumer Lending,"
in Costs and Benefits of Regulation of Consumer Financial Series, Interim Re-
port of Abt Associates to the National Science Foundation, August 1977.

Part Two
Microeconomic Effects (A)


Can Safety Be Mandated?

Nina Cornell

I HAVE BEEN DEBATING all morning whether I should open with my
favorite regulation cartoon. It shows people scrambling into a lifeboat
from a very obviously sinking ship; one turns to the captain and says,
"Captain, you are in deep trouble. Pollution laws do not allow you to
sink in these waters."
The topic I was handed by Dean Lanzillotti was "Can Safety Be
Mandated?" I was tempted to get up and say "no" and sit down
again. I decided that was not exactly what he had in mind, so I
thought I would go a little further. I think the answer is no-at least
I would like to advance the hypothesis that the answer is no if man-
dates are the sole or primary approach to all of these problems that
come in a sense under the rubric of safety. I think I would stretch it
to include environment, safety, and health-both safety and health
as applied to occupational safety and health in such matters as Food
and Drug Administration, CPSC matters, and the like. Looking at the
process as it has worked in these agencies, at the federal level at least
(presumably the same is true at the state level), we may judge that
reliance for safety, health, and environmental protection on a system
of standards enforced by penalties for noncompliance may, in fact,
be unrealistic. Such a system may be too cumbersome a mechanism
to effect those changes initially intended by legislation.
Going back to Professor Weidenbaum's example, it may be not
only that the medicine is very strong and has very potent side effects
that no one thought of. It also may be that too much use of it breeds
drug-resistant bacteria in a sense. While there is a lot of meat to con-
sider in the legislation proper, I would like to go one step further.

Microeconomic Effects (A)

Having passed laws, silly or not, if you rely overwhelmingly on the
approach of standards with penalties, which go along with a set of
standards, presumably three areas "bring the system down," i.e., in
terms of getting the real impacts where you would want them where
regulations make the most sense in the areas as outlined by Roger
Noll. Considering (1) the way agencies pick their priorities or targets
when handed the whole smorgasbord of problems to deal with, (2)
the process of promulgating standards, and (3) the compliance and
enforcement techniques, one can formulate a hypothesis that standard
setting, with compliance through penalties, may not be an effective
approach. In short, it means giving the medicine too many bacteria
to attack.
Considering priorities, I offer a couple of examples. The attempt
by the Consumer Product Safety Commission to develop its NIOSH
Index is perhaps one of the most systematic attempts that an agency
has taken to develop a system of priorities based on objective evi-
dence as opposed to relying on telephone tips. Various problems
came up almost immediately since the index has too many things to
look at. It had to develop a very arbitrary system of weights with
which to relate injuries and accidents that have very different con-
sequences. How do you decide what is a "more severe" impact? What
is a "less severe" impact? You come up to something very arbitrary
that got 2,516 points as a death and 10 points as a sprained ankle.
While I certainly view the 252 sprained ankles as worth one death,
I do not think that one death is worth 252 sprained ankles, which
illustrates the arbitrary nature of that particular kind of index. There
are other problems with the NIOSH Index.
Given that kind of index with its particular set of problems, you
end up with a choice of targets that may be the least amenable to
regulation. You end up as the Consumer Products Safety Commission
did, very interested in the regulation of matches, kitchen knives,
and stairways, since these seem to lead to the greatest number of
domestic accidents. That they are also perhaps the best known haz-
ards, and that people do the most they can privately to avoid those
hazards, seems to be irrelevant in the calculation. The attempt to for-
mulate objective standards leads into areas which may indeed be im-
possible for regulation to deal with. How do you regulate a safe
stairway other than making it horizontal? It is great for safety, but it
does not get you between floors. It seems to be a case that this picks
up well-known hazards and is least likely to pick up the kinds of

Can Safety Be Mandated?

cases where market failure and information failures are the greatest.
Likewise, the OSHA priorities lead to a very similar conclusion: a
situation in which OSHA promulgates regulations on those hazards
that are best known. However, OSHA still has not turned its atten-
tion to those hazards which may have the most severe consequences
down the road, but now are the least known and therefore the least
protected against by private actions.
In the health field, which it makes the most sense for OSHA to be
in, you still have the same problem of priorities. Thus, OSHA is
putting its biggest attention on coke-oven emissions, lead, asbestos,
and the like. The least amount of attention is given to synthetic chem-
icals, new on the market or new in occupational use, where least is
known and the most danger lies because of that total ignorance of
what those effects may be 20 years down the road-when the effects
will also be found to be irreversible if past experience is any guide.
Secondly, because of compliance problems down the road, the pres-
sure is put back on the priority system to choose hazards that are the
most easily seen and the most easily acknowledged. I think it is no
accident that the first proposed health standard OSHA thought of
promulgating, although it later backed off, was to require that toilet
seats be split instead of continuous across the front. It was something
very easy to check with compliance.
When you move on from choice of targets, having been given too
big a smorgasbord to deal with in the first place, to this process of
setting standards, there is another set of problems. For a variety of
legal reasons, we have a system in place of the Administrative Prac-
tices Act and all the case law that goes along with it that essentially
mandates a fairly lengthy and cumbersome process that has two prob-
lems related to something touched on earlier. You either get (1) a
set of standards that are obsolete or made obsolete by the time they
get them out because the products are so totally changed that there
may not even be a product to which their standard applies, or (2)
technological change that is blocked because of the standard. There
is a wonderful case of pressure vessels which are required to be cer-
tified by the ASME. At one point OSHA adopted wholesale the
ASME standards. ASME has since upgraded the standards, but
OSHA has not. In order to meet the OSHA standards, they have to
meet a code specification that is obsolete by ASME standards, in
which case nobody will buy it and people are caught in the cross fire
there. The process by which OSHA adopted many of their so-called

Microeconomic Effects (A)

industry consensus standards slowed the process of those standards.
The industry can change them much faster than OSHA can because
the industry is not subject to the Administrative Practices Act.
Another problem with the standard-setting process when you use
it as your sole or primary approach is that, because OSHA is re-
quired to address such a large number of issues and is not allowed to
do independent research of its own, you have to rely on outside in-
formation sources. You have no choice (even if you are allowed to
do research) but to turn to outside and usually interested parties for
information. You just have too many things to deal with. Therefore
you must take your data where it is available quickly, and that usually
means from interested parties. You cannot stand off and develop
very good checks, let alone very good independently generated in-
formation. You end up with the very likely situation, which I would
urge those people who are funding research to devote more attention
to, that you may have created a device for creating and enforcing
cartels. The CPSC household hazards example is one that has been
in the literature awhile. There is some evidence that the effect of the
OSHA coke-oven standard may be that you will reduce the number
of suppliers by a large number. I have heard it rumored that such is
true in the paper industry, and also that the use of standards limits
entry, cuts out small competitors, and puts into effect a cartel where
one was not supportable in the marketplace without the standards.
A third area I wanted to look at is the compliance enforcement
effort. Standards as an approach require inspections with penalties if
you are out of compliance. The number of inspectors and support
personnel likely to be budgeted by any kind of reasonable assessment
of Congressional action is going to be insufficient for the task of cov-
ering the whole waterfront, even though you have been given the
whole waterfront to cover. For example, EPA estimated that there
are 42,000 chemicals that they thought were worthy of investigation,
or suspected of occupational hazard, toxicity, and the like. Some of
these are present only in one work place; others like asbestos are al-
most ubiquitous now in work places all over. If you make a very
conservative assumption that for those 42,000 there will be an aver-
age of ten establishments you need to inspect for those chemicals
(assuming you have developed the standards for them), and it takes
one man week per inspection (the inspector, getting the lab reports
back, analyzing the monitor samples and the like, filling out the pa-
perwork, checking on the medical compliance records at the plant,

Can Safety Be Mandated?

