FINAL REPORT                       MAY 15,1971
    AN ECONOMIC MODEL SYSTEM FOR
       THE ASSESSMENT OF EFFECTS
       OF AIR POLLUTION ABATEMENT


       DEVELOPMENT AND DEMONSTRATION PHASE
                 Prepared for:

             OFFICE OF AIR PROGRAMS
          ENVIRONMENTAL PROTECTION AGENCY
 by

-------
FINAL REPORT
AN ECONOMIC MODEL SYSTEM
FOR THE ASSESSMENT OF EFFECTS
OF AIR POLLUTION ABATEMENT
Volume I:  The OAP Economic Model System Development
          and Demonstration
Prepared for:

Office of Air Programs
Environmental Protection Agency
Air Pollution Control Office
Prepared by:

CONSAD Research Corporation
121 N. Highland Avenue
Pittsburgh, Pennsylvania  15206

-------

-------
                              PREFACE
      To economists,  optimal resource allocation decisions require
knowledge of the costs, effects and benefits of alternatives.  Because
the real world is complex,  models or simulations are used in economic
analysis to test the outcome of alternatives strategies.

      This study by the Office of Air Programs and CONSAD Research
Corporation represents the first systematic effort to determine the in-
direct economic effects of air pollution control costs and to  test the
possible results and implications of some alternative approaches to the
sharing of costs of control among producers,  consumers and the  public
generally.

      Because it is a new modeling effort and the data inputs somewhat
old, the results must  not yet be considered anything but illustrative of
the potential capability of economic modeling on such a wide scale.
The present exercise  utilizes engineering estimates of costs of control
and makes specific assumptions regarding the level of benefits  attain-
able from air  pollution control and the manner in which those benefits
are introduced to and  experienced by the system.

      It will quickly be obvious to the careful  reader that better control
cost estimates are needed together with a more flexible method of intro-
ducing benefits  if the model is to  come  closer to describing  reality.
The model obviously requires new information on the facts of economic
life and of the control requirements, technology and costs implicit in
the Clean Air  Act Amendments of 1970.

      It will be  for the users and  future developers of the model to sup-
ply the necessary improvements needed in order to make it a usable
contributor to policy analysis and formulation.  Despite its limits, how-
ever,  we  believe that  this economic modeling effort represents a most
important first step and that it should be viewed at in that light.
Paul H. Gerhardt
Chief Economist
Office of Program Development
Office of Air Programs

-------
                          ACKNOWLEDGEMENTS
     The research reported here has been carried out by CONSAD Research Corporation
for the Office of Air Programs* Environmental Protection Agency (EPA), under subcon-
tract to TRW S.ystems, Inc. The CONSAD team that developed and demonstrated in a
preliminary manner the Regional Economic Model System include:

             Robert J. Anderson              Consultant
             Robert F. Byrne
             Penny Globus
             T. R. Lakshmanan               Project Director
             Fu-Chen Lo                     Project Manager
             Kathryn C. Mason
             Arthur Silvers
             Nat Simons, Jr.
             John M. Thompson, Jr.
             Venkaiah Yedla

     CONSAD could not have prepared the analytic and conceptual tasks without the
encouragement, guidance, and review of the staff of APCO and TRW, in particular:

             APCO Staff                     TRW Staff

             Allen Basala                     Michael Frankel
             Larry Barrett                   Donald Lewis
             Ronald Campbell
             Paul H. Gerhardt (Project Officer)
             Henry Kahn
             Ken Woodcock

     Any opinions expressed in this report are those of CONSAD and do not neces-
sarily reflect the views of the individuals cited above.
*Formerly Air Pollution Control Office

-------
                         TABLE OF CONTENTS
                                                              Page

EXECUTIVE SUMMARY                                         ix

1. 0   AIR POLLUTION AND THE
      ECONOMY:  AN OVERVIEW                                  1

      1. 1  The Context                                            1
      1. 2  The APCO Economic Model System                      10

2. 0   THE RAP A PROGRAM AND THE
      REGIONAL ECONOMIC MODEL SYSTEM                     13

      2. 1  Purpose of the RAPA Program                          13
      2. 2  The Role of the Economic Model in RAPA                16
      2. 3  The APCO Economic Model System (Phase III)          20

3. 0   MODEL DEVELOPMENT
      AND FORMULATION                                       27

      3. 1  The Model in General                                 27
      3.2  The APCO Economic Model System                     30
          3.2. 1  The Regional Model                            30
          3.2.2  I-O Model and Interregional Feedback           36
      3.3  Empirical Estimation                                 40

4. 0   PRELIMINARY STUDY OF A MODEL
      FOR NATIONAL ECONOMIC EFFECTS
      ASSESSMENT                                             43

      4. 1  The Transition from Regional to
          National Policy Analysis:  Background                  43
      4. 2  The OBE Quarterly Model                             46
      4. 3  Desired Attributes in a National Model                   52
      4. 4  Data Availability and Reliability                        56
      4. S  Alternative Approaches to

-------
                         TABLE OF CONTENTS (continued)
5. 0   SIMULATING THE IMPACT OF AIR
      POLLUTION CONTROL STRATEGIES:
      A SIMULATION OF THREE STRATEGIES                    61

