united States
Environmental Protection
Agency
Office of Pesticides
and Toxic Substances
Washington. DC 20460
EPA-560/12-80-002
October 1980
Toxic Substances
Supporting Innovation
A Policy Study
-------
EPA-560/12-80-002
September, 1980
SUPPORTING INNOVATION:
A POLICY STUDY
Contract No. 68-01-5878 ,
REGULATORY IMPACTS BRANCH
ECONOMICS & TECHNOLOGY DIVISION
OFFICE OF TOXIC SUBSTANCES
WASHINGTON, D.C. 20460
U.S. ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF PESTICIDES AND TOXIC SUBSTANCES
WASHINGTON, D.C. 20460
-------
DISCLAIMER
This document is a contractor's study done with the
supervision and review of the Office of Pesticides and Toxic
Substances of the U.S. Environmental Protection Agency. The
purpose of the study was to develop and evaluate policies to
support innovation in the chemical industries.
The report was submitted in fulfillment of Task Order 1 of
Contract Number 68-01-5878 by the subcontractor, MIT's Center for
Policy Alternatives. Work was completed in August 1980.
The study is not an official EPA publication. The document
can not be cited, referenced, or represented in any court
proceedings as a statement of EPA's view regarding the chemical
industries, or the impact of the regulations implementing TSCA.
-------
Acknowledgements
The authors wish to acknowledge and thank Edward Greenberg of
Washington University in St. Louis, who served as an economic consultant
to the project, and Renee Ford who edited the final report. We also wish
to thank Berenice Hanrahan, Senior Staff Assistant at CPA for her
extraordinary efforts in supporting this project and in producing several
versions of this report under tight time constraints. She was ably
assisted by Patricia Harrision, Robert Kaplan, Ellie Phelan, and Carole
Richmond.
-------
3.
TABLE OF CONTENTS
Page
Acknowledgement
List of Figures
List of Tables
i. EXECWTiVE. sumKRY 1-1
1.1 The Background and Purpose of the Report 1-1
1.2 Effects of TSCA Regulation on Chemical Innovation 1-2
1.3 Design and Analysis of Policy Options to Offset
Unnecessarily Restrictive Impacts of TSCA on Innovation. . 1-5
1.4 Results of the Policy Analysis I-11
1.5 A Comprehensive Program Opportunity 1-16
1.6 Conclusion !-18
2. THE EFFECTS OF REGULATION ON TECHNOLOGICAL INNOVATION:
A BACKGROUND REVIEW 2-1
2.1 Effects of Regulation on Innovation for Main
Business Purposes 2-2
2.1.1 Changes in Expected Profitability 2-2
2.1.2 Changes in the Number of Innovations that Fail
' for Environmental, Health, and Safety Reasons . . 2-3
2.1.3 Changes in Investment Opportunities Due to
Increased Environmental, Health, or Safety Risks. 2-4
2.1.4 Diversion of Managerial Personnel 2-4
2.1.5 Diversion of R&D Resources 2-5
2.1.6 Ancillary Innovation from Redirected R&D 2-6
2.1.7 Regulation-Induced R&D and Process Improvement. . 2-6
2.1.8 Rechanneling Creativity 2-7
2.1.9 Change in Industry Structure 2-8
2.2 Effects of Regulation on Compliance Innovation 2-10
2.2.1 Redirection of Technological Capabilities for
Environmental, Health, and Safety Purposes Only . 2-10
2.2.2 Saleable Compliance Technologies 2-11
2.2.3 Compliance Technologies with Ancillary Benefits . 2-12
2.2.4 Joint R&O Efforts for Compliance 2-12
2.2.5 Reorganization of Firms to Meet Compliance
Requirements 2-12
2.2.6 Information Sources for Compliance Technology . . 2-13
2.3 Effects of the Kefauver-Harris Amendments on Drug
Innovation 2-14
2.4 Implications for the Probable Impacts of TSCA on
Technological Innovation 2-17
References 2-19
ii
-------
4.
Page
3. A FRAMEWORK FOR ASSESSING THE EFFECTS OF TSCA ON CHEMICAL
INNOVATION -. 3-1
3.1 Legislative Background 3-1
3.2 Administrative Discretion 3-2
3.3 Understanding the Issue: A Framework. 3-3
3.4 The Origins of Unnecessarily Restrictive Impacts on
Innovation , 3-9
3.5 Small Firms, Small Volume Chemicals, and New Entrants. . . 3-10
3.6 The Stimulation of Safer New Products and Processes. . . .3-11
3.7 Conclusion 3-12
4. POLICY OPTIONS FOR CONSIDERATION 4-1
4.1 Developing Policy Options 4-1
4.2 The Options 4-5
5. A PROCESS FOR ASSESSING THE POLICY OPTIONS. . 5-1
5.1 Criteria for Assessment 5-1
5.2 A Screening Process 5-4
5.2.1 A Brief History of the Use of Group Judgment
Methods ,.5-5
5.2.2 The Procedure Followed in This Project 5-5
5.2.3 Group Judgment Elicitation Sheet 5-6
5.2.4 Group Judgment Integration Model 5-8
References 5-12
6. AN ASSESSMENT OF THE POLICY ALTERNATIVES 6-1
6.1 Results of the Magnitude Estimation Procedure 6-1
6.1.1 Overall Rating and Rank Orders 6-1
6.1.2 Policy Ratings on Each Criterion 6-7
6.1.3 A Consistency Check on the Overall Rankings 6-9
6.1.4 Political Feasibility of the Policies. . 6-9
6.2 Discussion of the Top Seven Ranking Policies 6-13
6.3 Discussion of the Seven Lower Ranking Policies 6-16
6.4 Relationship of Policy Ranking to General Policy Areas . . 6-18
6.5 Conclusions Regarding the Assessment of Policy
Alternatives 6-23
7. FINANCING THE POLICY OPTIONS 7-1
7.1 Available Financing Modes 7-1
7.2 Matching Policy and Financing Options 7-5
7.2.1 Budget Outlays 7-5
7.2.2 Off-Budget Financing 7-8
iii
-------
5.
Page
8. A COMPREHENSIVE PROGRAM OPPORTUNITY 8-1
8.1 The Top Seven Policy Options 8-1
8.2 Discussion of the Comprehensive Program 8-4
8.3 Conclusion 8-6
APPENDIX A. THE CHEMICAL INDUSTRY, WELFARE ECONOMICS AND GOVERNMENT
ACTIVITY A-l
A.I Purpose and Approach A-l
A.2 Fundamental Reasons for Government Responsibility and
Intervention A-3
A.3 Nature of Low-Volume Chemicals A-6
A.4 Nature of Chemicals Relative to Other Goods A-7
A. 5 Nature and Extent of TSCA Impacts A-8
A.6 Demand/Supply Dynamics A-11
A.7 Microeconomic Welfare Effects A-l8
A.8 Micro-Macroeconomic Welfare Effects A-22
A.9 Examples of Chemical Products Illustrating Differing
Market Structures A-24
References A-32
APPENDIX B. INNOVATION IN THE CHEMICAL INDUSTRY B-l
B.I A Model of Technological Innovation B-2
B.2 The Structure of the Chemical Industry B-5
B.3 Inputs to the Innovation Process in the Chemical Industry. B-13
B.4 Evidence on Trends in Chemical Innovation B-20
B.5 Chemical Industry Trends That May Affect Innovation. . . . B-29
B.6 Major Findings B-31
References B-33
Appendix C. DETAILED LIST OF CRITERIA FOR ASSESSING INDIVIDUAL
REGULATORY PROGRAMS C-l
Appendix D. SOURCES USED TO IDENTIFY CANDIDATE POLICY OPTIONS . . . . D-l
IV
-------
6.
Page
LIST OF FIGURES
3.1 Some Scenarios For The Future Course Of Chemical Innovation
If TSCA Had Not Been Passed. 3-5
3.2 Some Scenarios For The Course of Future Innovation With TSCA
In Place (Compared With The No-TSCA Business As Usual Scenario). 3-7
5.1 Rating Sheet 5-7
5.2 Anchor Program Description . 5-9
5.3 Criteria For Initial Assessment of General Programs 5-10
6.1 Trend Of Overall Score With Rank Order For 32 Policies 6-5
6.2 Correlation Between Two Measures Of Overall Policy Ranking . . . 6-10
6.3 Relationship Of Rank Orders Of Policies On Political
Feasibility And Overall Ratings 6-12
A.I Production Economics For A Single Firm A-12
A.2 Effect Of Testing Costs On Single Firm Production Economics. . . A-12
A.3 Impact Of TSCA Requirements On The Supply And Price Of A
Chemical Good A-15
A.4 Marginal Analysis Of The Shift Of The Supply Curve For A
Chemical Good Under TSCA A-15
A.5 Change In Consumers' Surplus Under TSCA A-16
A.6 Short-Run Change In Producers' Surplus Under TSCA A-16
-------
7.
Page
LIST OF TABLES
1.1 Policy Options for Consideration 1-6
1.2 Matching Policy and Financing Options 1-9
1.3 Results of the Policy Assessment 1-12
1.4 Ranks on Individual Criteria for Seven Policies Ranked
Highest Overall 1-15
4.1 Policy Options for Consideration 4-3
6.1 Results of the Policy Assessment 6-2
6.2 Policy Ratings on Each Criterion in Rank Order 6-8
6.3 Policy Ratings on the Political Feasibility Criterion 6-11
6.4 Ranks on Individual Criteria for Seven Policies Ranked
Highest Overall 6-14
6.5 Ranks on Individual Criteria for Seven Policies Rankeo
Lowest Overall 6-17
7.1 Matching Policy and Financing Options 7-3
8.1 Some Aspects of the Top Seven Policy Options 8-2
8.2 Ranks on Individual Criteria for Seven Policies Ranked
Highest 'Overall 8-3
A.I Examples of Chemicals in Various Categories A-26
B.I Standard Industrial Classification System for the Chemical
Industry B-8
8.2 Chemical Industry Concentration Ratios Based on the Value of
Industry Shipments in 1972 B-10
B.3 Measures of Chemical Sector Concentration B-12
B.4 Chemical Industry R&D Expenditures B-15
B.5 Number of Full Time Equivalent R&D Scientists and Engineers
1957 to 1976 B-18
B.6 A Summary of Studies of Chemical Innovation B-21
VI
-------
1. EXECUTIVE SUMMARY
1.1 The Background and Purpose of the Report
After several years of debate the Toxic Substances Control Act
(TSCA) was enacted into law in 1976. Its purpose is to protect health
and the environment from unreasonable risk of injury resulting from the
production, use and disposal of chemical substances. The Act was
designed to fill the gaps in existing regulation of toxic substances by
establishing a framework within which new chemical substances can be
assessed for potential hazard before they are marketed and widely
distributed, and by establishing a structure that broadens the authority
of EPA to regulate existing and new chemicals.
The framers of this legislation recognized that it had the potential
to influence the process and the outcomes of technological innovation in
the chemical and related industries. Therefore, the Act states at
Section 2(b)(3):
Authority over chemical substances and mixtures'should be
exercised in such a manner as not to impede unduly or create
unnecessary economic barriers to technolgical innovation while
fulfilling the primary purpose of this Act to assure that such
innovation and commerce in such chemical substances and
mixtures do not present an unreasonable risk of injury to
health or the environment.
The concern for innovation has a number of origins. First, a major
purpose of the Act is that the industries involved should be encouraged
to develop and market safer, more healthful new chemicals that can
substitute for existing hazardous chemicals now on the market. Thus, the
law encourages a systematic shift in industry priorities for new product
development. Second, the development and marketing of new chemical
products is the basis for the historic pattern of rapid growth in the
chemical industry and for its contributions to meeting social needs and
to the growth of the economy. Innovation in these industries includes
8.
-------
1-2
not only a wide variety of minor variations on existing chemicals, but
also major new chemicals that can address social problems or that might
be the basis for future industry growth. In this regard, Section 2(c)
makes clear that Congress is concerned that EPA carry out the Act with
due regard for its environmental, economic and social impacts - an
indication of concern for the economic strength of the industry and for
its contributions to society.
The purpose of the present project is to design, analyze, and assess
alternative policies that might be used to address the problems TSCA
might create for innovation, while maintaining the dominant thrust of
TSCA to protect health and the environment from unreasonable risk of
injury and disease. Such policies might be implemented by EPA through
administrative actions, or they might require Congressional action to
amend the Act or to establish new complementary authorities elsewhere in
EPA or other government agencies. The ongoing policy discussions and the
literature on technological innovation yield many suggestions of such
policy options, and a major concern of this research has been to assess
and analyze the potential of those options to contribute to the solution
of the problem at hand in a cost-effective and responsible manner.
1-2 Effects of TSCA Regulation on Chemical Innovation
It is useful to think of technological innovation as a process whose
outcomes are new, commercially successful products, processes, systems or
services. The process involves inputs of human and financial resources
to such activities as research, invention, development, testing,
marketing and diffusion.
There is not a good understanding of the nature and sources of
chemical innovation. For example, there are no sound data on the number
of new chemicals marketed each year, or on the contributions of small and
large firms or new entrants to chemical innovation. A few studies have
found that large firms are more innovative than small ones, but even
9.
-------
1-3
these results are open to serious question. A variety of factors in the
scientific, financial and competitive environment of the chemical
industry are changing, and even if TSCA had not been passed historic
trends in chemical innovation are unlikely to be followed in the future.
Contributing to this is the fact that developments in products liability
and in other environmental and occupational health and safety regulation
are influencing chemical innovation quite apart from TSCA's effects.
Superimposed on the uncertain future of chemical innovation are the
variety of effects that the TSCA regulatory requirements may have.*
Environmental, health, and safety regulation can act through a variety of
mechanisms to inhibit, stimulate or redirect technological innovation,
depending on the circumstances. Since TSCA features several different
regulatory stimuli, and since the "chemical industry" is in fact a
combination of many very different kinds of industries in various stages
of maturity whose products differ greatly in the hazards they present; it
is to be expected that inhibition, stimulation, and redirection will all
occur at the same time. However, the current understanding of the
interaction of regulation and innovation does not allow one to predict
the quantitative impact of TSCA on the rate of chemical product
innovation.
The inhibition of innovation by TSCA could arise, for example, from
the marketing delays, testing costs, resource diversion, and commercial
uncertainties it would introduce into the innovation process. The
stimulation of innovation could arise, for example, from the increased
staff diversity and revised corporate decision-making process required to
comply with TSCA. Redirection could arise from firms electing to seek
safe substitutes or abandon lines of research into chemicals expected to
pose a high risk to health. The inhibition of innovation is more likely
to occur in small firms, new entrants, and makers of innovative specialty
products, while stimulation is more likely to occur in large,
established, mature firms that use highly-integrated process technology.
*Regulation.in this context includes .both the procedural requirements
such as premanufacturing notification and substantive requirements such
as use restrictions or testing requirements.
10.
-------
1-4
At the same time, regulation can also stimulate innovation in some small
firms and new entrants and inhibit it in some mature firms. Thus,
redirection due to TSCA can occur in both the nature and sources of
chemical innovation.
The main purposes of TSCA are to slow the rate of introduction
and/or encourage the more prudent use and operation of products and
processes that pose unreasonable risks of injury to health and the
environment. Thus, some inhibition and some re-direction of chemical
innovation was expected due to TSCA - it was part of the social bargain
struck by Congress. Therefore, an observation that the rate of chemical
innovation has declined, or that the nature of chemical innovation has
shifted is not, by itself, grounds for determining that EPA has acted
"unduly" or "created unnecessary economic barriers to innovation."
Offsetting policies should not attempt to return the rate and direction
of chemical innovation to some hypothetical pre-TSCA baseline.
Nonetheless, despite EPA's best intentions, the implementation of
TSCA may unnecessarily restrict technological innovation. Two concepts,
regulatory fine tuning and transition phenomena, can help explain the
origins of the unnecessarily restrictive impacts of TSCA on chemical
innovation.
First, TSCA gives the Administrator of EPA considerable discretion
in carrying out the purpose of the Act. Despite the many special
provisions and wide latitude for decision making embodied in the Act,
rules to implement TSCA are likely to bear more heavily on some parties
than on others in ways which are unnecessary to accomplish the regulatory
goals. This is likely to happen as a result of the need to compromise
fine tuning of the rules and procedures for political and administrative
feasibility. Furthermore, the very complexity of TSCA may cause
unnecessary burdens for some regulated parties, such as costs, delays or
uncertainties, that would unnecessarily restrict innovation.
Second, when a new law is passed that is intended to influence
industrial behavior, a finite period of time elapses while the rules and
11.
-------
1-5
procedures to implement the law are adopted. During this period, firms
and investors may perceive a high level of uncertainty in making business
decisions. Also, for a time after the law is passed and implemented, the
infrastructure necessary to respond to the law's requirements may not be
in place. During this period, newly-regulated firms can be seriously
disrupted, and smaller firms may even disappear with the result that
innovation declines, even for safer chemicals.
Congress and the agency do not intend to burden industry with these
transition phenomena or to use rules and procedures that are inadequately
tuned to the needs of industry, yet some problems are inevitable if a
vigorous new regulatory program is to be put in place to accomplish the
primary goal of controlling unreasonable risk. To the extent that the
regulations unduly inhibit or create unnecessary barriers to
technological innovation, and to the extent that these undesirable
effects can be corrected by policies whose costs are commensurate with
the benefits they offer, EPA and/or Congress may wish to take action to
put such policies into action.
1.3 Design and Analysis of Policy Options to Offset Unnecessarily
Restrictive Impacts of TSCA on Innovation
- Based on a comprehensive review of the literature and on a general
understanding of the influence of regulation on innovation, thirty-two
policy options for offsetting the unnecessarily restrictive impacts of
TSCA on chemical innovation were developed. These options, which are
discussed in detail in chapter 4, are intended to be widely
representative of the possibilities open to government and to reflect the
options proposed by various interest groups. Table 1.1 lists the
options, categorized by the general approach they use.
The policy options were judged using a structured approach by each
of the project team members who rated each of the thirty-two policies on
seven separate-criteria. The individual judgments were combined into an
overall rating of each policy relative to the others. The procedure was
12.
-------
1-6
TABLE 1.1 Policy Options for Consideration
Reducing the Cost of New Chemical Development
A. Direct cost subsidy for general new chemical development via
grant mechanism.
B. Direct cost subsidy for general new chemical development via
loan mechanism (or loan guarantee).
C. Direct cost subsidy for testing/compliance costs of new chemical
development via grant mechanism.
D. Direct cost subsidy for testing/compliance costs of new chemical
development via loan mechanism (or loan guarantee).
E. Indirect cost subsidy for chemical innovation generally via tax
mechanism.
F. Indirect cost subsidy for testing and compliance costs via tax
mechanism.
Increasing the Financial Rewards for New Chemicals
G. Increased patent life for new chemicals.
H. Strengthened trade secret protection by limitations on EPA
authority to release information.
I. Decreased taxes on sales of new chemicals.
'3&~-
Increase the Availability of Capital for New Chemicals --
-i.jj,
J. Increased capital availability for new chemical development via
government supported venture capital company.
K. Increased capital availability for new chemical development via
tax changes or via SEC rules.
Reduce the Commercial Risk Associated with New Chemicals
L. Reduce risk through goverment financed insurance for regulatory
losses.
M. Reduce risk through government procurement of new chemicals.
N. Reduce risk from products liability actions by establishing
limits on liability.'
Reduce the Cost of Testing ^m
0. Establish government testing for TSCA requirements.
13.
-------
1-7
TABLE 1.1 Policy Options for Consideration
(continued)
Reallocations of Cost within the Private Sector
P. Sharing of test data with reimbursement.
Q. Facilitate private sector joint R&D or joint testing.
Information-based Strategies
R. EPA dissemination of chemical information—test results and/or
labeling.
S. Chemical technology extension service, including dissemination of
information on test and compliance methods.
Changing Market Structure
T. Antitrust action to favor new, small firms in the chemical
industry.
U. Tax adjustment to favor small firms or new entrants in the
chemical industry.
Improving the Technology Necessary for Compliance
V. Government support to develop new, better test methods.
W. Government support for education and training programs.
Regulatory Changes
X. Actions against existing substitutes for new chemicals.
Y. Fixing time periods for regulatory actions.
Z. Post-market surveillance of PMN's.
AA. Regulatory exemptions for low volume, new chemicals.
SB. Regulatory exemptions for small firms.
CC. Regulatory exemptions for "low risk" chemicals.
DO. "Fast track" PMN's for safe and/or major innovations.
EE. Generic PMN for classes of new chemicals.
FF. Improve EPA staff capability to assess impact of regulatory
actions on innovation.
GG. "No-intervention" policy; (i.e., no change from existing TSCA
regulation).
14.
-------
1-8
used to gain semi-quantitative insight into the relative merits of each
po.licy and to stimulate structured discussion of each option by the
project team.*
1
The seven criteria used to formulate the overall policy ratings are:
1. Capacity to countervail
2. Private costs
3. Public costs
4. Administrative feasibility
5. Time to implement
6. Supportive of TSCA's aims
7. Other side effects.
An eighth criterion, political feasibility, was also used, but not inlfni'
forming the overall ratings. Subsequently, two other criteria were
added: criterion 9, "initial policy rating," which is an initial estimate
of the overall rating of a policy made without reference to the detailed
criteria, and criterion 10, "effectiveness," which represents a
combination (the product) of a policy's capacity to countervail and its
administrative feasibility.
The question of financing the public costs of each policy was
treated separately from assessment of the relative rating of each
policy. In general, on-budget financing is regarded as a superior
alternative to off-budget financing. On-budget financing is both more
predictable and more reviewable than off-budget expenditures. In
addition, it fits well within existing budgetary and institutional
structures, while off-budget financing often requires the establishment
"' ,r
of new institutions. In examining the 32 policy options, the arguments
for on-budget financing have appeared to be especially persuasive when
the public costs of new programs are small, or when they require little
in the way of new institutional structures. Matching the policy and
financing options is illustrated in table 1.2. No attempt was made to
select the best financing method for each policy option.
*An economic model could not be used to analyze the options because
available models do not adequately address the dynamics of technological
innovation, the data needed for such an assessment are lacking, and
non-economic-factors such as administrative feasibility and effects on
the primary goals of TSCA must be considered.
15.
-------
0\
TABLE 1.2
Hatching Policy and financing Options
Policy
A. Grant
Subsidy
* <
B. Loan Subsidy
Loan Guarantee
C. Testing Grant
0. Testing Loan
Testing Loan Guarantee
£. General Tax Subsidy
F. Testing Tax Subsidy
G. Patent
H. Trade Secret
1 . Decreased Taxes
J. Venture Capital
Company
K. Increased Capital
L. Insurance
N. Procurement
N. Liability
Limits
). Government
Testing
'. Sharing Test
Data
Budgetary Outlays
Real location of
Discretionary
Funds
New EPA
Programs
X
X
X
X
X
X
X
New Non-EPA
Authority
X
X
X
X
X
X
X
X
Off-Budget Outlays
Tax
Expenditures
X
X
X
X
New
Taxes
X
X
X
X
X
X
X
X
New Financial
Entitles
X
X
X
Contingent
Liabilities
X
X
X
X
-------
TABLE 1.2
continued
Policy
Q. Joint USD
ft. Information
Dissemination
S« Technology
Extension Service
T. Antitrust
U. Tax
Adjustments
V. Better Tests
W. Education
X. Action Against
Substitutes
V. Fixed Tl«e
Periods
Z. Post-Market
Surveillance
AA. Low Volume
Exemption
BB. Small Firm
Exemption
CC. Low Risk
Exemption
DO. Fast-Track
EE. Generic PHN
FF. Improve EPA
Budgetary Outlays
Real location of
Discretionary
Funds
X
X
X
X
New EPA
Programs
X
X
X
X
X
New Non-EPA
Authority
X
X
X
Off -Budget Outlays
Tax
Expenditures
X
UMJ
now*
Taxes
X
X
X
New Financial
Entitles
Contingent
Liabilities
I
o
-------
1-11
1.4 Results of the Policy Analysis
The 32 policy alternaties are listed in table 1.3 in descending
order of their overall rating. These ratings represent the geometric
means of the individual ratings assigned by each of the six members of
the project team, using the seven criterion model. Very generally, the
ratings can be interpreted as indicative of a measure of the ratio of
effectiveness to costs for each policy. However, the ratings in table
1.3 have no quantitative significance as benefit/cost or
effectiveness/cost ratios, since a rating of 1.0 on each criterion was
assigned to an arbitrarily chosen option, against which the other options
were compared..
The ratings range from a high of 1.212 for the top ranking policy to
a low of 0.175 for the lowest ranking policy. There is a sharp drop in
ratings around the seventh ranked policy, and policies R, EE, V, C, DD,
W, and D appear to be substantially superior tq the others overall.
Policies that rank highest overall do so by virtue of a combination
of their weighted ratings on several criteria, so it is not necessary for
them to rank high on all criteria. (Table 1.4 shows the rank orders of
the seven highest ranking options on each of the criteria.) For example,
each high ranking policy alone is not able to offset the unnecessarily
restrictive impacts of TSCA on innovation. This may also require using a
somewhat lower ranking policy that is more effective, or using a
combination of policies. This situation is analogous to that faced by a
stock market investor who wishes to invest in the stock with the highest
rate of return, but who may have to invest in lower return stocks as well
if the number of shares of high return stock is limited.
18,
-------
Table.].3 Results of the Policy Assessment
Rank
Order
1
2
3
k
5
6
7
8
9
10
11
12
13
Overall
Rating
1.212
1.114
1.073
1.061
1.003
0.902
0.817
0.612
0.562
0.531
0.463
0.463
0.453
Policy
Number
R
EE
V
C
DO
W
D
F
X
S
Y
0
FF
Pol Icy Name
EPA dissemination of chemical information - test resul
Generic PMN for classes of new chemicals
Government support to develop new, better test methods
ts and/or labeling
Direct cost subsidy for testing/compliance costs of new chemical development
via grant mechanism
"Fast track" PMN's for safe and/or major innovations
Government support for education and training programs
Direct cost subsidy for testing/compliance costs of new chemical development
via loan mechanism (or loan guarantees)
Indirect cost subsidy for testing and compliance costs
Actions against existing substitutes for new chemicals
via tax mechanism
Chemical technology extension service, including dissemination of information
on test and compliance needs
Fixing time periods for regulatory actions
Establish government testing for TSCA requirements
Improve EPA staff capability to assess Impact of regulc
itory actions on
innovation
O.MS AA Regulatory exemptions for low volume, .iew chemicals
ro
-------
Table 1.3 continued
Isj
O
Rank
Order
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
Overall
Rating
0.440
0.362
0.354
0.353
0.332
0.316
0.288
0.287
0.268
0.263
0.239
0.225
0.221
0.219
0.208
Policy
Number -
BB
P
B
Q
K
T
1
U
N
J
A
L
M
Z
E
Pol Icy Name
Regulatory exemptions for small firms
Sharing of test data with reimbursement
Direct cost subsidy for general new chemical development via loan mechanism
(or loan guarantee)
Facilitate private sector Joint RSD or joint testing
Increased capital availability for new chemical development via tax changes
or via SEC rules
Antitrust action to favor new, small firms in the chemical industry
Decreased taxes on sales of new chemicals
Tax adjustments to favor small firms or new entrants in the chemical industry
Reduce risk from products liability actions by establ ishlng limits on llabll
Increased capital availability for new chemical development via government
supported venture capital company
ity
Direct cost subsidy for general new chemical development via grant mechanisms
Reduce risk through government financed Insurance for regulatory losses
Reduce risk through government procurement of new chemicals
Post-market surveillance of PMN's
Indirect cost subsidy for chemical Innovation generally via tax mechanism
CO
-------
Table:). 3 continued
Rank
Order
30
31
32
Overall
Rating
0.200
0.188
0.175
Policy
Number :
H
G
CC
Pol Icy Name
Strengthened trade secret protection by limitations on
release Information
Increased patent life for new chemical sr
Regulatory exemptions for ulow
-------
TABLE }.k
Ranks on Individual Criteria For
Seven Policies Ranked Highest Overall
Overall Policy Rank
Policy Identifier
Short Pol Icy Name
Criterion
Capacity to Countervail
Private Costs
Public Costs
Administrative Feasibility
Time to Implement
Support TSCA Aims
Other Side Effects
Effectiveness
Political Feasibility
1
R
information
Dissemination
31
24
7
1
5
1
1
16
5
2
EE
Generic
PMN
3
15
5
12
3
15
10
k
k
3
V
Better
Test
Methods
Support
22
5
12
3
13
3
3
6
3
k
C
Grants
for
Testing
Costs
it
22
17
5
1
5
5
3
8
5
DD
PMN
Fast
Track
15
13
1
17
6
6
7
11*
6
6
W
Education
and
Training
Support
9
k
19
6
19
k
2
7
2
7
D
Loans
for
Testing
Costs
6
28
15
8
4
10
8
5
1
NJ
cn
-------
1-16
1.5 A Comprehensive Program Opportunity
It is suggested that EPA consider a comprehensive program to offset
the unnecessarily restrictive impacts of TSCA on technological
innovation. The program would include six of the top seven policy
options; R, EE, V, DD, W, and either C or D:
1. (R) EPA dissemination of chemical information - test results
and/or labeling
2. (EE) Generic PMN for classes of new chemicals
3. (V) Government support to develop new, better test methods
4. (C) Direct cost subsidy for testing/compliance costs of new
chemical development via a grant mechanism
5. (DD) ."Fast track" PMN's for safe and/or major innovations
6. (W) Government support for education and tra-ining programs
7. (D) Direct cost subsidy for testing/compliance costs of new
chemical development via a loan mechanism (or loan
guarantee)
The top seven policies include policies designed to reduce the costs
of new chemical development directly (C and D), to provide more
information (R), to improve the technology needed for compliance (V and
W), and to modify the administrative procedures for managing PMN's (DD
and EE).
It is important to consider more than one policy for a comprehensive
program because no single policy is expected to be able to offset all the
unnecessarily restrictive impacts of TSCA on innovation. There is no
analytic way to determine how much, or how many programs would be
sufficient. However, a package of six policy elements, appropriately
designed, should go a long way toward offsetting these effects at
reasonable cost while supporting TSCA's primary goal of preventing
unreasonable risk of injury.
Two of the top seven policies, C and D, would reduce the cost of new
chemical development. They are nearly equivalent, with the grant program
rating higher in most respects. Either of these policies would be
particularly helpful to small firms and new entrants, but there is no
reason to include both of them.
23.
-------
1-17
Policies DO and EE are changes in the administration of the
regulatory process under current law. They are complementary programs,
designed to address somewhat different issues. The generic PMN, policy
EE, recognizes that certain classes of chemicals are very similar, and
can reasonably be reviewed as a group, perhaps even including contingent
clearance of future developments in the class, subject to certain
procedural requirements. This option, which is currently under
consideration at EPA, would address the problems of highly specialized
chemical producers. The "fast track," policy DD, recognizes the
additional public interest in the rapid processing of PMN's for safer or
major innovations. It addresses all parts of the industry, with a bias
toward new products that meet major social needs; i.e., safe substitutes,
major innovations, or other criteria.
Policy R, information dissemination, is designed to improve the
market demand for safer new chemicals and thereby offset the commerical
bias that TSCA creates in favor of chemicals already on the market. It
would favor innovative firms throughout the industry and could complement
the other policies.
Policies V and W are designed to improve the technology for
compliance with TSCA and ultimately to reduce the costs of compliance.
Government is already devoting substantial funds to developing new,
improved test methods, so option V is essentially already in place.
Government support for education and training programs, policy W, could
help meet the current high demand for professionals and technicians for
industry and for testing laboratories, which would help all segments of
the industry.
If six of the policies were adopted, the total budget costs are
expected to be in the range of $3 to $30 million per year. A program
involving relatively limited commitment to the more costly elements
(policies C, D, W, or V) could cost in the neighborhood of $7 million per
year, with about $2 million per year for each element. An experimental
or trial program could be implemented at less expense and with a lower
chance of disrupting the regulatory process. However, it will be
difficult to evaluate an experimental program's effectiveness in view of
the uncertainty in data on chemical innovation.
24.
-------
1-18
1.6 Conclusion
A comprehensive program to offset the unnecessarily restrictive
impacts of TSCA on technological innovation need not be very large,
expensive, or disruptive. The analysis in this study suggests that very
expensive programs such as grants or tax incentives for all chemical
innovation in general are neither necessary, nor cost-effective.
Furthermore, this analysis has shown that in order to address
unnecessarily restrictive impacts of TSCA on technological innovation it
is not necessary to consider programs such as regulatory exemptions for
new, small volume chemicals, for low-risk chemicals, or for small firms
that would seriously compromise EPA's efforts under TSCA to protect human
health and the environment from unreasonable risk of injury and disease.
25.
-------
2. THE EFFECTS OF REGULATION ON TECHNOLOGICAL INNOVATION
A BACKGROUND REVIEW
It is widely understood that environmental, health, and safety
regulations can change the process of technological innovation.
Regulation acts to alter the innovative climate both within and outside
individual firms. It also acts to change the structure of industries,
thereby systematically changing the proclivity and need to innovate in
both existing firms and potential new entrants.
Despite the wide agreement that regulation affects technical
innovation, there are surprisingly little data on which to form sound
opinions about whether the overall effect is positive or negative. (Hill,
1979) There is even less of a basis for understanding how the effects
depend on the characteristics of the regulation, the industry, the firm
or the technology. (Hijl, et al., 1975; Ashford, et al; 1979)
Environmental, health, and safety regulations can affect the
development and use of technology in a number of ways. They can:
* require product safety to be demonstrated prior to marketing;
* require product efficacy to be demonstrated prior to marketing;
* require safety to be proved or require the control of product use
after products have been marketed;
* require the control of production technology to reduce workplace
safety and health risks;
* require effluent, emission or waste control; or
* require safe transportation of hazardous materials.
The Toxic Substances Control Act (TSCA) acts most directly to
require product safety to be examined, if not demonstrated, prior to
marketing; and it can also act directly to require safety to be
demonstrated even after products have been marketed. Less directly,
activities under TSCA may trigger regulatory controls in such additional
areas as workplace safety and health, environmental quality control, or
hazardous materials transportation.
26,
-------
2-2
Regulation can affect innovation either for "main business" purposes
or for "compliance" purposes. In the former case, regulation affects a
traditional, ongoing activity of the firm; in the latter, regulation
demands technological changes not previously within the scope of a firm's
ordinary activity. This review, focuses on the impacts of regulation on
innovation for "main business" purposes, since that is the primary
concern of the present study.
This chapter summarizes a framework for understanding the effects of
regulation on innovation that has been developed in greater detail
elsewhere by Ashford, Heaton, and Priest. (1979) It reviews some of the
empirical literature that illustrates the framework, including literature
in the related topic of the impacts of the Food, Drug, and Cosmetic Act
on pharmaceutical innovation. The chapter closes with some implications
for the effects of TSCA on chemical innovation.
2.1 Effects of Regulation on Innovation for Main Business Purposes
2.1.1 Changes in Expected Profitability
Regulation may change the profitability expected from a portfolio of
R&D investments by affecting either the expected rate of return or the
perceived risk. Profitability may decrease as a result of
regulation-induced costs, delays, or uncertainty. As a result, a firm
may modify its level of investment in R&D. Its response to the market
pull stimulus for innovation will be modified by actual changes in R&D
costs. In addition, if R&D is perceived to be less profitable or to be
less certain to pay off, investment may be cut back and fewer main
business innovations may be produced.
Perhaps the greatest costs imposed by regulation on new products
have occurred in the pharmaceutical industry, where testing for both
safety and efficacy are required. Schwartzman has cited several studies
that indicate an increase of 100-1000* in R&D costs per new chemical
27.
-------
2-3
entity. (Schwartzman, 1976) These costs are attributed to toxicological
testing, premarket testing, and the increased paperwork required by the
Food and Drug Administration (FDA) for registration and approval.
Agency actions, other than promulgating individual standards, can
change the profitability of investment in innovation in subtle ways. For
example, requiring the submission of confidential data may disturb trade
secret protection. This may penalize technological innovation because it
decreases the legal protection available to new technologies, and may
dampen the desire to develop new products or processes, especially if
they are not patentable. On the other hand, it must be recognized that
some technologies present enough risk to the public that their components
must be disclosed. In such cases, whatever dampening effect occurs
toward innovation may be justified by the public benefit of disclosure.
Moreover, such disclosure is likely to provide an incentive to redirect
innovation along competing, but safer, technical lines.
2.1.2 Changes in the Number of Innovations that Fail for
Environmental, Health, and Safety Reasons
Otherwise successful innovations may ultimately fail if they are
found to pose unacceptable environmental, health, or safety problems.
For example, pesticides in use for some time have often been removed from
the market for environmental and health reasons.(Wechsler, et al. 1976)
Regulation can increase the number of such failures by imposing new
requirements on products. Regulation that requires premarket testing can
eliminate such failures of fully developed products by catching problems
early.
Care must be taken to distinguish observations of decreased
innovation during the period of transition to new regulatory demands
(when existing, but never-before-scrutinized, products are taken off the
market), from an equilibrium or final state (when the developer
scrutinizes products more thoroughly for possible problems during the
development process). Overall, the change in failure rate is likely to
reduce the output of harmful new products.
28,
-------
2-4
2.1.3 Changes in Investment Opportunities Due to
Increased Environmental, Health, or Safety Risks
The chance that a new product or process might be unable to enter or
remain on the market due to the regulation of environmental, health, or
safety problems may discourage investment in innovation, especially for
products with limited market potential, such as specialty chemicals.
Schwartzman suggests that a shift is occurring in the nature of
pharmaceutical innovation, with new applications for demonstrably safe
technologies being preferred to open-ended searches for new
concepts.(Schwartzman, 1976)
In some cases, the lack of defined standards under a regulatory
program can deter new investment. For example, the development of new,
high-risk, large-scale processes such as shale oil production, may be
hindered by the fact that environmental or work-place regulations are
undefined. Here, regulations that specify acceptable emission targets
are needed to reduce uncertainty.
Regulation undoubtedly changes investment opportunities; however,
the ultimate effect is not generalizable across all industries.
Industries that historically have been highly innovative may merely shift
the type of products developed. On the other hand, noninnovative
industries may find themselves competing with more innovative new
entrants.
2.1.4 Diversion of Managerial Personnel
To the degree that important advances in marketing, financing,
strategic planning, and corporate organization depend on actions by
management, the diversion of management to regulatory tasks can have
serious implications for the innovative performance of firms. It is
important to distinguish between transition diversion and administrative
diversion. Transition diversion occurs as the emergent regulations
create new problems with which management must deal. Once management has
decided on a strategy to address these problems, the transition diversion
29.
-------
2-5
will disappear. What remains is a need to monitor the compliance efforts
and the regulatory developments that are likely to follow. This
administrative diversion will accompany management as long as the
problems to which regulation is addressed remain. Much of the literature
about the diversion of management is addressed to the transition problem,
rather than to the long-run effects of regulation on management.
2.1.5 Diversion of R&D Resources
Regulation causes some firms to redirect resources away from
conventional innovative activities into compliance-related activities,
which will tend to reduce main business innovation. If, as some say, the
long-term marginal rates of return on R&D investment are as high as 30-50
percent, (Mansfield, 1976) this highly productive use of resources is not
likely to be significantly reduced by firms. Instead, other ways to
reduce spending will be found, and the opportunity cost of regulation
will be more likely reflected in a cutback in outlays for expansion and
acquisition than in outlays for R&D. Moreover, early scrutiny of the
environmental effects of new technologies can offset much of the
diversionary impact on R&D. For example, Mansfield points out that in
the chemical industry, 83 percent of the costs of new product development
occur after the applied research stage and 57 percent occur after the
pilot plant stage. (Mansfield, et al., 1971) This finding implies that
earlier rejection of potential new products would be less costly than
rejection nearer to commercialization.