etc.), then this means that there will be a total of 52 inspections per
inspector per year, and-unless my arithmetic is wrong-it will re-
quire about 8,075 inspectors just to inspect each establishment once
a year. Chemical exposures, of course, occur daily, often all through
the day.
This estimate for the chemical inspection does not include any-
thing else, neither health problems nor safety problems of which
there are many, probably more potential ones than there are health
problems. Thus you end up with an enforcement problem whereby
one cannot possibly visit the work places as often as might make
sense to make the system work. There is an alternative: inspect much
less frequently, assess very large fines, and ensure that the probability
of getting caught is very high. Unfortunately, the evidence from
OSHA to date indicates that it is not willing to make that kind of
tradeoff. Experience reported is that the average fine for a citation by
OSHA between July 1972 and March 1974 was $25 per violation.
It was also estimated that a very large firm can be expected to be
visited an average of once every ten years. The average fine was for
any kind of citation. The average for the most severe penalty-for
willful or repeated imminent danger violation-would be $1,400.
When you put together the fact that for a large firm the cost of
compliance at that time was estimated to be $350,000, making some
assumptions about interest rates, a large firm would have to expect
more than 400 willful repeated or imminent danger citations per visit
before they were indifferent as to complying to the regulations in the
first place. We are clearly nowhere near a penalty structure that sub-
stitutes for adequate inspection. We are not willing to budget re-
sources for the kind of inspection that goes along with penalty struc-
ture, and we are left with a system that in fact does not have very
much impact. It is not very surprising that OSHA has not had much
effect on accident rates. It could be argued that this comes about be-
cause of silly legislation by Congress, and I think that is a large part
of it. It is a sad story because there is a role for standards. You can
get some safety by standards legislation, but I think you are not going
to get much of it if that is your sole reliance.


OSHA and U.S. Industry

John F. Morrall m

PROBABLY NO OTHER governmental regulatory body has evoked more
controversy and emotion among both industry and labor than the
Occupational Safety and Health Administration. A recent article on
the front page of the Washington Post was entitled "OSHA: A 4-
Letter Word, Safety Unit Scorned by Business, Labor."1 Of course,
business and labor usually criticize OSHA from different standpoints.
But the article goes on to point out that over the last six years OSHA
has only promulgated standards to control seventeen out of the 22,000
toxic substances in use in American industry and yet has twelve entire
pages of regulations in force on the proper construction of portable
wood ladders, including a long passage on the spacing and size of
This type of criticism-that OSHA spends too much effort on
safety and not enough on health-is old hat, having been made by
just about everyone that has evaluated OSHA over the last five years.3
However, despite the consensus that the need for action and the po-
tential for effectiveness is greater in the occupational disease than the
industrial safety area, the number of health inspectors is not predicted
to equal the number of safety inspectors until 1979.4
It does not seem unusual that when business and labor criticize an
agency so vehemently from different standpoints, the bureaucracy
becomes paralyzed. But what explains the paralysis, or at least snail-
like pace, when the critics are relatively united that OSHA should
move in a given direction, that is, toward more standards develop-
ment and enforcement in the occupational disease area than in safety?
The simple explanation is that it was easiest for OSHA to move in

OSHA and U.S. Industry

the wrong direction and therefore bureaucratic momentum was hard
to reverse.
Shortly after passage of the act, OSHA was permitted by Section
6(a) of its legislation to adopt without any public discussion 4,400
"consensus standards," many of them established by industrial ad-
visory groups and not meant to be compulsory or necessarily designed
to promote health and safety. Much of the criticism and many of the
problems with OSHA resulted from this action. Once these standards
were on the books, the quickest way that an inspector could complete
his daily task was to report the most easily identifiable violations,
those that can quickly be solved by the employer-employee bargaining
process. Of course, it is still easier to cite safety violations than health
violations. In 1975 only one of the hundred most cited violations was
health related.
Although OSHA has already started training programs for health
professionals, probably OSHA can only escape this entanglement in
safety regulations by rationalizing and simplifying the old complex
consensus standards, thereby releasing enforcement effort for the oc-
cupational disease area. The informational and administrative require-
ments needed to develop and enforce the complex specification stan-
dards that OSHA has relied upon in the past are so large as to hamper
severely the achievement of safer working conditions. To make prog-
ress OSHA probably must shift its efforts from attempting to dictate
the inputs and production functions that are used in four to five mil-
lion establishments to determining the proper level of safety and
health demanded by our collective society. The Public Health Service
estimates that there are 390,000 new cases of occupational disease
and as many as 100,000 deaths each year, and the Bureau of Labor
Statistics estimates that there are about five million injuries and ill-
nesses and 5,300 deaths in 1975 in private industry." The enormity
of the task and the great number of other important social goals and
objectives that are competing for our scarce resources require that
we as a society pursue our health and safety goals through cost-effec-
tive and equitable procedures.
This paper will propose some reforms of OSHA procedures that
should promote the goals of the OSHA Act at a more rapid pace and
at less social cost than is probably being generated now. In making
recommendations, the paper explores what has gone wrong: first, it
analyzes why the market has failed to provide the desired amount of
workplace health and safety; second, it draws out the implications of

Microeconomic Effects (A)

market failure for public policy; and finally, it evaluates OSHA's at-
tempt to deal with these problems through the standard-setting process.

The Under-Provision of Health and Safety
Nina Cornell, Roger Noll, and Barry Weingast in their chapter on
"Safety Regulation" in Setting National Priorities emphasize the im-
portance of information and uncertainty (which is a lack of informa-
tion) to the pure market system.6 With complete and free information,
product prices and compensating wage differentials along with factor
mobility should lead to the optimum level of health and safety-and
at least total hazard and hazard prevention costs. This leads them to
recommend public policy measures that reduce information costs and
uncertainty, a recommendation that I am sure most of us would sup-
port. Based on this analysis they also recommend that OSHA con-
centrate its standard setting on uncertain areas and areas of high
information cost that the private market cannot handle. These areas
are mainly in the complex occupational health field. An injury tax or
strengthened workmen's compensation system is recommended for in-
dustrial accidents.7 Robert Smith in an American Enterprise Institute
evaluation study arrives at the same recommendations for health
standards and injury taxes but adds lack of labor mobility to lack of
information as a rationale for government intervention to increase
the provision of health and safety.8 Knowing one is facing a health
hazard does little good if one has no alternatives.
Despite the importance of these two sources of market failures,
there are several others that may be just as important. Even with
complete information, no uncertainty, and free factor mobility, it is
likely in our society that not enough health or safety will be provided
in the workplace. This results because the parties to the transaction,
labor and management, do not bear the full costs of injuries and ill-
There are at least three types of costs not borne by the employee
or the employer. First, there are the psychic externality costs of
bereavement that a community feels for its injured or ill members."
The magnitude of this dimension is hard to gauge, but as reflected in
the political area it appears to be substantial and is certainly growing.
Second, since we live in the world of second best, the opportunity
cost or shadow price of the medical resources used to treat the in-
jured or ill employee is normally much higher than the actual cost

OSHA and U.S. Industry

paid by either the employee or the employer. Workmen's compensa-
tion, Medicare, Medicaid, Blue Cross-Blue Shield, and other insur-
ance plans and government health and income supplement programs
have reduced the marginal out-of-pocket medical costs to a small
fraction of total costs. Martin Feldstein and Amy Taylor have esti-
mated that for hospital care in general consumers only spent directly
21.4 percent of the total cost in 1975.10 For workplace illnesses and
injuries this fraction is likely to be much smaller.
This phenomenon is quite complex, having aspects of the "moral
hazard" problem (for both hazard prevention and medical services
demanded) and provider's choice whereby the physician determines
the quantity and quality of service. Although there are many other
costs of injury and illnesses-such as pain and suffering, the lost in-
come of workers not refunded fully by workmen's compensation, some
increase in some workmen's compensation premiums for some firms,
and lost productivity-the fact remains that a large part of these
costs is not borne directly by the firm or its employees, and thus not
enough preventive health and safety will be demanded. Any preven-
tive program that reduces injury and illnesses could produce impor-
tant cost savings for society. Note that this argument holds also,
although with less force, because of the absence of workmen's com-
pensation for health and safety problems caused by hazards in the
environment and the home, and that this effect is likely to grow over
time unless major progress is made in reforming our health care sys-
Another argument for intervention, especially in the case of health
hazards with long latency periods such as cancer, is that the average
worker's rate of discount or time perference is likely to be higher than
the social rate of discount. Our income supplement and health insur-
ance programs designed for the elderly and education programs de-
signed for the young are in part probably a manifestation of society's
recognition of this fact. Thus, additional programs that reduce occu-
pational disease that appears twenty to thirty years after exposure are
desirable on these grounds.
Nicholas Ashford has made the case that governmental interven-
tion into workplace health and safety matters can be justified on
equity or Rawlsian fairness grounds (i.e., that it is unjust that some
must bear hardships so that in the aggregate a greater good can be
obtained).'1 That is, a potential Pareto move is not just. However,
if the other conditions described above have been satisfied (i.e., per-

Microeconomic Effects (A)

fect information, free mobility, and all externalities internalized) then
we have a true Pareto optimum. Moreover, if the distribution of in-
come is still not equitable, then incomes not hazards should be trans-
ferred. Thus the equity argument is seen as an expression of the fact
that one does not like the current imperfect state of affairs, and it
may be a less scientific way of describing the results of market failure.
If the market failures are attacked directly and successfully, the equity
problems, if any, are more easily resolved.

