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.

-------