When R&D resources are diverted as a result of regulation, it does
not follow that there is a corresponding proportional decrease in total
innovative output. In small firms, incremental reductions in R&D could
have significant results, especially if they have limited access to
capital. On the other hand, such incremental decreases may not lead to
particularly dramatic results in large firms, which may already have
surpassed the advantages of economies of scale in R&D.(Schmookler, 1972)
30.
-------
2-6
Innovation in large firms may even increase if organizational red
tape and communication barriers decrease with personnel real location.
The productivity of R&D for innovation may be improved in other ways if
regulation encourages the more efficient use of resources. In sum the
effect of resource diversion on innovation is not well-established.
2.1.6 Ancillary Innovation from Redirected R&D
Redirection of R&D may result in more innovation. A study of
governmental effects on the innovation process in five foreign countries
found that innovations for ordinary business purposes (not necessarily
for compliance) were much more likely to be commercially successful when
environmental, health, and safety regulations were present as an element
in the planning process than when they were absent.(CPA, 1975) In
addition, compliance-related technological changes often led to product
improvements far beyond the scope of the compliance effort. In an
example from the five-country study, a textile manufacturer developed a
new dye in order to minimize worker exposure to toxic fumes. In so
doing, he arrived at a dye which was also more colorfast and, hence, a
better, more saleable product.
Ancillary innovations often appear to be the unexpected or
serendipitous results of regulatory compliance efforts, which may occur
because of the necessity (brought on by regulation) to rethink
established and previously unquestioned modes of operation.(Allen, et al.
1978) There are enough of such innovations, however, to suggest that
they may be predictable phenomena.
2.1.7 Regulation-Induced R&D and Process Improvement
Regulation creates opportunities for firms to make process
improvements unrelated to compliance. These appear to occur more
frequently as a greater technological change is required to comply. Two
examples from a study of chemical innovation illustrate the
pattern.(Ashford, et al., 1979) In one instance, the petroleum refinery
31.
-------
2-7
industry developed improved catalysts, and, consequently, a more
efficient system, as a result of the R&D that went into the effort to
comply with regulations to control lead in gasoline. Similarly, the need
to limit employee exposure to vinyl chloride monomer led to the creation
of a more efficient production system and to some increase in output.
Similar phenomena have been uncovered in other studies. Iverstine
reported that 33 percent of his study's respondents cited process
improvements resulting from regulatory changes; these included the
development of closed systems and better process instrumentation.
(Iverstine, et al., 1978) The Denver-Research Institute similarly found
that regulation provides an opportunity to make process improvements in
areas not related to regulation.(Boucher, 1976)
Because it is less expensive and disruptive to make multiple changes
simultaneously, businessmen naturally take the opportunity of regulation
to introduce other improvements. Such improvements are often
complementary to the regulatory purpose (e.g., safer closed systems with
greater yields). They may often be suggested by the R&D that was
necessitated by regulation. Although these improvements might have
occurred eventually, regulation can be viewed as accelerating normal
business innovation.
2.1.8 Rechanneling Creativity
The innovative potential of a firm is, in large part, a function of
the creative energies and abilities of its personnel. While one effect
of regulation is to divert personnel from the normal business of the
company to regulatory problems, there also appears to be an opposite
effect - the creative potential of the firm is rechannelled, augmented,
or enhanced.
Because compliance involves a large component of technical
expertise, many of the people brought into firms to assist in compliance
are highly trained professionals—typically, environmental scientists or
32.
-------
2-8
engineers When this new source of expertise enters the normal R&D
process, innovative products and processes are likely to result. Some of
the companies interviewed in the study of chemical innovation by Ashford,
et al., (1979) mentioned the recent need for sophisticated analytical
chemistry expertise in order to assess the health and environmental risks
of both new and existing products. They felt that the sophisticated
analyses gave the companies a better knowledge of the properties of their
products and suggested new uses for them. They believed that this new
analytical capability would be important in developing new products and
processes. The same phenomenon was noted by 33 percent of the
interviewees in Iverstine's study.(Iverstine, et al., 1978)
An explanation for increased creativity under regulatory conditions
was offered by Allen and colleagues, (1978) who argued that regulation,
besides adding new dimensions to older problems, "increases the problem
space of the engineer." The need to optimize along several new
dimensions is likely to foster more creative solutions than those that
prevailed under less complex conditions. This effect is especially
likely to occur in older, more rigid, industries where few external
stimuli have demanded creative responses.
2.1.9 Change in Industry Structure
Regulation may have different impacts on firms within an industry
and may change the composition of that industry. The mix of size of
firms or the competitive environment may change. The structural
alterations brought about by regulation will in turn affect innovation.
Regulation influences the quality and quantity of innovation insofar as
it changes barriers to the entry of new firms into an industry, the
balance of firm size, and the extent of monopoly power. Distinguishing
between the effects of regulation and of other influences (e.g., changing
technology, or inflation) is a difficult methodological problem, and no
consensus has emerged as to the relative importance of these factors.
33.
-------
2-9
Barriers to the entry of new firms into an industry are introduced
when compliance measures are expensive and subject to economies of
scale.(Leone, 1977) This may be offset to some extent if new entrants
are able to meet regulatory requirements at lower costs, since they do
not have to engage in costly retrofit activites. Regulation can create
market opportunities that attract new entrants, especially those with a
new technology. For example, there are now several competing substitutes
for PCBs; whereas, before the regulatory ban, there was only one
manufacturer and no accepted substitutes in major applications. (Ashford,
et al., 1979) In general, regulation is likely to make it somewhat
harder for firms and industries to compete and survive in a more dynamic
market.
There is general agreement that small firms are harder hit by
regulation, although the evidence is primarily qualitative.(Iverstine, et
al., 1978, Boucher, et al., 1976) The effects of resource diversion
mentioned earlier may fall more heavily on the small firm due to its
limited resources. Regulatory agencies tend to concentrate their
enforcement efforts on the larger firms. However, small firms are
limited in their ability to influence policy, and regulations are not
typically designed with their special problems in mind.(Charleswater
Associates, 1975) Some regulatory agencies now have programs directed at
the needs of small firms.
Innovation may decrease as a result of increased entry barriers and
decreased competition caused by regulation. The effect will be reduced
if regulations contain provisions, say, for variances of financial
assistance for small firms. There may be a compensating effect, however,
since regulation provides new market opportunites for new entrants,
especially those with new technologies. Existing firms may try harder to
retain their market share through innovative competition. These
long-term effects may have the most important influence on innovation.
(Eads, 1979) Quantification of the net effects is speculative due to the
presence of additional influences, the diversity of industry s'tructures,
and the lack of suitable aggregate measures of innovation that capture
both quantity and quality differences.
34.
-------
2-10
2.2 Effects of Regulation on Compliance Innovation
Regulation clearly encourages technological changes for compliance
purposes. However, such changes will not necessarily be innovative.
Regulatory mandates often elicit the adoption of technologies that are
fully developed, but on the shelf. Moreover, in cases where the
regulatory standard is set at the level of the best practice in a few
leading firms, the major effect that occurs is diffusion of an existing
technology to the lagging firms—a noninnovative, although important,
response.
It is in the best interest of both firms and society to encourage
the development of innovative compliance technology. To the firm, such
technology is likely to reduce the cost of meeting regulatory goals. To
society, the adoption of better, safer technologies resulting from
innovation is an important addition to the benefits of regulation.
Moreover, to the extent that the overall impact of regulation is to
demand a long-term and widespread alteration in the nature of industrial
technology, innovation is crucial.
2.2.1 Redirection of Technological Capabilities for Environmental,
Health, and Safety Purposes Only
Most compliance technologies adopted by a sample of 50 chemical
firms in a study by Ashford, et al.(1979) were found to be in a late
stage of development when the regulatory signal was acted upon by the
firm to meet regulatory demands. Similarly, the great majority of
responses were based on well-established technologies. Responses
involving modifications of an industrial process as opposed to a product
tended to be more comprehensive in scope; there were, however,
significant exceptions.
35.
-------
2-11
2.2.2 Saleable Compliance Technologies
The great majority of compliance technologies are developed in the
firms that are directly subject to regulatory requirement.(Iverstine, et
al., 1978, Boucher, et al., 1976) However, there is also a large market
for the sale of goods and services to meet compliance requirements. Some
of these sales are made by regulated firms seeking to market the
technologies developed to solve their own in-house control problems.
Although it is not clear from the existing research whether recognition
of sales potential most often predates the development of compliance
technology, or is actually an after-the-fact appreciation of market
potential, the attempt to sell compliance technologies appears to be
fairly common. Most often, the sales are to other firms in the same
industry. One study showed that a developer of a less-polluting process
for the production of chlorine tried to market it to other firms in the
chloralkali industry. Iverstine's study documented that a large
percentage of firms were able to sell the pollution control technology
they had developed.(Alien, et al., 1978) On the other hand, the study
also indicated that the uniqueness of each firm's environmental, health,
and safety problems often makes such sales difficult.
The need to create new compliance technologies, and the dynamic
relationships between the regulated and pollution control industries,
have restructured the innovative effort in many industries. For example,
the regulations to control lead in gasoline appear to have encouraged
diversification among lead additive manufacturers, including the
development of some highly innovative new automotive
technologies.(Ashford, et al., 1979) Similarly, the suppliers of
automobile parts, perhaps more than the automobile manufacturers, have
been a major responder to regulatory demand, thereby changing the balance
of innovative activity within the industry.(Rubenstein and Ettlie, 1977)
36,
-------
2-12
2.2.3 Compliance Technologies with Ancillary Benefits
Technological change to comply with regulation can provide ancillary
benefits to the complying firm. These benefits are more likely when
compliance responses are innovative and/or comprehensive in
scope.(Ashford, et al., 1979)
Typically, ancillary benefits result from the ability of the firm to
transfer the technologies developed for compliance purposes to other
uses. For example, the use of microprocessors in automobiles to regulate
fuel consumption and emissions has opened the door to other applications
of this technology for improving performance.
From the point of view of the firm, the ancillary benefits of
regulation may result from serendipitous, unpredictable events. However,
the documentation of such effects in several studies leads to the
conclusion that this phenomenon is, on the whole, to be expected.
2.2.4 Joint R&D Efforts for Compliance
Firms within an industry often share the results of their research
on compliance methods—especially with respect to difficult regulatory
problems--even when they do not undertake it jointly. One study found
such sharing in 53 percent of its sample.(Iverstine, et al., 1978)
Although this phenomenon is not likely to have a major impact on the
development of new compliance technologies, it may have an important
impact on the diffusion of appropriate solutions.
2.2.5 Reorganization of Firms to Meet Compliance Requirements
It has been widely reported that regulation has*fostered
organizational change in companies. A study of innovation in the
chemical industry found that about 65 percent of the chemical firms
interviewed had formal environmental affairs groups (Ashford, et al.,
1979) and the Conference Board recently reported that 78 percent of a
37.
-------
2-13
sample of chemical and pharmaceutical firms have government relations
units.(C&E News, 1979) These groups often serve primarily as a liaison
between the regulators and their company. They participate regularly in
the regulatory process, often indicating to the regulatory agencies the
technical limits of existing compliance capability.
This interaction is seen by some as a way of tempering potentially
strict technology-forcing regulatory standards by considerations of
"feasibility," and also as a way by which the holders of certain
compliance technologies can "capture" the regulatory standard for their
particular compliance method. (Eads, 1979) The environment affairs unit
often functions within the firm in a manner very similar to a regulatory
agency. Environmental review procedures are frequently established, with
the environmental affairs unit able to "pass" on the acceptability of
various products or processes, particularly in their early stages of
development. Thus, these groups will encourage the development of safer
technologies.
Environmental affairs units are more common in large than in small
corporations. They are typically located in the central corporate
headquarters, rather than in production facilities. They may be staffed
with young environmental scientists rather than engineers. As such, it
appears they often do not play a major role in the development of new
compliance technology or in the engineering aspects of compliance. These
functions are more typically within the realm of the engineers at the
plant level, or R&O personnel.(Ashford, et al., 1979)
2.2.6 Information Sources for Compliance Technology
One barrier to effective regulatory compliance is a lack of
knowledge about the best technical solutions, especially in smaller
firms. Government agencies such as Occupational Safety and Health
Administration (OSHA) have programs to assist firms in developing
appropriate compliance responses. In fact, the OSHAct mandates a program
of assistance for small business.
38.
-------
2-14
The efficacy of information programs sponsored by the government is
open to some question. One study reported that less than one percent of
all solutions to environmental problems originated with the
government.(Iverstine, et al ., 1973} Another study reported a widespread
perception that EPA is deterred from establishing closer working
relationships with industry for fear of legal or political reaction by
environmental groups. (Boucher, et ;1., 1976)
In addition to government action, a firm's personnel assigned to the
regulatory compliance function plays an important informational
role.(Boucher, et al., 1976) In the regulatory area, these individuals
provide liaison between the firm and outside technical knowledge, which
can be a force for innovation in both compliance and noncompliance areas.
2.3 Effects of the Kefauver-Harris Amendments on Drug Innovation
The impacts of the 1962 Kefc'jver-Harris amendments to the Food,
Drug, and Cosmetic Act on drug innovation have been widely studied, and
they may provide son^ insight into the effects of TSCA on chemical
innovation. Of course, there is no doubt that TSCA is a different type
of regulation with some qualitatively and major quantitatively different
barriers to innovation. Therefore, the analogy to drugs, while useful,
is not an exact replication of the chemical industry. This discussion
focuses on changes in corporate strategy and on industry in response to
the new regulations.
In the drug industry, Jadlcw (1976) found that "the rate of
innovation was greatest in the period 1955-1960 in those drug markets
where small firms were introducing new products and taking market shares
away from the largest sellers." He showed that the smaller firms in the
actively innovating market segments played a significant role prior to
the 1952 drug amendments. In the /eriod 1963-1972 he showed that the
large firms were able to obtain economies of scale in research. These
economies of scale, along with regulatory barriers and the large capital
39,
-------
2-15
requirements to buy into research, constituted major disincentives for
new, small firms to consider entry into the industry.
Jadlow attributed some of the economies of scale in research to
advantages in meeting compliance requirements. He concluded that: 1)
any policy that reduces seller concentration decreases innovation, and 2)
it is desirable to reduce cost disadvantages to small firms in order to
increase innovation.
Today the drug industry still contains many small firms, but
according to Schwartzman (1976), it appears that large firms devote
proportionately more effort to R&D and produce more products than small
firms. Their role apparently has changed since the 50's when small firms
accounted for a large percentage of product innovation.
Randall (1972) carried out a series of interviews with drug firms
that resulted in an understanding of the strategies that firms had
adopted in response to regulation:
o continue "in an already established line of research mode", use
new technology to increase the sophistication of the research
effort and concentrate efforts on long range breakthrough
products."
o continue in special areas where the firm had demonstrated
high innovation capability. Extend research efforts in
technologically related product areas in order to ease
the burdens of long periods without breakthrough products.
o concentrate in short run, developmental areas, acquire
products from other firms in an effort to improve a
product line rapidly.
These strategies were adopted by those companies still devoted to
R&O. Other companies ceased doing R&D and went into the production of
generic products; that is, products on which patents have expired and
that are produced by other manufacturers. Generic drugs offer the
potential for market penetration and survival for small firms through
cost advantages and lower prices. Sales advantages can be obtained
40.
-------
2-16
through undercutting the pricing policies of the firm that developed the
product. The firms that develop the drug set prices to cover a
substantial portion of their total R&D; they take advantage of the
monopoly position offered by the patent system by setting a high price.
In the chemical industry, product turnover is more rapid than in
Pharmaceuticals, substitutes are more readily available, and the largest
firms dominate the production process, so that the generic product
strategy used by small drug firms may not offer survival to the smaller
chemical companies.
Most researchers attribute the changes in the market structure of
the drug industry to the imposition of the 1962 Drug Amendments. There
is little question that the regulations encouraged such changes, however,
several other changes in the nature of the industry were underway before
the 1962 amendments.
One of these changes was the type of discovery process used in the
industry. Discovery shifted from molecular modification and screening to
a more rational process, that requires more intensive research efforts
(Ashford, Butler and Zolt, 1977; Hattis, et al., 1980). At the same
time, competition from around the world was increasing, putting
particular pressure on small firms. A final argument often presented is
that the drug industry goes through cyclical perids of discovery and fast
growth followed by consolidation and slower growth. The 50's are
characterized as a period of rapid growth in which the technical
capabilities of the industry combined with basic research to produce new
products. The interim period, the 60's and 70's, was a time of
consolidation, process change and reinvestment in basic science.(Ashford,
Butler and Zolt, 1977) The 80's appear to be offering another turn of
the cycle that will return this industry to the dynamic growth
experienced in the 50's. The evidence for this comes from a recent
upturn in drug product submissions, and very optimistic remarks by
leaders in the drug industry.(Bloom, 1978)
41.
-------
2-17
Following passage of the Kefauver-Harris amendments in 1962, small
firms in the drug industry were acquired by other firms, dropped out,
produced generics, or specialized in a market niche in which they had
strength. The precise strategy used depended on the company's market
position, anticipated market position, and internal technical skill. In
the chemical industry, TSCA is not expected to impose as dramatic an
entry and survival barrier as the 1962 amendments did for drugs.
2.4 Implications for the Probable Impacts of TSCA on Technological
Innovation"
TSCA is a major new regulatory stimulus that may have significant,
but currently uncertain impacts on technological innovation in the
chemical and related industries. At minimum, the procedural requirements
of TSCA's pre-manufacturing notification system will introduce some
additional delays and costs in the process of innovation. For a portion
of both new and existing chemicals, TSCA may lead to additional costs and
delays to meet requirements for additional testing, or it may even lead
to more restrictive actions such as limitations or prohibitions on the
manufacture and use of chemicals.
As a product regulation, TSCA can be expected to have its primary
effect on product innovation. However, because TSCA can also lead to
controls on manufacture and use practices, and because product and
process innovation are often intimately intertwined, TSCA may affect
process and systems innovation as well.
TSCA may not be limited to inhibitory effects on product and process
innovation. Experience with pre-market product regulation in the
pharmaceutical industry and with effluent regulation in the chemical
industry shows that regulation can sometimes accelerate or even stimulate
technological innovation. These effects are seen not only for
technologies required to comply with regulation, but also for
technologies not directly related to compliance.
42.
-------
2-18
The evidence presented In this chapter and in Appendix B suggests
that TSCA may seriously affect the ability to innovate of small firms, of
new entrants, or of firms whose products are inherently sold in small
volume. This effect may be especially severe during the period of
transition from the pre-TSCA to post-TSCA state. Furthermore, the
unequal treatment of existing and new chemicals under the Act may also
inhibit innovation in these cases.
43.
-------
2-19
References for Chapter 2
Allen, Thomas, et al., 1978. "Government Influence on the Process of
Innovation in Europe and Japan, Research Policy 7, April, pp. 41-47.
Ashford, N.A., et al., 1979. "Environmental/Safety Regulation and
Technological Change in the U.S. Chemical Industry," Center for Policy
Alternatives (CPA), Massachusetts Institute of Technology, Cambridge, MA;
see also Nicholas A. Ashford and George R. Heaton, Jr., "Effects of
Health and Environmental Regulation of Technological Change in the
Chemical Industry: Theory and Evidence," in Federal Regulation and
Chemical Innovation, ed. C.T. Hill, American Chemical Society Symposium
Series No. 109, American Chemical Society, Washington, O.C., 1979.
Ashford, Nicholas, A., S. Butler and E.M. Zolt, 1977- "Comment on Drug
Regulation and Innovation in the Pharmaceutical Industry," Center for
Policy Alternatives, Massachusetts Institute of Technology, Cambridge,
MA, February. Presented to HEW Review Panel on New Drug Regulation,
January 10, 1977.
Ashford, N.A., G.R. Heaton, Jr., and W.C. Priest, 1979. "Environmental
Health and Safety Regulation and Technological Innovation," in
Technological Innovation for a Dynamic Economy, C.T. Hill and J.M.
Utterbacic, eds., Pergamon Press, N.Y.
Bloom, B. M., 1978. "Remarks to the New York Society of Security
Analysts," March 8.
Boucher, Wayne, et al., 1976. Federal Incentives for Innovation, Denver
Research Institute, University of Denver, Denver, CO, January.
Center for Policy Alternatives, 1975. "National Support for Science and
Technology, An Examination of Foreign Experience," Working Paper No.
75-12 of the Center for Policy Alternatives, Massachusetts Institute of
Technology, Cambridge, MA.
Charleswater Associates, 1975. The Impact on Small Business Concerns of
Government Regulations That Force Technological Change, Charleswater
Associates, Washington, O.C., U.S. Government Printing Office, September,
pp. 58-59.
Chemical and Engineering News, 1979. "More Firms Set Up Government
Relations Units," £7, July 2, p. 8.
Eads, George C., 1979. "Chemicals as a Regulated Industry: Implications
for Research and Product Development," in Federal Regulation and Chemical
Innovation, C.T. Hill, ed., American Chemical Society Symposium Series,
no. ioy, American Chemical Society, Washington, D.C.
44.
-------
2-20
Hattis, D., R. Andrews, J.W. Estes and S.T. Owen, 1980. "Relationship
Between Aspects of Pharmaceutical Regulation, Innovation and Therapeutic
Benefits," report to the National Science Foundation by the MIT Center
for Policy Alternatives.
Hill, C.T., editor, 1979. Federal Regulation and Chemical Innovation,
American Chemical Society Symposium Series No. 109, ACS, Washington, O.C.
Hill, C.T., E. Greenberg, D.J. Newburger, G.R. Whitaker et al., 1975. A
State of the Art Review of the Effects of Regulation on Technological
Innovation in the Chemical and Allied Products Industries, "CAP!
Project," Center for Development Technology, Washington University, St.
Louis, MO.
Iverstine, Joe C., Jerry. L. Kinard, and William S. Slaughter, The Impact
of Environmental Protection Regulations on Research and Development in
the Industrial Chemical Industry, (Grant No. PRA 76-21321, Division of
Policy Research, National Science Foundation, Washington, O.C., May
1978). See also Iverstine and Kinard in Federal Regulation and Chemical
Innovation, C.T. Hill, ed., American Chemical Society Symposium Series
NO. iuy, Washington, O.C., 1979.
Jadlow, J.M., 1976. "An Empirical Study of the Relationship Between
Market Structure and Innovation in the Therapeutic Drug Market," report
to the National Science Foundation.
Leone, Robert A., 1977. "The Real Cost of Regulation," Harvard Business
Review, Vol. 55, No. 6, Nov/Dec.
Mansfield, Edwin, et al., 1971. Research and Innovation in the Modern
Corporation, W.W. Norton, Co., N.Y., p. 118.
Mansfield, Edwin, 1976. "Federal Support for R&D Activities in the
Private Sector," in Priorities and Efficiency in Federal Research and
Development, Joint Economic Committee Compendium, Washngton, O.C., U.S.
Government Printing Office, pp. 97-99.
Rubenstein, Albert and John Ettlie, 1977. "Analysis of Federal Stimuli
to Development of New Technology by Suppliers to Automobile
Manufacturers: An Exploratory Study of Barriers and Facilitators,"
Submitted to the U.S. Department of Transportation, Washington, D.C.,
March.
Schmookler, Jacob, 1972. "The Size of Firm and the Growth of Knowledge,"
in Patents, Invention and Economic Change, eds., Z. Griliches and L.
Hurwicz, Harvard University Press, Cambridge, MA.
Schwartzman, David, 1976. Innovation in the Pharmaceutical Industry,
John Hopkins University Press, Baltimore, MD.
Wechsler,, A.E., et al., 1976. Incentives for Research and Development
in Pest Control, prepared by Arthur D. Little, Inc. for the National
Bureau of Standards, Washington, D.C., December.
45.
-------
3. A FRAMEWORK FOR ASSESSING THE EFFECTS OF TSCA
ON CHEMICAL INNOVATION
3.1 Legislative Background
The fundamental purpose of the Toxic Substances Control Act (TSCA)
is to prevent chemicals from presenting an unreasonable risk of injury to
health and the environment. The Act gives the Environmental Protection
Agency a variety of authorities to require testing of both existing and
new chemicals, to require notification before new chemicals are
manufactured or before chemicals are used in significant new uses, and to
regulate the production, use, and disposal of chemicals that are found to
pose an unreasonable risk.
The implementation of TSCA by EPA can be expected to modify the
constraints and opportunities under which both new and existing chemicals
are developed, marketed, or used. Thus, TSCA can be expected to affect
the economic performance of the chemical industry in such areas as
profitability, growth, imports, exports, employment, and technological
innovation. The model and literature reviewed in chapter 2 of this study
make it clear that regulatory programs analogous to TSCA have had a
variety of impacts on economic performance and especially on innovation.
This review also makes it clear that regulation can enhance, inhibit, or
redirect technological innovation, depending on the exact form of the
regulatory requirement and the nature and stage of development of the
industry and its products and process.
Congress was mindful that TSCA might affect the economic performance
of the chemical industry and that technological innovation could be
especially affected. Thus, in the statement of policy in Section 2(b)(3)
Congress said:
46.
-------
3-2
Authority over chemical substances and mixtures should be exercised
in such a manner as not to impede unduly or create unnecessary
economic barriers to technological innovation while fulfilling the
primary purpose of the Act to assure that such innovation and
commerce in such chemical substances and mixtures do not present an
unreasonable risk of injury to health or the environment.(emphasis
added)
This section of TSCA makes it clear that Congress fully anticipated
that implementation of TSCA might "impede...or...create...economic
barriers to technological innovation" as the Agency carried out its
primary purpose "to assure that...chemical substances do not present an
unreasonable risk." Thus, the framers of the Act expected some negative
effects on innovation and directed the Agency to take care not to inhibit
innovation any more than necessary to carry out its primary regulatory
charge.*
3.2 Administrative Discretion
Under TSCA, the EPA Administrator has considerable administrative
discretion over which chemical substances to regulate and over the
detailed design and implementation of the procedural and regulatory
authorities he is to exert. He also has authority to provide various
kinds of technological and administrative assistance to regulated
parties. Within the scope and range of this discretion the Administrator
could have significantly different effects on innovation while regulating
effectively.**
*That TSCA may also stimulate technological innovation in certain sectors
of the industry and especially of safer chemicals than those now in use
is not addressed explicitly in the Act but is also of concern, as
discussed in Section 3.3
**Many of the rules and procedures to implement TSCA are being developed
by EPA at the present time, and Section 2(b)(3) is one consideration that
guides EPA in this work. To provide a reasonable basis for comparison of
a wide variety of policy options, we have included among the options
assessed in this study several actions which EPA has, or intends to
implement. As a result, this report tends to overstate the potential
restriction of innovation by TSCA regulation.
47,
-------
3-3
In addition to addressing the innovation problem directly by careful
design of its regulatory programs under TSCA, EPA could consider two
other routes to execute its charge to be concerned for chemical
innovation. It could encourage other agencies of government to pay
special attention to the needs of the chemical industry, such as research
and education authorities (e.g., National Science Foundation or National
Institutes of Health) or economic development authorities (the Small
Business Administration, the Economic Development Administration, or the
Patent Office). As a third approach, EPA could seek additional
authorization from Congress to address innovation directly through, for
example, authorization to offer financial assistance to firms for
innovative activities or for regulatory compliance. All three of these
approaches are considered in this study.
3.3 Understanding the Issue: A Framework
If EPA is to take actions so as not to unduly impede or create
unnecessary economic barriers to technological innovation, there is a
need to understand how and where such undesirable effects might arise and
what forms they might take. Only then can sensible policy options be
designed and assessed.
This report is devoted to the design, assessment and analysis of
cost-effective policy options that could offset the unnecessarily
restrictive effects of TSCA on technological innovation while not
jeopardizing fulfillment of the primary purpose of TSCA. The policy
options are designed to be responsive to the following considerations.
£f_the_f uture ourseof
the emTcaldusfrry re
It is useful to think of technological innovation as a process whose
outcomes are new, commercially successful products, processes, systems or
services. The process involves inputs of human and financial resources
to such activities as research, invention, development, testing,
marketing and diffusion.
48.
-------
3-4
Appendix B discusses the process and outcomes of technological
innovation in the chemical industry in some detail. There it is noted
that a variety of factors in the scientific, financial and competitive
environment of the chemical industry are changing and that even in the
absence of TSCA, historic trends in chemical innovation are unlikely to
be followed in the future. Contributing to this is the fact that
developments in products liability and environmental and occupational
regulation are influencing chemical innovation quite apart from TSCA's
effects. It is also pointed out in Appendix B that there is not yet a
good understanding of the nature and sources of chemical innovation. For
example, there are no sound data on the number of new chemicals marketed
each year, or on the contributions of small and large firms or new
entrants to chemical innovation. A few studies have found that large
firms are more innovative than small ones, but even these results are
open to serious question.
Since TSCA is primarily concerned with regulating products, it is
reasonable to focus our attention, though not entirely, on product
innovation. Figure 3-1 illustrates the difficulties in projecting
chemical product innovation, even if TSCA had not been passed. Data on
past chemical innovation are not available. For the future, a wide range
of possibilities exist, depending on which of the current forces
dominate. Furthermore, in the absence of TSCA, the concern for safety
might decelerate innovation if it causes producers to be less aggressive
in marketing new products, or it might accelerate innovation if consumer
demand causes a shift to new safer products as substitutes for old ones.
ii) lS£A_may_s_timuJ[ate_,_ jnhij>it,_or jredire£t
Innovation^ o_r_iT may_ 5o_aTl_t][ree_sTmuTCane£uTly_
Superimposed on the uncertain future of chemical innovation are the
variety of effects that the TSCA regulatory requirements may have.* As
*Regulation in this context includes both the procedural requirements
such as premanufacturing notification and substantive requirements such
as use restrictions or testing requirements.
49.
-------
3-5
u
2
u
r
o
£
LU
tr
UJ
CD
2
THE PAST IS NOT
WELL UNDERSTOOD
/
/
/EMPHASIS ON
/ SPECIALTIES
/BUSINESS
AS USUAL
MATURING
INDUSTRY
PAST
PRESENT
FUTURE
FIGURE 3.1 SOME SCENARIOS FOR THE FUTURE COURSE OF
CHEMICAL INNOVATION IF TSCA HAD NOT BEEN
PASSED
50.
-------
3-6
noted in. chapter 2, environmental, health, and safety regulation can act
through a variety of mechanisms to inhibit, stimulate or redirect
technological innovation, depending on the circumstances. Since TSCA
features several different regulatory stimuli, and since the "chemical
industry" is in fact a combination of many very different kinds of
industries in various stages of maturity or rigidity whose products
differ greatly in the hazards they present; it is to be expected that
inhibition, stimulation, and redirection will all occur at the same time.
At our current level of understanding of the interaction of
regulation and innovation, it is not possible to predict the quantitative
outcome of TSCA for, say, the rate df chemical product innovation. This
situation is outlined in Figure 3.2, assuming that innovation in the
absence of TSCA would have followed the "business as usual" scenario.
The inhibition of innovation by TSCA would arise, for example, from
the marketing delays, testing costs, resource diversion, and commercial
uncertainties it would introduce into the innovation process. The
stimulation of innovation would arise, for example, from the increased
staff diversity and rejuvenated corporate decision-making process
required to comply with TSCA. Redirection would arise from firms
electing to seek safe substitutes or abandon lines of research into
chemicals expected to pose a high risk to health.
The research reviewed in chapter 2 and appendix B suggests that the
inhibition of innovation is more likely to occur in small firms, new
entrants, and makers of innovative specialty products, while stimulation
is more likely to occur in large, established, mature firms that use
•
highly-integrated process technology. At the same time, regulation can
also stimulate innovation in some small firms and new entrants and
inhibit it in some mature firms. Thus, redirection due to TSCA can occur
in both the nature and sources of chemical innovation.
iii) Spme_degre£ of_inhibitjpn £f tescjvrujl^g^cal i nnovatjoji
amd Tso7 n
Historically, some of the new products and processes manufactured or
used by the chemical industry and its customers have posed unreasonble
51.
-------
3-7
CO
/
/ STIMULATION
/ BY TSCA IS
/ DOMINANT
/NO-TSCA
/ BUSINESS
AS USUAL
INHIBITION
BY TSCA IS
DOMINANT
SOME REDISTRIBUTION
OCCURS TOWARD
SAFER CHEMICALS
PAST
PRESENT
FUTURE
FIGURE 3.2 SOME SCENARIOS FOR THE COURSE OF FUTURE INNOVATION WITH
TSCA IN PLACE (COMPARED WITH THE NO-TSCA BUSINESS AS
USUAL SCENARIO)
52.
-------
3-8
risks of injury to health and the environment. One of the main purposes
of TSCA is to slow the rate of introduction and/or encourage the more
prudent use and operation of such products and processes. Thus, some
inhibition and some re-direction of chemical innovation was expected due
to TSCA - it was part of the social bargain struck by Congress. Thus, an
observation that the rate of chemical innovation has declined, or that
the nature of chemical innovation has shifted is not, by itself, grounds
for determining that EPA has acted "unduly" or "created unnecessary
economic barriers to innovation." Therefore, offsetting policies should
not attempt to return the rate and direction of chemical innovation to
some hypothetical pre-TSCA baseline.
iv) !*_!! not. £ossible_t£ know from examining th£ historical record.
£r~"f rpm £ro JecTi cms_
~
.__
£esuTt_f?om TSlTA. ~~
As discussed in chapter 2, appendix B, and this chapter, there are
great uncertainties in both the historical record and future projections
for chemical innovation. There are not models which allow for
quantitative projection of the rate of innovation in an industry or for
the influence of such complex regulatory schemes as TSCA on that rate.
Thus, if in the future an. interested party alleges to EPA or to Congress
that the rate of chemical innovation is too low, or too high, as a result
of TSCA, there is no way to support or refute such a statement.
This does not mean that sensible actions cannot be taken, however.
Analysts can argue from empirical observation and on a priori grounds
about the impacts of different regulatory options or offsetting policies
on the process on innovation, and from these arguments can make informed
judgments about the relative effects and effectiveness of various
approaches. This is done in chapter 5 of this report.
53.
-------
3-9
v) Th£ jm£l£mejrvta£i£njrf TSCA_may_hav£ spme_unn£«ssarij_y
nnvat in.
Earlier, it was asserted that TSCA will inhibit chemical innovation
to some degree and that some part of that inhibition was intended by
Congress. Similarly, it can be asserted that the implementation of TSCA
may unnecessarily restrict technological innovation despite EPA's best
intentions. However, because of the great uncertainties about the
innovation process, it is not possible, or necessary, to attempt to draw
a clear distinction between the justifiable and the unnecessarily
restrictive impacts of TSCA.
3.4 The Origins of Unnecessarily Restrictive Impacts on Innovation
Two concepts, regulatory fine tuning and transition phenomena, help
explain the origins of the unnecessarily restrictive impacts of TSCA on
chemical innovation.
First, consider "regulatory fine tuning." TSCA gives the
Administrator of EPA considerable discretion in carrying out the main
purpose of the Act: to prevent unreasonble risk of injury due to
chemicals. Yet, despite the many special provisions and wide latitude
for decision making embodied in the Act, rules to implement TSCA are
likely to bear more heavily on some parties than on others in ways which
are unnecessary to accomplish the regulatory goals. This is likely to
happen as a result of the need to compromise fine tuning of the rules and
procedures for political and administrative feasibility. On the other
hand, the very complexity of TSCA may cause unnecessary burdens for some
regulated parties, such as costs, delays or uncertainties, that would
unnecessarily restrict innovation.
Consider next, "transition phenomena." When a new law is passed
that is intended to influence industrial behavior, a finite period of
time elapses while the rules and procedures to implement the law are
54.
-------
3-10
adopted. During this period, firms and investors may perceive a high
level of uncertainty in making business decisions. Also, for a time
after the law is passed and implemented, the infrastructure necessary to
respond to the law's requirements may not be in place. For example,
trained people and facilities required to do tests to comply with TSCA's
rules may not be available at reasonable cost, if at all, for several
years. During the period, newly-regulated firms can be seriously
disrupted, and smaller firms may even disappear with the result that
innovation declines, even for safer chemicals.
Congress and the agency do not intend to burden industry with these
transition phenomena or to use rules and procedures that are inadequately
tuned to the needs of industry, yet some problems are inevitable if a
vigorous new regulatory program is to be put in place to accomplish the
primary goal of controlling unreasonable risk. To the extent that the
regulations unduly inhibit or create unnecessary barriers to
technological innovation, and to the extent that these undesirable
effects can be corrected by policies whose costs are commensurate with
the benefits they offer, EPA and/or Congress may wish to take action to
put such policies into action.
3.5 Small Firms, Small Volume Chemicals, and New Entrants
Small firms, especially those that specialize in new products, may
be especially burdened by transition phenomena, as well as by problems
arising from the insufficient degree of fine tuning in, or complexity of,
TSCA. Similarly, new chemicals that will only be produced in small
amounts (a condition that can often be judged before marketing with fair
certainty) will also be heavily burdened by TSCA's requirements.
For example, large firms that sell many products can achieve
economies of scale in administrative costs of regulatory compliance, and,
inevitably, can receive more attention to their needs from regulators
than small firms. Chemicals produced in small volume may have to bear
55.
-------
3-11
the same costs of PMN submission and/or testing as larger volume ones
(although EPA's priority setting may consider production volume as a
factor in deciding to require testing), which may put them at an economic
disadvantage. Or, firms entering the chemical industry for the first
time with a new product may experience extraordinarily high costs of
learning how TSCA programs work. As discussed in detail in appendix A,
the seriousness of these problems for the firm depends heavily on the
detailed circumstances of supply and demand for the product, for its
substitutes, and for inputs to its production. Ideally, the regulatory
effort should be tailored to every such circumstance. One way to attempt
such tailoring is to adopt policies that offset the unnecessarily
restrictive impacts arising from the degree of regulatory fine tuning and
from transition phenomena.
3.6 The Stimulation of Safer New Products and Processes
Parallel to the concern for the unnecessarily restrictive impacts on
technological innovation is the interest in the development of safer
chemical products and processes. The innovation of safer substitutes is
a socially desirable outcome that the unregulated market will not produce
in sufficient numbers; a major reason for TSCA's existence.
As noted in the welfare economic discussion in Appendix A,
government intervention to increase the production of safe chemicals is a
separate policy issue from the concern for the possible unnecessarily
restrictive impacts of TSCA. However, the two concerns are related,
since the impacts on innovation could include a reduction in innovation
of safer products and processes. It is likely that programs intended to
offset unnecessarily restrictive impacts of TSCA on innovation could also
be designed to enhance the innovation of safe substitutes. This
possibility is included in the assessment of policy options in chapter 6.
56.
-------
3-12
3.7 Conclusion
For largely unavoidable reasons; TSCA and its implementation may
unnecessarily restrict technological innovation in the chemical and
related industries. It is not possible to quantify the extent of this
restriction, or even to know whether the net effect of TSCA on chemical
innovation will be expansive or reducing. In fact, some degree of
inhibition of innovation is intended by TSCA. However, our understanding
of the innovation process and of the inevitable compromises and
simplifications inherent in the regulatory process lead to the above
conclusion.
Where actions might be taken by EPA, by other agencies, or by the
Congress to offset these unnecessarily restrictive effects on innovation
without compromising the primary goal of TSCA, and where such actions can
be taken at a cost lower than the costs they create, such actions would
be fully in harmony with Section 2(b)(3) of TSCA. The next chapters
design and evaluate some thirty-two separate actions that might be
considered.
57.