Implications for Public Policy of Market Failures
As others have suggested, OSHA and the National Institute of Occu-
pational Safety and Health (NIOSH) should play an active role in
providing and disseminating information on the risks and uncertain-
ties of occupational hazards. Programs that increase worker mobility,
especially in hazardous occupations, should also be encouraged. Since
many of these hazards are newly discovered or made known to work-
ers, perhaps an OSHA adjustment assistant program patterned after
our trade adjustment assistance program but optional to workers who
wish to leave risky industries should be developed. Together these
reforms should increase the prices of goods and services produced in
hazardous industries and increase the compensating wage differen-
tial to workers in these industries, thus leading firms to increase ex-
penditures for hazard reduction and consumers to purchase fewer
hazard-intensive goods and services.
The National Commission on Workmen's Compensation has al-
ready suggested reforms that would increase experience rating, but
they have also recommended increases in benefit levels that would
intensify the moral hazard problems.12 I am not suggesting, therefore,
that benefit levels should not be raised, but only that the workmen's
compensation system cannot by itself effectively internalize injury and
illness costs, since the insurance aspect for firms can never and in-
deed should not be completely eliminated and since workers do not
bear the costs of their illnesses and injuries inflicted on third parties.13
The moral hazard problems of private and public medical and income
insurance programs also prevent the attainment of the desired level
of preventive health and safety and are also not amenable to reforms
that would internalize social costs since the insurance function of
these programs is socially desirable. Even if private and public in-
surance were eliminated and workers were held liable for their actions,

OSHA and U.S. Industry

not enough preventive health and safety would be produced since the
externality and social discount reasons for government intervention
would not be addressed.
Optimal public policy prescriptions must, therefore, search for
other mechanisms. Two such mechanisms appear to exist at polar
extremes, standards and taxes or user charges. Although politicians
appear overwhelmingly to endorse standards and economists to en-
dorse charges, the two are frequently intermixed and in fact may be
viewed as part of one continuum. Standards are enforced with fines
or outright prohibitions, which may be viewed as prohibitive fines,
and user charges are simply fines based on performance standards.
In the former case, the greatest detail is spent in specifying the
standard while the terms of the fine are left unspecified and are ar-
bitrary relative to the harm done. In the latter case, the terms of the
user charge are clearly specified relative to the harm done while the
standard is defined in very broad terms, ideally in terms of perform-
ance in meeting the social objective. Once this continuum is ad-
mitted, the debate between standards and taxes is seen as a debate
over what point to choose along the continuum and will presumably
depend upon the total costs and difficulty of enforcement. Note that
this dispute also merges with the dispute between specification and
performance standards.
In the OSHA area these considerations lead to recommendations
of different mixes in the safety and health area, mainly because caus-
ality of injuries is immediately apparent while the latency period for
most occupational diseases makes the connection between cause and
effect difficult to identify. Taxes or fines levied on firms in approxima-
tion to the noninternalized social cost of accidents, as discussed above,
would be more cost effective than specifying from a central point the
actual safety practices and equipment modifications that all firms
should follow in all of their production processes. Under the injury
tax, firms would compete among themselves to provide safe working
conditions and would have incentives not only to modify their equip-
ment and working practices but to provide incentives for their work-
ers to adopt safe practices and use care themselves. Contrast this situa-
tion to the present one where a firm's incentive from safety regulations
is to comply with or fight the legality of standards, not to reduce the
number and severity of injuries.
Providing the desired amount of health in a cost-effective manner
requires, because of the separation of cause and effect, that the stan-

Microeconomic Effects (A)

dard be based on a proxy for the reduction in occupational disease-
a step back along the continuum. If the substance of harm can be
identified, fines should be tied to these emissions. Firms would then
have the incentive to reduce emissions in the least costly way. Con-
trast this situation to the present one where firms have an incentive
only to reduce emissions down to the standard; and if they can't meet
the standard, they have little incentive to go part way. In the case of
most of OSHA's present health standards the only substantive change
that would be required to move in this direction is to levy fines ac-
cording to a formula that is proportional to the number, severity, and
length of exposure that workers receive and relates that proportional-
ity to an approximation of the noninternalized social cost of the oc-
cupational disease.14
In the case of carcinogenic and many toxic substances, many health
experts believe that there is no absolutely safe level of exposure ex-
cept zero. Thus the present OSHA standard-setting process frequently
develops unsafe standards because standards are limited by "feasibil-
ity," which in practice has come to mean "economic feasibility," al-
though not in the cost-benefit sense. Although the standards urge the
development of new technology that will reduce hazards further, at
present private firms have little government inspired incentive to do
so. Fines based on the degree of compliance with performance stan-
dards would offer that incentive. If there are difficulties with fining
firms for not doing something which is technically not feasible, the
fines could become negative, that is, subsidies, at given threshold
levels. This modification, of course, would require legislation.
Finally, note that all these policy recommendations can be pur-
sued in different degrees. More information, reduction in uncertainty,
encouragement of worker mobility, strong experience in rating of
workmen's compensation, and a movement toward fines tied to the
degree of compliance with performance standards should all move us
in a cost-effective manner toward the desired reduction in occupa-
tional illnesses and injuries.

The OSHA Standard-Setting Process
In this section, two OSHA health standards will be analyzed in order
to determine how their development and content differs from the fore-
going policy recommendations and how the standard-setting process
might be improved.

OSHA and U.S. Industry

Except for the 4,400 national consensus standards promulgated un-
der Section 6(a) of the Occupational Safety and Health Act soon
after the act became effective and also numerous minor modifications
to those standards, OSHA has promulgated (under section 6[b]) a
total of only fourteen standards in six years-four in the health area
and ten safety standards. One of the four health standards (the coke-
oven standard)15 and the potentially most expensive and far-reaching
standard pending at this time (the industrial noise standard) will be
discussed. The secretary of labor may promulgate permanent stan-
dards on the basis of petitions from interested parties, his own in-
formation, or on the basis of criteria documents submitted by NIOSH.
Presently it is the policy of the Department of Labor to proceed on
the basis of criteria documents from NIOSH that identify workplace
hazards and specific recommendations from industry-labor advisory
groups. Once the Department of Labor starts its development process,
the issue of economic feasibility will be faced for the first time.
Up until a few years ago, OSHA attempted to estimate the costs
of proposed standards but not the benefits of the standards since if
NIOSH developed a criteria document it meant that a hazard had
been recognized and the act mandated all hazard reduction that was
"feasible." However, on November 27, 1974, Executive Order 11821
was promulgated requiring an inflation-impact statement for all ma-
jor regulatory proposals. This requirement was defined by OMB Cir-
cular A-107 issued on January 28, 1975, to be a cost-benefit analysis
of a proposal and its major alternatives. Thus the Department of
Labor, as well as other executive-branch departments and agencies,
started or continued performing such studies. Nevertheless, many in-
formed individuals in this area still misinterpreted this requirement,
which has since been renamed "Economic Impact Analysis," as im-
posing a cost analysis that is by nature "not neutral" since "all price
rises are not inflationary.""'' However, the purpose of the program
was to correct just such biases as OSHA had been making and that
caused others concern.
The problem with the specification-standard/arbitrary-fine regula-
tory approach is that it forces governmental concentration on the cost
of compliance with little or no quantification of the benefits. Yet this
is just the reverse order of what the government does best. As Cor-
nell, Noll. and Weingast point out, the role of the government should
be to develop information on hazards and to attempt to turn uncer-
tainty into risk, since private firms have little incentive to do so.17

Microeconomic Effects (A)

Firms know far better than government what the true costs of alterna-
tive risk-reduction strategies are. Moreover, the way OSHA attempts
to gather cost information now is to hire contractors to ask industry-
a technique that is bound to lead to biased estimates and constant
conflict at administrative hearings.
A movement toward performance standards and fines tied to de-
gree of performance would shift the information needs in the desired
directions. Indeed, if OSHA knew the true benefits of standards and
their alternatives, it would have no need to know the costs to the
industry. A fine system could be set up that internalized all social
costs, and optimum prices and outputs would result even if certain
firms or industries went out of business.18 Thus, perhaps, the IIS pro-
gram should have been a benefit-impact-statement program.
In the absence of a fine system pegged to the degree of compliance
with performance standards, a properly done cost-benefit analysis of
alternative standards is required to determine the desired standard.
However, OSHA officials have expressed a great reluctance to per-
form and use such analysis.'1 Another hindrance to the use of cost-
benefit analysis in OSHA standard-setting may result from a con-
clusion of a recent important House Subcommittee on Oversight and
Investigations report entitled Federal Regulation and Regulatory Re-
form.20 The report concludes that "the limitations on the usefulness
of benefit/cost analysis in the context of health, safety, and environ-
mental regulatory decision-making are so severe that they militate
against its use altogether."21 Interestingly, the report offers no al-
ternative to cost-benefit analysis except for "conscious responsible
choice."22 It is not clear to me how "conscious responsible choice"
could be anything but a careful weighing of costs and benefits.23

The Coke-Oven and Industrial Noise Standards
In the standard-setting process and in particular the IIS requirement,
OSHA has declined to compare systematically the costs and benefits
of proposed and alternative standards. However, in the case of the
coke-oven and industrial noise standards, cost and benefit data were
supplied in the Inflation Impact Statements that could be used for
that purpose." In fact, the Council on Wage and Price Stability pre-
pared such analyses and testified on them at public hearings.25
On the cost side, the difficulty of OSHA's getting reliable estimates
of the cost of compliance is well illustrated by both of these cases.

OSHA and U.S. Industry

In the coke-oven case, OSHA hired a consulting firm to gather con-
fidential data from industry with the promise that the data on in-
dividual companies would not be released to OSHA. From these data
the consulting firm estimated that the total annual cost of the proposed
standard would range from $241 million to $1,280 million.26 Much
of the debate at the month-long hearings revolved around what the
proper cost figures were and whether the industry could survive the
impact. In promulgating the regulation OSHA stated its belief that
"total annual costs are likely to fall in the $200 million range rather
than the $1 billion range";27 however, this is only a guess and prob-
ably not even the industry knows what the costs will be.
A similar type of problem arose in the noise hearing when a con-
sulting firm produced two sets of estimates one year apart at $13.5
and $31.6 billion the first time for an 80 and 85 dBA standard and
$10.5 and $18.5 billion the second time.28 At the same time, the
Environmental Protection Agency testified before the Senate Select
Committee on Small Business that the true costs of compliance with
80 and 85 dBA noise standards would be only $7.5 and $13.5 bil-
lion.20 Clearly, the government's comparative advantage does not ap-
pear to lie in estimating industrial costs.
In contrast, at least in the case of these two standards, quantifica-
tion of the benefits of risk reduction-carcinogenic risks in the coke-
oven case and risks of hearing loss in the industrial-noise case-ap-
pear to be fairly reliable and objectively determined. The evidence
was gathered using scientific methodology by academic experts. There
is still some scientific disagreement, particularly between EPA and
OSHA over hearing risks, but this can probably be explained by
OSHA's concern with material harm and EPA's broader concern
under the Noise Control Act of 1972 with public health and welfare.30
Since OSHA declines to place dollar values on risk reduction, there
has been no disagreement on such figures despite the fact that OSHA
claims that IIS requires explicit dollar determinations on the value of
life or hearing.31 However, the IIS requires quantification of benefits
only to the extent feasible, and CWPS has testified to this effect.32
This misunderstanding probably has arisen because CWPS has pointed
out that the decision whether to promulgate a standard implies some-
thing about the policy-maker's implicit value determination for risk
reduction. Comparing the quantified benefits and costs systematically
will produce such implicit values for risk reduction. The policy-maker
should then use this information to help determine whether the regu-

Microeconomic Effects (A)

lation should be promulgated. However, OSHA still refuses to ac-
cept the validity of the implicit value proposition.33
The main problem is that OSHA interprets its act to mean that all
risks of material impairment must be eliminated if feasible. Thus any
risk reduction is desirable as long as the industry's viability is not
threatened.34 In the coke-oven case the annual cost of $1.3 billion
and $241 million, and the twenty-seven lives saved annually as esti-
mated by the IIS,3" imply a value of risk reduction per person from
$9 to $48 million. This figure should be reduced since morbidity risk
is not counted and yet it is increased since the benefits accrue 10 to 30
years after the costs due to the long latency period of respiratory
cancer. These adjustments are not likely to change the magnitude of
the numbers.
CWPS's position has been that these estimates should at least be
compared to estimates implied by other government actions or by
willingness to pay kinds of analysis.36 It is interesting to note that at
about the same time that DOL was promulgating the coke-oven regu-
lation, the Department of Transportation issued its ruling on the use
of airbags. If one accepts the benefit and cost estimates developed by
the National Highway Traffic Administration,37 then not promulgating
the airbag regulation implies that DOT valued saving a life at less
than $120,000. I conclude from this that probably both decisions
were wrong.
Let me hasten to add, though, that a cost-beneficial standard should
not be promulgated if there are more cost-effective alternatives. In
both cases there are. In the airbag case, a mandatory seat- and lap-
belt law with 70 percent effectiveness would save lives at the cost of
less than $40,000. From now on I will wear my seat and lap belt.
Also note that if respirators were worn by coke-oven workers at a
cost of $100 per year, lives could be saved at a cost of about $80,000.
Of course, at the present time respirators can only be used as an in-
terim procedure until the standards are met by engineering controls.
These examples illustrate the principle that as one moves toward
performance standards and regulation of output rather than input
more alternatives are opened up, some of which are likely to be more
cost-effective. Allowing the use of personal protective devices usually
results in large cost savings for the same performance and frequently
allows greater protection than engineering controls since the tech-
nology to reduce hazards through this technique may not be currently
available. This is true of both the coke-oven and noise standards.

OSHA and U.S. Industry

Respirators if properly worn reduce carcinogenic emissions by 10- to
30-fold,38 while hearing protectors if properly fitted may achieve a
30-decibel attenuation.39
Since there is no safe level of carcinogenic emissions and some
sensitive people will suffer hearing loss below the 85-decibel level,
additional protection is offered by personal protective devices. Since
both the promulgated coke-oven standard and proposed noise stan-
dard require interim usage of personal protective devices until engi-
neering controls are in place, the standards could have the counter-
productive effect of increasing risks of material impairment after full
compliance. These inconsistencies may be resolved in court since the
American Iron and Steel Institute has challenged the standard in a
law suit before the Court of Appeals for the Third Circuit.
The importance of cost-effective analysis is best illustrated by the
proposed occupational noise standard. Table 5.1 shows the number of
workers protected from a 25-decibel hearing impairment after twenty
years of exposure for various alternative regulations mixing engineer-
ing and hearing-protector approaches.40 The first number in column 1
indicates the decibel level for the use of engineering controls, and the
second the level for hearing protectors assuming a 75 percent effec-
tiveness. The rule in cost-benefit analysis is to accept the alternative
that maximizes the amount of net benefits. But since we do not have
an explicit value for the loss of hearing twenty years from now, we
can present the implicit values of hearing loss implied by the various
policy options.
The analysis proceeds by listing the options in order of increasing
effectiveness or protection offered and calculating the marginal costs
of protecting additional workers. In the cases where marginal costs
decline, the less protective standard is eliminated from the policy set
since more workers can be protected at less average costs. This pro-
cedure eliminates six options and leaves six. The marginal costs of the
set of cost-effective standards are then plotted in Figure 5.1. The
policymaker can now explicitly examine the marginal cost per worker
protected by tightening the standard. Note that this analysis elimi-
nates all the engineering-only standards and the standard originally
proposed by OSHA 90/85. From this analysis a performance standard
allowing the use of hearing protectors at 85 decibels implies an eval-
uation of hearing impairment (twenty years from now) of $23,000,
while the least expensive engineering option comes at a cost of over
$300,000 per additional worker protected. Although we may quarrel