-------
4-1
4. POLICY OPTIONS FOR CONSIDERATION
4.1 Developing Policy Options
A major task in this project was to develop a list of possible
policy options for eliminating or mitigating the potential effects of
TSCA on technological innovation. This list of options is intended to be
widely representative of the possibilities for government action and to
reflect the wide variety of policy options that have been proposed by
various parties. Some of the options could be implemented
administratively; others would clearly require new legislation. Both are
potentially important approaches to encouraging innovation.
The list is comprised of policy options to encourage innovation.
These proposed policies are directed at only one of the many legislative
goals of TSCA - not to unnecessarily restict technological innovation.
They are not strategies for the implementation of TSCA, but are intended
as only a part of such a strategy. The policy list was developed with
innovation foremost in mind. Thus, it may contain options that should be
rejected because they are inconsistent with TSCA's other goals, as
reflected in the evaluation of the policies in chapters 6 and 8.
The generation of the list began with an analysis of the process of
techno logical innovation described in chapter 2 - what drives it, what
impedes it, and how it progresses. It was then possible to develop a
construct of the barriers and incentives to innovation that government
can influence. For example, it can alleviate shortages of capital for
new investments (a barrier) or guarantee markets (an incentive), but
cannot affect management creativity to any great extent.
Based on the construct, we developed a series of programs that
government can use to encourage innovation by overcoming the barriers or
enhancing the stimuli. This was done by developing our own program
ideas, by examining those originating in the Office of Toxic Substances,
and by canvassing the extensive number of policy proposals that have been
suggested, both for TSCA and for other similar regulatory regimes (e.g.,
58.
-------
4-2
pesticides or drugs). This survey covered the positions of a wide range
of interest groups from chemical trade associations, to
environmentalists, to advocacy task forces for small business.
The sizable body of literature on government policy and
technological innovation on which this work is based is presented in the
bibliography accompanying this chapter in Appendix D. It includes
scholarly reports, Congressional hearings, agency publications, position
papers, and popular or trade publications, concerning the following
issues:
o government policy and technological innovation, in general
o studies of foreign government policies toward innovation
o reviews of existing U.S. policy
o the relationship between regulation and innovation
o regulatory reform
o small business and innovation
The policy options considered are presented in Table 4.1, arranged
according to the principal barrier or incentive toward which a policy is
directed or the type of policy mechanism employed. Each of the options
is a discrete concept. It is recognized, however, that groups of them
could be combined into coherent program packages, as is done in chapter
8. (Note that the options are assigned a letter code - A, B, C, etc. -
that is used consistently throughout the report.)
The detailed discussion of each of the thirty-two policy options
that follows adheres to the following format:
o A brief statement of each option with its purpose and its
rationale
o An explication of the details of the option
o A brief discussion of analogous existing policies, including
their success or failure, and
o The advantages and disadvantages of each option.
59.
-------
4-3
TABLE 4.1 Policy Options for Consideration
Reducing the Cost of New Chemical Development
A. Direct cost subsidy for general new chemical development via
grant mechanism.
B. Direct cost subsidy for general new chemical development via
loan mechanism {or loan guarantee).
C. Direct cost subsidy for testing/compliance costs of new chemical
development via grant mechanism.
D. Direct cost subsidy for testing/compliance costs of new chemical
development via loan mechanism (or loan guarantee).
E. Indirect cost subsidy for chemical innovation generally via tax
mec h an i sm.
F. Indirect cost subsidy for testing and compliance costs via tax
mechanism.
Increasing the Financial Rewards for New Chemicals
G. Increased patent life for new chemicals.
H. Strengthened trade secret protection by limitations on EPA
authority to release information.
I. Decreased taxes on sales of new chemicals.
Increase the Availability of Capital for New Chemicals
J. Increased capital availability for new chemical development via
government supported venture capital company.
K. Increased capital availability for new chemical development via
tax changes or via SEC rules.
Reduce the Commercial Risk Associated with New Chemicals
L. Reduce risk through goverment financed insurance for regulatory
losses.
M. Reduce risk through government procurement of new chemicals.
N. Reduce risk from products liability actions by establishing
limits on liability.
Reduce the Cost of Testing
0. Establish government testing for TSCA requirements.
60,
-------
4-4
TABLE 4.1 Policy Options for Consideration
(continued)
Real locations of Cost within the Private Sector
P. Sharing of test data with reimbursement.
Q. Facilitate private sector joint R&D or joint testing.
Information-based Strategies
R. EPA dissemination of chemical information—test results and/or
labeling.
S. Chemical technology extension service, including dissemination of
information on test and compliance methods.
Changing Market Structure
T. Antitrust action to favor new, small firms in the chemical
industry.
U. Tax adjustment to favor small firms or new entrants in the
chemical industry.
Improving the Technology Necessary for Compliance
V. Government support to develop new, better test methods.
W. Government support for education and training programs.
Regulatory Changes
X. Actions against existing substitutes for new chemicals.
Y. Fixing time periods for regulatory actions.
Z. Post-market surveillance of PMN's.
AA. Regulatory exemptions for low volume, new chemicals.
BB. Regulatory exemptions for small firms.
CC. Regulatory exemptions for "low risk" chemicals.
00. "Fast track" PMN's for safe and/or major innovations.
EE. Generic PMN for classes of new chemicals.
FF. Improve EPA staff capability to assess impact of regulatory
actions on innovation.
GG. "No-intervention" policy; (i.e., no change from existing TSCA
regulation).
61.
-------
4-5
For systematic evaluations of the virtues and drawbacks of the
options see chapters 6 and 8 in which the policies are evaluated and
compared, and in which combinations are examined.
4.2 The Options
A. Direct Cost Subsidy for New Chemical Development, in General, via a
Grant Mechanism
A direct cost subsidy for new chemical development is based on the
assumption that the social benefits of technological innovation often
exceed the private benefits. Because firms often cannot capture all of
these social rewards, they tend to "underinvest" in innovative activity
from a societal viewpoint. Thus, there is an appropriate role for the
government in encouraging innovation through subsidy efforts that will
reduce the costs to individual firms.
The subsidy program may be directed at different parts of the
innovation process: basic research, applied research, development, or
initial prototype manufacture. Most commonly, such programs are aimed at
either basic or applied research, on the theory that the government
should not be directly subsidizing commercial applications of new
technologies, but that its more appropriate role is in the earlier stages
of the Innovation process, which have less immediate commercial
importance.
Such a program may either discriminate among types of applicants or
be non-discriminatory. If it is discriminatory, distinctions can be made
on a variety of bases. Grants can be given based on a firm's
characteristics -- for example, only to small or to new firms. They can
also be given on the "value" of the project proposed, which can be
determined, for example, by assessing commercial feasibility, social
benefits, or the overall persuasiveness of the project proposal; or the
grant program may be restricted to the submitters of PMN's only.
62.
-------
4-6
If grants were to be administered on a non-discriminatory basis,
they would be open to all firms that submitted applications. In this
case, they would necessarily be very small. Alternatively, a
non-discriminatory program could be keyed to a percentage of the
applicant firm's R&D.
This subsidy program's development would clearly require new
authority from Congress. It would also represent a major re-orientation
in national policy with respect to technical development in industrial
firms. The program might or might not be lodged in EPA; however, its
mission is sufficiently different from the nature of current EPA
regulations that another institutional home would seem more appropriate.
Because of its diffuse nature, this program would require a minimum
of about $50 million a year to make any significant impact.
Many developed countries have major programs of this type. Japan,
Germany, the United Kingdom, the Netherlands, and France all have grant
programs. Although these countries have been committed for some time to
this type of subsidy, there is little evidence available to show the
overall success of their programs.
Nevertheless, the evidence does show the programs' encouragement of
individual projects that may be both commercially successful and of
significant social value. However, it can only be presumed that their
overall impact is positive. A prerequisite is a cooperative relationship
between the government and participating industrial groups .
On the negative side, a subsidy program of this type would be a
major departure from existing national policy. It does not particularly
favor the goals of TSCA because it is so broad-based. Further.it would
be hard to administer, particularly if it were set up on a discriminatory
basis. It is politically unpalatable to many, and would be expensive.
Lastly, it m-ight be subject to politicization and favoritism.
63.
-------
4-7
B. Direct Cost Subsidy for New Chemical Development, in General, via
Loans or Loan Guarantees
Loan or loan guarantees for new chemical development in general are
aimed at innovation and only relate to TSCA purposes indirectly. They
seek to spur innovation by making funds available for research and
development (R&D), thus reducing the financial risk assumed by a firm or
investors. As opposed to a grant program, however, the government loan
is repaid if the chemical is a commercial success. In the case of
guarantees, the government assumes a contingent liability that only
materializes when private firms fail to meet their loan obligations. For
the program proposed here, EPA would be authorized to administer, with
the Small Business Administration (SBA), a loan or loan guarantee program
for chemical companies seeking to develop new products. Applications
would be evaluated by both EPA and SBA.
Loans (or guarantees) could be made up to, for example, $50,000 for
basic R&D and $100,000 for product development and commercialization.
(These amounts, although reasonable from a political standpoint, may be
small relative to real financial need.) A loan could be repaid on a
schedule determined as a function of sales. If the venture were a
commercial failure it would not have to be repaid. Government loan
guarantees would extend to commercial banks, allowing reimbursement in
the event a company receiving money under the program defaulted.
The program could be started on a trial basis with a relatively
small authority of around several millions of dollars. To have any real
impact, however, it would have to be larger. Possible criteria for
evaluating applications might include the following:
o Whether the applicant had been refused a commercial loan first,
o The company's overall financial situation,
o Whether the product was for a specific purchaser,
64.
-------
4-8
o Whether the product was merely an attempt to make a "me-too"
(i.e. incremental advance from an existing product) or whether it
was a true innovation that might be less hazardous than an
existing commercial product.
From an environmental perspective, it would be preferable that companies
qualifying for loans or guarantees be small firms attempting to develop
new products that are low-risk substitutes for existing ones^ However,
this is obviously a difficult criteria to fulfill when the research has
not been done.
This program does not have many existing analogues. Although
similar in certain respects to the direct grant program (option A,) a
loan program would be more complex from an administrative and financial
viewpoint. One possible analogue, the Solar Energy Bank, has not yet had
any real experience on which it can be judged.
There are complicated and difficult issues associated with this type
of program; its administrative complexity (which EPA is not at this time
capable of handling), the need for close cooperation and coordination
between EPA and SBA, the size of the program necessary to have an effect
on the industry, the uncertain payoffs in actual innovation, and the need
for monitoring the loans. In the case of guarantees, while EPA may face
very low operating costs, there could be substantial costs later if many
companies defaulted.
On the plus side, it would cost less than a direct grants program
and, since SBA already has experience with loans, could be integrated
with an ongoing activity.
C. Direct Cost Subsidy for Testing and Compliance Costs of New Chemical
Development, Specifically, via a Grant Mechanism
This option is intended to directly offset the costs of testing and
compliance with TSCA, such as the costs of submitting a premanufacturing
65.
-------
4-9
notification (PMN). It proceeds on the assumption that some
indeterminate fraction of these costs is potentially "undue," thus
appropriate for the government to underwrite. These costs may be
especially severe during a period of transition as TSCA regulations are
instituted. Small firms may be particularly burdened. An important
additional goal of this program is to encourage increased and improved
testing.
As envisioned, the program would offer grants to all firms
submitting an adequate PMN. Adequacy would be an EPA determination in
each case; however, most applications that are formally complete should
be considered adequate. Grants would vary in size, tied to the size of
the costs involved in submitting a PMN. Thus, firms that undertook more
testing would receive larger grants. For example, a minimal PMN
consisting only of a literature review and a physical property analysis
might receive a few thousand dollars; whereas a detailed submission,
including chronic toxicity tests on two animal species, might qualify for
as much as $100,000.
Grants would be taxable. They would be given automatically, and
their processing procedure could be separated from EPA's regulatory
activities. A ceiling on total individual firm grants - e.g. $100,000 -
should also be established. With this limitation, the program need not
be large. An appropriation of about $2 million should be adequate.
To focus this program more closely on innovative companies, the
grants might be restricted to new firms, those of a certain size, or
those submitting more than one PMN per year.
There are apparently no analogues to this program elsewhere in the
government. One of its principal virtues is that it concentrates on the
unnecessarily restrictive impacts of TSCA requirements. Moreover, it
would be of particular benefit to small firms and new entrants. It
should not be excessively costly so long as the number of PMN's and the
grant sizes remain small, nor would it be especially difficult to
66,
-------
4-10
administer. Because grants would be given to virtually all applicants, a
major incentive should be created for firms to undertake more testing.
This increase in PMN information should greatly benefit EPA, especially
since under current law there is no generally applicable testing
requirement.
The major problem is probably political; i.e., to make politically
acceptable the concept of direct grants to firms to help them fulfill
existing legal requirements. This is particularly so for TSCA, since
Congress specifically rejected government-funded testing during debate on
the bill. Endorsement of this concept for TSCA might also be difficult
without extending similar benefits to companies affected by regulations
in other areas.
D. Direct Subsidy for Testing and Compliance Costs of New Chemical
Development via Loans or Loan Guarantees
The goals of this option, as those of the preceding option, are to
reduce a company's immediate cost of complying with TSCA regulations and
to promote appropriate compliance efforts. Specifically, the aim is to
reduce the cost of testing and compliance by providing loans that would
be repaid after the PMN had been accepted and sales started.
The proposed program is basically an extension of the Section
7(b)(5) -loan program under the Small Business Act, and it could be
administered by the SBA, with technical assistance from EPA. Companies
planning to submit PMN's would apply to SBA, which under Section 7(b)(5)
of the Small Business Act has authority to grant loans for regulatory
compliance purposes. The applications would be evaluated on the
following bases (consistent with the existing goals of 7(b)(5)):
o the company must have been refused a commercial .loan;
67.
-------
4-11
o the protocol it intends to follow must satisfy EPA guidelines for
health and safety evaluations and PMN information requirements;
and
o the company must have made good faith efforts in the past to
comply with TSCA regulations, including previous PMN's it has
submitted.
The loans would be for up to 90% of the cost of compliance or
$500,000, whichever was less, and would have a maximum maturation of 30
years. The Section 7(b)(5) loans are not forgivable: even if the
product fails commercially, the loan must be repaid. However, the
program proposed here could also be based on forgivable loans; for
example, in cases where testing results were unfavorable enough to
dissuade the firm from further development. Loan guarantees could be
substituted for loans or could supplement them. They would be for the
same amount and would be evaluated on the same basis as the loan program.
Under the Section7(b)(5) program, the appropriate Federal agency
(EPA, Occupational Safety and Health Administration (OSHA), Consumer
Product Safety Commission (CPSC), etc.) must certify prior to a loan
approval that the firm's actions will bring it into compliance, and after
the work is done the agencies must certify that the firm is then actually
in compliance. This program now offers loans for OSHA, CPSC and Mine
Safety and Health regulations. In approximately 10 years of operation,
the SBA has granted 571 loans for a total of only $110.8 million. The
major reason for the program's small size appears to be that it is
actually easier to get a disaster assistance loan or loans guarantee
under Section 7(a) than one of the Section 7(b)(5) loans, primarily, it
seems, because of the double certification requirement noted above.
In order to avoid this problem in the case of TSCA loans, the second
certification could be waived. Because certification is most in issue
where compliance technology needs to be reviewed, this requirement could
be modified under TSCA. Thus, EPA might review PMN's in advance of the
loan to determine at least their procedural sufficiency, rather than
attempt to "certify" compliance.
68.
-------
4-12
Although this option is closely focused on the goals of TSCA, there
are still difficult administrative problems associated with it. For
example, while the SBA grants a great deal of autonomy to its regional
offices, EPA's Office of Toxic Substances is becoming increasingly
centralized, thus making the coordination between the two agencies
difficult on a day-to-day operational level. There is also the question
of whether the firms will find the requirements for getting a loan so
difficult and time-consuming (especially the certification process) that
they may find it easier to borrow at higher commercial interest rates.
This option could alter the perceived and actual barriers of the
cost of regulatory compliance, and might encourage companies to do
adequate health and safety evaluations of new chemicals, thus furthering
TSCA's goals.
E. Indirect Cost Subsidy for Chemical Innovation, in General, via a Tax
Mechanism
This option is aimed at reducing the costs of technological
innovation generally. It is based on the same assumption as the other
programs within this category, i.e., that reducing the costs of new
chemical development will encourage innovation. It also presumes that
the social benefits of innovation may exceed the private benefits, and
that therefore there is a role for government in encouraging innovation.
It works differently from the other programs however, in that it relies
on an indirect incentive, the tax mechanism.
This option would change section 174 of the Internal Revenue Code,
which governs research and experimental expenditures. Either one of the
following approaches could be taken. The current deduction for R&D
expenditures could be increased to, for example, a 200 percent deduction
rather than the present 100 percent in the year of the expenditure or
capitalization and amortization over the lifetime of the investment. A
second approach would be to provide a tax credit for R&D. This would
69.
-------
4-13
reduce a firm's tax in proportion to the amount spent for R&D. For
example, if there were a 10 percent R&D tax credit, this would mean that
for every $ 100 spent, the firm's tax would be reduced by $10. Either of
these two approaches could be specifically addressed to the chemical
industry.
This option has been used by several countries. Canada and Japan
use variations of the basic idea, and for many years Germany had an
accelerated or increased deduction for R&D, but recently abandoned it.
The efficacy of these programs has never been clearly established.
On the positive side, the tax mechanism is presumed to be effective
in encouraging R&D. To the extent that R&D promotes technological
innovation, so will the tax mechanism. Furthermore, it is equitable and
easy to administer once the program is in place.
On the negative side, tax programs tend to be costly. However,
because their costs are not directly reflected in new budgetary
expenditures, they are difficult to track. The chief criticism of this
proposal is that it is not sufficiently aimed at the unnecessarily
restrictive effects of TSCA. Also, its effect on new entrants and
unprofitable firms is minimal because such firms do not have taxable
income against which to use the credits or deductions.
F. Indirect Cost Subsidy for Testing and Compliance Costs via a Tax
Mechanism
The purpose of this option is to reduce the costs of testing and
compliance by indirect subsidy through the tax mechanism. This option
would be used to offset the costs that TSCA imposes as a result of
testing and record-keeping requirements, or other compliance efforts. It
could apply to both new and existing chemicals. There are three
principal approaches to accomplish this purpose. First, it would be
possible to increase the current deduction for costs of this type. Under
existing law, TSCA-related costs are an ordinary and necessary business
70.
-------
4-14
expense and are, therefore, entitled to full deductibility in the year in
which they are incurred. Under the approach proposed here, an increased
deduction would be given; for example, a double deduction for those same
costs in the year in which they are incurred.
A second approach would be to have a testing and compliance cost tax
credit that would apply to the same kinds of expenses as the deduction.
This would be a percentage credit that would reduce a firm's taxes
dollar-for-dollar in proportion to its expenses. The third approach is
to allow a tax-deductible testing "reserve." This would allow firms to
put money into a tax-deductible fund in one year and draw upon it in a
future year when substantial testing or compliance costs would be
incurred.
Any of these options would require specific legislative approval and
amendments to the Internal Revenue Code. This means that the House Ways
and Means Committee and the Senate Finance Committee would be involved in
creating this policy option.
One of the virtues of this program is that it could be relatively
simple to administer since it does not involve a large bureaucracy. On
the other hand, the experience with Sections 169 and 104 of the current
Tax Code suggest that there may be some problems with determining exactly
what kinds of costs qualify for special tax issues. A second virtue is
that it will indeed defray a large amount of the costs of TSCA. A third
is that because of the tax financing mechanism, the more the firms incur
testing costs, the more tax benefits they derive. This is consistent
with the goals of TSCA because it tends to encourage testing.
On the negative side, this option would necessitate a major change
in the existing tax laws. It violates the principle of tax neutrality
and, therefore, would face predictable political difficulty. It would be
especially difficult under the tax laws to justify special treatment for
the chemical industry when many other industries are also experiencing
increased regulatory costs.
71.
-------
4-15
G. Increased Patent Life for New Chemicals
This option is aimed at increasing the rewards for new chemicals by
lengthening the patent life that may be granted to them. It is based on
the assumption that one of the motivating forces for innovation is the
size of the financial reward that can be captured by the innovating
firm. The program responds to the allegation that the delay necessary in
order to comply with TSCA's regulatory requirements effectively decreases
the patent life of new chemicals.
This policy option wou.ld require new legislation to amend existing
patent laws. One approach would be to increase the period of patent
protection, for example, by increasing the patent life from 17 to 20
years. This period could apply to all patented products, or only to new
chemicals. Another approach would be to provide for an "add on" period
to patents for new chemicals. This period would equal the amount of time
that the regulatory agency takes to complete its decision making. A
third approach would be to start the period of patent protection running
when the PMN notification period is completed.
Another approach would not involve changes in patent legislation,
but rather, changes in enforcement practices. For example, EPA might
attempt to reduce patent infringements. This could be done by using TSCA
Section 8 reporting and recordkeeping requirements to identify potential
infringement cases.
Other countries have programs analogous to these in some respects.
For example, several countries provide for longer periods of patent
protection than the U.S. It should be noted that the beneficial effects
of patent protection, or of changes in patent protection, on techno-
logical innovation have never been clearly demonstrated even though they
are widely assumed, both here and abroad.
One of the.chief advantages of this option is that it would be very
simple to administer once the necessary legislation is in place.
72.
-------
4-16
However, it has a nunber of drawbacks. For example, it appears to have
little capability to offset TSCA's impacts and its effect would be felt
too far into the future (i.e., 17 years) to create much of an incentive
to innovate in the present. If the increased patent protection were
restricted to the chemical industry only, it would be difficult to
justify politically. If it were not so restricted, the legislative
change required would be so major as to render the option politically
infeasible at this time. Lastly, the effectiveness of EPA use of TSCA
reporting to enforce patent rights is questionable, and an attempt to do
so might even hinder the reporting it seeks to encourage for its main
regulatory purposes.
H. Strengthened Trade Secret Protection by Limitations on EPA Authority
to Release Information
ihe rationale for this option is similar to that for increased
patent life. Trade secrets are a form of protection for intellectual
property, which are presumed to encourage innovation by increasing the
financial rewards. To the extent that regulation decreases trade secret
protection, it creates a disincentive to innovate. TSCA, in particular,
may result in such a disincentive because of its comparatively strong
provisions that allow EPA to release confidential information to the
public. This program would counteract those provisions.
Currently, EPA has regulations governing the use and protection of
trade secret information that is submitted to it under TSCA. In
addition, there has been litigation between EPA and the Polaroid
Corporation on this issue. As a result, the current policies of the
agency are fairly firmly established. The only really feasible way of
strengthening the protection of trade secrets would be to amend the
section of the Act that governs such protection. The kind of amendment
envisioned would make TSCA more like other environmental statutes in that
EPA would be prohibited from releasing trade secret data to the public.
73.
-------
4-17
There are several analogues to this proposal. In most of the other
environmental statutes the agencies may not release to the public the
trade secret data to which they have access. This option would be
modeled on such statutes.
This option would encourage innovation if, in fact, trade secret
protection has this effect. On the other hand, it would also
substantially weaken TSCA by eliminating a major section of the Act
relating to the release of trade secret data to the public when necessary.
I. Decreased Taxes on the Sales of New Chemicals
This option rests on the assumption that increasing the potential
financial rewards for new chemicals will, in turn, encourage innovation.
It recognizes that regulation may increase costs, and attempts to offset
that effect to some extent by decreasing the taxes on profits from new
chemicals.
Many measures could be enacted to accomplish this purpose. The
simplest would be to decrease the tax rate on profits accruing to new
chemicals. But, singling out the chemical industry for such favorable
treatment is probably politically infeasible. A more realistic approach
would be to provide for deferral of taxes on new chemicals for a period
of one, two, or three years when they first are marketed. As a
variation, the tax deferral might only be made available if the
manufacturer who realizes the income reinvests it in the production of
new chemicals for new business. Another approach would be to change the
tax treatment of royalties and other licensing fees that are charged for
the use of chemical products. These fees are now considered to be
ordinary income. This approach would allow license fees to be taxed as
capital gains.
74.
-------
4-18
There are a number of analogues to these proposals. The Tax Code
contains many instances of tax deferral. For example, income taxes are
deferred on foreign income until it is repatriated, and capital gains
taxes on the sale of personal residences are deferred so long as the
gains realized are reinvested in other residential property. Many other
types of income are also awarded capital gains status.
These kinds of programs could be expected to be fairly effective in
that they substantially increase the awards accruing to new chemicals and
thus provide an incentive to innovate. Low or no capital gains tax and
investment incentive programs are widely employed abroad.
On the other hand, this option would probably be very costly. In
addition, it is so broad-based that it is unlikely to counteract whatever
negative impacts TSCA may have on innovation. Lastly, there are equity
and political problems associated with special treatment under the tax
code for the chemical industry that might make the program unacceptable
to Congress.
J. Increased Capital Availability for new Chemical Development via a
Government Supported Venture Capital Company
This option in intended to reduce the cost, and increase the
availability, of risk capital to firms that undertake new chemical
development. It should be tailored to those firms that are most
restricted in their access to capital; i.e., small and new firms.
The idea is to establish a new quasi-public corporation that would
provide venture capital (either loans or equity interests) to firms in
the chemical industry. Its total investment portfolio should probably be
in the range of $10 to $100 million. The corporation, which would be
funded initially by the government, would be run on a private, non-profit
basis, and should be self-sustaining over the long term. It would make
investments in new and hopefully innovative chemical companies. It would
75.
-------
4-19
be able to provide this capital cheaply because of the inital funding
base from the government, and also because it would not be under the same
kind of profit constraints as an ordinary private corporation. Over the
long term, the venture capital company would derive its revenue
principally from the capital gains (i.e., appreciated equity interests)
that would arise from successful! new chemical firms. Those gains could
be used to reimburse the government for the initial capital base. Excess
profits could be used to make more loans or stock purchases in innovative
ventures.
The corporation would establish criteria on which to judge its
investment options. The primary criterion should be the one that all
commmercial venture capitalists apply; that is, the commercial
profitability of the investment. Because of the special social purposes
of this corporation, additional criteria for its investments should
include the safety of the chemical products that are being proposed for
new development and the size of the firms that apply for the equity
investments.
Most other developed countries have programs of this type. For
example, in the United Kingdom, the National Research and Development
Corporation makes equity investments in firms and makes loans to new
ventures that are too risky for a private sector venture capitalist to
fund. Similar programs exist in West Germany and France. There is some
controversy about whether they have been successful; however, in at least
some instances their investments have been financially rewarding.
One of the principal benefits of this option is that it solves, at
least in part, the real problem of access to capital that many small and
new firms experience. A second advantage is that it is likely to be
relatively inexpensive over the long term. The third advantage is that
it is directed at new entrants, which would be expected to be more
innovative.
One of it§ drawbacks is that it represents a major policy departure
from existing government functions in the U.S. It would establish the
76.
-------
4-20
government as a venture capitalist in competition with existing venture
capital firms, which may not be necessary or desirable. In addition,
because this option is directed at new chemical development in general,
it cannot be expected to mitigate many of the potential undue impacts
that may be associated with TSCA. Lastly, it is possible that the
corporation might make so many poor investments that it would lose money.
K. Increased Capital Availability for New Chemical Development via
Changes in Tax or Securities and Exchange Commission Rules
This option attempts to decrease the cost and increase the
availability of risk capital in general in order to encourage new
entrants and innovative firms. It may offset to some extent, the
restrictive effects that regulation has, or is said to have, on the
availability of capital.
Fiscal, monetary, or money market regulatory policy could be used.
Fiscal policies are implemented through the tax system. Capital may be
attracted to the money markets via loosened rules for capital gains
treatment, a decreased capital gains tax rate, liberalizing the
qualifications for Small Business Investment Corporation status, or
allowing more lucrative employee stock options.
Monetary policies that affect the supply and cost of capital are
generally implemented by the Federal Reserve through changes in the
discount rate.
Money market regulatory policies are the province of the Securities
and Exchange Commission (SEC). The SEC policy that is generally
considered to bear most closely on the problem of innovation and venture
capital is contained in Rules 144 and 146. These prescribe a holding
period before new privately held stock issues can be resold to the public
and, in addition, dictate the rate of resale. As a result, they affect
the expected rate of return of a venture capitalist from investments in
new firms. The principal policy change that has been suggested in this
77.
-------
4-21
regard is to change the holding period and rate of resale provisions to
allow for a faster turnover and, therefore, a higher return.
All of these policy options have been tried occasionally in this
country. Many of them represent only small changes to existing policies
or reversions to previous policies. A considerable amount of controversy
surrounds all of them, as reflected in several series of Congressional
hearings on this problem.
Taken as a whole, these options can be expected to increase the
incentive to innovate, even though they are only modest incremental
changes from existing policies. Perhaps the most powerful arguments
against them, however, are that they do not deal specifically with the
problems presented by TSCA. In particular, they cannot be directed at
one industry such as the chemical industry. Coordination among the
various legislative mandates and implementing would also be difficult.
L. Reduce Risk Through Government-Financed Compensation for Losses due
to Regulation
This proposal seeks to lower the risk perceived by a firm or
investor at the early stages of product research and development. It
would do this by guaranteeing compensation for losses incurred as a
result of government regulation, which should encourage investment in
risky ventures.
The program envisioned would be a government-supported insurance
fund. Initial funding could originate either from general tax revenues
or from a specific tax on chemicals (see the discussion of financing
options in chapter 7). If EPA regulated a product under Sections 5, 6,
or 7 of TSCA, the fund would then reimburse the firm for direct losses
due to the regulation (the costs of testing, the market value of the
current inventory, and perhaps some percentage of anticipated future
losses).
78.
-------
4-22
There are very few similar programs at EPA or any other regulatory
agency, but the Office of Regulatory Analysis has a major review of this
policy alternative underway. The Overseas Private Investment Corporation
(OPIC), which ensures against political risks abroad, can serve as a
model.
This proposal would certainly encourage companies to research and
market products that might be regulated in the future because of
potential risk to consumers or the environment. It has two major but
possible contradictory effects; while it encourages the development of
new products, it could also encourage companies to market less safe
products because they would be protected against losses due to
regulation. This clearly contradicts the spirit of TSCA.
M. Reduce Risk Through Government Procurement of New Chemicals
This option would guarantee companies with new products a certain
minimum market by directing government procurement policies toward new
chemicals. The government would create a "market-pull" toward new
chemicals that were safer than existing chemicals.
Procurement in the government is carried out primarily by the
General Services Administration (GSA), although individual agencies
(particularly the Department of Energy and the Department of Defense) do
have large internal procurement programs. Under this proposal, the GSA
might work with EPA to develop a list of the chemical products purchased
by various agencies. It could be authorized to consider substituting new
products that appeared to be safer, even if they were more expensive that
existing ones. The authorization for this effort could either be made
through the annual authorizations of various agencies and departments or
as an experimental policy, through amendment of the Federal Procurement
Regulations.
79.
-------
4-23
The Experimental Technology Incentives Program (ETIP) administered
by the National Bureau of Standards, is similar in concept to this
proposal. ETIP was started in 1973 as an "effort by the Federal
government to learn how to better stimulate the application and use of
technology in commercial products and services." Rather than
concentrating on one type of product or industry, it sought to perform a
number of diverse experiments in technology-forcing procurement. At best
it can be said that ETIP's procurement effort has drawn mixed evaluations.
Another type of program, which has no precedents, would not try to
substitute safe compounds for existing products but would, in fact,
introduce a large new budget item specifically for the procurement of new
chemicals. This type of program is obviously less focused, but might be
easier to administer. It is certain to have a more widespread effect.
There are several obstacles to the procurement strategy. First, the
most effective procurement program is one that is as well-defined as
possible, but this would also be the most difficult to administer. For
example, the first proposal above might be very helpful to a small
company trying to market a new product that would be substitutable for an
organic solvent implicated as a possible mutagen. Identifying that
company and insuring that it conformed with all of the other procurement
requirements might prove so burdensome that the GSA would go back to the
large manufacturer of the implicated solvent. There is also the problem
of ensuring that the other goals of the GSA—low cost, reliable delivery,
and other statutory obligations—are being met at the same time.
Finally, in the case of the second option discussed for general new
chemical procurement, it is not clear that the government could influence
a large enough segment of the overall chemical industry market to make
such a broad approach worthwhile. In any event, it would be very costly.
N. Reduced Risk from Products Liability Actions by Establishing Limits
On Liability
The possibility of a products liability suit is one of many market
risks that the entrepreneur faces. This threat may, to some extent,
80.
-------
4-24
deter innovation. Recently, it has been argued that the uncertainty in
the product liability laws makes it extremely difficult, especially for
small or new firms, to contract for adequate product liability insurance,
which also is likely to deter innovation. This proposal may encourage
innovation by providing an upper bound to the risks involved in product
liability actions, and by reducing uncertainty may make it easier to
obtain insurance.
Products liability doctrines arise from state court decisions; there
is no Federal law of products liability. Therefore, in all likelihood,
the policy option considered here would have to be carried out on a state
by state basis. A prototype of a uniform Federal products liability law
has recently been circulated by the Department of Commerce. However, the
possibility of its passage seems quite remote.
There are at least three major alternatives within this policy
option. One is to place a dollar value limit on plaintiff recovery in
those products liability actions that are based on defects in chemical
products. This limit on liability could be coupled with the government
assumption of damages exceeding that amount. Another alternative is to
enact statutes of repose or statues of limitation that fix the time
period over which products liability actions are allowed. A third is
that the government could actually assume liability itself, as in the
case of the swine flu vaccine.
There are some analogues to this option in other areas. One is the
Price-Anderson Act, a Federal law that establishes limits on liability
for nuclear accidents. Another is a series of state laws that limit
plaintiff recovery to specified dollar amounts in airline accident
fatalities.
A favorable aspect of this option is that it increases the certainty
with which the commercial risk can be calculated and, therefore,
eliminates some of the deterrent that products liability actions may pose
to innovation.
81.
-------
4-25
One major drawback Is that it works a substantial injustice on
harmed plaintiffs, because it denies them recovery. However, this can be
mitigated somewhat if the government assumes some or all of the
liability. Another drawback is that it decreases the incentive to
manufacture safe products. Many believe that the threat of products
liability actions is, in fact, one of the chief incentives for improving
the safety of existing chemical products—precisely because of the
uncertainty and high risk associated with products liability recoveries.
These risks are likely to be highest for the least safe products. This
policy option can also be criticized because it would be very difficult
to implement since it would necessitate changing the laws in all 50
states.
0. Establish Government Testing of Health Effects
Another proposal that would reduce the cost of testing and obtaining
data as well as increase the industry's confidence that EPA will accept
test data, is to have the government establish national test facilities.
Such facilities would also increase the EPA's control over test protocols
and results.
Because many large firms now have their own testing facilities, a
national laboratory would tend to serve small and medium-sized companies
lacking in-house testing capability. These firms would send samples of
their compounds with a request for the types of tests to be performed.
The lab would charge a fee (probably subsidized, in order to compete with
private testing labs) for each procedure. The results and sample would
be sent only to the manufacturer; EPA would not see the results except as
they appeared on the PMN.
This program, which would substantially alter EPA's role in the
administration of TSCA, would probably necessitate new Congressnioal
authority. The testing could take place either in a government lab
82.
-------
4-26
(v;ithin or outside EPA) or in a quasi-public corporation. In both cases,
the fee structure could be arranged so as to allow the facility to be
financially self-supporting.
While there are government programs that evaluate products for
health and safety, there are no facilities that could adequately perform
the quantity of tests that would be required for the implementation of
this program. Thus, it is not now posible to compare the experiences of
these existing programs in order to see how such a facility might be run.
There are a number of objections to this proposal. Companies
fearful of jeopardizing trade secrets might be reluctant to entrust their
new products to such a facility. Determining the most appropriate size
for a facility would be difficult, given the present uncertainty over the
extent to which such testing capacity may be required. It is possible
that because of both the recent surge of testing facility construction
and the uncertainty over the volume of testing to be done, the lab would
be underutilized.
The program does, however, offer several advantages: it allows for
more scrutiny of protocols, procedures, and test conditions; it is a
direct method of controlling the cost of testing; and it increases
manufacturers' confidence that test protocols, procedures, and results
will be accepted by EPA.
P. Sharing Test Data with Reimbursement
TSCA Sections 5(h}(2)(A),(B), and (C) allow the EPA Administrator to
set a fair and equitable reimbursement price when one company uses
another company's health and safety data for new chemicals. (We are not
concerned here with the reimbursement provisions of Section 4, applicable
to existing chemicals.) This helps to insure that the company that did
the testing will not suffer a competitive disadvantage because it
performs the tests. As now written, these sections apply only to a
83.
-------
4-27
narrow set of substances: chemicals on the Section 5(b)4 list. The
manufacturer of such a chemical may petition EPA for an exemption from
the testing requirements if the chemical substance is equivalent to one
for which data have already been sjbm'tted. The Administration will
grant the exemption if additional testing would be "duplicative."
The policy option proposed here would expand the existing
reimbursement authority under Section 5, to allow manufacturers of new
chemicals to share test data whenever they exist. Procedures for
obtaining reimbursement would remain as currently constituted in the Act.
This proposal would be similar to the current pesticide program.
EPA might catalogue the PttN's by co-pound and indicate whether health and
safety studies had been performed. This list would be widely distributed
so as to be readily available to any company investigating a new
compound. The company could then petition EPA and would receive from it
the names of all the manufacturers that had either carried out health and
safety studies or sought and received exemptions. The company would then
contact all of these firms and offe^ to reimburse them. If the parties
could not reach an ain'cable understanding about appropriate reimbursement
terms, the Federal Mediation Service might be able to arbitrate.
(Section 5(h)(2)(B) of TSCA, which calls for the Administrator to arrive
at a fair and equitable reimbursement by rule when the parties cannot
agree, may already permit this.) Ar.other approach would be to key the
reimbursement to a percentage of sales (e.g. 5%), in a manner reminiscent
of percentage depletion allowances.
EPA has had previous experience with data compensation programs with
pesticides, stemming from the Federal Insecticide, Fungicide, and
Rodenticide Act of 1972 (FIFRA). Its experience suggests that the
administrative determination of "fair and equitable" reimbursement is
extremely difficult, so much so tha- one of the major goals of the 1978
FIFRA amendments was to transfer that authority from EPA to an
arbitrator. A similar system under TSCA might require an amendment of
the Act.
84.
-------
4-28
This approach may be valuable, but its implementation could easily
become mired in administrative delay. Even the arbitration arrangement,
although apparently an improvement over the previous system, can be quite
lengthy. In particular, EPA authority to compel sharing and prescribe
reimbursement rates may be necessary to force cooperation from companies
reluctant to share their test data. The equity and cost-saving aspects
of the program are perhaps its best features. Its potential to promote
innovation is probably limited to chemicals that are closely related to
those already on the market and for which test results would be
transferable. This implies that its effects will probably be greater for
existing chemicals than for new products.
Q. Facilitate Private Sector Joint R&D or Joint Testing
This policy option is responsive to the argument that many
regulatory costs are "unproductive" from the viewpoint of individual
firms. Joint testing of new chemicals or joint R&O on compliance options
can reduce regulatory compliance costs to individual firms and perhaps
increase the efficiency of the resulting technologies. These benefits,
if they in fact exist, may result in more or better testing and
compliance methods and perhaps more innovation. The need to allow joint
testing or joint R&O may be especially pressing in the case of new and
small firms that lack the technical resources to perform tests and to
develop new compliance technologies.
This option is basically a question of antitrust policy.
Implementing it would involve both the Department of Justice (DOJ) and
the Federal Trade Commission (FTC), and possibly new legislation from
Congress. There are two separate issues that need to be addressed. One
concerns joint testing activities. Here the antitrust problems do not
seem to be particularly severe, although there is always the possibility
of collusion and monopoly. The second issue is somewhat more
complicated. It involves joint RW3 on compliance technology development
85.