0l 0 n 0
0 o 0\ eo en

m *
0 i I 0
0-> d-o In
0 oo-oo

> U


* w *
0e 0 * *

-g .IS0 oO-e 0 N'? t00

ou |


m r, 0 00

Co o oo ow

z u

coo o

a 0 02% 00 l *
r^ to
ooS ooooo~<^-mt-i'oor-~ u ^

OSHA and U.S. Industry

with the exact numbers,41 the methodology of the analysis is logical
and illustrates the importance of examining a wide variety of options,
particularly with respect to the level of stringency.
The noise-control standard also illustrates another area where the
use of cost-effectiveness analysis might result in large cost savings
with no reduction in benefits. If marginal costs per worker protected
were equal across industrial categories, more workers could be pro-
tected at the same or lower costs. For example, an 85-decibel, engi-
neering-only standard will cost about $263,000 per worker protected
in SIC 22 (textiles) but only $26,000 for SIC 36 (electrical ma-
chinery).42 Clearly a tighter standard for electrical machinery and a
weaker one for textiles could protect more workers at lower costs.
OSHA's argument that this is not equitable since different workers
will be subject to different degrees of risk neglects the fact that pro-
tecting more workers overall at lower costs increases the amount of
nonrisk benefits, such as income, that all workers could share. OSHA's
concept of equity is too narrow.4"

300 90/80

S200 costs marginal
0 net

/ 80

50 hypothetical marginal benefits
curve implied by choosing -/85 standard

200 220 240 260
Number of workers protected

Figure 5.1
SOURCE: column (7) from Table 5.1
Only "cost-effective" options are considered.

Microeconomic Effects (A)

Final Considerations
Although OSHA's refusal to use cost-benefit analysis or even cost-
effectiveness analysis is usually justified by OSHA because its man-
date is to eliminate all risks in the workplace to the extent feasible,
such a strategy is actually counterproductive to the promotion of
health and safety since the opposition to OSHA standards is directly
proportional to the potential costs of compliance. Since OSHA's act
and administrative procedures in both standard setting and enforce-
ment allow endless legal challenges and delays, little progress has
really been made. Although OSHA painstakingly attempts to develop
and promulgate standards that survive legal challenge, the law is
probably flexible enough in this area that the legal muscle of the
opposition, which is probably proportional to economic impact, is
more important than current case law. Thus, apart from the nice eco-
nomic arguments, simple pragmatics argue that benefit-cost analysis
(as well as full balance sheet analysis of who gains and who loses)
should be a major consideration in standard setting.44
The political, lobbying, and legal opposition to OSHA's standard
setting would also be reduced if OSHA shifted toward the perfor-
mance-standard approach since most businessmen prefer performance
standards, at least until they have complied with specification stan-
dards. Going further along the continuum and tying penalties and
incentives to compliance with performance standards should also re-
duce industrial opposition at the same time that cost-effectiveness and
greater risk reduction are promoted, especially in the long run. The
large fixed costs of complying with a standard lead instead to legal
fights and noncompliance at the margin. A penalty system tied directly
to the degree of compliance leads to decisions to improve health and
safety at the margin and long-run incentives to develop risk-reducing
technology. And a logical extension of incentives tied to going beyond
the standard promotes further risk reduction.
In the case of the noise standard such a system might work as fol-
lows. A schedule of risk of hearing loss per worker (which is a func-
tion of level and length of exposure) must be multiplied by an ex-
plicit estimate of the social cost of hearing loss. Since promulgation
of a standard implies an implicit determination of this amount as dis-
cussed above, this is not an unreasonable request. A threshold stan-
dard could then be promulgated, and firms could be fined per worker
exposed according to the schedule and rewarded by direct subsidy or

OSHA and U.S. Industry

tax credit again based on the schedule. The threshold standard is set
by equity, financial viability, or political considerations since fines and
tax credits should produce the same incentives if tied directly to per-
The same principle could be applied to the coke-oven standard
with the fines, of course, much higher and possibly prohibitive in
some cases. It is important that such systems be truly performance- or
result-oriented so that personal protective devices may be used as an
option. In this case a firm would be fined if employees were not ef-
fectively wearing personal protective devices. Thus the firm will have
an incentive to ensure that employee behavior does not lead to ex-
cessive risk-taking, thereby combatting the problem of moral hazard.
Employees would be rewarded according to their total employment
record, including health and safety practices. One possibility is that
employees could be given bonuses for effectively wearing personal
protective devices.
Finally, let me stress that although these ideas may seem radical
or impractical, I believe that any progress in these directions will in-
crementally reduce workplace illnesses and injuries and at lower over-
all costs to society. And I believe that such progress is not only pos-
sible but likely.

1. See Washington Post, February 12, 1977.
2. Ibid.
3. For example, see Nicholas Ashford, Crisis in the Workplace: Occupa-
tional Disease and Injury (Cambridge: MIT Press, 1976); Robert Smith, The
Occupational Safety and Health Act (Washington: AEI, 1976); James R.
Chelius, "Expectations for OSHA's Performance: The Lessons of Theory and
Empirical Evidence" (Paper presented at a DOL Conference, Annapolis, Md.,
March 19, 1975); and Nina Cornell, Roger Noll, and Barry Weingast, "Safety
Regulation," in Setting National Priorities: The Next Ten Years, ed. Henry
Owen and Charles Schultze (Washington: The Brookings Institution, 1976).
4. Morton Corn, "Status Report on OSHA," reprinted in The Occupational
Safety and Health Reporter 6 (January 20, 1977):1096.
5. From "The President's Report on Occupational Safety and Health," GPO
No. 2915-0011 (May 1972), as quoted in Ashford, Crisis in the Workplace,
p. 47, and OSHA Reporter 6 (December 16, 1976):908-12.
6. See Cornell et al., "Safety Regulation," pp. 464-70. Note that uncertainty
is defined as not knowing the probabilities. Risk, on the other hand, involves
known probabilities and can be ensured against.
7. Ibid., p. 503.

Microeconomic Effects (A)

8. See Robert Smith, Occupational Safety and Health Act, pp. 73-85.
9. For a formal discussion of this point, see E. J. Mishan, Cost-Benefit
Analysis, rev. ed. (London: George Allen, Unwin Ltd., 1975), pp. 315-20.
10. See Martin Feldstein and Amy Taylor, "The Rapid Rise of Hospital
Cost," CWPS Staff Report (January 1977), p. 31. Total medical care losses
amounted to about $120 billion (8.3 percent of GNP) in 1975 ("Economic
Report of the President, 1975," p. 118).
11. Ashford, Crisis in the Workplace, p. 360. Also see John Rawls, A Theory
of Justice (Cambridge: Harvard University Press, 1971).
12. See National Commission on State Workmen's Compensation Laws,
"The Report" (1972).
13. Chelius, in "Expectations for OSHA's Performance," summarizes the
results of studies that attempt to determine the extent that workers contribute
to accidents. He reports that the best study sponsored by the State of Wis-
consin found that approximately 45 percent of accidents were due to workers'
behavioral problems. See Wisconsin State Department of Labor, Industry, and
Human Relations, "Inspection Effectiveness Report" (1971).
14. OSHA presently has the power to levy civil fines of up to $1,000 per
day per violation; thus there do not appear to be any operative legislative re-
straints on such an incentive structure.
15. The other three health standards promulgated by OSHA are vinyl
chloride, asbestos dust, and the fourteen carcinogens. All three were originally
promulgated under the temporary emergency standards Section 6(c) after
petitions by labor and public interest groups. See Ashford, Crisis in the Work-
place, pp. 249-52, for a discussion of these standards.
16. For example, see Nicholas Ashford, "Regulatory Occupational Health
and Safety: The Real Issue," Challenge (November/December 1976):39-42.
17. Cornell et al., "Safety Regulation," p. 465.
18. For an application of such a proposal to the problem of abating air-
craft noise complete with actual tax rates for airplane types, airport, and time
of day with the rates based on property-value impacts, see John Morrall, "A
Proposal to Reduce Airport Noise through the Application of a Decibel
Charge" (April 1976).
19. For example, Morton Corn, the past assistant secretary for OSHA, stated
that the only question that cost-benefit analysis should be allowed to address
is "the time period allocated to the highly impacted sectors to permit institu-
tion of the required controls to ensure safe and healthful environment." See
Morton Corn, "Status Report on OSHA," p. 1099.
20. See "Federal Regulation and Regulatory Reform," Report by the Sub-
committee on Oversight and Investigations of the House of Representatives,
94th Congress, 2d Session (October 1976), pp. 505-15.
21. Ibid., p. 515.
22. Ibid., p. 555.
23. Cornell et al. argue that in situations of unknown hazards where "cata-
strophic mistakes" of public policy might be made, a decision rule that should
be used is "minimax regret"-that is, to adopt strategies that avoid the worst
logically possible outcomes ("Safety Regulation," p. 469). However, once it is
realized that policy choices are not discrete and that a nonzero probability
multiplied by an infinite cost would lead cost-benefit analysis to the same con-
clusion, the two decision rules become one. Also if "minimax regret" is ap-
plied to noncatastrophic cases, a decision rule must be developed to rank the
order of decision-making and draw the line before resources are exhausted.
Such a decision rule should logically be cost-benefit analysis.

OSHA and U.S. Industry

24. See "Inflationary Impact Statement: Coke Oven Emissions," U.S. De-
partment of Labor, OSHA (February 27, 1976), and "Economic Impact Analy-
sis of Proposed Noise Control Regulation" prepared by Bolt Beranek and
Newman, Inc. for U.S. DOL-OSHA (April 21, 1976).
25. See "Statement of John F. Morall III on Exposure to Coke Oven
Emissions before OSHA" (Washington, May 11, 1976), CWPS-149, and "State-
ment of John F. Morrall III on Occupational Noise Exposure before OSHA"
(Washington, September 22, 1976), CWPS-187. Much of the following is
based upon these two papers.
26. "IIS Coke Ovens," pp. 2-3.
27. "Exposure to Coke Oven Emissions," 41 Federal Register 46749 (Oc-
tober 22, 1976). A lower estimate of $160 million was derived by CWPS by
comparing the cost estimate of the steel firm that would likely gain from the
standard, Inland Steel (since it had already complied the most to the proposed
standard), with the consulting firm's estimate of that steel firms remaining
costs of compliance. The consulting firm's estimate was actually lower than
Inland's. This comparison was then used to adjust the whole industry's cost-
estimate down to $160 million. This procedure should produce a better esti-
mate since Inland has had the most actual experience in complying with the
standard and because it has had no incentive to overstate costs, indeed it may
have had an incentive to understate them. (See Morrall, "Statement on Coke
Ovens," p. 4.) Note that OSHA misunderstood this point and rejected CWPS's
estimate because "Inland's situation is atypical of the steel industry because it
already has more new equipment than many other employers" (41 Federal
Register 46749, October 22, 1976). But this is the very reason that this esti-
mate is more reliable.
28. "EIS Noise Control Regulation," pp. 3-31.
29. See "Cost to Cut Job Noise Drops," Washington Post, May 21, 1975,
pp. F-9, 10, and "Noise Control and Its Economic Implications" by G. Warnaka
et al. (paper presented to the 89th Meeting of the Acoustical Society of Amer-
ica, Austin, Texas, April 7-11, 1975).
30. See "Occupational Noise Exposure," OSHA, 40 Federal Register 72337.
31. See Morton Corn, "Status Report on OSHA," p. 1099, and "Exposure
to Coke-Oven Emissions," 41 Federal Register 46750.
32. See Morrall, "IIS Coke Ovens," pp. 9-12, and Morrall, "EIA of Noise
Control," pp. 11-21.
33. See 41 Federal Register 46751.
34. Note that standards implemented by cost-benefit analysis could very
well result in firms and industries going out of business. These results would
be viewed as necessary costs of internalizing externalities.
35. These figures are from the IIS although they are never explicitly com-
pared in this manner. OSHA used a figure of 240 lives saved per year and an
annual cost of $241 million. However, the 240 figure results only after 45
years of compliance and 45 years of expenditures. The proper cost figure to
compare to the 240-lives estimate is about $2.4 billion at a 10 percent dis-
count rate or $10 million per life.
36. See Morrall, "IIS Coke Ovens," p. 11. The estimates implied by willing-
ness to pay analyses range from $200,000 for Thaler and Rosen ("The Value
of Saving a Life: Evidence from the Labor Market," NBER Conference, No-
vember 30, 1976) to $1.5 million for Robert Smith, Occupational Safety and
Health Act, using labor market data and occupational risk differentials. Also
Glenn Bloomgast, using what he believes is a data set with fewer problems,
estimates that automobile drivers implicitly value their lives when they buckle

Microeconomic Effects (A)

or do not buckle up at $257,000 ("Value of Life: Implications of Automobile
Seat Belt Use," Ph.D. thesis draft, University of Chicago [November 16, 1976],
p. 46). These numbers, of course, do not reflect the externality effects dis-
cussed above and therefore must be adjusted upwards.
37. See 41 Federal Register 24078-79 (June 14, 1976). However, these cal-
culations are the author's.
38. 41 Federal Register 46773-74.
39. "EIA for Noise Control," pp. 2-34.
40. This table and the following analysis is based on Morrall's "Statement
on Noise Control Regulations," pp. 11-16 and in particular Table 3. The
basic data for that study are all from the "EIA for Noise Control Regulations."
The data chosen in each case represent, according to the EIA, the best point
estimates. Additional sensitivity analysis based on this methodology should be
done before final standards are promulgated.
41. For example, the marginal-net-cost line in Figure 5.1 also illustrates the
effect of subtracting estimates of intangible benefits, such as the annoyance
costs of excessive noise and the reduction in employee absenteeism that might
result from a quieter work environment. These estimates are thought to be
biased upwards and are not applied to the hearing protector standards. These
adjustments do not affect the results since the same standards as before are
eliminated. See Morrall, "Statement on Noise Control Regulation," pp. 15-21.
42. See ibid., pp. 21-25, for details of this type of analysis.
43. This problem has also come up in the cotton-dust standard where dif-
ferent standards for different sectors of the industry (yarn production, cotton
ginning, cotton weaving, and waste processing) could result in greater protec-
tion at lower costs. However, OSHA again refuses to consider this approach.
(See "Inflationary Impact Statement: Cotton Dust," U.S. Dept. of Labor,
OSHA, October 1976.) However, as pointed out above, OSHA might consider
varying the length of time for compliance based on economic impact.
44. For a full exposition of the pragmatic benefits of benefit-cost analysis
in forming policy options, see Harold Luft, "Benefit-Cost Analysis and Public
Policy Implementation: From Normative to Positive Analysis," Public Policy
24 (Fall 1976):437-62.