-------
4-29
among different firms. This raises a stronger possibility of collusion
or monopoly because It Involves activities closer to commercial
development. In both cases, different policy options are raised when the
joint efforts are undertaken by a trade association and when they are
undertaken among individual firms.
There are two approaches to deal with these various problems. One
would be to provide for antitrust exemptions for all or some of these
kinds of activities. This might involve new legislation from Congress,
or it might be simply a matter of DOJ or FTC policy. The second approach
would be to Inaugurate an advisory opinion mechanism from the OOJ and FTC
in circumstances of this sort. This would allow the firms who wish to
engage in joint R&O the opportunity to be reasonably certain about the
antitrust implications of the venture before they begin, and might be an
Improvement over existing antitrust policy, which some people allege is
uncertain in this regard.
Joint R&O for compliance technology, and joint testing, are common
practices abroad. The antitrust laws in most developed countries are
quite different from those in our own, and the practices described in
this policy option are not seen to pose any serious social danger. The
one analogue in this country is the experience of the automobile
Industry. In 1969, the OOJ brought an action against the automobile
Industry for collusion and conspiracy with intent to frustrate the
pollution control laws. This action resulted in a consent decree between
the government and the Industry that prohibited automobile firms from
engaging in joint R&O. IN 1979, as a result of another court action,
this restriction was lifted and cooperation among members of the industry
is now allowed.
This policy has the virtue of not requiring any new financial
commitments. In addition, it would probably be of most use to the new
and/or small firms that are most in need of cooperating in regulatory
compliance efforts.
86,
-------
4-30
On the other hand, this option may contravene the goals of antitrust
policy and contradict established procedures of the antitrust enforcement
agencies. It should also be recognized that cooperative activities among
regulated firms raise the possibility of collusion, not only to
monopolize certain portions of the industry but also to frustrate the
intent of different regulations. Lastly, this option appears to be more
valuable when the costs of compliance technology are extremely high (for
example, for air pollution devices or major chemical process changes)
than in the case of the individual regulatory requirements under TSCA, no
one of which is likely to be very costly.
R. EPA Dissemination of Information: Test Results and/or Labeling
The main goal of this proposal is to create favorable market
environments for safer chemicals. By publishing the results of health,
safety, and exposure studies and/or labeling products with health and
safety information, consumers can be encouraged to purchase products with
safety as one criterion. This should provide a "market-pull" for safe
chemicals and encourage product substitution.
For the labeling option, EPA would issue guidelines specifying the
kinds of information that must be shown on the product's label.
Information like acute toxicity levels, effects observed during animal
studies (including statistically significant increases in tumor
formation), exposure studies, and suggested limitations on use might be
included. A simpler alternative is to allow for an EPA approval on the
label. These options might require new statutory authority because they
go rather far beyond existing policy -- both in terms of information
disclosure requirements and in terms of EPA product endorsements.
The option to disseminate test results would involve EPA itself in
such activity. Essentially, it means a much more vigorous use of TSCA
Sections 6 and 14 authority. To minimize any problems that might arise
87.
-------
4-31
with respect to data confidentiality, EPA could release test data and the
product names only without disclosing any confidential identities.
However, companies often claim confidentiality for this information. The
second option, labeling, would not create confidentiality problems. It
has the obvious advantage of immediacy to the consumer, but is probably
more difficult to implement than the dissemination of test results,
because of its product endorsement aspect.
One analogue is the Food and Drug Administration's (FDA) labeling
program. There has been consistent opposition to FDA labeling rules in
the past. One important lesson that the EPA can learn from the FDA's
experience is that the form the information takes can play a major role
in the amount and kind of opposition engendered. It can be argued in the
case of drugs that much of the information printed on the label is
incomprehensible to the average consumer. However, the greater the
simplification, the more interpretive of the data the labels will-
become. Striking a balance satisfactory to all parties will probably be
the most difficult part of any labeling program adopted by EPA. EPA and
OSHA are currently in the process of developing generic labeling programs
that will, of necessity, address this issue.
These policy options have the virtue of being inexpensive to the
government. They probably do not require new legislation, and are
entirely consistent with the purposes of TSCA. They may be effective in
promoting new safer chemical products, and they work through the
establishment of market incentives on the part of consumers. Their chief
drawback, however, is the difficulty the EPA would face in balancing the
public "need to know" with the desires of manufacturer's to protect their
trade secrets. The inability to achieve a proper balance of these
factors could have the effect of impeding the flow of health and safety
information from the companies to the agency.
88.
-------
4-32
S. Chemical Technology Extension Service, Including Dissemination of
Information on Test and Compliance Methods
This option is targeted at the variety of problems that arise in the
development of new technology, both generally and for compliance with
regulation. Its objective is to reduce costs, either those associated
with developing new technology or those resulting from compliance with
regulation. Furthermore, this option assumes that the existing body of
technical knowledge can be better utilized. It would, therefore, attempt
to bring lagging firms up to the state-of-the-art in various technical
areas by disseminating the latest knowledge and providing consultation.
It is particularly targeted at the differences in the knowledge base
among various firms in the private sector, and would emphasize the
special needs of new and small firms. Lastly, it assumes a lack of
knowledge about applicable regulations and how to comply with them, and
would seek to remedy both of these deficiencies by the dissemination of
regulatory information.
The basis for this program would be a network of technical centers
and technical extension agents. The centers would be responsive to
inquiries from firms in the private sector, and would provide an
institutional affiliation for the technical agent. The agents would
comprise an out-reach program to visit and consult with individual
firms. Both would attempt to foster technological innovation in the
private sector. Their special mission would be to disseminate regulatory
information and information about compliance technology.
The centers and agents would be federally funded. However, because
of the decentralized nature of the program's services, it would have to
be implemented and run on a state and/or local level. The funding base
would have to be quite large in order for the program to make any
significant impact. It is estimated that approximately $50 to $100
million a year would be needed for a major impact, although smaller
experimental efforts could be undertaken.
89.
-------
4-33
There are three analogues to this program in use today in the United
States. One is the Agricultural Extension Service, which has been
extremely successful in disseminating agricultural technology to farmers
in this country throughout the last century. The second is a group of
approximately 30 state technical centers, authorized by the State
Technical Services Act. Thus far this program has been implemented on
only a small scale, and has not dealt at all with regulatory problems.
The one similar program that deals with regulatory issues is OSHA's
state-level consultation program, which is restricted to a single
regulatory area and is funded at a modest level.
The principal virtue of this proprosal is that it is likely to be
quite effective in improving the knowledge base of some firms in the
private sector and in improving the uses of existing technologies. Its
chief negative aspect is that it would be very expensive. In addition,
it would be difficult to implement, since it relies both on a federal
program and a multiplicity of state or local programs. Lastly, it can be
faulted because it is not targeted well at the particular problems that
TSCA raises and would do little to offset its impacts on innovation.
Indeed, the problems of compliance with TSCA, especially for new
chemicals, may not be as susceptible to technological solutions as other
regulatory problems.
T. Antitrust Action to Favor New Small Firms in the Chemical Industry
This option is based on the assumption that monopoly power, perhaps
more than regulation, is a significant barrier to innovation in the
chemical industry. It assumes that innovation will be increased if
competition in the industry is increased. It is intended to favor
smaller and newer firms that would probably find it easier to survive if
the large existing firms in the chemical industry were broken up.
90.
-------
4-34
This policy option could be implemented in several ways. First, it
could be accomplished by new antitrust legislation. For example, firms
of a certain size could be legislatively specified as per se violators of
the antitrust statutes. This kind of proposal has been considered
intermittently by Congress during the last five years or so, with very
little legislative success. Another approach would be to have more
vigorous enforcement of existing antitrust legislation, which would have
to be pursued by the DOJ and FTC. A third approach would be that various
legislative and/or administrative mechanisms could be used to make
mergers more difficult, for example, strong pre-merger clearance
procedures.
As far as can be ascertained, there are no analogues either in this
country or abroad to this kind of program. There have been no major
recent antitrust actions taken against an industry quite so large and
complex as the chemical industry. To the extent that competition
actually does foster innovation, and antitrust action against the
chemical industry would indeed increase competition, this option would be
beneficial. It might also have favorable political or social side
effects. On the other hand, it would be extremely difficult to implement
properly. It would not be easy to determine, for example, on exactly
what basis to break up the existing firms and what kind of new entities
would be most advantageous. Furthermore, antitrust actions are usually
undertaken on a case-by-case basis and can involve years of litigation.
This option could also be criticized on the basis that its connection to
the goals of TSCA is very tenuous, therefore, it would have little
capability of offsetting the unnecessarily restrictive impacts of TSCA.
U. Tax Adjustments to Favor Small Firms or New Entrants in the Chemical
Industry
This policy option is based on the assumptions that small and new
firms tend to be more innovative, and that regulation affects them more
severely than it does large firms. It should be recognized that neither
of these assumptions has been proven. (See chapter 2 and appendix B.)
91.
-------
4-35
The program would operate through the tax mechanism, which has the
advantage of minimal government intervention and is likely to be
efficacious. A variety of specific tax measures could be implemented to
reduce the tax rate for new and/or small firms. In order to favor small
firms, one approach would be to loosen the requirements necessary to
qualify for subchapter S corporation benefits. A second approach would
be to provide a period of tax deferral for new entrants into the
industry. Both approaches would require new legislation and would have
to originate in the committees having jurisdiction over taxation rather
than in those having jurisdiction over TSCA.
None of the mechanisms proposed here is new; they are essentially
incremental variations or additions to existing tax programs. There has
been substantial policy debate about all of them for several years.
Congress has not yet seen fit to carry many out, although the loss
carry-forward provisions have been liberalized in recent years. This has
allowed greater profits to new firms after the first few years of
operation in which they usually experience losses.
These programs, considered as a package, would probably be helpful
to small and new firms, and even be effective in changing industry
structure to some extent. On the other hand, they are almost irrelevant
to the aims of TSCA. In addition, because they substantially reduce
taxes, they are likely to be very expensive.
V. Government Support to Develop New, Better Test Methods
This option provides government support to improve test methods
relevant to TSCA. It casts the government in a traditional role that is
widely acceptable. Its justification is that private firms do not fully
benefit from all the social rewards that improved testing would yield,
therefore government support is warranted. Such support is not likely to
decrease the cost of compliance substantially, but it could improve
signifcantly the quality of compliance efforts.
92.
-------
4-36
Government support can be provided via a variety of mechanisms. In
theory, grants could be given to private chemical firms, but this would
be quite controversial. Pore reasonable alternatives include research
grants and contracts to universities and non-profit institutes or work in
government laboratories.
Currently, there is a considerable amount of work going on along the
lines proposed here, funded for example, by EPA and the National
Institutes of Health (NIH). An assessment of the quality and quantity of
this existing activity would be needed to appropriately judge the
potential effectiveness of the policy options suggested here.
W. Government Support for Education and Training Programs
Since all environmental regulatory programs must ultimately be
implemented through the efforts of highly trained individuals, a new
regulatory regime creates new personnel needs in both the public and
private sectors. This is particularly true of TSCA, which relies to such
an extent on product testing. The government has traditionally assumed
the role of funding education and training (either directly or through
the research contract mechanism) in areas related to new national needs.
The space program is one outstanding -example. The proprosal here is to
apply traditional modes of educational support to the needs created by
TSCA.
This policy option would be implemented primarily through the
university system. Government support could be for internships,
fellowships, or contract research in TSCA-related disciplines --
toxicology, biology, genetics, epidemiology, etc. Outside of the
universities, programs could include retraining and internships. These
might be funded directly or through special tax provisions for the
educational expenditures of private firms.
93.
-------
4-37
It would be expected that this policy option will generate little
controversy. It exhibits a clear special benefit, and is consistent with
longstanding notions of the appropriate governmental role. One argument
against such training programs is that they do not show results until
after the national need is past because of the long lead-times involved
in university education. In addition, it has been argued that the
availabilty of scholarship or research funding in a particular field
attracts more students to it than the actual demand warrants, resulting
in eventual oversupply. These disadvantages may be overcome, however, by
shorter-term retraining programs.
X. Actions Against.Existing Substitutes for New Chemicals
This policy option attempts to increase the market incentive for
new, safer chemicals in the chemical industry. It recognizes that there
are always substantial market barriers to the introduction of new
products, especially for those that are safer but more expensive than
existing substitutes. While one of the purposes of TSCA is to increase
the safety of chemical products on the market, the regulatory regime that
the Act establishes is more rigorous for new chemicals than for existing
chemicals. This proprosal would provide one means to equalize new and
existing chemicals. It relies on regulating the existing substitutes .for
new, safer chemical products rather strictly. It is hoped that thereby
the development of safer and more innovative new chemicals would be
fostered.
The program would begin when a PMN is filed. Action would then be
triggered against an existing product (or products) substituting for the
product covered by the PMN. This would take place in addition to
whatever action is taken on the PMN. There are a variety of actions
available to EPA in this regard. One approach would be for EPA to
establish a kind of presumption of unreasonable risk associated with the
existing substitute whenever a PMN is filed for a safer new chemical.
94.
-------
4-38
Procedures to require testing could then be commenced. Another approach
might be to release whatever test data EPA has access to on existing
chemicals when a PMN is filed for a product that could be a substitute.
A third would be to release a briefing to the public whenever a PMN on a
new, safe substitute for an existing chemical is filed.
As far as can be ascertained, no analogues exist in any of the
existing regulatory programs either in this country or abroad.
This option has the virtue of providing a strong incentive for
safety. In this reg'ard, it is highly consistent with the aims of TSCA.
Also it requires no new legislation. On the other hand, it poses rather
severe difficulties in implementation. For example, it will be very
difficult for EPA to make relative risk decisions, to discriminate
properly between new and existing products, and to be certain about its
determination that the new substitute is actually safer than the existing
product. Lastly, political controversy may be engendered, particularly
by those firms which have been adversely affected in the market by the
publicity that the new substitute products have received from EPA.
Fixing Time Periods for Regulatory Action
This option is addressed to the problem of uncertainty in regulatory
actions. One of the criticisms that business interests have leveled
against regulatory agencies is that their requirements change over time,
thus making it difficult for business to plan. Innovation may be impeded
in this climate of uncertainty.
The policy option considered here would set a fixed time period for
individual regulatory actions. For example, the agency might agree not
to contact an applicant firm or review its product, except in
extraordinary circumstances, for a fixed period of time following
submission of a PMN, presuming no action were taken by the agency on the
PMN within the normal review period.
95.
-------
4-39
As far as can be ascertained, this policy option has no analogue in
other regulatory areas. It has the advantage of increasing the private
sector's perception of certainty in regulatory decision-making. On the
other hand, it has a variety of disadvantages. For one, it is not likely
to be especially effective in offsetting whatever undue impacts TSCA may
have. For another, it tends to deprive the regulatory agency of needed
flexibility, especially in circumstances where new environmental, safety,
or health hazards come to light. When such problems arise there is an
imperative to protect the public. Therefore, it is difficult to see how
such a policy option could provide sufficient degree of certainty while
still protecting the public from unreasonable risks.
Z. Post-Market Surveillance of PMN
This policy is addressed at two distinct problems in the TSCA
regulatory system. For one, the decisions on PMN's are often seen as
"all or nothing," in which a positive decision on the PMN is often the
final action that the agency takes against the chemical. This has the
drawback of hindering the agency's reconsideration of its action when new
health and safety data come to light. Often, in the face of this
dilemma, it will be forced to resolve its doubts about the PMN
determination in favor of safety, and keep new products off the market.
The post-market surveillance option allows EPA not to take action on
products that it otherwise might while continuing to watch them over a
period of years until it is certain about their environmental, health, or
safety consequences.
There is a clear analogue to this program in the post-market
surveillance efforts that have taken place under the Food, Drug, and
Cosmetic Act in the regulation of new drug applications.
This option has the virtue of allowing innovations to be marketed
that otherwise would not have been. On the other hand, it poses some
health, safety, and environmental risks to the general population and, to
the extent of these risks, may be inconsistent with the goals of TSCA.
96.
-------
4-40
AA. Regulatory Exemptions for Low-Volume Chemicals
This is the first of a series of three proposals that would exempt
different classes of chemical products or manufacturers from certain
provisions of TSCA. These exemptions could apply to the submission of
PMN's or to any other regulatory requirements under the Act.
The assumption underlying this proposal is that for chemicals that
are produced in small volumes the risks are less than the potential
benefits. These exemptions would be applied to any company that produced
less than a certain volume of the chemical in question, regardless of the
size of the company involved. Thus, both large and small producers would
benefit. However, since EPA's concern is not only with the possible
exposure of workers but also with general human and environmental
exposure, it might be necessary to impose a total production ceiling over
which producers would have to start submitting PMN's.
Although there have been proposals to exempt small businesses from
regulatory requirements under different environmental laws (see option
BB), this program has no exact analogue. All R&O chemicals are exempted
under TSCA.
This exemption is likely to stimulate the innovation of new
chemicals, since producers would be confident of being able to market
without regulation or mandatory reporting of known health and safety
data. This would apply particularly to specialty chemicals for unique
purposes. EPA would also be confident that if the chemical turned out to
be a major innovation, it could always be regulated at a later time when
the volume ceiling was exceeded.
However, this exemption could be a major impediment to the effort to
control environmental exposure to potential chemical hazards. The
effects of a highly persistent chemical that was produced in small
volumes for several years could be significant and long-lasting. In
97.
-------
4-41
addition, without the submission of PMN's, even if EPA discovered that a
new chemical was a potential hazard, there would be logistical problems
in locating and regulating its manufacturer. Lastly, the program's aim
could be circumvented by the proliferation of small companies, each
producing volumes just below the exemption limit.
Regulatory Exemptions for Small Firms
This exemption from all of the TSCA regulatory requirements would
apply to small firms, on the assumptions that they may be more innovative
than large ones and are less able to bear the costs of regulation.
The designation of firms as small could be based on the number of
employees (as the SBA does), on the volume of sales within the firm's
manufacturing cohort, (e.g., its four-digit SIC code or on some other
measure of size).
There have been efforts to exempt small businesses from the
requirements of Acts such as the Occupational Safety and Health Act, but
there is no good precedent on which to gauge the possible effects of this
exemption from TSCA. Also, TSCA itself exempts small firms from certain
requirements, notably those of Section 8.
Such an exemption would result in lower regulatory costs for small
businesses , and would probably promote innovation. However, there are
several arguments against it. First, there is no evidence that small
companies produce only small volumes of chemicals. This is particularly
true when the criterion for size is number of employees. Second, there
is some reason to believe that the working conditions in small companies
may be less safe than in large firms. This could lead to an actual
increase in human exposure that would be proportionally greater than if a
firm's exemption was based on the volume of a particular chemical it
produced (option AA).
98.
-------
4-42
CC. Regulatory Exemption for Low-Risk Chemicals
This last exemption from the TSCA requirements would be applied to
chemicals that were judged by EPA to be low risk. As such, it would be
related to the generic PMN (option EE) and the fast-track option (option
DO). The assumption is that it is possible to predict on the basis of
structure-function relationships, chemical family classification, and
prior experience, whether a new compound that has not undergone extensive
health/safety evaluations will present an unreasonable risk of injury to
human health or to the environment.
EPA would prescribe guidelines (as in the preceding two cases) for
exemptions classified according to chemical family, molecular weight, or
class (e.g., definite structure or indefinite structure). This would
have the effect of driving innovation in the direction of safer
categories.
As in each of the two preceding exemptions, EPA could expect major
controversy over the definition of the exempted category, particularly in
this case, however, since EPA would often be basing determinations about
"safety" on very limited actual or theoretical evidence. Therefore,
while this basis for exemption is more consistent with the overall goals
of TSCA in that it seeks to rank chemicals on the basis of perceived
risk, it is questionable whether EPA or anyone has the necessary
knowledge and skill to perform such a ranking. In addition, these
determinations, which would be very controversial initially, could become
even more so if they ultimately prove incorrect. Thus, they could be
seriously damaging to the agency's credibility.
00. "Fast-track" PMN's for Safe and/or Major Innovations
The fast-track program is a method of cutting down on the time
involved in the PMN process. Compounds that qualify would receive rapid
99.
-------
4-43
reviews and actions. In order to qualify, a potential producer would
declare that a new compound is low-risk, or that it is as safe or safer
than one already in use. Since the notification period under TSCA is a
maximum of 180 days, delay is not generally considered as important a
factor under TSCA as it is for approval of a new.drug or pesticide.
Nevertheless, some specialty compounds or other products that require
short turn-around times may benefit from this policy, There is, however,
no way to estimate the number that would be affected.
The fast-track would not be expected to play a significant role in
accelerating innovation overall, but could be useful in some cases. As
with the exemption for low-risk chemicals, there is likely to be strong
disagreement about whether a specific chemical should be fast-tracked.
EE. Generic PMN for Classes of New Chemicals
This program would greatly simplify the PMN procedures for those
chemicals and innovations judged by EPA to be low-risk. It would allow
classes of new chemicals to be reviewed as a group by EPA. Rather than
exempting these compounds, however, this proposal would decrease the
reporting requirement on the PMN. For example, lists of all known safety
and health data, volume, or manufacture, and so on could be reported for
the class rather than for each individual chemical.
This option has been suggested in connection with the FDA approval
of drugs. It has the advantage over the regulatory exemptions (see
option CC) of providing EPA with a data base in the event it were to
become necessary to regulate the chemical at some future time.
Furthermore, it does not tie EPA's hands as an exemption might. The same
problem exists, however, of determining which categories of innovations
are low-risk. Thus EPA could anticipate facing similar (but less
intense) controversies than with the regulatory exemptions.
100.
-------
4-44
FF. Improved EPA Staff Capability to Assess the Impact of Regulatory
Actions on Chemical Innovation
A number of criticisms have been leveled against regulatory agencies
by business interests. One is that these agencies are not sufficiently
sympathetic to the problems of business. A second is that they do not
have adequate knowledge of business practices. A third is that agency
personnel do not have sufficient technical expertise. And a fourth is
that insufficient analysis is undertaken before regulatory actions are
concluded. This option proposes a mechanism to respond to these kinds of
criticisms.
The proposal is based on the assumption that better informed action
on the part of the regulatory agency will result in a smaller negative
impact on innovation. There are many actions that agencies can take in
this regard. For example, industry - government personnel exchange
programs have been suggested. This might Involve amending of the
Intergovernmental Personnel Loan Act to apply to exchanges of personnel
between the public and private sectors'. Another possibility would be to
provide more training on the analytical assessment of the consequences of
each regulatory action on technological innovation.
There are a variety of analogues to these kinds of programs 1n other
regulatory areas. For example, there are already several different kinds
of impact statement requirements that have been imposed on regulatory
agencies within the last few years. These tend to delay regulatory
action and to impose new administrative costs. On the positive side, it
can be presumed that the kinds of programs proposed here can only help
EPA's decision-making. However, the degree of improvement may not be
very great. Lastly, the delay in regulatory action resulting from
additional analysis could impede the achievement of the aims of TSCA and
other regulatory legislation.
101.
-------
5. A PROCESS FOR ASSESSING THE POLICY OPTIONS
In chapter 3 a basis is developed for understanding how TSCA might
have unnecessarily restrictive impacts on the process of technological
innovation. Chapter 4 then presents a wide variety of policy options for
offsetting these impacts. In this chapter the process by which these
options were assessed by the project team is explained.
There are three reasons that the construction of an elaborate
economic model to estimate the costs and benefits of the options was
neither feasible nor appropriate: 1) currently available economic models
generally cannot adequately address the dynamics of technological
innovation, 2)the data needed for such an assessment are lacking, and 3)
factors other than relative costs and benefits, such as administrative
feasibility and effects on the primary goals of TSCA, are necessary
inputs to the decision process. Therefore, the assessment procedure used
in this study is based on a less rigorous, although structured, analytic
approach known as Magnitude Estimation Group Scaling.
5.1 Criteria for Assessment
Any proposed policy option must be evaluated on the basis of certain
general criteria: whether it can it be implemented; whether it will work
if implemented; what it costs will be; and whether its implementation
will lead to any further problems. Each of these broad criteria can be
subdivided; e.g., both private and public costs must be taken into
account.
For this project, a detailed set of evaluation criteria was first
developed. (They are presented in appendix C.) From these, eight
fundamental criteria for policy assessment were selected:
102.
-------
5-2
1. Capacity to countervail
2. Private costs
3. Public costs
4. Administrative feasibility
5. Time to implement
6. Supportive of TSCA's aims
7. Other side effects
8. Political feasibility
Subsequently, two other criteria were added. Criterion 9, "initial
policy rating," which is an initial estimate of the overall rating of a
policy made without reference to the detailed criteria, and criterion 10,
"effectiveness," which represents a combination (the product) of a
policy's capacity to countervail and its adminstrative feasibilty.
£ap_aci jiy to_ Countervail
This criterion captures the ability of a policy option to offset the
unnecessarily restrictive impacts of TSCA on innovation. As with all the
criteria used in assessing the policies, the capacity to countervail was
judged independent of cost or any other factors. However, it does take
both the direct and indirect effects of policies into consideration.
Pr1vate_ Cpsts_
While it might appear that policies to stimulate innovation should
not create new private costs, this criterion is designed to capture
whatever private costs may accrue from taking advantage of or complying
with a policy. Many innovation stimulating policies could require
additional record keeping by a firm. These costs could offset some of
the expected benefits to the firm, or might discourage smaller firms from
utilizing the program.
103.
-------
5-3
>ub1cCSts
This criterion is designed to assess only the direct public budget
outlays for a policy, without regard to a cost's possible mutliplier
effects elsewhere in society.
This criterion assesses the ease with which a policy option can be
effectively administered. Restraining factors are limits on the
administering agency's number and quality of staff, the level of
information needed to effectively carry out a policy program, the
expertise necessary to assess this information, and whether the program
complements or conflicts with the existing operations of the
administering agency.
J_ime_t£ J[mpJ enent
If it is decided to adopt a particular policy, there is generally a
time delay until it actually can be implemented and another time delay
until the benefits are enjoyed. This criterion captures the overall time
delay between the decision to adopt a policy and the point at which It is
effectively operating to counter any unnecessarily restrictive impact on
innovation.
Sugporti ve_oj[ JSCA|_s_Aims_
Some policies to stimulate innovation (e.g., the exemption for low
volume chemicals) act to thwart the main purpose of TSCA, while others
may encourage the innovation of safe chemicals. This criterion captures
the degree to which a policy option is supportive or non-supportive of
the primary TSCA goal - protecting against unreasonable risk of injury.
104.
-------
5-4
Othe Sde Ef f ects1
Policies to stimulate innovation may have other effects unrelated to
innovation per se. Some policies, for example, may give a competitive
edge to large firms, while others may give rise to improved toxicological
testing methods. While judging the net positive or negative nature of
these side effects is somewhat subjective*, the extent to which they
support other publicly mandated purposes may be thought of as positive,
and side effects that act conversely may be thought of as negative.
This criterion is difficult to evaluate independently from the
others because the level of public expenditure as well as other criteria
will also be reflected in political judgments. Nonetheless, policies
that require new legislative authority may be less feasible than those
that do not. Among those requiring Congressional action, some may be
inherently more acceptable. Furthermore, other Executive branch agencies
may be more or less supportive of a policy initiative. This criterion
reflects the views of the project team members regarding the objective
political reality to the extent feasible, rather than their personal
preferences for political action. Because it is felt to be more
subjective and less analytical than the other criteria, political
feasibility was not considered in the overall policy ratings discussed in
section 5.2 and in chapters 6 and 8.
5.2 A Screening Process
A procedure to elicit judgments from the project team was employed
in assessing the policy options. The purposes of the procedure were to
gain insight into the relative merits of each policy, and to stimulate
structured examinations and discussions of the policies. These purposes
are reflected in the qualitative discussion of options in chapter 4 and
in the rankings discussed in chapters 6 and 8.
105.
-------
5-5
5.2.1 A Brief History of the Use of Group Judgment Methods
Group judgment procedures are now in common use as aids to
decision-making in business, government, and other institutions. While
there are many variations in practice, group judgment procedures have
many common elements. Usually, combinations of elicitation forms and
group discussions are used. It is common to keep choices limited to a
small set of questions, options, or alternatives. Often, the procedure
is carried out in several phases. For example, there might be an
open-ended elicitation phase where participants are queried for options,
a closed-ended evaluation phase, and a second round of elicitation and
evaluation to facilitate greater closure on a final set of evaluations.
While some equate group judgment procedures with the Delphi process,
it is actually only one of many processes in use. Dalkey and Helmer
originally used the Delphi process as a means of minimizing conforming
influences by eliminating face-to-face discussions and having respondents
remain anonymous. Others, such as Delbecq, et al. (1975) have employed a
procedure called the Nominal Group Technique as a means of augmenting
discussion approaches. Priest (1978) developed a factor analytic*
variation of group judgment techniques that is suited to complex problems.
5.2.2 The Procedure Followed in this Project
The group judgment procedure used is a factor analytic Magnitude
Estimation Group Judgment technique that employs several criteria for
judging the policy options. The advantage of a factor analytic approach
is that individuals are often better able to judge the merits of
alternatives if their decision process considers every one of the several
factors that determine the merits of each choice.
*In a factor analytic approach to decision-making, the decision rule is
decomposed into several sub-rules, each of which is decided separately.
Then the separate decisions are recombined to yield the overall decision.
106.
-------
5-6
The magnitude estimation approach involves judging the number of
times one alternative is better or worse than another. It is
particularly useful for setting up a rating scale having a wide range.
In this analysis, one policy could be judged as anywhere from 10,000
times worse to 10,000 times better than another, along each criterion.
While it was not expected that participants could make very meaningful
distinctions at the extremes of the scale, these were made available for
expressions of emphasis.
Another characteristic of this method is the use of an anchor
policy, which can be any reasonably straightforward policy. It was
discussed at length so that the participants understood and agreed upon
the criteria as applied to the anchor. All other policies were then
judged in comparison with the anchor policy, which was assigned an
arbitrary rating of 1.0 for each criterion. Thus, a rating of 1.0 for a
policy option on one of the criteria means that it was judged equivalent
to the anchor option on that criterion.
The policy options were discussed by the group before they were
rated. Each participant was then asked to conceptualize the best
possible specific policy under each of the more general policy
categories, and to rate that policy option in comparison with the anchor
option. Using this approach, each policy was placed in the best possible
light by each participant. (The raters wrote notes on their best
conceptualized policies, that were incorporated into the discussions of
the policy options in chapter 4.)
5.2.3 Group Judgment Elicitation Sheet
The eight criteria listed in section 5.1 were used to assess the
policy options in chapter 4. These thirty-two policies were rated by six
project participants using the rating sheet shown in figure 5.1. Every
rater was asked to assess how each policy stood with respect to the
anchor policy on each of the eight criteria. In addition, an overall
rating compared with the anchor was requested. The anchor policy was
107.
-------
5-7 Figure 5-1 Rating Sheet
Rater
Program to be Rated
PROGRAM RATING
Anchor Rating
Program Rating
=
«0
u.
4)
+j
c
3
O
O
0
4-1
^^
4-1
•0-
o
m
Q.
m
o
1
(A
^J
U>
o
o
0)
^J
(0
^
*_
^
Q.
1
tn
*••
4.J
(0
4-1
(A
M»
C
••V
E
<
1
4-1
0)
E
^»
a.
E
O
0)
E
•~
1
(A
E
TJ
'^J
O
w^
H*
U
^
o
a.
a.
3
">
1
4J
0
<*.
-------
5-8
described in a document circulated with the evaluation sheets, which is
shown in figure 5,2. The anchor policy was assigned an arbitrary value
of 1.0 on each criterion. The individual raters then assessed how much
better or worse a policy was on each criterion relative to the anchor
program. For example, if the individual believed that a "Direct cost
subsidy for general new chemical development via a grant mechanism" was
likely to be twice as feasible administratively than the anchor policy,
the rater would enter a 2.0 irv the appropriate column. The criteria
sheet used in the rating process is reproduced in figure 5.3.
5.2.4 Group Judgment Integration Model
To integrate the judgment of the six raters on the eight criteria, a
heuristic, quantitative model was developed. The model cannot be derived
from first principles; instead, it represents a plausible formulation of
the relationships involved. The calculations were done using a simple
computer program.
Because the rating system involved a multiplicative judgment, the
model entailed taking a geometric mean of the ratings on the different
criteria. Furthermore,, since the criteria were judged by the group to be
of varying importance, their ratings were weighted differently in the
model. The judgments of all participants were weighted equally.
First, the rating of each policy on each criterion was calculated by
taking the geometric mean of the ratings for the six raters, as follows:
1/6
* \. r_ . . \. JL. . . \.
Rating for
policy n on
criterion j
»
/Rating by \ /Rating by \ /Rating by
( person 1 on j x I person 2 on) x...x( person 6 on
\criterion jJ \criterion y \criterion j
Next, an overall rating for each policy was calculated by taking the
geometric mean of the ratings on the first seven criteria, weighted by
various factors as follows:
109.
-------
5-9
FIGURE 5.2 Anchor Program Description
ANCHOR PROGRAM: #3 - DIRECT COST SUBSIDY FOR TESTING/COMPLIANCE
COSTS OF NEW CHEMICAL DEVELOPMENT VIA GRANT MECHANISM
Program Description
One best program of this type might have the following features:
1. Grants would be offered to firms submitting an adequate PMN.
2. The size of the grant would be tied to offsetting a substantial frac-
tion of the costs of testing, literature review, and PMN submission.
(Better PMN's would earn bigger grants.)
3. These costs and grant levels would be established by EPA based on a
cost analysis, updated periodically.
4. Typically, grants would range from, say $5000 for a minimum PMN with
only physical properties and a bare literature review, to say, $100,000
for an extensive PMN including chronic toxicity test results on two
animal species.
5. EPA would offer the grants automatically, once the adequacy of the sub-
mission for each grant level were determined. The grant award process
would proceed separately from all test ing/regulatory actions taken by EPA
based on the PMN.
6. Each firm would be limited to annual total grants of, say $100,000,
which would include one or more awards.
7- The EPA would seek an annual appropriation of two million dollars for
grant awards. Supplementary appropriations could be sought to cover
overruns.
8. No other limit on grant availability such as "first come-first served"
or a lottery would be used.
Discussion
1. This program would tend to offset directly the costs of testing and PMN
submission. Some, essentially indeterminate, fraction of these costs
are undue, arising from transition effects and possible unfair and in-
efficient costs for small firms.
2. The aggregate grant limit tends to favor small firms, while the taxable
feature tends to favor new entrants and marginal firms (little or no
tax) while recovering some portion of the grant from profitable firms
that need it less.
3. By giving larger grants for more complete PMN's, the program tends to
encourage firms to do more testing and analysis for new chemicals.
110.
-------
5-10
Figure 5.3 CRITERIA FOR INITIAL ASSESSMENT
OF GENERAL PROGRAMS
(each criterion, ceterls paribus)
1. Capacity to Countervail - the inherent capacity of the programs to offset
the undue innovation damping effects of TSCA regulations once it is put
in place. This rating should reflect the capacity of the programs to
mitigate these effects in total, not only for one specific class of chemi-
cals or one type of innovation barrier.
2. Industry Costs - relative to the base program, are the private Industry
costs of the program higher (less than 1) or lower (greater than 1).
3. Public (Government) Program Costs - relative to the base program, are the
budget costs of the program higher (less than 1) or lower (greater than 1).
k. Public Administrative Feasibility - relative to the base program, how
feasible is it to make the program work (e.g., obtain and process the in-
formation necessary to implement the program, obtain the necessary size
and mix of staff capabilities, etc.).
5. Time to Implement - relative to the base program, to what degree does the
time required to make the program operational make the program better
(greater than 1) or worse (less than 1).
6. Supportive of TSCA Aims - how well does the program support the main pur-
pose of the TSCAct.
7. Other "Side" Effects - are the net other effects of the program (e.g.,
market changes; distributional inequities; effects on consumers, workers;
and the public) more or less beneficial than those of the base program.
8. Political Feasibility - how likely is it that new legislation can be ob-
tained if needed, or that the program will be acceptable to the legislative
branch (authorizing and appropriation committies, etc.) and the Administra-
tion (e.g., other executive departments, White House, OMB, CEQ, CEA, RARG),
or that the program will be acceptable to interest groups.
9. Overall Rating of Program - relative to base program, how "good" is
this program.
111.
-------
5-11
Overall
rating for
policy n
/Mean rating V-° /Mean RatingX0-1 /Mean rating ^':
[ on Capacity \x /on Private \x...x( on Other \
I to Countervail) I Costs for I I Side Effects J
\for policy n/ N^policy n / \for policy n/
The group weighting factor for each criterion was taken to be the
geometric mean of all of the judgments of that factor by the six raters.
The weighting factors are:
Criterion Weight
Capacity to Countervail 1.0
Private Cost 0.1
Public Cost 0.9
Administrative Feasibility 1.0
Time to Implement 0.28
Support TSCA Aims 0.75
Other Side Effects 0.51
The group agreed that the total cost and the capacity to countervail
of a policy option should be weighted equally. Thus, the sum of the
weights of private and public costs should equal 1.0. For the anchor
policy, however, private costs are small compared with public costs.
Consequently, the relative public costs of other policies, which account
for more of the total cost on an absolute basis, are weighted more
heavily. This explains why public and private costs, which the group
felt should be weighted equally in principle, were ultimately weighted
0.9 and 0.1, respectively, in the model.
The model was used to calculate an overall rating for each option
based on only the first seven criteria. It should be noted that the
model does not take into consideration ratings on criterion 8, political
feasibility, or criterion 9, initial policy rating. Political
feasibility was always treated as an independent judgment that did not
influence the overall rating of the options. The initial policy rating
represents an overall judgment of the rating of a policy compared with
the anchor policy, made without reference to the individial criteria.
Criteria 8 and 9 are considered further in chapters 6 and 8.
112.
-------
5-12
REFERENCES FOR CHAPTER 5
Delbecq, Andre L., Andrew H. Van de Van and David H. Gustafson, 1975,
Group Techniques for Program Planning, Scott, Foresman and Co., Glenview,
III.
W. Curtiss Priest, 1978, "Interactive Decision-Making Using Factors,"
Humanic Systems Co., Lexington, MA.
113.
-------
5-1
6. AN ASSESSMENT CF THE POLICY ALTERNATIVES
The results of the assessment cr the 32 policy options that are
described in chapter 4 are presenter in this chapter. The discussion is
strongly based on an examination of the ratings and rank orderings of the
various criteria developed in chapter 5. Also included are a qualitative
discussion of these criteria and several checks of the consistency of the
method.
The quantitative ratings and rankings presented here are only
relative, therefore, small differences in rank or rating are not
particularly meaningful.
6.1 Results of the Magnitude Estimation Procedure
6.1.1 Overall Rating and Rank Orders
The 32 policy alternatives are listed in table 6.1 in descending
order of overall rating on the policy assessment model. These ratings
represent the geometric means of the individual ratings assigned by each
of the six members of the project tc-am,* using the seven criterion model
developed in chapter 5.**
The ratings range from a Iv'gh of 1.212 for the top ranking policy
(R, EPA dissemination of chemical information) to a low of 0.175 for the
lowest ranking policy (CC, Regulatory exemptions for low risk
chemicals). Figure 6.1 shows that the overall policy RATINGS vary
regularly with the rank of the policies. However, there is a sharp drop
six were Andrews, Frenkel, Beaton, Hill, Mitchell, and Priest.
"""Political feasibility" and "Initial rating" were not included in
these scores.
114.