James W. Ford
FIRST, I WANT TO RAISE a question about the presumption stated by
Mr. Morrall, which has been reflected otherwise, that it is socially
desirable for the government to invest in the discovery and dissemina-
tion of information on occupational hazards. One point Mr. Noll
made is that knowledge about conditions that create hazards is a
public good because it is of value to people other than those who
make the discoveries. It seems to me that this is not a sufficient cri-
terion for its being a public good. The same thing could be said about
the value of knowledge of the technological processes, for example.
There are clearly incentives for people to make such discoveries to
obtain property rights in them and sell them to others, and I think the
same incentives exist for employers and employees with respect to
knowledge about occupational hazards and safety and health hazards.
In other words, both employee and the employer have strong incen-
tives to make such discoveries, and the case for a government doing
so is one which, in general terms at least, escapes me. I would not
argue, though, that those incentives are necessarily or always suf-
ficiently strong or well expressed that there is no case for government
intervention; but it seems to me that intervention-and this will per-
haps seem curious coming from someone in business-ought to take
the harder form of coercion, which I distinguish from simply the pro-
vision of information that people may or may not use as they choose.
By coercive intervention I mean, of course, such things as standards
or alternatively taxes of one kind or another.
The approach that ought to be taken to increase or reduce the occu-
pational and safety hazards in business is to identify the areas in

Microeconomic Effects (A)

which, and the reasons for which, incentives to discover and apply
this knowledge of the hazards are weak. Thus, where there are hang-
ups in the incentives, some form of coercive intervention ought to be
What about taxes? Perhaps the burden of discovering the method
of dealing with the hazards-discovering what they are and their
costs-ought to be in the hands of those who directly benefit from
such knowledge, and that is the people who themselves may suffer
the damage or the costs of it (i.e., the employer or the employee
himself). So, what I recommended in this area is the general approach
to government intervention which says we ought to begin by identify-
ing the areas in which the desired results are not achieved or only
weakly achieved by private action; and where there are such areas-
for reasons of externalities, transactions costs, or information costs-
we should ask whether there ought to be intervention to establish
some form of alternative cost leading to incentives for the people
directly concerned to generate and use the knowledge.

Peter Barth
In general, I found much of John Morrall's paper to my liking. My
comments then are primarily aimed at correcting relatively minor
flaws in the paper or at elaborating on the points he makes.
Morrall, like many others before him, cites the figure of a hundred
thousand fatalities due to occupational diseases each year that was
initially reported in the first annual report by NIOSH. The figure is
totally inappropriate, having almost no basis in fact. While the "true"
level may be higher or lower-depending significantly on how one
defines an occupational disease-Morrall does himself a disservice by
simply repeating this totally fictitious "fact."
In citing reasons why it is appropriate for the government to inter-
vene in safety and health programs, Morrall might also have iden-
tified frictions that make the system even more imperfect. A major
friction, though not necessarily of the sort to cause the government
to become involved, occurs in workers' compensation insurance, where
40 to 45 cents of every premium dollar goes for administering the
insurance rather than for compensation benefits. In such cases em-
ployers payout a hefty sum that workers never receive.


While I agree with Morrall on his arguments regarding labor mo-
bility, several points need to be made. First, one can hardly be san-
guine about the prospects of the government assisting in such mobility.
Existing programs of retraining or moving manpower have hardly
been very successful. Moreover, in the case of carcinogens, moving
people rapidly into and out of jobs where they are exposed to such
hazards may be most dangerous. It is not clear that by moving peo-
ple out of contact with such hazards after only a short exposure you
reduce the risk of their contracting cancer. However, the technique
is being employed already in uranium mines where workers cannot
continue to work after they have been exposed to a threshold level of
Economists rarely dispute the need for providing information on
health and safety, subject to a cost constraint. But information should
be provided not only by government but by employers and unions in
order that workers may be better informed as to what they will en-
counter at the workplace.
The amount of relevant information in this field now known and
provided to workers is negligible. The National Cancer Institute needs
three years and over a hundred thousand dollars in order to test the
carcinogenicity of a single substance in only one type of test animal.
NIOSH has turned out only three dozen criteria documents (at huge
costs) since 1970-71, yet it lists over 22,000 entries on its toxic sub-
stance list.
Morrall argues that fines, tied to the extent to which a health or
safety violation deviates from some norm, may be more reasonable
than a standard. While I agree, the fine should also be a function of
the number of workers being exposed to that level of the hazard.
Such an approach has the salutary effect of encouraging firms to re-
duce the number of employees placed at risk, even if it cannot or will
not eliminate the risk entirely.
While the government may not be very good at calculating the costs
of imposing standards, there is no reason to suppose, as Morrall does,
that the government is good at calculating the benefits. It is naive to
argue that benefits of proposed standards are well measured, much
less objectively determined, simply because they are done "using sci-
entific methodology by academic experts."
How can benefits be well measured by anyone when we still do not
know the number of workers being exposed to certain risks? The

Microeconomic Effects (A)

techniques used by NIOSH in this regard have to date been incredibly
inadequate and may not approach the true extent of risk by an enor-
mous factor.
Benefits are also generally very badly estimated in the area of oc-
cupational health because our data on morbidity are even worse, if
that is possible, than on mortality.
I can hardly quarrel with the cost-benefit approach or Morrall's
view that sound judgment involves no more than evaluating costs and
benefits. Yet I am persuaded that the formal technique of analysis
cannot be explicitly used in the setting of policy in this arena, par-
ticularly where lives are at stake. Thus the critical need is for the
technocrats to provide policy people with cost-effectiveness analyses,
so that we can more efficiently produce the ends that are set by politi-
cal and social criteria.
Morrall argues that employers will resist OSHA less if a perfor-
mance-standard approach is used. My view of the world is more prag-
matic. The extent to which employers will fight is primarily a function
of what they believe it will cost them. A performance standard set at
a level that will be perceived as expensive will be vigorously resisted,
while a cheap engineering standard (with no concentration on out-
comes) will get very little attention.
While many of OSHA's past practices seem extreme and/or absurd,
it is hardly fair to criticize the agency for carrying out the bidding of
Congress. As one who has witnessed the problems of evaluating vari-
ous manpower programs because no clear legislative purpose is stip-
ulated, I marvel at the forthright manner in which Congress specified
the purpose of OSHA. The law requires that the Department of Labor
eliminate injuries and illnesses at the workplace, irrespective of cost
and rejecting an assumption of risk approach by the worker. While
economists can lament the way the act was written, it hardly seems
fair to indict the bureaucrats who are charged with carrying out the
law for doing so.

H. Michael Cushinsky
My comments will be limited to two specific areas of John Morrall's
paper and Nina Cornell's remarks: the relationship of mandated
standards and level of moral hazard, and the comparative institu-
tional choice between the government and the insurance industry in
providing incentives for safety.


Moral Hazard and Mandated Standards
Critics of OSHA here and elsewhere point to the cost-effectiveness
deficiencies in the system of mandated standards. However, when one
accounts for differential moral hazard levels prior and subsequent to
OSHA, mandated standards can be shown deficient in benefit re-
spects, that is, in delivering the desired level of safety.
Pauly (4) states that moral hazard can be manifested through any
of the following: overuse of insured services, excessive exposure to
hazard, and reduction in risk-mitigating activities. Obviously, in
workers' compensation insurance these behavioral manifestations can
be exhibited by both the employer and the employee; workers' com-
pensation is an exception in this regard.
In their chapter on safety regulation, Cornell et al. (1) state that
legislative assumptions underlying OSHA include that the principal
cause of inadequate occupational safety consists of the bad acts of
businessmen. This legislative posture is seemingly insensitive to the
employee's behavior as a determinant of accident levels. However, a
study sponsored by the State of Wisconsin (7) found that approxi-
mately 45 percent of accidents were due to workers' behavioral prob-
lems; this determinant of accident levels can hardly be ignored.
With these facts in mind, a closer examination of the employee's
side of moral hazard seems appropriate. Moral hazard can be re-
flected in two ways. One is through falsifying injury. Lommele and
Sturgis (2) have demonstrated significant correlation between the un-
employment rate in the previous year and the current year's workers'
compensation claims. This reflects the overuse of the insured-services
aspect of moral hazard; the impact of mandated standards in this area
is questionable.
The second way moral hazard can be reflected is through the level
of carelessness or horseplay, which can be viewed as the workers'
compensation analogue to reduction in risk-mitigating activities and
excessive exposure to hazard. This topic has been somewhat de-
emphasized in discussion of mandated standards. In evaluating vari-
ous public-policy alternatives and in choosing mandated standards,
policy-makers could implicitly be holding the level of moral hazard
constant (especially this aspect). There is no basis for this assump-
tion in either theoretical or empirical research.
Adaptation of an analysis of automobile safety regulation by Peltz-
man (5) might enhance the discussion. Figure 6.1 graphs the prob-