-------
Table 6.1 Results of the Policy Assessment
Rank
Order
1
2
3
4
5
6
7
8
9
10
11
12
13
Overall
Rating
1.212
1.114
1.073
1.061
1.003
0.902
0.817
0.612
0.562
0.531
0.463
0.463
0.453
Policy
Number
R
EE
V
C
DD
W
D
F
X
S
Y
0
FF
Pol icy Name
EPA dissemination of chemical information - test results
Generic PMN for classes of new chemicals
Government support to develop new, better test methods
Direct cost subsidy for testing/compliance costs of new
via grant mechanism
"Fast track" PMN's for safe and/or major innovations
Government support for education and training programs
Direct cost subsidy for testing/compliance costs of new
via loan mechanism (or loan guarantees)
and/or labeling
chemical development
chemical development
Indirect cost subsidy for testing and compliance costs via tax mechanism
Actions against existing substitutes for new chemicals
Chemical technology extension service, Including dissemination of information
on test and compliance needs
Fixing time periods for regulatory actions
Establish government testing for TSCA requirements
Improve EPA staff capability to assess impact of regulat
ory actions on
innovation
14 0.445 AA Regulatory exemptions for low volume, new chemicals
I
ro
-------
Table 6.1 continued
Rank
Order
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
Overall
Rating
O.VtO
0.362
• 0.354
0.353
0.332
0.316
0.288
0.287
0.268
0.263
0.239
0.225
0.221
0.219
0.208
Policy
Number
BB
P
B
Q
K
T
1
U
N
J
A
L
M
Z
E
Po 1 1 cy Name
Regulatory exemptions for small firms
Sharing of test data with reimbursement
Direct cost subsidy for general new chemical development via loan mechanism
(or loan guarantee)
Facilitate private sector joint R&D or joint testing
Increased capital availability for new chemical development via tax changes
or via SEC rules
Antitrust action to favor new, small firms in the chemical industry
Decreased taxes on sales of new chemicals
Tax adjustments to favor small firms or new entrants in the chemical industry
Reduce risk from products liability actions by establ ishlnp 1 imits on liability
Increased capital availability for new chemical development via government
supported venture capital company
Direct cost subsidy for general new chemical development via grant mechanisms
Reduce risk through government financed insurance for regulatory losses
Reduce risk through government procurement of new chemicals
Post-market surveillance of PHN's
Indirect cost subsidy for chemical innovation generally via tax mechanism
I
CO
-------
Table 6,1 continued
Rank
Order
30
31
32
Overall
Rating
0.200
0.188
0.175
Policy
Number
H
G
CC
Policy Name
Strengthened trade secret protection by limitations on
release information
Increased patent life for new chemicals
Regulatory exemptions for 'Mow risk" chemicals
EPA authority to
-------
6-5
1.2
1.0
0.9
o
p 0.8
cr ~
> < 0.7
O (T
2 ^ °-6
£*- 0.5
(E ^*
LU
O 0.4
0.3
0.2
O.I
0
_O
°0
— O
— O
OQO
oo
oo
10 20 30
OVERALL POLICY RANK
(7 CRITERIA)
40
FIGURE 6.1 TREND OF OVERALL SCORE WITH RANK
ORDER FOR 32 POLICIES
118.
-------
6-6
in ratings around the seventh ranked policy, so that policies R, EE, V,
C, DO, W and D (see table 6.1 for descriptions) appear to be
substantially superior to the others overall.
Very generally, the ratings can be interpreted as indicative of a
measure of the ratio of effectiveness to costs. However, the overall
ratings in table 6.1 have no quantitative significance as benefit/cost or
effectiveness/cost ratios. Rather, they represent the overall rating on
seven weighted criteria as compared with the overall rating for the
anchor policy option C, which ranks number fourth.*
The relative values of the ratings have limited quantitative
significance; for example, a policy rated 1.0 is not necessarily exactly
twice as effective as one rated 0.5. The exactness of the ratings is
uncertain for a number of reasons.
*
The top ranked policy options in table 6.1 represent a favorable
balance of effectiveness, costs, and effects on TSCA and on various
interested parties. However, the individual high ranking policies are
not necessarily particularly capable of offsetting the unnecessarily
restrictive impacts of TSCA on innovation. That may require adopting a
somewhat lower ranking policy that is more effective, or adopting some
combination of policies.** This subject is explored further in section
6.2 and in chapter 8.
*Policy option C has an overall rating of 1.061. Its rating is greater
than 1.000 because a few project members believed that this type of
policy could be slightly improved over the detailed anchor policy and,
could therefore be assigned a higher rating.
**This situation is analogous to that faced by a stock market investor
who wishes to invest in the stock with the highest rate of return, but
who may have to invest in lower return stocks as well if the number of
shares of high return stock is limited.
119,
-------
6-7
in ratings around the seventh ranked policy, so that policies R, EE, V,
C, DO, W and D (see table 6.1 for descriptions) appear to be
substantially superior to the others overall.
Very generally, the ratings can be interpreted as indicative of a
measure of the ratio of effectiveness to costs. However, the overall
ratings in table 6.1 have no quantitative significance as benefit/cost or
effectiveness/cost ratios. Rather, they represent the overall rating on
seven weighted criteria as compared with the overall rating for the
anchor policy option C, which ranks number fourth.*
The relative values of the ratings have limited quantitative
significance; for example, a policy rated 1.0 is not necessarily exactly
twice as effective as one rated 0.5. The exactness of the ratings is
uncertain for a number of reasons.
The top ranked policy options in table 6.1 represent a favorable
balance of effectiveness, costs, and effects on TSCA and on various
interested parties. However, the individual high ranking policies are
not necessarily particularly capable of offsetting the unnecessarily
restrictive impacts of TSCA on innovation. That may require adopting a
somewhat lower ranking policy that is more effective, or adopting some
combination of policies.** This subject is explored further in section
6.2 and in chapter 8.
Policy option C has an overall rating of 1.061. Its rating is greater
than 1,000 because a few project members believed that this type of
policy could be slightly improved over the detailed anchor policy and,
could therefore be assigned a higher rating.
**This situation is analogous to that faced by a stock market investor
who wishes to invest in the stock with the highest rate of return, but
who may have to invest in lower return stocks as well if the number of
shares of high return stock is limited.
120.
-------
TABLE 6.2 Policy Ratings on Each Criterion In Rank Order
Rank
Order
1
2
3
4
5
S
7
8
9
10
II
12
13
14
\t>
17
IB
19
20-
21
22
23
2k
25
28
27
28
29
30
31
32
Capacity to
Counterval 1
Rating Policy
3.550 AA
2 . 392 BB
1 . 308 EE
1 . 260 C
1.225 0
.843 D
.619 F
.531 L
.442 W
.442 X
.426 B
.398 1
.385 K
.368 H
.3D DD
.305 V
.290 U
.269 T
.265 S
.2fa9 A .
.242 H
.215 V
.215 CC
.198 J
.171 E
.107 FF
.076 P
.061 Q
.054 G
.050 z
.0*1 5 R
.041 N
Private
Costs
Rating Policy
4.347 E
4.263 K
4.107 N
2.667 U
2.587 V
2.000 V
.979 Q
.979 CC
.919 H
.886 S
.743 C
.648 AA
. 587 00
.468 BB
.454 EE
. 260 FF
.214 J
.193 U
.159 1
-132 U
.013 B
.000 C
.849 H
.794 R
.751 A
.739 F
.678 o
.586 D
.398 2
.231 P
.165 T
.068 X
Public
Costs
Rating Policy
8.849 OD
7.023 P
6.257 V
6.027 N
5.635 EE
3.942 BB
3.798 R
3.047 Q
2.587 FF
2.418 AA
2.154 H
1.452 V
1.260 I
1.145 CC
1.103 0
.992 F
.901 C
.765 G
.765 W
.681 x
.548 B
.541 S
.521 T
.464 J
.414 L
.251 K
.180 0
.171 A
.163 U
.121 1
.109 H
.069 E
Administrative
Feasibility
Rating Policy
2.418 R
1.800 N
1.680 V
1.648 E
.875 C
.79* W
.690 K
.586 0
.585 Q
.557 F
.557 1
1541 EE
.521 Z
.521 BB
.464 U
.398 G
. 394 DD
.380 H
.368 AA
.101 S
.292 V
.269 FF
.242 T
.239 B
.215 L
.215 0
.178 A
.165 J
.137 X
.135 N
.089 P
.073 CC
Time to
Implement
Rating Policy
.963 C
.950 A
.871 EE
.818 D
.765 R
.630 DO
.607 BB
.539 B
.488 F
.478 1
.464 H
.464 AA
.362 v
.342 V
. }29 U
.326 S
.301 x
.298 L
.298 W
.290 K
.259 E
.242 CC
.215 N
.215 I
.205 J
.171 N
.171 FF
.152 Q
.131 0
.116 P
.061 G
.654 T
Support
TSCA Goals
Rating Policy
5.848 R
5.020 X
2.513 V
1.956 W
1 . J08 C
1.260 DD
1.244 0
1.000 Q
.989 S
,849 D
.729 P
.649 H
.531 F
.468 FF
.426 EE
.368 T
.342 J
.217 B
.208 U
.17) 1
.155 A
.131 K
.116 G
.112 Z
.092 Y
.071 E
.044 L
.041 CC
.015 N
.01? AA
.069 H
.008 BB
Other
"Side Effects"
Rating Policy
3.732 R
2.980 w
2.655 V
2.466 S
1 . 308 FF
1.157 C
1.122 DD
.872 D
.858 x
ftefl EE
.737 T
.707 0
.497 A
.461 P
.410 F
.338 U
.326 B
.251 K
.251 V
.247 J
.242 1
.135 M
.107 L
.103 G
.100 CC
.089 E
.089 AA
.079 Z
.056 BB
.046 N
0.23 H
.022 Q
Effectiveness-1
Rating Policy
1 . 306 AA
1.246 BB
1.103 C
. 708 EE
.494 D
.361 V
.351 W
.345 F
.282 E
,266 K
.263 0
.222 1
.135 U
.123 DD
.114 L
.109 R
.102 B
.092 H
.089 Y
,080 S
.074 N
.065 T
.061 X
.050 H
.044 A
.036 Q
.033 J
.029 FF
.026 Z
.021 G
.016 CC
.007 P
Ot
CO
'Effectiveness - (Capacity to Countervail) X (Administrative Feasibility)
-------
6-9
ability to regulate unreasonable risks to health and the environment.
The six with the lowest ratings (L, CC, N, AA, H and BB) can be seen as
actually weakening the achievement of TSCA's primary goals, the top ten
or so might significantly strengthen it, and the remainder would have
only marginal effects.
6.1.3 A Consistency Check on the Overall Rankings
To test the validity of the ranking process and the assessment
model, each of the six members of the project staff was asked to assign
an initial overall rating to each of the policies in comparison with the
anchor policy, independent of the ratings calculated on each of the
separate criteria. The correlation between the rank orders of the 32
policies on these initial overall ratings and on the calculated overall
ratings based on the seven criteria is shown in figure 6.2. (The initial
overall ratings are the geometric means of the ratings of each staff
member.)
If the initial and calculated overall rank orders had been
identical, all the points in figure 6.2 would have fallen on the 45°
line. As the data actually fall, they show a remarkable consistency that
suggests that the assessment technique used here is valid.
6.1.4 Political Feasibility of the Policies
The project team members were each asked to rate the political
feasibility of the policy options based on their judgments of the
relative acceptability of the proposals to interested parties, to the
various affected Executive agencies, and to Congress. These ratings are
shown in table 6.3. The relationship of the rank order of the policies
on the political feasibility criterion to their overall rankings is shown
in figure 6.
Several points need to be made about table 6.3 and figure 6.3. The
ratings on "political feasibility," which are the geometric means for six
122.
-------
6-10
4O
I 30
UJ
0 H 20
o c
UJ °
< K
«1 *^
D
O
-J
< 10
o
ft / 45°LINE
_ ° Xo
o Ox
/ o
o /
^ o /
0 y
O
o /
Q *
~~" x^3
o /
/ o
o /
0
^ o
y 0
O '
^^ O '
/* °
0 X
0/'°0
X 0
X O
,0 1 1 1 1
10
20
30
ESTIMATED OVERALL RANK
(CRITERION ^9 ONLY)
40
FIGURE 6.2
CORRELATION BETWEEN TWO MEASURES
OF OVERALL POLICY RANKING
123.
-------
2.873
2.493
1.979
1.886
1.849
1.458
1.299
1.238
1.103
1.058
D
W
V
EE
R
00
Z
C
AA
CC
6-11
TABLE 6.3
Policy Ratings on the Political
Feasibility Criterion
Rank Rating Policy
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
.918
.780
.645
.607
.442
.394
.368
.358
351
.301
.298
.265
.242
.147
.132
.131
.112
.100
.094
.075
.056
.013
B6
U
Y
Q
K
S
B
FF
0
M
I
L
P
N
F
X
H
J
G
E
A
T
124.
-------
6-12
30
3 20
UJ
u.
g 'o
CL
O
/ O
X O
0 X
V
X O
X O
/
>
10 20 30
OVERALL RANK
(7 CRITERIA)
40
FIGURE 6.3 RELATIONSHIP OF RANK ORDERS OF POLICIES
ON POLITICAL FEASIBILITY AND
OVERALL RATINGS
125.
-------
6-13
project team members, do not represent the personal political views of
the project members; rather, they are intended to reflect the best
judgments by the team members of the objective political situation.
The ratings on political feasibility are separate judgments from
those on which the overall ratings based on the seven criteria were
calculated. Thus, the strong correlation shown in figure 6.3 can be
viewed as fortuitous. The project team made no assessment of whether any
particular rating or rank on "political feasibility" would render a
proposal politically unacceptable.
Most importantly, figure 6.3 shows that the first seven ranking
policies with respect to their overall ratings are among the first eight
ranking on political feasibility. Only policy Z (post-market
surveillance of PMN's), which ranks 28th overall and seventh politically,
is not in this group. This lends credence to the top seven ranking
policies as potentially good alternatives for action.
6.2 Discussion of the Top Seven Ranking Policies
The previous discussion has highlighted the seven highest ranking
policies on an overall ratings basis as well as on the basis of political
feasibility. This section examines those seven policies in greater depth.
Table 6.4 shows how each of the seven top overall policies were
ranked on the seven analytic criteria, on the "effectiveness" measure
discussed in section 6.1.2, and on political feasibility. Several points
about this table are noteworthy.
Policies that rank highest overall do so by virtue of their weighted
ratings on several criteria, so it is not necessary for them to rank high
on all criteria; table 6.4 shows that they do not.
126.
-------
TABLE 6.*»
Ranks on Individual Criteria For
Seven Policies Ranked Highest Overall
Overall Policy Rank
>olicy Identifier
Short Pol Icy Name
Criterion
Capacity to Countervail
Private Costs
Public Costs
Administrative Feasibility
Time to Implement
Support TSCA Aims
Other Side Effects
Effectiveness
Political Feasibility
1
R
Information
Dissemination
31
2k
7
1
5
1
1
16
5
2
EE
Generic
PMN
3
15
5
12
3
15
10
k
k
3
V
Better
Test
Methods
Support
22
5
12
3
13
3
3
6
3
<*
C
Grants
for
Testing
Costs
k
22
17
5
1
5
5
3
8
S
DD
PMN
Fast
Track
15
13
1
17
6
6
7
1
-------
6-15
The top seven policy options include those ranked 3, 4, 5, 6, 7, 14
and 16 on "effectiveness," which is the best measure of their ability to
offset any unnecessarily restrictive impacts. The two most effective
policies, AA and BB (exemptions for low volume new chemicals and for
small firms, see table 6.2) rank lower overall (14th and 15th, see table
6.1), largely because they are ranked low on their ability to support
TSCA's aims and to create side effects.
All of the top seven policy options rank relatively low on the
"private cost" criterion except for policy W, education and training.
However, this should not be cause for their rejection, because only the
policies ranked 29th through 32nd on private costs are seen to cause any
significant private costs.
Policies V, D, C, and W can be seen as having significant "public
costs," with C and 0 being grant and loan programs that could be
reasonably expensive. (See chapters 7 and 8 for a further discussion of
public costs.)
Most of the top seven ranking policies rank highly on "time to
implement." The exception is policy W, education and training, which
would obviously require several years before results could be expected in
terms of reduced costs of testing and more effective innovation.
A11 of the top seven policies, but one, rank in the top ten on
"supportive of TSCA'S aims." The exception is policy EE, (generic PMN),
which can be seen as weakening the ability of EPA to assess all new
chemicals adequately. Similarly, policy EE is the lowest ranking of the
top seven ranking policies on "other side effects," although here it only
slips to tenth place. This marginally low rank results from the somewhat
enhanced potential harm to health and the environment resulting from the
generic PMN. To repeat, the top seven policies overall are among the top
eight on "political feasibility."
128.
-------
6-16
The top seven policies Include policies designed to reduce the costs
of new chemical development directly (C and D), to provide more
information (R), to improve the technology needed for compliance (V and
W),and to modify the administrative procedures for managing PMN's (DO and
EE). As a group, the seven top ranking policies nicely cover several of
the major categories of policies discussed in chapter 4, and thus may, as
a group, offset the unnecessarily restrictive effects of TSCA on
innovation at both the firm and the industry levels. (See chapter 8 for
an elaboration of this point.)
6.3 Discussion of the Seven Lower Ranking Policies
As figure 6.1 shows, there is no special group of lower ranking
policies upon which to focus. Furthermore, figure 6.3 does not show a
good correlation between overall rank and political feasibility rank for
the lower ranking policies. Therefore, it has been arbitrarily decided
only to comment briefly on the seven lowest ranking policies (reference
table 6.1).
The ratings of the seven lowest ranking policies on the various
criteria are shown in table 6.5. (Recall that the criteria were given
unequal weights in computing the overall score; see chapter 5.) This
table shows that the lowest ranking policies tend to score low across the
board, with one exception.
The exceptional policy is the 29th ranked policy E, (indirect cost
subsidy for chemical innovation via a tax mechanism), which ranks first
on "private costs", fourth on "administrative feasibility, and ninth on
"effectiveness." These rankings might suggest that policy E would be a
good candidate for adoption. On the other hand, it ranks as the most
expensive, 32nd, on "public costs", 25th on "capacity to countervail",
and 26th on both "supportive of TSCA's aims" and "other side effects."
Furthermore, it was ranked quite independently as 30th on "political
feasibility." Thus, policy E can be viewed overall as a very expensive,
129.
-------
TABLE 6.5
Ranks on Individual Criteria For
Seven Policies Ranked Lowest Overall
U)
o
Overall Policy Rank
Policy Identifier
Short Policy Name
Criterion
Capacity to Countervail
Private Costs
Pulblc Costs
Administrative Feasibility
Time to Implement
Support TSCA Alms
Other Side Effects
Effectiveness
Political Feasibility
32
CC
Exempt
Low Risk
Chemicals
23
a
Id
32
22
28
25
31
10
31
G
Increase
Patent
Life
29
II
18
16
31
23
2
I
-------
6-18
unpopular policy that tends not to support the alms of TSCA and Is of
intermediate effectiveness. Instead of attacking the unnecessarily
restrictive impacts of TSCA on innovation specifically, it represents a
gross subsidy to'all chemical development. Finally, there are substitute
policies that are similar but more carefully targeted, such as C, D, and
F, which would provide grants, loans, or tax breaks for TSCA-related
costs of new chemical development only, and which rank much higher
overall.
6.4 Relationship of Policy Ranking to General Policy Areas
Chapter 4 categorized the 32 policy options into ten categories, as
noted in table 4.1. This section reviews the relationship between the
policy rankings and these categories.
Redu£irig_the_C£Sjt £
These policies use grants, loans, or tax incentives to reduce the
costs of new chemical development, in general (policies A, 8, and E), or
of the testing and compliance costs of new chemical development
specifically (policies C, 0, and F). The first group rank rather low
overall (25, 17, and 29 respectively), and the second group rank
considerably higher (4, 7, and 8 respectively). One or more of the second
group are high priority possibilities for adoption.
These three policies use extended patent life (G), strengthened
trade secret protection (H), and decreased taxes on sales of new
chemicals (I) to increase the financial rewards for marketing new
chemicals. However, they rank low overall as a group (31, 30, and 21
respectively). This occurs for G because the increased returns are
expected so far into the uncertain future. Strengthening trade secret
protection (H) ranks low on a number of counts: effectiveness,
131.
-------
6-19
supportive of TSCA's aims, and other side effects. Policy I (reduced
taxes on new chemical sales) ranks near the middle on most criteria, but
30th in terms of public costs, leading to a low overall rating.
Policies J (government supported venture capital company) and K
(increased capital through tax or SEC rules changes) rank 24th and 19th
respectively. Policy J ranks low on effectiveness, public costs, and
time to implement, making it generally unattractive. While policy K
ranks next to the best on private costs and reasonably high on
effectiveness, it ranks low on public costs and somewhat low on time to
implement, and near the bottom of the middle third on other criteria.
Thus, K is only of passing interest.
ReduetheCnmeCia i skAsscate vvi thNewChemias
Policies L, M and N rank at 26, 27, and 23 overall, therefore, are
unlikely alternatives for adoption. Policy L, reducing risk through
government financed insurance for regulatory losses, is near the middle
on effectiveness, private costs, and time to implement. However, it
ranks low on public costs and is seen as low on its ability to support
TSCA's goals and might create undesirable side effects. Thus, it ranks
low overall.
Policy M, reducing commercial risk through government procurement of
new chemicals, ranks low on most criteria except for a rank of 12th on
supporting TSCA's goals. This ranking probably reflects the thinking of
some raters that this policy would direct government procurement toward
safer substitutes. However, in most cases, government purchases of this
type have been found to be inadequate in changing the civilian sector.
•
Policy N, reducing risk due from products liability actions by
legislated limits on liability, ranks 23rd overall, despite its high
scores on private and public costs and administrative feasibility. (It
132.
-------
6-20
would rank very low on public costs as well if this program involved
government in covering losses in excess of the liability limitation.)
This occurs due to its low score on capacity to countervail (32nd), and
because it is seen as undermining the purposes of TSCA as reflected in
low scores on supporting TSCA's aims and on other side effects, since it
would limit the ability of injured parties to recover damages.
fte
-------
6-21
Policies R (dissemination of information) and S (chemical technology
extension service) rank 1st and 10th overall. Thus, policy R is a high
priority alternative for inclusion despite its low rank on capacity to
countervail. Policy S ranks near the middle on most criteria but near
the top on other side effects. The latter reflects perceptions about the
commercial and health-related values of such an activity that are not
related to TSCA per se.
Policy T, ranked 20th, and policy U, ranked 22nd, would use the
antitrust laws (T) or tax laws (U) to favor new and/or small firms in the
chemical industry. Policy T ranks very low on private costs and time to
implement and low on public costs and administrative feasibility. Since
it was seen as having only an average capacity to countervail, it ranked
relatively low overall as a means of offsetting the unnecessarily
restrictive impacts of TSCA on innovation. Policy U (use of the tax
system to favor small firms or new entrants) ranks low on public costs,
high on private costs, and near the middle on all other criteria; so it
is ranked relatively low overall.
the
Policies V and W, which ranked third and fourth overall, would
increase the supply of factors needed to improve TSCA compliance
technologies. As high priority policies in the top seven, they were
discussed in some detail in section 6.2.
These two structural policies are evaluated here only on the basis of
their abilities to offset the unnecessarily restrictive impacts of TSCA
on innovation. No attempt has been made to assess whether these policies
might stimulate chemical innovation in general, or whether they would be
cost-effective or desirable for other purposes.
134.
-------
6-22
This category includes a variety of administrative actions that
could be taken by EPA, or with the cooperation of Congress through
changes in TSCA, which would be expected to smooth the way for new
chemical devlopment and marketing. In terms of overall ranking, these
fall into three groups: EE and DO; X, Y, FF, AA and BB; and Z and CC.
EE and 00, which ranked second and fifth overall, are the
"generic PMN" and the "fast track." These are in the top priority list
of policies, and at least the generic PMN is currently under study by EPA.
X, Y, FF, AA, and BB rank 9, 11, 13, 14, and 15 respectively.
Thus, they rank just behind the top priority set of seven. Policy X,
taking action against substitutes when a PMN is submitted, ranks high on
some criteria but at or near the bottom on private costs and on
administrative feasibility. Policy Y, fixing time periods for regulatory
action, ranks high on both private and public costs, but at or below the
middle on other criteria. Policy FF, improving EPA staff capability, is
seen as relatively difficult to carry out (administrative feasibility and
time to implement) and to have a low capacity to countervail, despite
average or above ratings elsewhere. Policies AA and BB, which are
regulatory exemptions for low volume, new chemicals or for small firms,
are ranked highest on effectiveness and above average on private and
public costs and time to implement. However, their tendency to counter
the goals of TSCA directly and indirectly brings their overall rankings
down to 14th and 15th.
Policies Z and CC are ranked low at 28th and 32nd respectively, but
for different reasons. Policy Z, post-market surveillance of PMN's,
ranks low on capacity to countervail, private costs, time to implement,
support for TSCA's goals, and other side effects. Policy CC, regulatory
exemptions for low risk chemicals, ranks much higher than Z on private
costs, but much lower on administrative feasibility, and the same as Z on
other rankings.
135.
-------
6-23
6.5 Conclusions Regarding the Assessment of Policy Alternatives
The use of the magnitude estimation scaling technique has yielded
sets of ratings and rank orders for the 32 policy options under
consideration that appear to be both internally consistent and consistent
with the intuitive judgments of the team members regarding the various
policies and their attributes. It is, therefore, believed that the
assessments presented here should be of considerable assistance to EPA
and to other interested parties in addressing the problem of the
unnecessarily restrictive impacts of TSCA on technological innovation.
In section 6.2, a set of seven high ranking policies was identified
and discussed. Following a discussion of financing methods for the
various options in chapter 7, chapter 8 develops and analyzes a
comprehensive policy opportunity based on a selection from these seven
top policies.
136.
-------
7. FINANCING THE POLICY OPTIONS
7.1 Available Financing Modes
Any one of several different financing modes could reasonably be
employed to underwrite the cost of most of the policy options discussed
in chapters 4 and 6. The question of which modes are preferable is a
qualitatively different matter from the question of which policy option
is preferable on the basis of ability to achieve TSCA's goals. Different
criteria for judging merit apply in the two situations: for example, the
level of private cost is an important determinant of a program's merit,
but is largely irrelevant as a public financing concern. In addition,
beyond the differences in the ways they are analyzed, there are different
political issues as well as a different decision process associated with
the selection of a financing option. For these reasons, policy and
financing options are here considered as independently as possible,
recognizing, however, that at some point they must be dealt with jointly.
The first, most basic financing issue to be addressed is whether the
policy in question requires new public financing. Some of the policy
options require only minimal public expenditures, and in certain cases,
none at all. This may be either because the policies are'inherently
inexpensive or because the private sector, rather than the government,
bears their costs. In many instances, a policy's objective can be
achieved entirely within existing agency budgets, simply by altering the
agency's mandate or operating procedures. Although such policies are not
likely to be without cost to society as a whole, no significant new
public financing requirements are associated with them. Policies of this
type include:
G. Increased patent life for new chemicals
H. Strengthened trade secret protection by limitations on EPA
authority to release information
P. Sharing of test data with reimbursement
137.
-------
7-2
Q. Facilitating private sector joint R&D or joint testng
T. Antitrust action to favor new, small firms in the chemical
industry
X. Actions against existing substitutes for new chemicals
Y. Fixing time periods for regulatory actions
Z. Post-market surveillance of PMN's
AA. Regulatory exemptions for low volume, new chemicals
BB. Regulatory exemptions for small firms
CC. Regulatory exemptions for "low risk" chemicals
DO. "Fast track" PMN's for safe and/or major innovations
EE. Generic PMN's for classes of new chemicals
Although none of the above policies requires new expenditures, some
do require a realignment of personnel and emphasis within the agencies.
For example, post-market surveillance would be a new EPA function. These
new emphases may represent an opportunity cost to the extent that they
draw resources away from valuable existing programs.
When public financing is required, two basic alternatives are
available: financing via outlays from the federal budget, and financing
via a variety of "off-budget" outlays. The principal options within
these alternatives include:
o budget outlays
- reallocation of EPA discretionary funds within the agency
- new Congressionally approved EPA programs
- new authority and/or outlays for programs outside EPA
o off-budget outlays
- "tax expenditures," utilizing the Internal Revenue Code
- new taxes
- establishment of new off-budget financial entities
- government assumption of contingent liabilities
The following paragraphs discuss the virtues and drawbacks of these
financing options and consider how they may be applied to the policy
alternatives presented in chapter 4. While no attempt has been made to
138.
-------
TABLE 7.1
'Matching Policy and Financing Options
u>
vo
Policy
A. Grant
Subsidy
B. Loan Subsidy
Loan Guarantee
C. Testing Grant
D. Testing Loan
Testing Loan Guarantee
E. General Tax Subsidy
F. Testing Tax Subsidy
G. Patent
H. Trade Secret
1 . Decreased Taxes
J. Venture Capital
Company
K. Increased Capital
L. Insurance
H. Procurement
N. Liability
Halts
0. Government
Testing
P. Sharing Test
Data
Budgetary Outlays
Real local Ion of
Discretionary
Funds
•
New EPA
Programs
X
X
X
X
X
X
X
New Non-EPA
Authority
X
X
X
X
X
X
X
X
Off-Budget Outlays
Tax
Expenditures
X
X
X
X
New
Taxes
X
X
X
X
X
X
X
X
New Financial
Entitles
X
X
X
Contingent
Liabilities
X
X
X
X
-------
TABLE 7.1
centInued
Policy
Q. Joint IUD
ft. Information
OlsseMl nation
S. Technology
Extension Service
T. Antitrust
U. Tax
Adjustments
V. Better Tests
W. Education
X. Action Against
Substitutes
V. Fixed Time
Periods
2. Post-Harket
Surve 1 1 lance
AA. Low Volume
Exemption
BB. Small Firm
Exemption
CC. Low Risk
Exemption
DO. Fast-Track
EE. Generic PHN
FF. Improve EPA
Budaetarv Out lavs
Real location of '
Discretionary
Funds
X
.X
X
X
New EPA
Programs
X
X
X
X
X
New Hem -EPA
Authority
X
X
X
Off-Budaet Out lavs
Tax
Expenditures
X
New
Taxes
X
X
X
New Financial
Entitles
Contingent
Liabilities
-------
7-5
construct a formal set of criteria by which to judge each financing
option, the following objectives are believed to be generally desirable
for any financing scheme. Therefore they have been used as guidelines
for this assessment:
o equity
o ease of administration
o revenue raising ability
o predictability of cost
o accountability for expenditures
7.2 Matching Policy and Financing Options
A summary of how policy options and financing options have been
matched is shown in table 7.1. The entries in this table are discussed
throughout this section.
7.2.1 Budget Outlays
On-budget financing is generally regarded as superior to off-budget
financing because it is both more predictable and more reviewable. In
addition, it fits well within existing budgetary and institutional
structures. In contrast, off-budget financing often requires the
establishment of new institutions. For the 32 policy options suggested
in this report, the arguments for on-budget financing are especially
persuasive when the public costs of new programs are small, and when they
require little in the way of new institutional structures.
The option to reallocate internal EPA funds without the need to
consult the Congress is available for only a few of the policies and can
only be done when EPA has a large amount of funds over which it exercises
independent discretion, such as for R&D support. Therefore, this
financing option would be feasible only for policies such as the
development of new and improved test procedures (V) or the dissemination
of chemical information (R). It has the clear virtue of being quick and
simple to implement. Because it is entirely internal to the agency,
141.
-------
7-6
however, it can be criticized on the grounds that it allows programs to
proceed without public debate. Real location of funds by EPA might also
hamper other valuable agency activities whose funding is reduced.
At times, Congress may insert new line items in an agency's budget
without changing its legal authority. In other circumstances, new agency
initiatives will require new legislative authority. The public financing
issues in these two contexts are similar. New budget line-items,
however, usually tend to be of shorter duration than authority for new
programs. In addition, for new line-items only the legislative
appropriations process needs to be considered, as legislative authority
is not an issue. Both of these avenues provide for a measure of public
debate about the new program being considered, both keep new expenditures
within the realm of EPA programs and both keep the expenditures visible
as part of the federal budget. Policies that could be financed as new
line items in the EPA budget or as newly authorized EPA programs include:
A. Direct cost subsidy for new chemical development, in general,
via a grant mechanism
B. Direct cost subsidy for new chemical development, in general,
via a loan mechanism (or loan guarantee)
C. Direct cost subsidy for testing/compliance of new chemical
development via a grant mechanism
D. Direct cost subsidy for testing/compliance costs of new
chemical development via a loan mechanism (or loan guarantee)
0. Establish government testing for TSCA requirements
R. EPA dissemination of chemical information - test results and/or
labeling
V. Government support to develop new, better test methods
W. Government support for education and training programs
FF. Improve EPA staff capability to assess the impacts of
regulatory actions on innovation
While policies A and B could be kept within EPA, they have the
potential to become very large. In this case, they might be located
elsewhere, perhaps in a new agency or in another agency whose existing
142.
-------
7-7
mission is more closely tied to their goals, such as the National Science
Foundation (NSF). On the other hand, there is a much stronger reason to
keep policies C and D within EPA, since they are so closely related to
its regulatory programs. Policy 0 necessitates a new government or
quasi-public entity, which may or may not be attached to the EPA budget.
Policies V and W may be inside or outside of EPA, or both. For example,
existing versions of these policies now are housed in EPA, the National
Institutes of Health (NIH), and NSF, among others. Lastly, policies R
and FF must be implemented by EPA.
Financing these new policies outside EPA would require discussion of
changes in other agencies. This approach to financing an option would
also divorce its implementation from EPA's mission to some extent; a
strategy more appropriate in certain cases than in others. The following
policies might reasonably be implemented outside EPA:
A. Direct cost subsidy for new chemical development, in general,
via grant mechanism
B. Direct cost subsidy for new chemical development, in general,
via a loan mechanism (or loan guarantee)
J. Increased capital availability for new chemical development,
via a government supported venture capital company
M. Reduce risk through government procurement of new chemicals.
0. Establish government testing for TSCA requirements
S. Chemical technology extension service, including dissemination
of information on test and compliance needs
V. Government support to develop new, better test methods
W. Government support for education and training programs
Because policies A and B are likely to be rather large and not
closely related to EPA's mission, there is a valid argument for their
location elsewhere. An even stronger argument can be made for policy J,
because EPA has no expertise in providing venture capital. Policy M
would fall naturally within the authority of the General Services
Administration (GSA), not EPA. Policies 0 and S would involve the
government in substantial, new functions, and they may be more successful
143.
-------
7-8
if implemented outside EPA. Moreover, the confidence that private sector
participants would have in the independence of these programs might be
reduced if they were attached to EPA. As mentioned above, funds for
activities similar to policy option V already exist in EPA and other
environmentally-oriented agencies. Additional funding could be added to
provide for the education and training options contained in option W.
However, other existing agencies, such as NSF, may find such programs
more consistent with their existing missions than would EPA. This would
especially be the case if education and training programs become large.
7.2.2 Off-Budget Financing
Although off-budget financing is generally regarded as an inferior
option, it is appropriate in certain circumstances, such as when a
program requires a new institutional home, when it can be financially
self-sustaining, or when it involves a tax expenditure.
Tax expenditures are appropriate for the following policies:
E. Indirect cost subsidy for chemical innovation, in general, via
a tax mechanism
F. Indirect cost subsidy for testing and compliance costs via a
tax mechanism
I. Decreased taxes on the sales of new chemicals
K. Increased capital availability for new chemical development via
changes in the tax code or in SEC rules
U. Tax adjustments to favor small firms or new entrants in the
chemical industry
In general, tax expenditures are rather easy to administer and do not
require new bureaucratic programs. While new tax laws often spawn
considerable litigation, settlement of the legal questions usually
results in a predictable set of rules. On the other hand, however, it is
difficult to predict the size of tax expenditure programs. Furthermore,
it is often quite difficult even to estimate the size of the resulting
revenue loss to the Treasury, which can be large. Policies E, I, K, and
144.
-------
7-9
U would all be expected to result in substantial revenue losses. Policy
F would be much less costly. Because tax expenditures are difficult to
measure, they have the tendency to persist over many years without public
scrutiny. Tax expenditures should therefore be considered carefully
before enactment - as indeed they usually are. It is also noteworthy
that the debate about the enactment of tax expenditures necessarily
involves the Department of the Treasury and other proponents of the
principle of tax neutrality.
Another off-budget financing option is to enact a new tax, keyed to
chemical production. This could be levied in proportion to the physical
production volume or to sales dollars. (In addition, the tax level might
depend on the toxicity of the chemical in question. This strategy,
however, presents formidable administrative difficulties and is unlikely
to be successful.) The revenues from such a tax could be applied to
finance the cost of a number of the policies including:
A. Direct cost subsidy for new chemical development, in general,
via a grant mechanism
B. Direct cost subsidy for new chemical development, in general,
via a loan mechanism (or loan guarantee)
C. Direct cost subsidy for testing/compliance of new chemical
development via a grant mechanism
D. Direct cost subsidy for testing/compliance costs of new
chemical development via a loan mechanism (or loan guarantee)
J. Increased capital availability for new chemical development via
government supported venture capital company
L. Reduce risk through government financed insurance for
regulatory losses
M. Reduce risk through government procurement of new chemicals
0. Establish government testing for TSCA requirements
S. Chemical technology extension service, including dissemination
of information on test and compliance needs
V. Government support to develop new, better test methods
W. Government support for education and training programs
FF. Improve EPA staff capability to assess the impact of regulatory
actions on innovation
145.
-------
7-10
In considering financing with a special tax on chemicals, the
difficulties encountered in enacting any new tax should be kept in mind.
These are both logical - i.e., policy design - and political. (The
current, lengthy debate about the financing of the "Superfund" for clean
up of hazardous waste sites illustrates this problem.) As mentioned
above, tax policies are usually very long-lived and expensive. It does
not seem to be worth the difficulties involved to establish a new tax
just for financing many of the small programs that have been considered
in this study. While new taxes are frequently advocated as a means of
correcting for market failures, a chemical production sales tax would
not efficiently do so, unless more hazardous chemicals were taxed at a
higher rate.
A third off-budget financing option is to establish a new financial
entity such as a public or quasi-public corporation that could become
self-financing after an initial infusion of public monies. A new entity
might appropriately implement the following policies:
J. Increased capital availability for new chemical development via
a government supported venture capital company
L. Reduce risk through government financed insurance for
regulatory losses.
0. Establish government testing for TSCA requirements
S. Chemical technology extension service, including dissemination
of information on test and compliance needs
Policies J and L would include functions similar to those undertaken
by existing private firms, so it is reasonable to expect that they could
become self-sustaining. They could be operated on a non-profit basis,
re-investing revenues in program opportunities rather than distributing
dividends to shareholders. The same possibility exists for policy 0,
although more government subsidy would probably be required due to the
nature of the activity. Policy S is an unlikely alternative for a
self-sustaining new entity because it probably could not charge for its
services and would need continual public support.
146.
-------
7-11
The last off-budget possibility is government assumption of
contingent liabilities, which could be appropriate for these options:
B. Direct cost subsidy for new chemical development, in general,
via a loan mechanism (or loan guarantee)
0. Direct cost subsidy for testing/compliance costs of new
chemical development via a loan mechanism (or loan guarantee)
L. Reduce risk through government financed insurance for
regulatory losses
N. Reduce risk in products liability actions by establishing
limits on liability
Although government assumption of liability requires virtually no new
commitments of funds, it has the potential to expand into very large
expenditures. For example, government assumption of damage liability can
create a large, never-ending drain on the Treasury if there are numerous
damage suits. Accordingly, a major difficulty with this kind of
financing is that it has a low predictability.
147.
-------
8. A COMPREHENSIVE PROGRAM OPPORTUNITY
Previous chapters have described, analyzed, and assessed 32 policy
options that could be considered for offsetting the unnecessarily
restrictive impacts of TSCA on technological innovation. This chapter
focuses on the top seven ranking options identified in chapter 6, and
builds a comprehensive program on them. Consideration is given to
possible overlaps and conflicts among the program elements.
8.1 The Top Seven Policy Options
The seven highest ranking policies are identified in chapter 6, and
discussed in detail in section 6.2. Certain of their aspects are shown
in table 8.1, where these policies are listed in the order of their
descending overall rank. These top policies are:
V
1. (R) EPA dissemination of chemical information - test results
and/or labeling
2. (EE) Generic PMN for classes of new chemicals
3. (V) Government support to develop new, better test methods
4. (C) Direct cost subsidy for testing/compliance costs of new
chemical development via a grant mechanism
5. (DO) "Fast track" PMN's for safe and/or.major innovations
6. (W) Government support for education and training programs
7. (D) Direct cost subsidy for testing/compliance costs of new
chemical development via a loan mechanism (or loan
guarantee)
Two of these policies, C and D, are from the category, "Reducing the
Cost of New Chemical Development." They are almost equivalent, with the
grant program appearing preferable in most respects. (See table 8.2,
which is table 6.4 reproduced here for convenience.) If designed as
discussed below and in chapter 4, either of these policies would be
particularly helpful to small firms and new entrants. There is no
reason, however, to adopt both of these approaches.
148.
-------
TABLE RI
Some Aspects of the Top Seven
Policy Options
Rank Option
R- Information
. Dissemination
EE- Generic
2 PMN
V- Government
3 support for
test method
development
C- Grants to
k offset testing
and compl lance
costs
DO- Fast track
S for safer or
major Innova-
tions
W- Government
6 support for
education and
training
D- Loans to
7 offset testing
and compliance
costs
Node of Action
Improves demand for
safer chemicals
and offsets TSCA
bias In favor of
existing products
Approves PHN's for
certain classes of
chemicals, perhaps
In advance
Reduces testing
costs
Offsets costs
Reduces regula-
tory delay
Reduces testing
and compliance
costs
Offsets costs
Industry Segment
Affected
All
Highly specialized
custom producers of
generally safe
products
All
Small firms and
new entrants
more than others
Products that
meet major social
needs
All
Small firms and
new entrants
favored
Overlap or
Conflicts
Complements
other options
Complements
other options
Complements
other options
Overlaps
policy 0
Complements
other options
Complements
other options
Overlaps
policy C
Legislative
Action Needed
None
None
Appropriation
Authorization
and
appropriation
None
Authorization
and
appropriation
Authorization
and
appropriation
Estimated Annual
Cost* (ml II Ion
dollars)
0.1-1
0.1-1
1-tO
1-10
0.1-1
1-tO
1-10
Financing
Options
Real location
of agency
funds
None needed
Appropriation
Increase
New budget
authority,
perhaps
funded by
new tax
None needed
New budget
authority
New budget
authority or
revolving
fund, per-
haps funded
by new tax
Status
None
Under con-
sideration
by EPA
Underway in
several
agencies
None
None
Similar
efforts by
OSHA
None
vo
OD
I
IV)
*0rder of magnitude range of public budget costs
-------
TABLE 8 .2
Ranks on Individual Criteria For
Seven Policies Ranked Highest Overall
en
o
[Overall Pol icy Rank
Policy Identifier
Short Pol icy Name
Criterion
Capacity to Countervail
Private Costs
>ubl ic Costs
Administrative Feasibility
Time to Implement
Support TSCA Aims
Other Side Effects
Effectiveness
Political Feasibility
1
R
Information
Dissemination
31
2k
7
1
5
1
1
16
5
2
EE
Generic
PMN
3
15
5
12
3
15
10
k
l\
3
V
Better
Test
Methods
Support
22
5
12
3
13
3
3
6
3
l*
C
Grants
for
Testing
Costs
k
22
17
5
1
5
5
3
8
5
DD
PMN
Fast
Track
15
13
1
17
6
6
7
14
6
6
W
Education
and
Training
Support
9
1«
19
6
19
A
2
7
2
7
u
Loans
for
Test, ing
Costs
6
28
IS
8
M
10
3
5
1
00
CO
Source: Same as table 6.4
-------
8-4
Two other policies, DO and EE, are changes in the administration of
the regulatory process under current law. They are complementary
programs, designed to address somewhat different issues. The generic
PMN, policy EE, recognizes that certain classes of chemicals are very
similar, and can reasonably be reviewed as a group, perhaps even
including contingent clearance of future developments in the class,
subject to certain procedural requirements. This option, which is
currently under consideration at EPA, would address the problems of
highly specialized chemical producers. The "fast track," policy DO,
recognizes the additional public interest in the rapid processing of
PMN's for safer or major innovations. It addresses all parts of the
industry, with a bias toward new products that meet major social needs;
i.e., safe substitutes, major innovations, or other criteria.
Policy R, information dissemination, is designed to improve the
market demand for safer new chemicals and thereby offset the commerical
bias that TSCA creates in favor of chemicals already on the market. It
would favor innovative firms throughout the industry and could complement
the other policies.
Policies V and W are designed to improve the technology for
compliance with TSCA and ultimately to reduce the costs of compliance.
Government is already devoting substantial funds to developing new,
improved test methods, so option V is essentially already in place.
Government support for education and training programs, policy W, could
help meet the current high demand for professionals and technicians for
industry and for testing laboratories, which would help all segments of
the industry. -
8.2 Discussion of the Comprehensive Program
It is suggested that EPA consider a comprehensive program to offset
the unnecessarily restrictive impacts of TSCA on technological
innovation. The program would include six of the top seven policy
151.
-------
8-5
options; R, EE, V, DO, W, and either C or D. The characteristics,
details, advantages, and disadvantages of each policy option are
discussed in chapters 4 and 6. Chapter 7 discusses financing options for
each program option, and notes that some of the programs might be
administered by agencies other than EPA. The issue of interagency
cooperation has been addressed only as part of the "political
feasibility" criterion used in chapter 6.
It is necessary to consider more than one policy in a comprehensive
program because no single policy is expected to be able to offset all the
unnecessarily restrictive impacts of TSCA on innovation. There is no
analytic way to determine how much, or how many programs would be
sufficient. However, a package of six policy elements, appropriately
designed, should go a long way toward offsetting these effects at
reasonable cost while supporting TSCA's primary goal to prevent
unreasonable risk of injury.
Policy V, support for test method development, represents only a
commitment to maintain or expand on-going activities in EPA and
elsewhere. It needs little further attention here.
Other than the overlap between policies C and 0, all the policies in
the comprehensive program would complement each other in terms of mode
and locus of action and in terms of their administration by EPA or
another agency.
If six of the policies were adopted, the total budget costs could be
expected to be in the range of $3 to $30 million per year. A
comprehensive program involving relatively limited commitment to the more
costly elements (policies C, D, W, or V) could cost in the neighborhood
of $7 million per year, with about $2 million per year for each element.
152.
-------
8-6
An experimental or trial program could be Implemented at less
expense and with a lower chance of disrupting the regulatory process.
However, it will be difficult to evaluate an experimental program's
effectiveness in view of the uncertainty in the rate of chemical
innovation, with or without TSCA or the offsetting programs, as discussed
in chapter 3.
8.3 Conclusion
A comprehensive program to offset the unnecessarily restrictive
impacts of TSCA on technological innovation need not be very large,
expensive, or disruptive. The analysis in this study suggests that very
expensive programs such as grants or tax incentives for all chemical
innovation in general are neither necessary, nor cost-effective.
Furthermore, this analysis has shown that in order to address
unnecessarily restrictive impacts of TSCA on technological innovation it
is not necessary to consider programs such as regulatory exemptions for
new, small volume chemicals, for low-risk chemicals, or for small firms
that would seriously compromise EPA's efforts under TSCA to protect human
health and the environment from unreasonable risk of injury and disease.
153.
-------
APPENDIX A:
THE CHEMICAL INDUSTRY.WELFARE ECONOMICS
AND GOVERNMENT ACTIVITY
A.I. Purpose and Approach
A. 1.1 Effects of Regulation
The Toxic Substances Control Act (TSCA) is likely to affect the
chemical industry and the economy in general in ways that go beyond its
primary purpose of reducing chemical hazards. In particular, there is
concern in the chemical industry about the potential impacts of TSCA on
chemical innovation arising from notification, testing-related, and
regulatory costs imposed on the industry.
The nature and extent of the impacts on the chemical industry and on
the economy depend on a great many factors. And, while it is clear that
performing a general equilibrium economic analysis of the impacts lies
beyond the state-of-the-art, the major impacts can be separately
identified and explored. The purpose of this appendix is to explore the
application of welfare economic concepts to an analysis of the impacts.
A. 1.2 Utility of a Welfare Economic Approach
the utility of a welfare economic approach to the impacts of TSCA's
requirements is threefold: 1) to systematically identify the impacts to
help ensure that the appropriate, more important impacts are the focus of
the overall study; 2) to identify and clarify areas of government
responsibility for addressing the effects of TSCA; and 3) to better
estimate the likely effects of remedial government mechanisms that might
be adopted to countervail unwanted impacts of actions under TSCA.
154.
-------
A-2
A.1.3 The Framework and Its Limitations
A general framework was developed using classical welfare economics
and was augmented using a public policy perspective. While the
state-of-the-art of welfare economics provides the structure and the
tools with which effects can be understood, it still has only a limited
capacity to provide quantitative estimates. As Anderson (1974) has
noted, measurement with the general equilibrium approach "is far from our
reach because of the interactibility of practically estimating responses
of all prices and quantities in an economy."
A.1.4 Separation of Expected and Unnecessarily Restrictive
Effects
A crucial consideration in this framework is the separation of the
cost-benefit consideration of regulation from the consideration of the
internalization of external economies. Specifically, the intent of
action under the TSCA is to internalize external costs such as those for
future ill health. It is expected that these newly internalized costs
will reduce consumption of chemicals in relation to other goods as higher
prices must be charged for them. This is an expected effect of
regulation. What remains is the question of whether there are any
short-run or long-run unnecessarily restrictive impacts of TSCA on the
industry that the government might decide to attempt to ameliorate.
A. 1.5 Use of the Welfare Economic Framework to Trace Effects of
Countervailing Programs
Once remedial mechanisms are identified it is necessary to trace the
incidence of their economic impacts. For example, if input factor supply
elasticities are sufficiently low, a subsidy might be passed back as
increased rent to the owner of a basic material. While this might be the
desired incidence for the subsidy, it may not be what was anticipated.
Only when the remedial mechanism countervails with the same incidence as
the regulatory costs can success be assured.
155.
-------
A-3
A.2. Fundamental Reasons for Government Responsibility and Intervention
In our market-based economy, the rationales for government
intervention relate to the existence of market distortions, market
failure, and distributional inequities. Public sector resource
allocation is economically justified when the value of the public sector
goods exceeds their opportunity cost; that is, the value of alternative
private sector production that would employ the same resources.
Furthermore, when allocative efficiency can be improved by government
intervention such that the benefits more than offset the costs of
intervention, government action may be called for. Finally, when the
distribution of wealth, goods, and other valued items is "improper" on
equity grounds it may be government's responsibility to intervene.
A.2.1 External Economies and Diseconomies
There are various situations in which the market does not operate
optimally. One common circumstance is the existence of external
economies. Negative externalities exist when the consumption of some
good is accompanied by a detrimental impact on the welfare of others that
is not captured in the market price for the good. For example, the cost
of the health impacts of toxic substances on workers is not fully
captured in the price of the goods they produce. The result is an
overproduction of the good and a misallocation of scarce resources.
Positive externalities are also common. A person who paints his house
usually cannot capture the benefit derived by others who appreciate its
appearance. Such a good will generally be underproduced. Thus, in some
localities, governments intervene to set standards for the appearance of
the houses.
A.2.2 Distortions in the Decisions of Consumer/Producers
Market failure can also occur due to distortions in the decisions of
consumers or producers. For example, consumers may lack information
about the value of flood insurance and underconsume it. The federal
156.
-------
A-4
government provides subsidized flood insurance to counteract this
underconsumption. Also, producers may lack sufficient knowledge or may
be too rigidly based in some technology to take advantage of new
technologies. Government remedies in this situation include such
policies as guaranteed purchases of goods that meet specifications or
direct funding for pilot plants and the like.
A.2.3 Public Goods
There are situations in which producers cannot capture the benefits
of basic scientific and technical information. When information is
difficult to keep private, not all the benefits of R&D will accrue to the
private firms that performs it, so such firms tend to underinvest in it.
This is a major rationale for government support of universities for
basic chemistry and the like. To the extent that testing of toxic
chemicals requires a better scientific basis, government support for it
may be justified.
A.2.4 Investment Risk
Producers may also underinvest in and underproduce potentially
socially desirable chemicals when the magnitude of investment risk is
greater than they can afford. However, this is an important
consideration only when the magnitudes of risk are considerably greater
than those typically facing an industry It may be that the risk from the
regulation of toxic substances is greater than that commonly faced within
the chemical industry. If government cannot reduce regulatory
uncertainty and if these socially desirable chemicals are underproduced
due to risk, there may be reason for government to assume some of the
risks. However, such risk assumption is not usually economically
efficient, and it should be shown that the industry does not have
alternative, lower risk areas of equivalent welfare potential for
development. (Ashford, Heaton, and Priest, 1979)
157.
-------
A-5
A.2.5 Equity Consideration of Impacts on Industry Over A
Relatively Short Time Period
While longer-run effects of regulation on industry may not justify
remedial action, shorter-run effects may. It would be unfair to industry
to force the internalization of large external costs in a short period of
time. Out of concern for the "era-period" distributional consequences of
regulation for industry, government may wish to subsidize industry at
least during a transition period. (Government might wish to recoup some
of the subsidy by taxing the industry at a higher rate over some long
time period.)
A.2.6 Indivisibilities
Another instance of market failure and underinvestment occurs for
very low volume goods. The marginal cost curve of production rises not
only from increasing production with fixed assets but also for decreasing
levels of production at low volume where a good may not be produced due
to problems of input factor indivisibility. Since all large volume
chemicals often start as low volume chemicals, society might be deprived
of socially desirable goods if additional start-up costs are added by
regulation or other causes. The role of capital, and in particular
venture capital, is to support the production of a new product until its
price covers average costs. In general, the market should adjust to
added fixed costs and no negative welfare impact would result; however,
if it were shown that changes in the availability of investment capital
are sluggish, transitional investment capital from the government could
be justified until the money market adjusted to the new situation.
A.2.7 Monopolistic Pricing Behavior
For industries such as utilities that face decreasing costs of
production, government intenvention is required to prevent monopoly
profits. It can be shown that a monopolist can price a good higher than
is socially desirable, and that the good will be underproduced. Price
158
-------
A-6
regulation is then justified, in which the government sets prices at
levels equal to average costs plus some level of return for investors.*
For some industries such as the chemical industry, oligopolies are common
for the production of many goods. Price regulation is not usually
possible in the case of ologopolies but antitrust regulation in the form
of penalties and criminal sanctions is often used to reduce collusion
between "rival" firms.**
A.2.8 Comparative Trade Advantages
Another area where positive externalities exist and government
intervention may be warranted is international trade. The implications
of a trade deficit for the national economy may warrant government
supports to improve our "comparative trade advantage" beyond that which
the private market would provide. Also, because innovation is integrally
tied to productivity, and productivity is directly linked to the standard
of living, government support of means to increase innovation may also be
justified. These issues are further discussed in section A.8.
A.3. Nature of Low-Volume Chemicals
Low-volume chemicals are of particular concern under TSCA because
testing-related costs are likely to be a larger percentage of their costs
than for chemicals whose volume will grow over time. While almost every
chemical was at one time a low-volume chemical, the concern here is for
those that remain low-volume, or are perceived as having only a
low-volume potential when first marketed.
*This form of regulation, while meeting certain welfare goals, itself
causes allocative inefficiency.
**01igopolies are an unusual situation for government regulation for in
addition to protection against collusion, government action is used not
only to prevent collusion between firms, but also to protect oligopolies
from each other. Thus, for trucking and railroads the ICC sets rules for
entry and rates to control price wars that have been ruinous to all
involved.
159.
-------
A-7
As long as development costs are low, chemicals can be produced
haphazardly and the marketplace can be the test of final demand.
However, if testing-related costs are large in comparison to other
start-up costs, a different tack is required. More market research will
be required if the marketplace itself cannot be used as a proving ground
for chemicals.
A chemical is likely to remain a small volume chemical under two
conditions: (1) the chemical does not enter the final product or does so
only in trace amounts—this is the case, for example, of catalysts and
solvents that are recycled, or (2) the chemical is used in a small volume
product.
The welfare economic implications of these two situations are that:
(1) a chemical that does not enter a final product is likely to
constitute a small percentage of the costs of production and could
therefore withstand a large increase in price without appreciably
affecting the supply curve for the final product, and (2) a chemical used
for a small volume product is likely used for unusual purposes that are
often associated with a lower elasticity of demand—an increase in its
price could probably be passed through to the consumer.
A.4. Nature of Chemicals Relative to Other Goods
In understanding the welfare impacts cf increased costs it is
necessary to examine the cross-elasticities among chemicals and between
chemicals and other goods.
A.4.1 Possible Cross-Elasticities Between New and Existing
Chemicals
It is important to distinguish between existing chemicals and new
chemicals. It is asserted here that high cross-elasticities would
generally exist between existing chemicals and new chemicals. Therefore,
160.
-------
A-8
the price of a new chemical Is important in its ability to capture sales
for existing ones. This suggests that where TSCA requirements are more
costly for new chemicals than existing chemicals, the number of new
chemicals entering the market would be greatly diminished. Whether this
has a negative impact on welfare is not immediately obvious. It is
argued below that a policy that affects new chemicals disproportionately
will have a net negative impact on welfare for two reasons: new chemicals
would have a positive income effect and therefore a positive impact on
the standard of living and, new chemicals may be needed to displace
existing, more dangerous chemicals.*
A.4.2 Possible Cross-Elasticities Between Chemical Goods and
Other Goods
The cross-elasticities between chemicals (and materials) and all
other goods are low. This 1s because consumers are more inclined to
substitute one chemical for another than for a non-chemical good. For
example, latex paint might substitute for an oil-based paint as prices
change, but paint is less likely to be substituted for by going to the
movies, when prices change. The welfare implications are that a greater
proportion of test-related costs, if they evenly affect the entire
chemical industry, will be passed through to the consumer than if they
affect only certain chemicals.
A.5. Nature and Extent of TSCA Impacts
A.5.1 Costs Due to TSCA
Chemical testing and notification requirements raise the cost of
introducing a new chemical and thus raise its selling price. Also,
direct and indirect effects of TSCA may increase the number of tests of
*To the extent that new chemicals are riskier than existing chemicals
there could be positive welfare effects disproportionately affecting new
chemical production.
161.
-------
A-9
existing chemicals and add to their costs. These costs are viewed by the
firm primarily as fixed costs that are generally offset by a firm's
income across all of its sales.
The effects of TSCA on an individual firm will depend on the firm's
size, its mix of new products and existing products, and TSCA's general
impact on new versus existing chemicals. It is assumed that the
additional costs associated with new chemicals, as a result of TSCA, will
be greater than those for existing chemicals by at least 1) the cost of
notification plus 2) the additional testing required by the greater
uncertainties associated with new, unfamiliar chemicals versus existing,
better understood ones.*
A.5.2 Relation to Firm Size
TSCA impacts will be greater for small firms if there are economies
of scale for testing-related costs. It is generally claimed that large
firms have easier access to capital, thus can more easily respond to
testing-related requirements. As an alternative, if smaller firms could
contract for testing with outside testing laboratories that achieve the
same efficiency as those in large chemical companies, the size advantage
of large firms would be greatly reduced. While there may be additional
delays and secrecy problems associated with the use of an outside
laboratory, these costs may not outweigh the benefits achieved through
economies of scale.
A.5.3 Relation to the More "Innovative" Firm
The degree to which a firm's revenues are derived from the
introduction of new chemicals is of more serious concern in an analysis
of TSCA. If all firms developed the same percentage of new chemicals
there would be no no differential impacts. However, firms that derive a
*Some might question the assumption that existing chemicals are better
understood.
162.
-------
A-10
greater than average proportion of their revenues fom new chemicals could
be seriously affected if the demand for their new products is not highly
elastic.
A.5.4 Possible Excessive Exits or Mergers -- Concentration
Effects
Long run effects of TSCA could be serious if firms that emphasize
introduction of new chemicals are responsible for the overall pattern of
innovation in the chemical industry. If TSCA were to cause these firms
to exit from the market, the long-run impacts could be great. In
contrast, if the economic impact of the requirements is smaller and these
firms are not forced to exit, or if the innovative potential of the
industry is associated primarily with firms that are diversified between
existing chemical production and the introduction of new entities,
long-run impacts may be low.
If diversified firms find it to their competitive advantage to
acquire firms that produce high percentages of new chemicals, market
restructuring would occur. The welfare effects of these actions would
then stem from increased industry concentration. Greater industry
concentration leads to a more oligopolist market and attendant market
distortions. Consumer prices will rise as more monopolistic marketing
occurs. Also, the indirect effects of industry concentration on
productivity, innovation, and quality of working life are generally
believed to be negative. Smaller firms, in general, are often found to
contribute a disproportionately higher number of innovations within an
industry. Also, the "small is beautiful" literature suggests that
quality of working life decreases as firm size increases for various
structural/ sociological reasons.
163.
-------
A-ll
A.6. Demand/Supply Dynamics
TSCA can be viewed as adding to the fixed and variable costs of
producing chemical goods.* Since the increased value of chemicals —
lower risk to health -- is unlikely to be substantially captured in their
selling prices, the demand for them will remain constant in the short-run.
A.6.1 Cost-Sharing Provision of TSCA
The "cost sharing" provisions of TSCA are designed to allow the
spreading of testing costs over all firms that produce the chemical for
which testing is required.** Thus, production volume may be important
since fixed testing costs that are spread out over many units of
production raise average fixed costs only slightly.
A.6.2 Single Firm Production
Consider the case of a low volume chemical that is produced by a
single firm. Such a firm faces the demand schedule shown in figure A.I
and is unable to influence the price at which its product is sold. The
fixed costs of testing for low volume chemicals can add substantially to
the average costs of production.
The firm will make a profit by producing any quantity between Ql and
Q2 and will, in fact, produce the profit maximizing quantity (as in the
case of a monopoly, price does not necessarily equal marginal cost at the
profit maximizing point). When fixed costs are increased by testing
costs for example, the average cost curve will move to AC' as in figure
A.2.
*Costs due to TSCA need not all be fixed costs. As production rises,
the chemical is likely to cause greater exposure and additional testing
costs are likely.
**It should be noted that this provision provides for reimbursement of
testing related costs for the successful chemical but not necessarily for
all the unsuccessful chemicals a firm considers. Thus, reimbursement may
actually defray only a small part of the total costs to the firm.
164.
-------
A-12
in
O
o
*»
Ul
O
IT
Q.
AVERAGE
COST
DEMAND
01
02
QUANTITY
FIGURE A.I PRODUCTION ECONOMICS FOR A SINGLE FIRM
O
o
LU
O
cr
a.
DEMAND
QUANTITY
FIGURE A.2 EFFECT OF TESTING COSTS ON SINGLE
FIRM PRODUCTION ECONOMICS
165.
-------
A-13
In deciding whether to produce a new chemical a firm will consider
the testing costs as a fixed cost (albeit a one-time cost), and will
augment the average cost curve by this amount (adjusted for the likely
"life" of the chemical). Too large a testing cost may make the average
cost of production too high for some chemicals, and consequently, they
will not be produced. The importance of testing costs is reduced by
higher expected demand, inelastic demand, longer chemical "life," and
patent protection.
For an existing chemical, the decision to perform mandatory testing
will be decided on the basis of the average variable costs of
production. Since the costs of existing plant and equipment used in
producing a chemical are sunk costs, they must be covered regardless and
should not affect the decision. Testing will be performed if revenues
are adequate to cover pre-existing average variable costs plus the
testing costs as amortized over the remaining commercial life of the
chemical.
As the sunk costs are reduced,-the criterion for deciding whether to
continue producing an existing chemical approaches the criterion for
deciding whether to produce a new one.
A.6.3 Multi-Firm Production
For a multi-firm market the situation is more complex. Cost sharing
and relatively high dollar sales volume can mean that the impact of the
testing costs on per-unit production costs is likely to be small.
However, for-chemicals produced in moderate volume, testing costs force
supply prices up, and lead to reductions in demand. The result will be a
new supply/demand equilibrium at a lower quantity of production (and
higher price) and the accelerated exiting of marginal, less efficient,
firms from production.
166.
-------
A-14
A.6.4 Multi-Firm Production with Variable Cost Changes
Increasing variable costs add to the marginal cost of production and
shifts the supply curve, much like a tax per unit of production acts to
shift a supply curve. This situation is depicted in figure A.3. The
increase due to TSCA is added vertically to the supply curve, moving it
from S to S1. The quantity sold decreases and the price increases, but
not by an amount equal to the added cost.
In terms of marginal analysis, the reduction in supply is caused by
the fact that each firm's marginal cost is now greater at each level of
output. This means that the profit maximizing position of each firm is
at a lower level of output. Hence, the total market supply curve shifts
upward and to the left. This is shown in figure A.4. In a competitive
market, firms set the quantity such that MC=P-
A.6.5 Consumers' and Producers' Surplus
The welfare economic effects of these shifts can be traced from the
changes in consumers' surplus and producers' surplus accompanying them.
At an elementary level it can be said that there has been a loss to the
consumer, owing to the price rise. This loss, which corresponds to the
difference in consumer surplus before and after the price change, is
shown in figure A.5.
The change in consumers' surplus is the area of rectangle A plus the area
of triangle B. Area A depicts the situation where the good is still
consumed but with less surplus to the consumer (e.g., if consumer A was
willing to pay twice the price for the good at its original price, and if
the price increases by 40%, consumer A is now only willing to pay 60%
more than the price for the good. Thus, while the good is still
consumed, the surplus diminishes). Area B depicts the loss in surplus
from the quantity of the good that is no longer consumed and is the
result of the drop in consumption is from Q to Q1 — here the marginal
167.
-------
A-15
UJ P'
cc
a.
SUPPLY, S'
SUPPLY, S
VARIABLE COSTS DUE TO
TSCA REQUIREMENTS
DEMAND
Q1 Q
QUANTITY
FIGURE A.3 IMPACT OF TSCA REQUIREMENTS ON THE
SUPPLY AND PRICE OF A CHEMICAL GOOD
S' S
VARIABLE COST
DUE TO TSCA
REQUIREMENTS
Q' Q
Q' Q
QUANTITY
FIGURE A.4.MARGINAL ANALYSIS OF THE SHIFT OF THE
SUPPLY CURVE FORA CHEMICAL GOOD UNDER
TSCA
168.
-------
A-16
<**
U
P
P
I I
Q' Q
QUANTITY
FIGURE A.5 CHANGE IN CONSUMERS1 SURPLUS
UNDER TSCA
QUANTITY
FIGURE A.6 SHORT-RUN CHANGE IN PRODUCERS'
SURPLUS UNDER TSCA
169.
-------
A-17
utility of the good to the consumer was sufficiently low that the
consumer substitutes other goods for the consumption of this particular
good as a result of the change in price.
The loss of consumers' surplus is a real loss to consumers and a
welfare economic impact. But, looked at from the other direction, it
partly represents the subsidy provided by those individuals who suffered
health losses before TSCA requirements were added. This is part of the
welfare economic exchange that Congress has bargained for in passing
TSCA, and should not be considered to be an uncompensated economic loss.
A complementary concept to consumers' surplus is producers'
surplus. This is the value represented "behind" the supply curve as
shown in figure A.6.
In general, the producers' surplus decreases as the fixed cost
increases and as area A minus area B becomes less than zero. (The
producers' surplus is the difference between the total revenue and the
total cost).
The welfare interpretation of the producers' surplus is not as
simple as that of the consumers' surplus. In the short-run, a decrease
in producers' surplus reduces the profitability of the firm. In the long
run, where there is time for firms to enter or leave a competitive
industry because of changes in price and profit, the producers' surplus
goes to zero. In this longer time period, the market supply curve is not
the sum of the supply curves of the individual firms but rather a
description of how costs in the individual firms will be affected by the
entry of firms: in a competitive market, profit levels are generally
fixed because of the entry and exit of firms. For example, if
profitability were to increase, other firms would enter the market,
increasing the supply of the product and decreasing the price of the
product until profits return to "normal." (Normal profits are those
levels of return on capital that are typical for the level of risk
involved.)
170.
-------
A-18
Thus, the actual impacts of changes in producers' surplus relate
more to the entry and exit of flrr.s than to changes within any particular
firm. The next section will further explore the welfare economic
implication of these changes.
A.7. Microeconomic Welfare Effects
A-.7.1 General Rules for Cost Pass-Forward and Pass-Back
The imposition of costs at a given point, in an economic system, is
felt thoughout the system to a degree that depends on the supply and
demand elasticities for the input factors and the outputs.
o for input factors -- the rore inelastic the supply of an input
factor, the more that costs can be passed back to the supplier
of that input factor.
o for outputs -- the more inelastic the demand for outputs, the
more that costs can be passed forward to consumers.
o for the producer -- the relative slopes of the producers'
supply curve ana the consumers' demand curve determine the
proportion of costs that is passed forward to consumers.
A.7.2 Multimarket Dynamics
In a multimarket situation, where suppliers and producers are
vertically integrated, recent welfare theory has shown that the concepts
of consumers' surplus and producers' surplus capture the aggregate
welfare effects back to the first suppliers and forward to the final
consumers. Under certain conditions required for the validity of
consumers' surplus measures in the final goods market and producers'
surplus measures in the initial resource market, the following was found
to hold:
The area behind a general equilibrium demand curve in an
intermediate market dees rot measure benefits to buyers in that
market alone, but rather measures the sum of rents to producers
selling in all higher markets (assuming no intervening market
has perfectly elastic demand) plus final consumers' surplus.
171.
-------
A-19
o The area behind the general equilibrium supply curve in an
intermediate market measures not only rents for producers
selling in that market, but also rents for all producers
selling in more basic markets (assuming no intervening market
has perfectly elastic supply) plus those of initial resource
suppliers. (Just and Hueth, 1979)
Thus, a practical approach to studying the distribution of welfare
effects over all other market groups in a sector is to estimate areas
behind general equilibrium supply and demand curves in the market of
interest. Then welfare effects on direct market participants are
separated out by subtracting areas behind ordinary supply and demand
curves. Further, if appropriate data exist, the distributional aspects
for other industries can also be studied by separating out the respective
values-added.
A.7.3 Captive Suppliers
When the elasticity of demand for a final product is high and the
elasticity of supply is low, costs will be passed back from the chemical
manufacturer to the resource supplier. This is often the case with
captive suppliers of major inputs. For example, the rent on cadmium or
in the manufacture of paint pigments may be primarily determined by the
price the final product can command in the market. Thus, added costs in
production will go to reducing the rent on the cadmium ore. (If cadmium
ore production is not fully captive, that is -- other producers who are
not affected by cost increases keep the price up, costs cannot be passed
back as easily.)
A.7.4 Worker Impacts
When the input under consideration is labor, and when labor mobility
and labor supply elasticity are low, there are serious economic welfare
consequences of increases in production costs. In these cases, costs
will be passed back to labor in the form of lower wages and the net
effect of regulation will be a negative transfer to the worker. It is
generally expected that this effect will be short-run and that long-run
171.
-------
A-20
elasticities would generally be high enough to reduce this effect.*
Nonetheless, if the effect is significant it argues for government
attention to reducing the impact on workers out of equity considerations.
A.7.5 Impacts Due to Reduced Producers' Surplus
Reductions in producers' surplus, as mentioned above, will in the
long run cause firms to exit or discourage new firms from entering the
market. These market changes result in losses due to idle capital and to
unemployment. These are important welfare considerations. In some
industries, such as the railroads, when changing market conditions
reduced the need for workers, additional problems were caused by
featherbedding and other practices that mitigated against displacing the
workers. Clearly it is not economically efficient to retain unneeded
labor, and thus it may be advisable to review our country's capacity to
retrain and absorb displaced firms arvd workers that may result from added
costs to chemical production through regulation.
A.7.6 Impact on Financing
A further impact of cost increases to a firm is a reduced ability to
raise capital. Capital is generated internally through a firm's earnings
or raised through equity issues or borrowing. The ability of a company
or industry to raise capital is tied to its earnings. An increase in
production costs that is not passed on reduces a firm's ability to raise
capital in three ways.
First, for a chemical firm to make purchases using debt, it must put
up around 20% of the purchase price. The equipment or facilities the
company is purchasing can serve as collateral to the vendor, who finances
the other 80% of the cost. When a chemical company must spend funds on
testing-related costs, its earnings are correspondingly reduced and the
*Higher elasticity will reduce the wage impacts, but the worker still
bears the cost of relocating, retraining, and job searching.
171.
-------
A-21
ability to make such payments is reduced. The effect of this reduction
is magnified five times; if the chemical firm cannot put up 20£ it cannot
receive returns on a $1.00 investment.
The second way a cost increase impacts raising capital relates to
the debt/equity ratio.* To a lender, the debt/equity ratio is a key
indicator of the risk involved in extending further credit to a company.
As the ratio rises (for example, because of the need to finance testing
facilities) the marginal lender may decide not to lend to the company.
This reduces the money available to the chemical firm for a given lending
rate.
Finally, reduced earnings and equity influence the market value of a
firm's stock. When the firm tries to raise capital through an equity
offer, the stock process is an important factor since the price at which
the new stock in tendered is closely tied to the current market price. A
production cost increase that decreases earnings also decreases the stock
price and thereby reduces the amount of capital the company can raise
through an equity offer.
The possible decreased ability of a chemical firm to raise capital,
at least over the short-run, adds to problems of stability, and more
importantly, hinders the firm's ability to finance new technology with
improved productivity. In response, government might be concerned about
increased industry instability due to regulation,.at least during a
transition period, and also be concerned about the possible reduction in
innovation and productivity growth over the long-run.
*Stockholders equity is the sum of the par value of the stock, the
paid-in capital (total capital raised through equity issues), and the
retained earnings of the company (earnings not distributed as dividends
to stockholders).
172.
-------
A-22
A.7.7 Positive Impacts Due to Regulation
Positive welfare effects will also accrue to the firm from
testing-related requirements. Chemical firms have said that information
they have gathered in the course of complying with a government
regulation have contributed positively to new product development in some
cases. (Ashford, Heaton and Priest, 1979; Ashford and Heaton, 1979)
Klein (1979) has suggested that for some industries that are rigidly
based, regulation provides an impetus to reducing industrial inertia. To
the extent that positive externalities due to regulation are evident in
the chemical industry, these will mitigate against cost increases and
reduce the negative welfare impacts described above.
A.8. Micro-Macroeconomic Welfare Effects
Considerations of microeconcmic welfare effects fail to capture the
larger scale dynamics that are believed by many to be critical to the
welfare of the U.S. economy. These include concerns for innovation,
productivity, labor/capital ratios, inflation, and international trade
advantage.
Improvements can be viewed in these micro-macroeconomic concerns as
positive externalities that arise from all economic/social activities,
including activities of the chemical industry. The limited resources
allocated to each sector are subsequently leveraged to attain these
positive micro-macroeconmic effects.
The welfare economic consideration here is estimating the benefits
of these effects and determining increases or decreases in these effects
caused by TSCA. Many of the components of this exercise have already
been discussed in prior sections. What remains is to consider the
overall sense of the effects.
173.
-------
A-23
A.8.1 Increased Productivity as a Significant Contribution to
Social Welfare
Bourdon (1979) has traced the effects of innovation and
technological change on labor productivity and compensation. Hill (1979)
has traced the linkages between innovation and technological change and
such social welfare measures as the quality of work and the quality of
life. These studies conclude that productivity increases are accompanied
by increases in the standard of living and may be a means of taming
inflation. The implications of these conclusions are that the promotion
of innovation is a primary means to achieve increases in social welfare.
At a welfare economic theoretical level the implications of
innovation can be examined as follows. Innovation can decrease costs or
increase value. By decreasing costs and holding demand constant,
personal disposable income increases -- this is an income effect. With
this additional income, more goods can be purchased and this is an
increase in welfare. If the value of a product is increased (such as the
development of a paint that wears longer or is more aesthetically
pleasing), and if demand is constant, consumer surplus is increased --
this too increases the net welfare of society.
A.8.2 Should the Chemical Industry be Singled Out for Attention
Vis a Vis Innovation in General?
There is no reason to prefer subsidizing one activity versus another
with regard to innovation (or any of the other micro-macroecnomic
effects) unless it is believed that (1) one sector of the economy offers
greater leveraging advantage than another, and that (2) the market system
would not naturally capture this advantage. Since innovation brings
benefits to both business and society, areas of greater innovation
potential will correspond in most cases with areas of industrial growth
and investment. Government intervention is justified only when the
positive externalities of innovation in one sector are greater than in
another.
174.
-------
A.-24
Thus, for the chemical industry, if positive externalities are
identified and if these external "-.ies are of a larger value than those
associated with other sectors (per resource dollar allocated), government
should preferentially intervene in the chemical industry. Furthermore,
government should intervene regc'-cless of whether TSCA imposes additional
costs. One positive externality d>eady indentified is the potential for
new chemicals to replace existing, r.ore hazardous chemicals. A negative
externality associated with the clerical industry, however, is its
dependence on petroleum sources end its demand for energy.
In general, if the chemical incustry offers greater increases in
value-added because of innovatior ir comparison with other pursuits, this
would lend support for government intervention to increase dynamic
economic welfare.
A.9. Examples of Chemical Products Illustrating Differing Market
StructureT
A.9.1 Factors Determini-g Inputs
In the preceding section the economics of TSCA-related impacts was
discussed from a theoretical vie.:c-nt. As was shown, the impacts of
chemical testing requirements va^y .,ith firm size, volume of production,
chemical price, and the extent t: ..r.ich cost increases can be passed
forward to consumers or back to me suppliers of the impact.
Firm size influences the irr.:aci of TSCA requirements because small
companies will, in general, not as easily be able to raise the resources
or perform the testing required.
High dollar sales insure tret TSCA related costs will only result in
a small percentage price increase. Cn the other hand, low dollar sales
mean that added production costs vsi'l have a proportionately larger
impact on prices.
175.
-------
A-25
The elasticity of demand determines the extent to which cost
increases can be passed on to the consumer (or intermediate user) without
incurring large sale losses. High elasticity implies that there will be
large sales losses if costs are passed on, while low elasticity implies
the reverse.
Production requires the input of raw materials, intermediate goods,
labor, and capital. When a raw material or intermediate good is used to
produce one good almost exclusively, the demand for that good will in
large part, determine demand for the input. If the price of the input
falls as the demand for the good falls, supply price elasticity will be
said to be high. If, on the other hand, the input is also used in other
production processes, demand for the good will not greatly influence its
price and supply elasticity will be said to be low. High supply
elasticity is borne by the suppliers of inputs, while low supply
elasticity implies the reverse.
A.9.2 Examples of Inputs
Any chemical product subject to TSCA regulation may be characterized
by its sales volume, by the size of firm(s) engaged in its production,
and by the elasticities of demand and the factor supply relevant to its
sale and manufacture. While these criteria do not fully determine the
likely impact of TSCA regulation, they are useful for an initial
assessment.
In table A.I, each of these criteria has been decided dichotomously,
resulting in 16 different industry "cases." An attempt has been made to
illustrate each case with an example from the chemical industry.* The
reasons for the choice of each example are discussed below.
*There is markedly little literature on measured demand and supply
elasticities. These examples were arrived at through the concensus of
the staff members familiar with chemical usage and should be considered
only suggestive here^
176.