Microeconomic Effects (A)

ability of occupational injury versus level of carelessness or horseplay.
Prior to OSHA the employee is confronted with trade-offs depicted
by ray I and chooses the desired level of carelessness or horseplay
with commensurate probability of injury represented by point A.
After passage of the Act the implementation of mandated safety
standards can be represented by a reduction in the probability of in-
jury for each level of carelessness or horseplay, illustrated by ray II.
If the moral-hazard issue is ignored, policy-makers are implicitly as-
suming that the elasticity of level of carelessness with respect to
changes in the relevant probabilities of injury is zero. This is de-
picted by point B, which reflects the same level of carelessness with
its decreased probability of injury. However, there is no reason to
expect that this elasticity is zero and that the level of horseplay or
carelessness remains constant. Confronted with a new set of trade-
offs, and assuming horseplay/carelessness is a positive good, the re-
sult could be a point such as C. There is no a priori reason not to





Level of carelessness or horseplay

Figure 6.1


believe that the result is a point such as D, which reflects a higher
probability of injury after safety standards are implemented. If the
benefits of mandated standards are calculated with the assumption of
point B, it should not be surprising if these benefits are not achieved.

Experience Rating vs. Mandating Standards
In employing the comparative cost approach in the choice of institu-
tions, John Morrall implies the coexistence of experience rating
through the private insurance system and government mandates. A
brief survey of the history of experience rating in workers' compensa-
tion will serve as a comment on this approach.
The initial experience-rating mechanism was developed early in
this century and was rather sophisticated. Termed experience and
schedule rating, the mechanism provided credits for management's
attitudes toward safety and its programs, as well as credits for better-
than-average loss experience. This system was short-lived. A report
(6) on this experience and schedule-rating system issued by the Penn-
sylvania Inspection Bureau stated that by emphasis on minor services
the experience and schedule-rating system actually impeded safety
work. Other reports (2) stated that morale factors broke down under
the pressures of competition and that the system was generally aban-
doned because it ceased to contribute to loss prevention-that is,
rating credits were applied rather arbitrarily, thereby losing their ef-
fectiveness as an incentive technique. My own observation is that the
failure of this type of experience rating was due to the interplay of
strictly regulated rates and the insurance distribution system (the
American agency system).
To relieve the rating mechanism of its arbitrariness, schedule rating
was abandoned and a system of computed experience credits without
any subjective credits for management policies was implemented. This
is basically the experience-rating system employed in workers' com-
pensation today. The benefits of such a system are limited by the
following: first, differing perceptions of risk by the insured and the
insurer, which are reflected in the insured's desire to have more cur-
rent data emphasized in the rate-making calculation and the insurer's
preference for a larger amount of data over a longer period; and
second, application of the experience-rating system to less than 20
percent of total workers' compensation risks. An insured must de-
velop $750 in premium to become eligible for the program, and this

Microeconomic Effects (A)

disallows the great majority of risks. This is not an arbitrary exclusion
but one that results from the application of accepted actuarial prin-
In their chapter on safety regulation, Cornell et al. (1) note that
the uniqueness of a particular occupational disease hazard makes the
probabilities underlying the risk difficult to estimate, thereby causing
high uncertainty levels, which result in market failure. The existence
of this same small-numbers condition in the number of exposure units
of smaller risks also makes their probabilities difficult to estimate,
resulting in the failure of the market to use experience rating as an
incentive device. In conclusion, I am not sanguine about the role of
experience rating in the insurance industry in providing incentives to

1. Cornell, N, W.; Noll, R. G.; and Weingast, B. "Safety Regulation." In
Setting National Priorities: The Next Ten Years, edited by H. Owen and C. L.
Schultze. Washington, The Brookings Institution, 1976.
2. Lommele, J. A., and Sturgis, R. W. "An Econometric Model of Work-
men's Compensation." Proceedings of the Casualty Actuarial Society 61
3. New York State Compensation Insurance Rating Board. "Memorandum
on Schedule Rating." New York: The Board, February 3, 1941.
4. Pauly, M. V. "The Economics of Moral Hazard: Comment." American
Economic Review 58 (June 1968):531-37.
5. Peltzman, S. "Effects of Automobile Safety Regulation." Journal of
Political Economy 83 (August 1975):677-727.
6. Pennsylvania Compensation Rating and Inspection Bureau. "Memorandum
on Schedule Rating." Philadelphia: The Bureau, July 21, 1933.
7. Wisconsin State Department of Labor, Industry, and Human Relations.
"Inspection Effectiveness Report" (1971).


General Discussion

Eric Zolt
I AM NOT HERE TO DEFEND OSHA, but it is time to stop swapping
stories about split toilet seats and to start investigating the effect of
OSHA and its current record rather than its past. The effectiveness
of OSHA has changed; the horror stories should be put aside and
analysis substituted. I agree with most of John Morrall's paper, but
a couple of areas bother me.
Although administrative costs are high, there has not been an in-
depth discussion of how they would vary on a continuum between
the proposed tax and the mandated-standards approach. If we take
Morrall's calculus (basing fines on number of workers affected, in-
tensity of harm, and length of exposure) and recognize how many
workers, workplaces, chemical hazards, and other hazards are in-
volved, and the number of inspections required, any system, be it tax
or standards, is going to be hopelessly bureaucratic and the costs will
be quite high. More empirical work is needed to determine where the
administrative costs are higher. My hunch is that it will be in the
"fine" system.
Morrall states that in the case of two standards, quantification of
the benefits of risk-reduction (carcinogenic risk and risk in the case
of hearing loss in industrial-noise cases) appears to be fairly reliable
and objectively determined. I have studied the occupational-noise
records at some length, and I disagree. Too often charges have been
leveled at Congress that they measure precisely the wrong thing. The
occupational noise standard may be a case in point. The Council on
Wage and Price Stability testimony measures the value of hearing
impairment based on the decline in value of the land around the air-

Microeconomic Effects (A)

port. Among things that make these data inappropriate to hearing
impairment is the level of noise around the airport at the 70-dB range,
whereas in the workplace it is in the 90- to 100-dB range. Also the
relationship is such that every 5-dB increase means double the level
of noise.
This was just the measure of the noise factor around the airport.
There is no hearing impairment from 70 dB. Hearing impairment
starts at a much greater level as you go up the decibel scale past 80
to 90, and this was not considered. A further element is that people
living around the airport often are there on a more voluntary basis
than workers in cotton mills or other noisy places. Other elements
tend to bias these the other way. The point I am trying to get across
here is that it is not easy to quantify the benefits.
At the Center for Public Policy Analysis we are currently conduct-
ing a study on the copper smelter industry similar to Dr. Heggestad's
study. We are looking at costs and benefits of regulations affecting
production of copper, including the sulphur oxide emission standards.
Noll earlier stressed the difficulty in getting from sulphate pollution
to sulphate emission level. We have had a similar experience, plus the
additional difficulty in getting from the pollutant to the biological im-
pact. It is not so clear any more that the level of sulphate in the
atmosphere is directly related to mortality rates.
The third point is that companies will not litigate and fight fines
the way they will fight mandated standards. For fines to be effective
(if they are through the workers' unemployment compensation sys-
tem) one must note the fact that a large number of worker injuries
are not reported.
Therefore, in order for the size of the fine to be economically ef-
fective, we have to consider the probability that the workers' com-
pensation would be recorded. If you go to an injury tax, you would
have to consider the probability of the firm actually being cited, tried,
and actually fined. You would have to adjust the size of the fine, if it
is going to be economically deterrent. There may be some legal prob-
lems with fines of this size. As the fines get bigger and more "crimi-
nal" in nature, there are likely to be more lawsuits testing their con-
Finally, John Morrall states that one of the reasons there is a mar-
ket imperfection is that individuals have a social discount rate dif-
ferent from the rate of the firm. If the burden is switched to the
firm to get the necessary equipment to comply, or to pay the tax,

University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - Version 2.9.7 - mvs