-------
TABLE A.I Examples of Chemicals in Various Categories
-o
-j
Sales Volume
Low
High
Compony Size
Large (highly
concentrated)
Many small with
some large
(med i urn to
un concent rated)
Large f
Many small (with
some largej
Demand
Elasticity
High
Low
High
Low
High
f Low
High
Low
Supply
Elasticity
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low x
High
Example
Auto Touch-Up Paint
Flame Retardants
Catalysts (used in production of ammonia, hydrogen
and ni tr? c acid)
Enzymes
Low Volume Consumer Adhesives
Exotic Non-Essential Intermediates
Airplane Adhesives
Exotic Essential Intermediates
Synthetic Rubbers for Tires
Some Colored Inks
Carbon Black
Phosphate Fertilizers
Plastic Colorants
Rendering of Animal Fats
Propane for Home Use
Sulfuric Acid as a Waste Product
I
ro
Source: CPA
-------
A-27
Auto Touch-Up Paint - these relatively low volume paints are
produced predominantly by a few medium-sized companies. The
elasticity of demand for any specific product is likely to be high
since substitute competitive formulations exist. The materials used
in the production of these paints are generally used in larger
quantities elsewhere; hence, the prices of these inputs are not
likely to change much in response to changed demand from auto
touch-up paint manufacturers. Products falling in this category are
likely to be quite sensitive to mandated costs increases since sales
are likely to fall and costs cannot be passed back to material
suppliers.
Flame Retardants - flame retardants are used extensively in
plastics and synthetic fibers, and some are relatively low volume.
Because of the number of competing formulations, the demand for any
specific flame retardant is likely to be somewhat elastic. On the
other hand, the input materials used in manufacturing a flame
retardant may be used almost exclusively for that purpose. It is
thus possible that supply prices are fairly elastic where the input
is not captively produced.
Catalysts Used in the Manufacture of Ammonia, Hydrogen, and Nitric
Acid - these catalysts are produced in comparatively low sales
volume but are important in the production of commodities sold in
much larger volume. Consequently, demand elasticity is low while
supply elasticity (for such chemicals as nickel, iron, copper, zinc,
platinum, and silver) is essentially non-existent for this end use.
The catalyst industry is characterized by considerable captive
manufacture, with only 20 firms producing catalysts for the market.
Enzymes - enzymes are used in the preparation of meats and dairy
products, in brewing, in laundry detergents, and in the preparation
of tobacco, paper, textiles, leather, and other products. Relative
to their cost, enzymes play an important role in the manufacture of
these high volume items, and in consequence demand for them is
178.
-------
A-28
likely to be inelastic (except in laundry detergents). Raw
materials for enzyme production include fungi and yeast, and such
meat by-products as hog and beef pancreases, stomachs, and glands.
Since enzyme production may be an important market for some of these
inputs, it is likely that supply price elasticity will be high
(i.e., the market price for beef pancreases will be sensitive to the
use of enzymes made, from this input).
Low Volume Consumers
Adhesives - this group of chemical products includes lower volume
adhesives placed directly on the consumer market, often for
specified applications. When the intended application is not
critical or when substitute formulations exist, it is likely that
consumer demand for any individual product will be elastic. Because
the inputs used in producing these products are also used in much
higher volume elsewhere, a slight change in factor prices will
result from reduced sales of a specified low volume adhesive (factor
supply elasticity is low). While adhesives are also produced by
large firms, this segment of the chemical industry is dominated by
small and medium size manufactures.
Fragrances and Flavors - while there are only about 150 natural
essential oils in commercial production there are approximately
3,000 synthetic aroma chemicals commercially available, and many of
these are produced or blended by small companies. With such a wide
range of chemicals available, it is likely that demand for any
particular aroma chemical is high. Supply price elasticity for
these low volume chemicals is likely to be low, however, as the raw
inputs are also extensively used to manufacture other products.
Because of the skill involved in developing and blending aromas and
fragrances, small firms thrive in this area, although some larger
firms produce and blend aromas for their own use.
179.
-------
A-29
Exotic Non-Essential Intermediates - while it is difficult to locate
an example of a chemical in this category, they undoubtedly exist.
Any low volume chemical intermediate for which economic substitutes
exist, and which is produced from low volume inputs, would fall into
this category.
Airplane Adhesives - adhesives are an important, if low volume,
input in airplane construction. Because of the importance of the
adhesives in maintaining structural integrity, and the strict
testing required to substitute an alternate product, it is likely
that there is little demand elasticity for these products. While
data on the production of these adhesives could not be easily
located, the adhesives industry as a whole is marked by small firms.
Exotic Essential Intermediates - essential, low volume,
intermediates are chemicals used in a production process that cannot
economically operate in their absence. Demand elasticity for such
chemicals will be low, and if they are in turn produced from low
volume chemicals, input prices will be elastic with respect to
demand.
Synthetic Rubbers for Tires - synthetic rubbers are high volume
chemicals produced predominantly in large plants. In the
manufacture of tires, natural rubber and synthetic rubbers may be
blended together. Because the price of natural rubber is now close
to the price of synthetic rubber, there is some demand elasticity
for the synthetic product. The raw material from which synthetic
rubber is produced, petroleum, is totally inelastic in price with
respect to changes in this source of demand.
Selected Colored Inks - many colored inks are high volume items with
ready substitutes. They are thus likely to face considerable demand
elasticity. On the other hand, some of the inputs used in manufac -
turing these inks may be specifically manufactured intermediates so
that supply factor elasticity may be relatively high.
180.
-------
A-30
Carbon Black - carbon black is an essential ingredient in the
manufacture of rubber tires, this application accounting for 90% of
their total production. Because of the importance of carbon black
in reinforcing rubber and the lack of economic substitutes, demand
elasticity is low. On the other hand, because carbon black is
produced from liquid or gaseous hydrocarbons, and constitutes only a
minor use of these inputs, input price is insensitive to demand.
All of the eight producers of this high volume chemical are large
companies.
Plastic Colorants - plastics come in many colors. Because specific
color is riot essential in many applications, and because substitute
color formation exist for some shades, demand elasticity for many
specific colors is going to be high (volume may either be low or
high). When raw material inputs are chemicals that are generally
available, supply elasticity will be low. It should be noted that
only some, not all, plastic colorants will fall in this category.
Rendering of Animal Fats - surfactants include both soaps from
natural fats and detergents produced from synthetic raw materials.
Detergents have replaced soap nearly everywhere except in the toilet
bars market, but even here mixed soap and detergent and pure
detergent bars have appeared. The first step in the process of
producing soap is to render animal fat, a process that is dominated
by small firms located close to the raw materials market (i.e., meat
processing plants). Demand for this product, rendering animal fat,
is highly demand elastic because of the alternate formulation of
soap with synthetic detergents, and because other oil (e.g.,
soybean, coconut) can be used instead. The raw inputs for rendering
oil are essentially waste products that could not find a market
elsewhere. Hence, supply price is highly elastic with respect to
demand by the rendering industry.
Propane for Home Use - while the bottling of propane for home use
does not involve a chemmical process, this is one clearly defined
181.
-------
A-31
(and regulatable) use of the material. Demand elasticity for
propane in this use is relatively low (because of the costs of
converting to an alternate energy source), and bottling is done by
small local companies. Supply elasticity for propane is low. The
price of natural gas will not be greatly influenced by changes in
demand from this quarter.
Sulfuric Acid as a Uaste Product - sulfur emissions are often
produced as a waste product of industrial production processes.
Sulfuric acid from waste gasses is often (in fact, because of clean
air regulations, most often) collected and disposed of. Demand
elasticity for sulfuric acid is low; there are often no economic
substitutes for the substance. Supply elasticity, however, is high
because the raw input - waste gas - is not useful in any other
application.
182.
-------
A-32
REFERENCES FOR APPENDIX A
Anderson, J.E., 1974. "A Note on Welfare Surpluses and Gains from Trade
in General Equilibrium," American Economic Review, Sept., pp. 758-62.
Ashford, Nicholas A., George R. Heaton, Jr., and W. Curtiss Priest,
1979. "Environmental, Health, and Safety Regulation and Technological
Innovation," in Technological Innovation for a Dynamic Economy, C.T. Hill
and J. M. Utterback, editors, Pergamon Press.
Ashford, Nicholas A. and George R. Heaton, Jr., 1979. "Effects of Health
and Environmental Regulation of Technological Change in the Chemical
Industry: Theory and Evidence," in Federal Regulation and Chemical
Innovation, ed., C.T. Hill, American Chemical Society Symposium Series
No. 109, American Chemical Society, Washington, D.C.
Bourdon, Clinton C., 1979. "Labor, Productivity, and Technological
Innovation: From Automation Scare to Productivity Decline," in
Technological Innovation for a Dynamic Economy, eds. C.T. Hill and J.M.
Utterback, Pergamon Press. »
Hill, Christopher T., 1979. "Technological Innovation: Agent of Growth
and Change," in Technological Innovation for a Dynamic Economy, eds. C.T.
Hill and J.M. Uterback, Pergamon Press, New York.
Just, R.E. and D.L. Hueth, 1979. "Welfare Measures in a Multimarket
Framework," American Economic Review, 69:5, pp. 947-954.
Klein, Burton H., 1979. "The Slowdown in Productivity Advances: A
Dynamic Explanation," in Technological Innovation for a Dynamic Economy,
eds., C. T. Hill and J. M. Utterback, Pergamon Press, New York.
183.
-------
APPENDIX B INNOVATION IN THE CHEMICAL INDUSTRY
To understand how TSCA might affect chemical innovation, it is
necessary to have some understanding of the history and future trends of
both the nature and origins of new chemicals. Among the significant
factors to be considered are the relative importance of small and large
firms, and of new entrants and established companies,; the contributions
of academic research; the importance of government and of private funding
for new chemical development; the relative emphasis on product and
process research and development; and the differences among the various
industry sectors for each of these factors at various times. It is also
necessary to understand what changes may be occurring in the structure
and in the environment of the U.S. chemical industry in order that such
changes are not mistakenly attributed to environmental, health, and
safety regulations in general, or to TSCA in particular-
This appendix begins with a model of the innovation process in
industry which is presented to provide a context for the subsequent
discussion. This model takes a long-term view of the innovation process;
it is not concerned with how a particular firm manages its innovative
activities at a particular point in time, but rather with how the pattern
of innovation changes as a firm, a productive unit, or an industry sector
evolves over time.
Next, data are presented on the basic structure of the chemical
industry that help to understand the response to regulation and the
nature of innovation in various sectors. Subsequent sections describe
trends in the allocation of resources to innovation in the industry, and
review what is known from the literature about the nature and sources of
new chemical products. Following an assessment of future trends in
competitive and other pressures on the industry, the appendix closes with
a summary of findings.
184.
-------
B-2
B.I A Model of Technological Innovation*
The process of technological innovation involves the creation,
design, production, first use, and diffusion of a new technological
product, process, or system. While the process of technological
innovation is sometimes viewed as synonymous with R&D, in fact, organized
R&D is only one of several kinds of innovative activity -- in some cases
R&D comes only after the fact of an innovation, or not at all. The
process of technological innovation can happen rapidly, or it may require
an extended period of time. The changes that occur in the costs, the
performance, or the characteristics of a technology may be major or
incremental, but the total of the incremental changes over a period of
time can be as important as the dramatic breakthroughs in improving the
quality or reducing the cost of a product.
Technological innovation involves matching, in a new way, a social
or economic need with capabilities drawn from science, technology, or
craft. Innovative activity can be a risky business. It is not possible
to know in advance whether any particular project will be successful.
The greater the advance sought, the more uncertain the outcome.
Four indirect approaches are used to measure technolgocial
innovation. The first, measures inputs to the process of technological
innovation, such as the R&D budget or the number of scientists working in
an area. The second, measures intermediate outputs, such as the number
of patents awarded, the number of technical papers published, or the
number of new chemical entities synthesized. The third, measures the
performance of a product or process, such as its weather resistance,
dispersability, durability, or cost. The fourth, measures the amounts of
various inputs required to produce a product, such as hours of labor,
*This section is drawn from (Hill, 1979) and (Hill and Utterback, 1979).
185.
-------
B-3
barrels of oil, or dollars worth of capital equipment. None of these
measures is entirely satisfactory, but all are used in various studies
discussed below.
In the market-oriented U.S. economy, technological innovation nearly
always occurs in private firms, although government plays many roles in
supplying the resources necessary for innovation and for creating the
environment within which it must happen. However, with rare exceptions,
government itself does not innovate. Instead, it must find ways to
influence the rate and direction of technological innovation in private
firms. Similarly, only rarely do other institutions in our society, such
as universities or special interest organizations, engage in
technological innovation. When they do, it is nearly always by
cooperating with or establishing a profit-making venture to commericalize
an idea or an invention.
Firms choose among the available products, strategies, and
technologies in order to maximize some measure of profitability. A major
concern of a firm is to increase the probable rewards to success and to
reduce the risk of failure in the marketplace. Therefore, firms must
consider the profits likely to result from an innovation and the
technological risk involved, when deciding to attempt it.
However, they must also consider another and more important risk --
the risk that in failing to innovate, an existing line of business may be
taken over by a competitor who does. Thus, an important incentive for
firms to attempt risky technological innovation is their desire to
survive in the face of effective competition or rivalry from other
firms. Such rivalry can be especially effective if the competitor is a
new entrant who has a new technological product or process that is
superior to those of the existing firms in the industry.
At the level of the firm, several factors tend to increase the
extent and success of innovative activity: a flexible organizational
structure; a high diversity of staff experience; an adequate financial
186.
-------
B-4
condition; a good recognition and understanding of market needs; a good
recognition and understanding of competitive and other environmental
pressures; a willingness, ability, and need to take risks; and a set of
technological possibilities. Innovation is facilitated if a firm is
experiencing, or can anticipate, a rapid growth in demand for its
products, and if the workers, owners, and managers can expect to earn a
financial reward for their efforts. Government policies such as fiscal
and monetary policy, regulation policy, labor policy, and industry
structure policy can act to increase or decrease the rate of
technological innovation in a variety of ways.
The continuing entry of new firms is an important determinant of the
degree of rivalry in an industry, because new entrants, in hopes of
becoming a major factor in an industry, have a greater incentive to take
risks. In the absence of such new entrants, there is a tendency for the
degree of risk taking in existing firms to decline. In existing firms,
the thinking habits of entrepeneurs and their patterns of search are
likely to be highly constrained by the forms of the technology and the
organization they already have. Therefore, they are likely to search for
new technologies more narrowly than do new entrants.
Small new ventures and larger firms entering a business for the
first time, introduce a disproportionate share of the innovations that
create major competitive threats and rivalry. Established firms often
respond to a new entrant's invasion of their product line with redoubled
creative effort and investment in what they already know well. Even
though it may be crude, the new technology may have greater performance
advantages in certain submarkets, and gain ground by competing there
first. Use of the new technology then expands as it captures a series of
submarkets. The new technology often has a much greater potential for
improvement and cost reduction than does the existing technology. Thus,
price cutting by established units as a defense may be ineffective.
Firms must become more specialized and efficient to exploit
innovations over an extended time. This drives them toward a more stable
187.
-------
8-5
production process and a more structured organization. Most established
large firms have evolved from small, disorganized, and highly innovative
beginnings. Demands for greater sophistication, uniformity, and lower
cost in the product create an ongoing demand for the development and the
improvement of both product and process. This means that product design
and process design become more and more closely interdependent as a line
of business develops. A shift from radical to evolutionary product
innovation usually occurs as a result of this interdependence, and is
accompanied by heightened price competition, and by an increased emphasis
on process innovation. Small-scale units that are flexible and highly
reliant on manual labor, and that use general-purpose equipment, develop
into units that rely on automated, capital-intensive, high-volume
processes. Thus, innovative patterns, production processes, and the
level of production capacity all change together in a consistent,
predictable way over time.
B.2 The Structure-of the Chemical Industry
The chemical industry includes many sizes and types of firms
producing many product types and using many different technologies. It
is one of the most vital sectors of the U.S. economy, whose sales growth
and profits have been larger than the average in manufacturing. (Keegan,
1977; Landau, 1979) The total output of the industry is expected to
increase at a rate of 4-6% in real terms through 1985. (A.D.Little, 1978;
Keegan, 1977; Boyden 1976) This rate is lower than in the past but is
still strong when compared with the U.S. industrial average. (C&E News,
1979a; A.D.Little, 1978; Keegan, 1977)
The U.S. chemical industry has a net positive trade balance,
presently contributing a balance greater than $5 billion per year. (C&E
News, 1979a) This is expected to continue even in light of the
anticipated competition from the oil producing nations that are now
constructing their own petrochemical plants. The predicted strength of
the U.S. industry has been attributed, among other things, to its heavy
188.
-------
B-6
use of natural gas as a primary chemical feedstock. Owing in part to the
history of price regulation, natural gas prices have not risen as rapidly
as that of oil, especially oil in other importing countries. The
chemical industries of most other nations rely more on oil and have
experienced greater price increases for their primary feedstock. It is,
therefore, likely that the industry in the U.S. will be able to maintain
competitive prices in the increasingly competitive world market.
(Business Week, 1980; C&E News, 1979b)
The chemical industry produces three fundamental product types:
basic chemicals, intermediates, and finished products. Basic chemicals
form the bulk of the industry output. This sector is dominated by large
producers, and is becoming more and more controlled by chemical
subsidiaries of the large petroleum firms. After an aborted attempt in
the early sixties to enter the chemical industry, their immediate access
to feedstock and to capital has allowed the petroleum firms to stake out
the basic chemical sector as their province over the last ten years.
Growth in intermediates is expected to exceed the overall real
growth rate of 4-6% anticipated for the years 1977-1985. Intermediates
include many specialty products that have high profit margins and
comparatively rapid turnover in the market. This sector contributes the
greatest proportion of new chemical products. (Foster D. Snell, Inc.,
1975)
The finished product sector is the most diverse of the three
sectors. Finished products use basics and intermediates as inputs. New
discoveries occur in this sector although with a lower frequency than for
intermediates. The need for improved intermediates often originates in
finished product firms. The identification of a market need that can be
filled by a finished product or the requirement for improved finished
product performance often stimulates the development of new chemical
intermediates. Finished products also have large public exposure, and
safe chemicals are becoming a very high priority for this sector due to
189.
-------
B-7
changing public expectations regarding chemical safety, and the growing
market for safety and government regulation of chemicals. (Business Week,
1979)
The chemical industry includes approximately 7,000 firms operating
in 10,000 establishments. Forty-seven percent of the firms are
manufacturers rather than processors or mixers; the manufacturers are the
firms that are presently primarily concerned with TSCA. The extent to
which processors and mixers will have to be concerned with TSCA is not
yet settled.
The Standard Industrial Classification (SIC) system recognizes 28
four-digit classes within the chemical industry (SIC 28) as shown in
table B.I. Nearly all of these sectors will be directly affected by
TSCA. In addition, the products of or the inputs to the following
two-digit sectors are also likely to be affected by TSCA:
20 Food and kindred products
26 Paper and allied products
29 Petroleum refining and related industries
30 Rubber and miscellaneous plastic products
31 Leather and leather products
32 Stone, clay, glass, and concrete products
33 Primary metals industries
This analysis is primarily concerned with SIC 28 and the chemical
activities of firms in SIC 29. Resource and data limitations preclude
examination of the other sectors. As with all analyses based on the SIC
system, there are problems of overlapping markets, firms that span
several sectors, and inappropriate aggregation of the data.
Concentration in the chemical industry as a whole is moderate
measured in terms of the value of shipments. The Kline Guide estimates
the four-firm concentration ratio at 35% and the ten-firm ratio at 50%.
(Keegan, 1977) It argues that in comparison with other capital intensive
industries such as automobiles, tires, or aircraft this concentration
ratio is low. The real concern, however, is concentration ratios within
190.
-------
B-8
TABLE B.I
Standard Industrial Classification System
for the Chemical Industry
2812
2813
2816
2819
2821
2822
2823
2824
2831
2833
2834
2841
2842
2843
2844
2851
2861
2865
2869
2873
2874
2875
2879
2891
2892
2893
2895
2899
Alkalies and chlorine
Industrial gases
Inorganic Pigments
Industrial inorganic chemical nee*
Plastics materials and synthetic resins
Synthetic Rubber
Cellulosic manmade fibers
Organic fibers, non cellulosic
Biological Products**
Medicinals and botanicals**
Pharmaceutical preparations**
Soap and other detergents
Polishes and sanitation goods
Surface active ag'ents
Toilet preparations
Paints and allied products
Gum and wood chemicals
Cyclic crudes and intermediates
Industrial organic chemicals nee*
Nitrogenous Fertilizers
Phosphatic Fertilizers
Fertilizer mixing only
Agricultural chemicals nee*
Adhesives and sealants
Explosives
Printing Ink
Carbon Black
Chemical preparation nee*
* nee - not elsewhere classified
**
Exempt from the TSCAct
191.
-------
B-9
the industry subsectors shown in Table 8.2 for four-digit groups. As
shown in the table, the most concentrated sectors are:.
2812 Alkalies and chlorine
2813 Industrial gases
2822 Synthetic rubber
2823 Cellulosic man-made fiber
2824 Non-cellulosic organic fiber
2841 Soaps and detergents
2861 Gum and wood chemicals
2892 Explosives
2895 Carbon black
This assessment agrees with those by Arthur D. Little (1978) and the
Kline Guide (Keegan, 1977) regarding the most concentrated sectors. The
least concentrated sectors are:
2851 Paints and allied products
2875 Fertilizers-mixing only
2891 Adhesives and sealants
2899 Chemical preparations (not elsewhere classified)
Table B.3 presents four-firm concentration ratios based on number of
establishments, on value added, and on total employment. (U.S. Bureau of
Census, 1972) These data confirm the findings regarding the most and
least concentrated sectors, reached from the concentration ratios based
on the value of shipments.
Concentration ratios provide some insight into the ability of
different sectors to cope with the costs of regulation. Firms in the
less concentrated sectors face a more competitive situation than ones in
the more concentrated sectors, and may not be able to pass on the costs
of compliance. In addition, sales of firms in the less concentrated
sectors may be more affected by unregulated substitutes than the larger
firms. Furthermore, the generally larger firms in the more concentrated
sectors are more likely to be able to influence the nature of the
regulations governing their behavior.
A brief review of the adhesives and sealants sector illustrates
through small firms the types of problems that may be encountered in the
191.
-------
2812 Alkalies and Chlorine
2813 Industrial Gasses
2816 Inorganic Pigments
2819 Industrial inorganic
chemical nee
2821 Plastics materials and
resins
2822 Synthetic rubber
2823 Cellulosic man-made
fibers
2824 Organic fibers non-
cellulosic
2841 Soal and other detergents
2842 Polishes and sanitation
goods
2843 Surface active agents
2844 Toilet preparations
2851 Paints and allied
products
2861 Gum and Wood chemicals
2865 Cyclic crudes and
intermediates
2869 Industrial inorganic
chemicals nee
2873 Nitrogenous fertilizers
2874 Phosphatic fertilizers
2875 Fertilizers mixing only
2879 Agricultural chemicals
nee
2891 Adhesives and sealants
2892 Explosives
2893 Printing Ink
2895 Carbon Black
2899 Chemical preps nee
B-10
TABLE B.2
icentration Ratios
ustry Shipments in
4 firm
72
65
52
34
27
62
96
74
5 62
43
8
38
22
68
34
43
35
29
24
39
19
67
39
74
16
8 firm
91
81
72
52
41
81
X
91
74
54
42
53
34
83
52
57
53
47
38
57
31
86
54
99+
26
Based on 1
1972
20 firm
99+
93
91
76
65
98
100
99+
85
65
64
74
51
94
77
74
84
83
57
76
52
98
75
100
41
50 firm
98
99
93
90
100
100
92
78
89
91
66
99
96
92
100
99
74
89
76
99+
58
Source: U.S. Bureau of the Census
192
-------
B-ll
less concentrated sectors. Adhesives and sealants is one of the least
concentrated sectors of the chemical industry. The largest firms in the
industry are H.B. Fuller, National Starch, 3 M, and U.S.M. Sales and
concentration data for adhesives and sealants do not reflect captive
markets. Georgia-Pacific dominates the plywood and wood products
industry and develops and manufactures all the adhesives it uses
internally. Georgia-Pacific's internal use of adhesives is equivalent to
a major fraction of all U.S. adhesive sales, and if included as a
manufacturer, it would top the list of major producers.
This sector is characterized by a high acquisition rate, which
reflects, in part, the high margins available in specialty adhesives, and
the growing penetration of adhesives into the fasteners market. These
acquisitions, which are dominated by large firms, both increase their
market share and also help them acquire new products. Large firms not
only acquire small ones, they also acquire divisions of other large
firms. Furthermore, start-up and entry of new firms has come to a
halt. An indication of the high acquisition rate is the increasing
concentration of this sector over time. (Frost and Sullivan, 1976)
According to Frost and Sullivan (1976) the problems faced by the
small firms in the adhesive and sealants sector, which may be similar to
the problems of the other less-concentrated sectors, are:
o a substantial proportion of large firms are involved in
acquisitions
o captive markets that cannot be penetrated
o increasing capital outlays due to rising costs in an
increasingly aggressive environment.
o an inability to acquire capital
Based on the concentration data, the sectors composed of small firms
that may have trouble dealing with regulatory costs are:
193.
-------
VO
TAULE B.3
Measures of Chemical Sector Contented! ion
SIC Code
28)2
281)
2816
2819
2821
2822
282)
2824
2841
2842
284)
2844
2851
2861
2865
2869
2873
2874
2875
2879
2891
2892
2893
2895
2899
Sector
Alkalies and chlorine
Industrial gases
Inorganic pigments
Industrial inorganic chemicals nee*
Plastics materials and resins
Synthetic rubber
Cel lulosic nun -made fibers
Organic fibers, non-eel lulosic
Soap and other detergents
Polishes and sanitation goods
Surface active agents
Toilet preparations
Paints and allied products
Gum and wood chemicals
Cyllc crudes and intermediates
Industrial organic chemicals nee*
Nitrogenous fertilizers
Phosphallc fertilizers
Fertilizer mixing only
Agricultural chemicals nee*
Adhesives and sealants
Explosives
Printing Ink
Carbon black
Chemical preparations nee*
Number
of Firms
28
105
77
166
193
50
12
)6
577
1022
151
59)
1)18
118
124
351
47
66
422
297
317
55
213
II
1485
Number of
Establish-
ments
48
50)
114
384
32)
59
IB
61
642
1108
178
645
1599
1)9
174
51)
73
145
627
388
46)
92
407
)7
1606
Percent of
Establish-
ments Owned
by Top 4
K 1 rnis
)8
59
17
8
7
15
56
31
4
2
6
2
)
9
12
6
16
17
7
*
12
19
22
62
1
Percent
Value Added
by To|> 4
F 1 1 HIS
71
67
i.4
39
)0
65
X
78
65
46
29
41
2)
70
49
49
)9
29
29
47
20
69
)5
74
16
Percent
Total
Employment
by Toj> 4
K i rins
79
52
50
)8
2'J
yj
X
69
46
19
23
18
19
62
30
3/
33
25
8
29
13
67
)9
69
9
* Not elsewhere classified
Source: U.S. Bureau of Census (1972)
i
rv>
-------
B-13
SIC
2581 Chemical preparations, n.e.c.
2899 Paints and allied products
2875 Fertilizers - mixing only
2891 Adhesives and sealants
Small firms can also be expected to have difficulty meeting
regulatory costs in the sectors that are dominated by large firms in
which there is also a significant portion of small firms in a very
competitive environment. These include:
SIC
2841 Soap and other detergents
2842 Polishes and sanitation goods
2844 Toilet preparations
2879 Agricultural chemicals n.e.c.
B.3 Inputs to the Innovation Process in the Chemical Industry
Three critical inputs to the process of innovation are R&D
expenditures, R&D personne-1, and corporate R&D strategies. There have
been major changes over time in both R&D strategies and R&D
expenditures. R&D personnel numbers have not changed significantly in
recent years; however, it is claimed that the productivity of research
personnel has . (Landau, 1979) The following sections outline the
changes in each of these areas.
B.3.1 Research and Development Expenditures
Industry spent about $1 billion annually on chemical R&D from 1969
through the early seventies. In 1978 expenditures surpassed $2.2
*The information in this section is largely drawn from "Facts and Figures
for Chemical R&D" in the July 23, 1979 issue of Chemical and Engineering
News. . .
195.
-------
B-14
billion, excluding those for pharmaceutical research. Government funds
only a small percentage of chemical research, contributing an additional
13% to this total. It should be noted that these figures represent
expenditures for chemical R&D in SIC 28 only and that they also include
non-chemicals R&D expenditures by SIC 28 firms.
Chemical industry R&D spending, not adjusted for inflation,
increased at an average annual rate of 9% throughout the seventies. In
real terms, however, spending for 1967 through 1972 fell approximately
25%; in recent years, R&D investment in real terms has been increasing at
2-3% per year. (C&E News 1979a) Since the 1979 inflation rate was
greater than 13%, R&D spending for this year may not show any real gain.
The small real gains of recent years may be even smaller in view of the
fact that they are estimated using the average rate of inflation, and
that costs of salaries and other contributions to R&D in chemicals have
been rising faster than average costs in the economy. (Keegan, 1977)
As shown in Table B.4, R&D expenditures averaged 3 - 3.8% of
chemical industry net sales during the seventies. This is about one
percentage point below the investment of 15 years ago, but is still
higher than the U.S. industrial average. This declining trend in
chemical R&D reflects the overall U.S. industrial pattern: U.S. industry
R&D as a portion of the GNP has fallen since 1964, and now averages about
2.2%, down from just under 3% at its high point. (Business Week, 1979)
The NSF (1977) survey of R&D expenditures divides chemical research
into basic, applied, and development. Industry funds have grown most
rapidly in recent years for applied research, more slowly for
development, and have fallen behind the inflation rate for basic
research. The proportions in 1978 were 11% for basic research, 41% for
applied research, and 48% for development (not including spending on
drugs). In basic research, the major portion of funding goes to the
physical sciences as opposed to engineering by a ratio of almost two to
one. Applied research funds are devoted predominantly to industrial
chemicals and synthetics. The breakdown on spending for development was
not available.
196,
-------
B-15
Table B.4
Chemical Industry R&D Expenditures
Year R&D as % Net Sales
1958 3.8
1963 4.3
1964 4.5
1965 4.3
1966 4.4
1967 4.6
1968 4.2
1969 4.2
1970 3.9
1971 3.8
1972 3.6
1973 3.5
1974 3.2
Source: National Science Foundation (1977)
Chemical research activities can also be categorized into defensive,
environmental, product, and process areas. An authoritative breakdown on
R&D spending for these areas is not available. However, it is known that
R&D for new product development has decreased, and that product and
process improvements are becoming the major goal of the industry. (See
section B.3.3.)
The Snell survey of the chemical industry examined 1972 R&D
expenditures. (Foster D. Snell, Inc., 1975) Their survey excluded
product maintenance costs and commercial development expenditures. The
remaining R&D expenditures were divided among new products, 44%; new
applications, 36%; and other, 20% Another source reported the allocation
of R&D expenditures for 1974 as 31% for new products and 51% for new
applications. (Chemical Week, 1975) The differences among these two
sources reflect different time periods, different measurement
conventions, and different sectors of the industry.
197.
-------
B-16
Landau (1979) estimates that recent chemical R&D funds are divided
as follows:
Expenditure
Topic Area Allocation (%)
1978 1979
improving existing products 58 62
new processes 16 20
new products 26 18
pollution control 5 4
energy related 4 3
These data show a 30% drop in new product investment by the industry from
1978 to 1979. Combined with the Snell survey and Chemical Week data,
they indicate that a major decrease is occurring in new product
investment by the chemical industry.
It is possible to conclude from statements by the industry reported
in the trade press and from the importance of new products to small
innovative firms, that large firms, more than small, have been cutting
back on new product development efforts. According to Landau's data,
these funds are being shifted toward product and process improvements
rather than being directed to meet environmental or regulatory demands.
The large diversified chemical firms dominate the statistics on
industrial R&O spending. More than 70% of R&O spending for the chemical
and allied products industries is spent by the top 20 firms; for
industrial chemicals the fraction is greater than 90%. (C&E News, 1979a)
Mansfield (1968) has demonstrated that historically R&D funding in
chemicals has increased more than proportionately with firm size. The
chemical industry is the only major industry in which this trend is
maintained through to the largest firms. In all other industries,
mid-size firms are proportionately more heavily engaged in R&D spending.
In a more recent analysis, Soete (1979) found an increasing relative
R&D expenditure with firm size in chemicals. However, he did not
consider firms with sales of less than $100 million per year, effectively
198,
-------
B-17
omitting small firms from the analysis. Soete linked R&D expenditures
with inventive activity. However, Chemical and Engineering News (1979a)
has pointed out that measures of innovation based on R&D funding:
...provide little insight on the quality of R&D or on the
ill-defined ties between R&D efforts and innovation and
economic development. It is increasingly evident that
scientific and technological problems are not necessarily
solved simply by throwing more money at them; more
research is not necessarily better research.
The model of technological innovation outlined in section B.I shows
that as firms enter a more mature phase, R&D becomes directed toward
improvements in products and processes and away from risky new ventures.
As firms grow more successful they become locked into particular
products. Investments in R&D increase not for new product developments
but for improvements in existing facilities and product lines.
B.3.2 R&D Personnel
Employment of R&D scientists and engineers in the chemical and
petroleum industries is shown in table B.5. Scientists and engineers in
these data are defined as individuals having had at least the equivalent
of a four-year college program in science or engineering The reported
equivalent numbers are adjusted to reflect time devoted only to R&D.
The data show an increase in chemical R&D employment in the late
fifties, reaching approximately 30,000 (excluding Pharmaceuticals and
petroleum) in 1960. From 1960 to 1976 employment ranged from 29,200 to
32,400. It was somewhat higher in 1976, following a period of decreased
employment during the early seventies.
Research employment is at best a rough measure of investment in
research efforts. Furthermore, the productivity and efficiency of
research have increased dramatically since the late fifties due, for
example, to improved instrumentation and to the use of computers.
Researchers today are able to accomplish substantially more than their
199.
-------
TABLE b.5 Number of Full Time Equivalent HKD SUentlsis and Engineers 1957 to 1976
K)
O
o
Industry
Total
Chemicals and
al 1 led products
Industrial chemicals
Druys and medicines
Other chemicals
Petroleum refining
and extraction
Industry
Total
Chemicals and
al lied products
Industrial chemicals
Druys and medicines
Other chemicals
Petroleum refining
and extraction
SIC CODE
28
781-82
283
284-89
29.13
SIC CODE
28
281-82
283
284-89
29.13
Thousands of Employees in January of Each Year
1957
229.4
29.4
IB.O
4.7
6.7
6.9
195«
243.8
31.0
18.8
5.1
7.1
7.4
1959
268.4
33.5
20.2
5.9
7.4
7.7
I960
292.0
36.1
21.8
6.0
8.3
9.2
1961
312.1
37.0
22.9
6.2
7.9
9.0
1962
312.0
36.5
21.6
6.8
8.1
9.1
1963
327.3
38.3
22.9
6.9
8.5
8.9
1964
340.2
37.8
23.6
7-3
6.9
9.0
1965
343.6
40.0
25.7
7.7
6.6
9.7
1966
353.2
40.0
24.7
8.0
7.4
10.2
Thousands of Employees In January of Each Year
196?
367-2
38.7
22.7
9.3
6.7
10.4
1968
376.7
40.8
23.2
10.0
7.5
11.2
1969
387.1
42.2
23.6
10.0
8.3
11.9
1970
384.1
42.1
22.9
11.6
7.6
11.5
1971
359.3
42.5
22.6
12.3
7.6
10.8
1972
349.9
40.9
19.7
11.8
9.5
».3
1973
356.6
40.7
19.8
11.2
9.7
8.2
1974
358.2
41.6
I'J.8
12.0
9.8
8.2
1975
360.8
44.9
22. 1
13.1
9.8
8.4
1976
361.6
46.4
22.4
14.1
10.0
8.9
CO
I
00
Source: NSF (1977)
-------
B-19
counterparts of the fifties and sixties. (Landau, 1979) Finally, R&D is
only one component of the innovation process. Thus, there is no way to
know how the level of R&D employment affects the overall rate of
technological innovation.
8.3.3 R&D Strategies
Industrial chemistry began in the mid-nineteenth century with the
creation of the synthetic dye industry. The major growth in chemicals
for the first 40 to 50 years occurred in Germany and the United Kingdom.
By the turn of the century, the U.S. industry had begun an expansion and
maturing that has continued to the present.
Davies (1978) has depicted three eras of industrial chemistry. The
first period he characterized as "science-push." In this period basic
discoveries led to the introduction of products, universities worked
closely with industry, and growth occurred rapidly. The second period
was characterized as "market pull." Market pull reflects the situation
where market needs are identified and teamed closely with the goals of
research. Research is actively pursued, pay-offs are high, and research
labs become institutionalized in firms. As this period evolved, the
market became a tougher place in which to survive and current and
competitive pressures grew. This led to the third era - "resource supply
and husbandry." In this period efficient use of feedstocks has become
essential, and management control of R&D is more stringent and
skeptical. Research efforts have turned inward in this period in efforts
to improve efficiency. Product and process improvements rather than
product discovery are high priority research efforts.
The chemical industry today is in its third period. The evidence
for this can be seen in trade journals and reports on research. Dupont
among other companies has introduced changes in its corporate planning
and research investments. (Pappas, 1978; Gubitosi, 1979; Burke, 1979)
Changes in the management of R&D are evident in many companies. R&D
management was originally an easy task, there was little pressure to
-------
B-20
prove the value of research. Today management of R&D is more rigid and
project selection is measured against successful products with strong
market positions. (Landau and Brown, 1978; Landau, 1979; Maisel, 1980)
The trade literature reports a major reevaluation of R&D efforts
among the larger companies in the industry. (Gubitosi, 1979; Verespecj,
1979) Efficiency and productivity are critical to corporate survival,
and this efficiency will be achieved through improvements in production
processes. (Verespecj, 1979, Kline, 1978) Old products have top priority
in chemical firms, and innovation is geared toward improving their
characteristics and production. (Chemical Week, 1977) Projects that
promise short pay-off horizons and low risk are being preferred to high
risk investments. (Landau and Brown, 1978)
B.4 Evidence on Trends in Chemical Innovation
The Foster D. Snell study (1975) estimated that up to 1000 new
chemicals are marketed each year. Seven hundred of these are intended
for R&D purposes, leaving about 300 new chemicals of potential interest
for TSCA purposes. In the following, discussion the classes of chemicals
into which these new products fall are described.
To date, a comprehensive study or collection of data to provide a
clear picture of chemical innovation has not been made. An attempt has
been made here to determine active areas of growth in chemicals and some
of the characteristics of that growth, by examining three perspectives:
previous studies of chemical innovation, the first round of
Premanufacturing Notice Data, and chemical patent activity. All of these
sources have limitations.
B.4.1 Previous Studies of Chemical Innovation
A number of studies of particular aspects of chemical innovation are
summarized in Table B.6. Most are older case studies that lack the
recent data necessary to understand the characteristics of chemical
202.
-------
TABLE B. 6 A Suuruary of Studies of Chemical Innovation
Study
Types of Study
Time Period
Measure of
Innovative ActIvly
Source of and Critical
Inputs to Innovation
Role of Small and
large Firing
IO
O
Arthur 0. Little
Impact of TSCA Proposed
Premaiiufacturlng~
Notification Require-
ments (1978)
Boyden. J. W.
A Study of the Innova-
tive Process In tHe
Plastic Additives
Industry (1976)
Enos, J. L.
Petroleum Progress
and Profits (1962)
Foster D. Snell
Study of the Poten-
tial Economic Im-
icts of The
roposed TSCAct
as Illustrated 'by
'
I. Interview
2, Output analysis
3. Patent exaniina-
I ion
Case study of 3 types
of plastic additives
Case study of the
petroleum cracking
process
1973-1978
Post WW II until
the present
1900-1960
Survey and
Interviews
1970-197'!
I, Patent activity
2. pliune survey
3. 25 "neu" chemical*
selected from
Chemical Week
Buyer's Guide
Selected product
outputs and their
achieved market
sales
Process discoveries
In the petroleum
Industry
Product Introductions
by selected firms
within particular
sectors of the indus-
try measured ayalnsl
per dollar sales
Innovation is occurring Not addressed but assumed
in all sectors of the small firms had a major
Industry. It's grow- role
Ing fastest In organiis
(Intermediates t cyclic
crudes)
In commerciali*ing firm
as opposed to the users
of the additives. De-
velopment was In re-
sponse to a perceived
need or new potential
I. Medium and smalI
firm BCD |dl>s
2. Some independent
inventors
3. Strong Individual
played a vlId I role
In every break-
through
Source of Innovation
is In all sectors with
a greater percentage
arising from smalI
f I rms, sma II f I mis
were defined as those
with less llun 30x10°
dollars In sales
The role of each ujs not
specifically addressed but
it was apparent trow the
study that larger firms
were believed to be more
able to Innovate
Small and medium firms
were the risk-takers in
the industry
Small firms in organ Ics
had up to 100 t hues tlie
activity of Ijiyer firing
in inorganics small firm
activity was 5 to 10
t hues greater
CO
i
-------
Table B.6 continued
Study
Type) of Study
Time Period
Measure of Source of and Critical Role of Small and
Innovative Actltlty Inputs to Innovation large firms
to
o
Freeman, C.A.
"The Plastics Industry:
A Comparative Study of
Research and Innovation"
(1963)
Case study of the
plastics Industry
1920-1960
Gibbons, Hlchael
"Factors Affecting
Technological Inno-
vation In British
Industry" (1973)
Greenberg, £•, C.T.
Hill and O.J. New-
burger, Regulation,
Market Prices, and
Process Innovation:
The Case of the Am-
monia Industry (1979)
Jewkes, J., D. Sawers
and R. St I Herman
The Sources of
Invention T.1969)
Case studies In
chemical, electrical
and mechanical
Industries
Case study and
economic mode.) Ing
1917-1972
Case til stories of
major industrial
innovations
ltfS-1969
Research expenditures Large firms dominated Not specifically addressed
patent activity, and
significant Innova-
tions
Products and process
innovations them-
selves
Productivity of each
Input factor, as
Mel I as qua)I tatIve
assessments of new
capabltitles
The actual innova-
tions themselves
R£0; plastic products
and patent*. German &
American doninduce
the Industry sector
Basic research In uni-
versities was a
critical Input tut In-
dustry was greatest
Input.
Discovery push and
market pull both
operate as stl.uu11.
Four types of Innova-
tions: science dis-
coveries, technologi-
cal discovery, cus-
tomer need, management
by objectives. The
market and management
control are best
stimulators of inno-
vation
Innovation from origi-
nal producing firms,
government laborator-
ies, and engineering
design and construc-
tion firms
but the large firms dom-
inated innovation of major
products In this sector
Not addressed
CD
i
Small firms own and operat
dunonla facilities, l>ut
play only a snull role in
innovation. Currently.
most innovation from en-
gineering design firms
Not addressed
-------
1ABU B.6
Study
Types of Study
Time Period
O
Ul
Measure of Source of and Critical Role of Small and
Innovative Activity Inputs to Innovation lanje firms
Langrlsh. J., M. Gibbons
W.G. Evans arid F. R.
Jevons
Wealth from Knowledge
71972)
Lelbhafsky. M.A.
SI Iicones Under the
Monogram (|u,/8)
Mueller, W.F.
"The Origins of the
Basic Inventions
Underlying DuPont's
Major Product and
Procebs, 1920-1950"
(1962)
Robertson, A.6.
Success and Failure
_ln_Industrial Inno-
vation; Report on
Project WPHO
Case studies of United
Kingdom Queen's award
winners for innovation
1966-196?
Case study of the
development of sill-'
cone diG.E. and an
overview of other
Industrial research
effort*
Case study of 18 pro-
duct, and 7 process in-
novation a.t Dul'vm
I974
1920-1950
A paired comparison of
4j successful and un-
successful chan I cat
and scientific Instru-
ment innovations
The products them-
selves and their
value measured as
a Queen' s av/ard
winner
Actual products and
process improvements
Hie actual product
and pioccss iuiprove-
meiits and the value
lo I lie company
Selected success-
ful Innovations
paired with pro-
ject fallures
There are mult I pie
sources of innovation,
but the entrepreneurial
Individual In whatever
setting Is cr11 ical.
There Is no major link
between basic dis-
coveries and technologi-
cal Innovation
The source of innovation
Is the Innovative Indus-
trial research lab whicli
combines independent
discoveries Into major
Innovations
Not addressed
Small firms operate lo
either promote or syri-
llicsi/e disLoveiies of
independent Inventors
13 out of 18 product Large firms arc teller
Improveuicnl s came from al impi ov in
-------
3-24
product innovation as it now occurs. Several of the studies used surveys
and interviews to derive a picture of innovation. The 1975 Foster D.
Snell study is the most recent comprehensive survey of innovation in
chemicals.
Both the Snell study and the ADL study also identified subject areas
of highest chemical innovation activity. The most active areas of
chemical development reported by the ADL study were chemical catalytic
preparations, soaps and detergents, surface active agents, flavors and
perfumes, and synthetic organic chemicals. These trends were determined
by conversations with industry representatives, a review of 25 "new"
chemicals, and an analysis of patent trends. The Snell survey found that
the soaps, plastics, and industrial organics sectors dominate in new
chemical development. Snell based their analysis on survey respondents
who provided a breakdown of product investment by class.
These studies were examined to determine the role of the small firm
in chemical innovation. Totally consistent results cannot be obtained
from these studies, since their definitions of small firms differed and
since several studies described the invention process rather than the
innovation process.
Large firms dominate the production of basic chemicals and play
major roles in finished products. They also control the largest number
of research personnel, and devote proportionately the largest amount of
funds to R&D. (Mansfield, 1968; Mansfield, 1972; Soete, 1979) It is not
clear, however whether large firms are responsible for the greater
portion of significant product innovations in the industry. This is
especially true if significant product innovations are thought of as
those that open up entirely new lines of business.
The largest firms in the chemical industry are typically diversified
into many product lines and businesses (C&E News, 1979a). This has
produced corporations that make R&D decisions based on expected
profitabiliy and assured pay-off. (Landau, 1979) In many cases, these
206.
-------
B-25
firms prefer to make existing products and to use production plants that
are fully depreciated and far down the learning curve, rather than to
risk making new products that require expensive plants and equipment.
New products and new process technologies that do not have a high
probability of achieving a profitable position in the market are
generally avoided. (Landau, 1979)
Small firms in the industry play three roles. First as
non-innovative, marginal producers; second, as non-innovative producers
of specialty chemicals, and third, as innovative producers of high
quality specialty products and intermediates. The third role of the
small firm is of greatest concern in this analysis.
A study of innovation at DuPont by Mueller (1962) showed that 13
out of 18 product innovations originated outside the firm, and that the
real successes of DuPont's research efforts were in making process and
product improvements. DuPont was found to be better at tailoring a
product to market demands and at efficiently producing that product than
it was at discovering new products.
Mansfield (1968, 1972) also found that large firms often develop
products that originally came from sources outside the firm. He
concluded that R&D in large firms is proportionately more productive than
in small ones. For firms with sales greater than $100 million per year,
.Soete (1979) found that the largest chemical firms devote a greater
percentage of their sales to R&D than do the somewhat smaller firms.
However, both Mansfield and Soete limited their studies to what would be
considered large firms for the purpose of the present analysis.
The Snell survey of the industry most closely approximates a
comprehensive analysis of innovation. It surveyed companies that
accounted for 24% of the total sales of the industry in all major
segments of the industry except Pharmaceuticals. The respondents were
divided as follows:
207,
-------
B-26
Sales Range ($/year) Number of Respondents
greater than 1,000 million 4
between 100 million and 1,000 million 27
less than 100 million 14
The role of the small firm was addressed with a sample of 9 firms,
each with less than $30 million in sales in 1972. The Snell study found
that for inorganic chemicals "new product activity appears to be five to
ten times as great among the small companies." In industrial organics,
for which new product innovation is predominately "custom chemical" work,
they found that activity is one hundred times as great in the small
firms. They stated, however, that because of the small number of firms
sampled, these findings are not statistically significant. The studies
of innovation examined in the current trade and professional press make
this doubtful. Small firms play an important, but not fully understood
or quantified, role in chemical innovation. Furthermore, small,
innovative chemica-1 firms have the following characteristics:
o they are more dependent on new products than large firms, (new
products are a larger proportion of sales in small firms),
o they obtain their new product ideas when their customers
describe special characteristic that they require,
o they have larger chemical firms as important customers who use
the small firms' products as production intermediates or
additives,
o they concentrate on specialties, often for a single buyer on a
short term contract basis,
o they depend in a large measure on their ability to respond
quickly to an order to obtain a contract,
o they do not have the staff to carry out toxicology testing and
other compliance requirements,
o they rely on trade-secrets to protect their products,
o they can usually pass on increased material costs to their
customers but may have trouble passing on increased regulatory
compliance costs if such costs fall unevenly on different
products and firms,
208.
-------
B-27
o they may rely on their customers to handle the requirements of
TSCA,
o they do not have the resources to appeal an EPA ruling,
o they fear take-over or loss of their markets if the different
treatment of old and new chemicals under TSCA creates an
incentive to they use of established chemicals in preference to
new, potentially safer ones.
B.4.2 Premanufacturing Notice Data
Premanufacturing notices (PMN's) may become a means for measuring
chemical innovation. Currently, there are too few PMN's to provide
statistically significant data, and, of course, there are no PMN's before
late 1979. The following information is derived from the first 57 PMN's
submitted to EPA. Information is not complete for all 57 because of
unreported data, claims of confidentiality, or inaccessible data. Totals
in some categories exceed 57 since some PMN's report data in more than
one category.
Type of Company Submitting
large (greater than $100 million in sales per year) 26
medium ($50 million to 100 million) 7
small (less than $50 million) 5
subsidiary of larger firm 11
Primary Product of Submitting Firm
plastics13
coatings and resins 16
flame retardants 1
detergents 1
adhesives 1
photographic supplies 1
Initial Production Volume
not reported or claimed confidential 13
less than 500 Ibs. 4
20,000 to 40,000 Ibs. 1
15,000 to 100,000 Ibs. 2
200,000 to 300,000 Ibs. 3
500,000 to 1,000,000 Ibs. 1
2,000,000 Ibs. 2
209.
-------
B-28
Type of Toxicity Data Submitted
acute animal studies24
subacute animal studies 1
eye and skin irritation tests 23
bacterial mutagenicity 12
chronic 0
no data 27
These PMN data may be useful in assessing chemical innovation, but
the data presently being provided to the public is not entirely adequate
for that purpose. (Chemical Week, 1980; Jellinek, 1980)
B.4.3 Patent Data
Patent data have been frequently used as an indicator of innovative
activity, although such data have serious limitations. For example,
Kamien and Schwartz (1975) have noted that many important innovations are
not patented, that the patent owner may not be the original innovator,
and that the patent data do not differentiate between product and process
changes. Furthermore, patenting practices differ by sector and change
over time within a sector. In the chemical industry, trade secrets are
often preferred to patents, although this may more frequently occur for
processes than for products.
These problems all distort the meaning of patent measures so that
they are not an accurate reflection of innovative output. However, as
noted by Kamien and Schwartz (1975), "systematic study of patenting
behavior has led Smookler, Scherer and others to conclude that the number
of patents granted to a firm is a usable proxy for inventive outputs."
If interpreted with care, patent data may be useful in assessing trends
in innovation in a sector over relatively short time periods and in
making gross comparisons between sectors.
Patents have been used recently as a measure of chemical innovation
by ADL and by the U.S. Patent Office. According to ADL's review of
patent activity classified according to SIC codes, industrial organic
chemicals dominate the new chemical innovation process. Cyclic crudes
210.
-------
B-29
and intermediates account for 28% of organic chemical production, which
is about 18% of all chemical innovation.
The U.S. Patent and Trademark Office (1979) has produced data on
patent activity for its Top 50 chemical subclasses ranked by the rate of
growth of the number of patents, the percent growth, and the foreign
share. Actual growth is dominated by Pharmaceuticals; after eliminating
these, growth is largest in organic chemicals (24%) including cyclic
intermediates and synthetic resins. However, the rate of increase of
patents for Pharmaceuticals, which account for the greatest number of
chemical patents, has leveled off. Fifty percent of the fastest growing
areas are carbon compound products - mostly cyclic intermediates.
Process improvements account for most of the remaining activity. The
largest foreign patenting area is in dying polyester fibers.
B.5 Chemical Industry Trends That May Affect Innovation
The chemical industry is presently experiencing major shifts in its
business environment. The industry is responding and adapting to many
changes in its environment of which TSCA is only one. These shifts
include increasing world-wide competition, rapidly escalating feedstock
prices, depletion of feedstocks, and the obligations a major industry has
to society. (C&E News, 1978) The strategies selected by the companies to
cope with these changes will in part, determine their potential for
innovative developments in the future.
Increasing world-wide competition in chemicals originates from two
sources: the large petroleum companies that are diversifying into
chemical production and the oil producing nations that are building
petrochemical plants. (Maisel, 1980; C&E News, 1979b) Some U.S. analysts
have predicted that oil producing countries will link chemical purchases
to oil purchases, and that they may dump ammonia and methanol on the
world market. (Webber, 1979) Two strategies are being adopted by the
U.S. chemical industry in response to these pressures.
211.
-------
B-30
The chemical and petrochemical producers are responding to pressure
from the oil producing nations by seeking greater efficiency through
process improvements for existing products. (Verespecj, 1979) For
example, developments in new catalysts are being pursued. (Burke, 1979)
The emphasis is on running firms efficiently and on planning and managing
research to obtain the best economic gain.
It is expected that the industry will grow more slowly for the
remainder of this century, and that new products will evolve from older
ones with fewer major new products being produced. (Chemical Week, 1977)
Some analysts believe that the growth areas will be in plastics and in
the replacement of natural products by synthetics. (Maisel, 1980)
Specialty products will continue to outperform the rest of the industry,
with emphasis on chemicals for the electronic and energy industries.
(Chemical Week, 1979a)
An economic downturn is currently not a major concern to chemical
company officials since they do not have excess capacity or significant
new production capacity coming on line in the next several years.
(Keifer, 1979) Also, the export market will continue to be profitable
since the major petroleum producing nations are not yet on the scene, and
since U.S. natural gas prices have not climbed as rapidly as petroleum
prices for European and Japanese producers. This feedstock price
difference gives the U.S. a world-wide price advantage. (Business Week,
1980)
The feedstock problem, however, will not go away. Petroleum is
becoming more expensive and scarce; natural gas, although plentiful at
present, will not sustain chemical producers in the U.S. forever.
substitutes for these feedstocks are available and will play a major role
in the industry over the next 30 years. (Landau and Brown, 1978) Coal,
although not predicted to be a major source of chemicals for 20 years or
more, is expected to play the largest role in replacing petroleum and
natural gas. Eastman Kodak has already announced plans to construct an
212.
-------
B-31
acetic anhydride plant using coal as the feedstock. (C&E News, 1980)
Biomass will also be explored both as a feedstock and as a natural source
of complex chemicals.
The chemical industry is also being influenced by the changing
public perception of the obligations of industry to society. This is
reflected, for example, in state and local recognition and regulation of
hazardous material handling, production, and disposal; and in the more
active role taken by public interest groups in legislation controlling
corporate activity that has developed over the last ten years. The
industry is also being affected by increases in product liability suits,
tort suits, workers compensation requirements, and efforts to increase
corporate disclosure.
To respond to these many changes, major capital investments and
improvements in chemical process technologies will be essential. Over
the next 30 years, the chemical industry will adopt new production
processes and will produce improvements on today's products. The future
of the industry will be characterized by slower growth and by more
careful research investment, regardless of regulatory requirements.
B.6 Major Findings
The complex process of technological innovation in industry
involves activities ranging from basic research and invention
to marketing and adoption. Organized R&D is only one element
in the process.
Despite the existence of several partial studies, there is no
authoritative source for the rate, direction, and nature of
chemical innovation.
The rate and nature of chemical innovation can be expected to
vary greatly among the sectors of the chemical industry. Small
firms and new entrants play an important, though largely
unmeasured role in innovation, especially in newer or more
rapidly developing sections.
213.
-------
B-32
Small firms, in sectors of the industry that have low
concentration ratios or that are dominated by large firms, may
encounter more trouble meeting TSCA cost increases than other
f i rms.
As firms become mature their R&D efforts become more
risk-averse, process change becomes more important, and they
face displacement if they do not recognize the need for
continued innovation.
Established chemical firms have demonstrated a shift toward
process change, product modification, and new uses for old
products.
Large firms in the industry have traditionally been best at
product modification and process innovation, but not at product
discovery.
Small innovative firms are more dependent on new products, and
are therefore more likely to be adversely affected if
regulation inhibits the development of new products than are
large firms.
Chemical firms face a variety of challenges including high
energy costs, foreign competition, vigorous entry by oil
companies, a maturing technology and capital, base, and changing
public perceptions of the social responsibilities of industry.
Regulation, and especially TSCA, is only a part of this
challenge.
214.
-------
B-33
References for Appendix 8
Arthur 0. Little, 1978. "Impact of TSCA Proposed Premanufacturing
Notification Requirements." Prepared for the Office of Planning and
Management, U.S. Environmental Protection Agency.
Boyden, J.W., 1976. A Study of the Innovative Process in the Plastic
Additives Industry. M.S. Thesis, Massachusetts Institute of Technology,
Cambridge, MA.
Burke, D.P., 1979. "Catalysts 1." Chemical *eek March 28.
Burke, D.P., 1973. "If It Doesn't Fit Forget It." Chemical Week Feb. 28.
Business Week. 1979. "R&D Spending at 633 Companies: Another Record
Year." July 2.
Business Week. 1980. "Chemicals Exports will be the Bonanza." Jan. 14:
54-58.
Chemical and Engineering News. 1978. "World Chemical Outlook." Dec. 18:
22-53.
Chemical and Engineering News. 1979a. "Facts and Figures for Chemical
R&D." July 23: 31-65.
Chemical and Engineering News. 1979b. "Chemistry in the 1980's." Nov. 26.
Chemical and Engineering News. 1980. "Eastman to Make Chemicals From
uoai." Jan. 14: b.
Chemical Week. 1975. May 14.
Chemical Week. 1977. "Old Products Remain Top R&D Priority." May 25:
39-40.
Chemical Week. 1979a. "Specialties Lure Chemical Market Researchers."
Feb. 21: 40.
Chemical Week. 1980. "Jellinek Pilots TSCA to 'edge of runway.1" Feb. 6:
45-6.
Davies, Duncan S., 1978. "The Changing Nature of Industrical Chemistry."
Chemical and Engineering News March 6: 22-27.
Fienman, S. and W. Fuentevilla, 1976. "Indicators of International Trends
in Technological Innovation." Prepared by Gellman Research Associates for
Directorate for Scientific, Technological, and Internationa1. Affairs,
National Science Foundation. April.
215.
-------
B-34
Enos, J.L., 1962. Petroleum Progress and Profits. Massachusetts Institute
of Technology Press, Cambridge, MA.
Foster D. Snell, Inc.. 1975. "Study of the Potential Economic Impacts of
the Proposed Toxic Substances Control Act as Illustarated by Senate Bill
S.776." Prepared for the Manufacturing Chemicals Association. Feb.
Freeman, C.A., 1963. "The Plastics Industry A Comparative Study of
Research and Innovation." National Institute of Economic Review No. 26,
Nov.
Frost and Sullivan. 1976. "Adhesives and Sealant Markets." New York:
Frost and Sullivan.
Gibbons, Michael, 1973. "Factors Affecting Technological Innovation in
British Industry." Industry Marketing Management Vol. 2, Feb.
Greenberg, E., C.T. Hill, and O.J. Newburger, 1979. Regulation, Market
Prices, and Process Innovation. Boulder, CO: Westview Press.
Gubitosi, J., 1979. "Oow's Oreffice Prescribes a Breather." Chemical
Business Dec. 10: 9-17.
Hill, C.T., E. Greenberg, D.J. Newburger, et a!., 1975. "A State of the
Art Review of the Effects of Regulation on Technological Innovation in
the Chemical and Allied Products Industries: CAPI Project." St. Louis,
MO: Center for Development Technology, Washington University, Feb.
Hill, C.T. and J.M. Utterback, 1979. "Summary and Policy Implications."
in Technological Innovation for a Dynamic Economy, eds. C.T. Hill and
J.M. Utterback. New York: Pergamon Press.
Hill, C.T., 1979. "Technological Innovation: Agent of Growth and Change,"
in Technological Innovation for a Dynamic Economy, eds. C.T. Hill and
J.M. Utterback. New York: Pergamon Press.
Jellinek, Steven D., 1980. Remarks before the Symposium on Regulations
and the Introduction of New Chemicals, 146th National Meeting of American
Association for the Advancement of Science, Jan. 5.
Jewkes, J., D. Sawers and R. StiHerman, 1969. The Sources of Invention.
New York: W.W. Norton, Co., Inc.
Kamien, M.I. and N.L. Schwartz, 1975. "Market Structure and Innovation: A
Survey." Journal of Economic Literature.
Keegan, Mary K., ed. 1977. The Kline Guide to the Chemical Industry.
Faifield, N.J.: Charles H. Kline & Co.
Kiefer, D.W., 1979. "Chemical Firms Face Business Slowdown in 1980." C&E
News, Dec.
216.
-------
B-35
Klein, B., 1979. "The Decline in Productivity Advance: A Dynamic
Explanation." in Technological Innovation for a Dynamic Economy.
C.T. Hill and J.M. utterback, New York: Kergamon Press.
Klein, R.C., 1978. "New Trends in Patents." Chemical Engineering Progress
April: 32-36.
Landau, Ralph and David Brown, 1978. "The Chemical Industry 2000 A.D.."
Chemical Engineering Progress Oct.: 27-31.
Landau, Ralph, 1979. "Chemical Industry Research and Innovation."
Presented at "Innovation and U.S. Research," a Symposium of the American
Chemical Society. Sept. To appear in the ACS Symposium Series.
Langrish, J., M. Gibbons, R.G. Evans, and F.R. Jevons, 1978. Wealth from
Knowledge. New York: Halstead Press Division, John Wiley & Sons.
Leibhafsky, H.A., 1978. Silicones Under the Monogram. New York: John
Wiley & Sons.
Maisel, D.S., 1980. "Trends in the Petrochemical Industry." Chemical
Engineering Progress Jan.: 17-23.
Mansfield, E., 1968. Industrial Research and Technological Innovation.
New York: U.Vi. Norton"To^
Mansfield, E., 1972. Research and Innovation in the Modern Corporation.
New York: W.W. Norton~UcH
Mueller, W.F., 1962. "The Origin of the Basic Inventions Underlying
DuPont's Major Product and Process Innovations, 1920-1950." in The Rate
and Direction of Inventive Activity. New Jersey: Princeton University
cress.
National Science Foundation, 1977. Research and Development in Inoustry,
Surveys of Science Resource Series,
Office of Management and Budget, 1972. Standard Industrial Classification
Manual 1972. Statistical Policy Division, Washington, u.u
Pappas, N., 1978 "Corporate Planning at DuPont." Chemical Engineering
Progress June: .---42.
Randall, Frederick, P., 1972. "Corporate Strategies in the Drug Industry:
A Study of Strategic Response and Political Pressures." Ph.D.
Dissertation. Cambridge, MA: Harvard University.
Robertson, A.B. Success and Failure in Industrial Innovation: Report on
Project SAPPHO. -
Sanders, H.J., 1978. "Flame Retardants." C&E News April 24: 22-28.
217.
-------
8-36
Soete, Luc L.G., 1979. "Firm Size and Inventive Activity." European
Economic Review ^2: 319-340.
The Economist, 1977. "Small and Beautiful." Dec. 17.
U.S. Bureau of the Census. 1972 Census of Manufacturers.
U.S. Patent and Trademark Office, Office of Technology Assessment and
Forecast. 1979. Technology Assessment and Forecast 9th Report,
Washington, O.C.
Verespej, M.A., 1979. "Future Focus." Industry Week April 16: 69-80.
Webber, David, 1979. "Methanol: A New Wrinkle or Two." Chemical Business
Oct. 15: 57-64.
Wilsey, K.C., 1979. "A New Life for Coal in Chemicals." Chemical Business
Nov. 12: 63-70.
218.
-------
APPENDIX C:
DETAILED LIST OF CRITERIA FOR ASSESSING
INDIVIDUAL REGULATORY PROGRAMS
1.0 Criteria I — Capacity to Countervail - Capacity of Alternative to Off-
set the Unwanted Impacts of TSCA Regulation
1.1 Criteria 1-1 — Amount of Resources Directly or Indirectly Allo-
cated - In Relation to Size of Unwanted TSCA Impacts
1.2 Criteria 1-2 — Locus of Countervailing Impact in Relation to Locus
of TSCA-Related Impact
1.2.1 Sectoral Locii
1.2.1.1 General Chemical Industry
1.2.1.2 Selected Firms (e.g., small firms or firms with
history of low volume production)
1.2.1.3 A Selected Firm (e.g., one chosen with certain
characteristics for a test case)
1.2.2 Chemical Locii
1.2.2.1 Class of Products (e.g., all catalysts)
1.2.2.2 Chemical Compounds (e.g., all synthetic resins from
phenols or phenoxides with resinifiable amine or
amide)
1.2.3 Firm and Market Cost/Impact Loci?
1.2.3*1 Premanufacturing Notification Costs
1.2.3.2 Testing Costs
1.2.3.3 Product Liability Costs (or savings)/Chemical Safety
1.2.3.4 R&D Costs
1.2.3.5. Delay Costs
1.2.3.6 Other Manufacturing Costs
1.2.3.7 Market Valuation of Safety and Cost Passthrough
219.
-------
C-2
1.2.3-8 Rate of Return/Profit on Regulated/Unregulated
Chemicals
1.2.3-9 Market Structure/Concentration (rates of entry and
exit)
1.2.3.10 Factor Prices (production factors)
1.2.3-11 Workforce Costs
1.2.3-12 Productivity decreases (increases)
1.2-3.13 Innovation (and attendant firm/market effects)
1.2.4 Public Cost/Impact Locii (welfare economic effects)
1.2.4.1 Employment (jobs)
1.2.4.2 Safety/Health
1.2.4.3 Changes in Prices due to Changes-in Market Structure
1.2.4.4 Changes in Prices due to Changes in Productivity
1.2.4.5 Changes in Prices Availabli1ity due to Changes in
Innovat ion
1.2.4.6 International Trade/Comparative Advantages
3 Specific Considerations Related to the Class of Alternative (four
general classes of alternatives which have distinctly different
countervailing characteristics)
1.3-1 Preventative - Alternative Acts to Negate the Effects of
TSCA Related Regulations by Bringing About Changes in the
Public Sector, to Change the Impact of the Regulations Them-
selves. (For example, if regulations of new entities occurred
such that existing products were overly favored, a preventa-
tive alternative would be one that redressed the balence in
the regulation of new and old chemicals.)
1.3.2 Direct Remedial - Acts to Negate a TSCA Related Effect by
Directly Countervailing that Effect Within the Private Sec-
tor (e.g., if lower revenues "cause" a decrease in innovation,
subsidizing RSD budgets within the firms constitutes a direct
mechanism).
220.
-------
C-3
1.3-3 Indirect Remedial - Aid Indirectly Offsets Impacted Regula-
tion. (It may be difficult of directly countervail against
a negative impact of regulation and more facile to aid some
other aspects of the chemical industry where the aid indirect-
ly offsets the regulative impact.
1.3-4 External Remedial - Mechanisms which Act to Ameliorate the
Welfare Impacts of TSCA Effects without Trying to Prevent the
Private Sector Changes which Give Rise to Them. (E.g., lower
revenues result in loss of employment. Increased employment
benefits for chemical workers constitutes an externally acting
program.)
2.0 Criteria 2 — Cost Effectiveness
2.1 Criteria 2-1 -- Degree of Reliance on Market Mechanisms
2.2 Criteria 2-2 — Size of Bureaucracy Needed to Administrate
2.3 Criteria 2-3 -- S?ze of Other Transaction Costs (e.g., private ad-
ministration costs)
2.4 Criteria 2-k — Lag Time and Startup Costs
2.5 Criteria 2-5 — Amount and Accuracy of Information Needed for Coor-
dination — How automatic is the mechanism
2.6 Criteria 2-6 — Extent to Which Resources can be Diverted from Tar-
get
2.7 Criteria 2-7 — Extent to Probable Effects on Target Area due to
Leveraging
3.0 Criteria 3 ~ Feasibility
3.1 Criteria 3-1 — Administrative Feasibility -- Public, Private
3.1.1 Budget Constraints
3-1.2 Personnel Constraints
3.1.3 Time Constraints
221.
-------
C-4
3.2 Criteria 3-2 — institutional Feasibility — Public
3.2.1 Legal Authority
3.2.2 Existing Modus Operand! and Focus of Regulatory Agency
3.2.3 Private Sector Behavior
3.2.4 Linkages with Other Institutions
3.2.5 Precedents
3-3 Criteria 3-3 ~ Political Feasibility
3.3.1 Feasibility of Obtaining New Legislation (if required)
3-3.2 Public and Industry Reaction to Legislation
(a) Level of Expenditures
(b) Restriction of Freedom, Other Rights
4.0 Criteria 4 -- Uncertainty and Risk
4.1 Criteria 4-1 — Possibility that Purpose may be Thwarted
4.2 Criteria 4-2 — Reliance on Unpredictable Behavior
4.3 Criteria 4-3 — Possible Delays/Litigation
4.4 Criteria 4-4 -- Probability of Political/Legisilative Modification
4.5 Criteria 4-5 — Number of Programs with Similar Target (size of
portfolio)
5.0 Criteria 5 ~ Equity
5.1 Criteria 5*1 — Distribution of Benefits Among Firms
5.2 Criteria 5-2 — Distribution of Costs Among Chemical Industry/Consu-
mers/Taxpayers
5.3 Criteria 5-3 — Distribution of Benefits (or Costs) Between Chemical
Producers and Supply Factors (i.e., multi-market equity)
222.
-------
C-5
6.0 Criteria 6 -- Dynamic Effects
6.1 Criteria 6-1 — Opportunity Costs of the Government Action - What
Other Goals and Programs will have to be Foregone as a Result of
Adopting the Program
6.2 Criteria 6-2 — Will the Program "Over Compensate" -Will the Stimu-
lus be so Large that the Original Safety and Health Aim of TSCA
Regulation is Thwarted
6.3 Criteria 6-3 — Serendipitous Effects -Will the Program Produce
Socially Desirable Changes in Other Areas
6.k Criteria 6-*t — Leveraging - Capacity to Produce Systemic, Long-Run
Positive Effects
6.5 Criteria 6-5 — Complementarity - To What Extent does the Program
Mesh with Other Regulatory Goals, Both of the EPA and Other Agencies
6.6 Criteria 6-6 — Other Long-Run Welfare Consequences
223.
-------
APPENDIX D
SOURCES USED TO IDENTIFY CANDIDATE POLICY OPTIONS
Abernathy, W.J., and Ginsberg, D.H., eds., Government, Technology, and
the Future of the Automobile, McGraw-Hill, Inc., New York, 1980.
American Bar Association, "Federal Regulation: Roads to Reform," prepared
by the Commission on Law and the Economy, Washington, D.C., August 5,
1978.
Ancker-Johnson, B. and D.B. Chang, "U.S. Technology Policy, prepared by
the Office of the Assistant Secretary for Science and Technology, U.S.
Department of Commerce, March 1977.
Boucher, W.I., et al., "Federal Incentives for Innovation," prepared by
the Denver Research Institute for the Natonal Science Foundation, Denver:
University of Denver, January 1976.
Cole, R.J. and P.O. Tegeler, "The Impacts of Government Requirements on
Small Busiess in Washington State," Battelle Human Affairs Research
Centers, Seattle, Washington, March 1979.
Center for Policy Alternatives, "National Support for Science and
Technology: An Examination of the Foreign Experience," Vol. 1, prepared
for the National Academy of Sciences, Cambridge: Massachusetts Institute
of Technology, May 15, 1976.
Center for Policy Alternatives, "Environmental/Safety Regulation and
Technological Change in the U.S. Chemical Industry," prepared for the
National Science Foundation, Cambridge: Massachusetts Institute of
Technology, March 1979.
Charles River Associates, "Subsidies, Capital Formation and Technological
Change," Boston, Massachusetts, 1978.
Charleswater Associates, Inc., "The Impact on Small Business Concerns of
Government Regulations That Force Technological Change," prepared for the
Small Business Association, Washington, D.C., Septembeer 1978.
Gerstenfeld, A., "Innovation: A Study of Technological Policy,"
Washington: University Press of America, 1977.
Hill, C.T., ed., Federal Regulation and Chemical Innovation, American
Chemical Society, Washington, D.C., 1979.
Hill, C.T. and J.M. Utterback, eds., Technological Innovation for a
Dynamic Economy. Pergamon Press, New York, 1979.
Industrial Research Institute, Research Management 22, November 1979.
224,
-------
D-2
The Institute for Innovation Enterprise, "Incentives for Innovation: A
Study of Innovation in Maine Industry," prepared for the National Science
Foundation, Portland: University of Maine, November 1976.
Kitti, C. and C.L. Trozzo, "The Effects of Patents and Antitrust Laws,
Regulations, and Practices on Innovation," Vol. 1: A State of the Art
Review, prepared by the Institute of Defense Analysis for the National
Science Foundation, Washington, O.C., February 1976.
Noll, R.G., et al., "Government Policies and Technological Innovation,"
Vol. 11-A and 8, State of the Art Surveys, prepared by the California
Institute of Technology for the National Science Foundation, October 1974.
Office of the Chief Counsel for Advocacy, U.S. Small Business
Administration, "Small Business & Innovation," Washington, D.C., May 1979.
Office of Technology Assessment, "Government Involvement in the
Innovation Process," prepared for the U.S. Congress by the MIT Center for
Policy Alternatives, 1978.
Office of the White House Press Secretary, "The President's Industrial
Innovation Initiatives," Press Release, October 31, 1979.
Robbins, M.O., et al., "Federal Incentives for Innovation," Parts I and
II, Denver Research Institute: University of Denver, January 1976.
Rubenstein, A.M., et al., "Management Perceptions of Government
Incentives to Technological Innovation in England, France, West Germany,
and Japan," prepared for the R&D Incentives (RDI) Program of the National
Science Foundaton, Research Policy, Vol. 6, 1977, pp. 324-357.
Schnee, J.E., "Government Programs and the Growth of High-Technology
Industries," prepared for the RDI Program of the National Science
Foundation, Research Policy, Vol. 7, 1978, pp. 2-24.
Science Policy Research Unit, "The Current International Economic Climate
and Policies for Technical Innovation ," prepared for the Six Countries
Programme, Sussex: University of Sussex, England, November 1977.
Toxic Substances Strategy Committee, "Report to the President by the
Toxic Substances Strategy Committee," Washington, D.C., August 1979.
U.S. Congress, House, "Capital Formation and Retention," Report of the
Committee on Small Business, 96th Congress, 1st session, 1980.
U.S. Congress, House, "FIFRA Extension," Hearing before the Subcommittee
on Department Investigations, Oversight, and Research of the Committee on
Agriculture, 96th Congress, 2nd session, 1979.
U.S. Congress, House, "Government Patent Policy," prepared for the
Subcommittee on Science, Research and Technology of the Committee on
Science and Technology, 95th Congress, 2nd session, May 1978.
225.
-------
D-3
U.S. Congress, House, "Productivity and Technical Innovation," Hearing
before the Task Force on Inflation of the Committee on the Budget,
Subcommittee on Science, Research, and Technology of the Committee on
Science and Technology, 96th Congress, 1st, session, 1979.
U.S. Congress, Senate, "Small Business and Innovation," Report of the
Select Committee on Small Business, 96th Congress, 1st session, 1979.
U.S. Department of Commerce, National Bureau of Standards, "Government
Procurement as an Incentive to Commercial Technology and Innovation,"
prepared for the Experimental Technology Incetives Program, Washington,
D.C., March 1973.
U.S. Department of Energy, "The Demonstration Project as a Procedure for
Accelerating the Application of New Technology," (Charpie Task Report),
Vol. 1, February 1973.
Wilson, A.H., Government and Innovation, prepared for the Science Council
of Canada, Ottawa, Canada, 1973.
« U. 8. GOVERNMENT PRINTING OFFICE : 1980 341-06S/M10
226.
-------
Sample C. Technical Report Data Sheet. EPA Form 2220-1
TECHNICAL REPORT DATA
{Please read Instructions on the reverse before completing!
i. REPORT NO.
EPA-560/12-80-002
2.
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Supporting Innovation: A Policy Study
5. REPORT DATE
9/80
6. PERFORMING ORGANIZATION COOS
7. AUTMORIS)
8. PERFORMING ORGANIZATION REPORT NO.
Christopher T. Hill, Richard A. Andrews, et
alf Center for Policy Alternatives. MIT
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Center for Policy Alternatives
Building W59-204
MIT
Cambridge, Mass 02139
10. PROGRAM ELEMENT NO.
B2CL2S
II. CONTRACT.GRANT NO.
Contract No. 68-01-5878
12. SPONSORING AGENCY NAME ANO ADDRESS
EPA/OTS/ETD
Regulatory Impacts Branch
401 M St., S.W.
Washington. D.C. 2Q46Q
13_T.YPE OF REPORT ANO PERIOD COVERED
final
(TS-794)
14. SPONSORING AGENCY CODE
IS. SUPPLEMENTARY NOTES
16. A8STRAC
tQT,
The autors conclude, from a review of the theoretical and
empirical literature and analysis of its application-to the chemical
industries, that the impact of TSCA on innovation is not predictable
For a number of reasons, TSCA is as likely to stimulate innovation
in some sectors as it is to discourage it in others. There are not
enough reliable data to separate the effects of TSCA from historical
trends and other factors.
The report recommends a cluster of six policies (chosen from
a group of thirty-three that were considered) that could be used
together to offset some of the negative impacts on innovation if the
government decides this is warranted. The recommended policies are:
—EPA dissemination of chemical information (in the form of test
results or labelling);
—Instituting generic pre-manufacturing notifications for certain
classes of new chemicals;
—Government support for developing cheaper and more reliable tes
methods;
(cont. on back of page)
17.
KEY WORDS ANO DOCUMENT ANALYSIS
DESCRIPTORS
b.lOENTIFIERS/OPEN ENDED TERMS
COSATI Field'Croup
N/A
N/A
N/A
18. DISTRIBUTION STATEMENT
Release Unlimited
19. SECURITY CLASS llhli Report I
21. NO. Or PAGES
226
20. SECURITY CLASS iTfiapaftl
22. PRICE
EPA Form 2220-1 (*-73|
12
-------
(Continuation of No. 16 (Abstract))
—A subsidy for testing or compliance costs for new chemical
development, either through a grant or a loan program;
—"Fast track" pre-manufacturing reviews for safe chemicals or
major innovations;
—Government support for education and training of toxicologists
and related professionals.
------- |