      5. 1  Introduction                                            61
      5. 2  The Scope and Nature  of Simulation
          of the Regional Model                                   62
          5. 2. 1  Nature of Simulation in General                  62
          5.2.2  APCO  Policies  and the Regional Model            64
          5. 2. 3  Regional Model System and Control Inputs         66
      5.3  Simulating Three Strategies:
          The Definition of Strategies                             70
      5.4  Preparing Inputs for Simulation                         82

6. 0   INTERPRETATION OF THE EFFECTS
      OF THREE TRIAL STRATEGIES                             93

      6. 1  The Three Strategies                                   93
      6. 2  An Approach to  Assessment of
          Effects of the  Three Strategies                          95
      6. 3  Changes in Unemployment Rates
          Under the Three Strategies                              97
      6.4  Changes in Profits  and Personal Income                106
      6. 5  Total Net Effects of Three Alternative
          Strategies for 91 AQCRs                               117
      6. 6  Concluding Comments                                  122

7. 0   AN ASSESSMENT OF THE  APCO
      REGIONAL ECONOMIC  MODEL SYSTEM:
      PROMISES AND PITFALLS                                125

8. 0   REFINEMENT AND UTILIZATION OF
      THE APCO ECONOMIC  MODEL SYSTEM:
      RECOMMENDATIONS                                      135

APPENDIX  A:   A MODEL TO ASSESS THE ECONOMIC
                EFFECTS OF AIR POLLUTION ABATE-
                MENT IN ST. LOUIS AQCR (PHASE I)           A. 1

-------
                        TABLE OF CONTENTS (continued)
                                                            Page

APPENDIX B:  31 AQCR MODEL (PHASE II)                    B. 1

APPENDIX C:  THE APCO REGIONAL ECONOMIC
               MODEL SYSTEM (PHASE III)                    C. 1

APPENDIX D:  INPUT-OUTPUT MODEL SYSTEM
               AND INTERREGIONAL FEEDBACK              D. 1

APPENDIX E:  DATA USED IN THE MODEL                    E. 1

-------
                          LIST OF FIGURES
Figure                                                         Page

 1. 1       Short-Term Effects of Standards                         3
 1.2       Long-Term Effects of Standards                         6
 2. 1       Regional Air Pollution Analysis Process                 14
 2.2       The APCO Economic  Model System:
           Development and Use                                  22
 3. 1       Economic Modelling for Policy Simulations              28
 3. 2       Major Components of the Model                         32
 3. 3       Notation of the Regional Model Variables                 33
 3.4       The Regional Model Formulation                        34
 3. 5       Regional Model  -- A More Detailed Look                 37
 4. 1       Effect of Tax Credit Policy Upon
           Implementation Plan                                  48
 5. 1       APC Policies and Model Inputs                          67
 5.2       Regional Model  and Control Inputs                       68
 5.3       Regional Model  and Control Policies:
           A More Detailed Look                                 71
 5.4       Supply Schedule                                         73
 5. 5(a)    Shift Due to Increase  in Disposable Income              74
 5. 5(b)    Shift Due to Increase  in Production Costs                74
 5. 6       Air Pollution Control Cost Per Unit Output              75
 5. 7       Price Freeze Demand Schedule                          78
 5. 8       Demand and Supply Relations Under Strategy 1           79
 5. 9       Demand and Supply Relations Under Strategy 3           81
 6. 1       The Three Strategies  at a Glance                        94
 6. 2       Geographic Distribution of Economic  Effects
           Under Strategy 1 (Measured by Change of
           Unemployment Rate)                                 102
 6. 3       Geographic Distribution of Economic  Effects
           Under Strategy 2 (Measured by Change of
           Unemployment Rate)                                 104
 6.4       Geographic Distribution of Economic  Effects
           Under Strategy 3 (Measured by Change of
           Unemployment Rate)                                 105

-------
                         LIST OF TABLES
Table                                                         Page

 5. 1       Incidence of Control Costs Under the
           Three Strategies                                      90
 6. 1       Sample Output of an AQCR:  A Summary Table           96
 6. 2       Change of Unemployment Rate from Simulation
           of Three APC Strategies                               98
 6. 3       Changes in Unemployment Rate, Profits and
           Personal Income  in the AQCRs under Strategy 1        108
 6.4       Changes in Unemployment Rate, Profits and
           Personal Income in the AQCRs under Strategy 2        111
 6. 5       Changes in Unemployment Rate, Profits and
           Personal Income  in the AQCRs under Strategy 3        114
 6.6       Median Values of the Distributions of the Changes
          in Unemployment Rate,  Percentage Changes in
          Profits and Personal Income in the Three Strategies     117
 6. 7       Net Effects of Three Alternative Strategies
          on 91  AQCRs                                          119

-------
EXECUTIVE SUMMARY







1.    INTRODUCTION







      In sustaining rapid population growth, high levels of consumption,




technological innovation,  and a quickening pace in the use of resources,




Americans have been hard on their environment -- the vegetation,  land,




air,  and water that sustain the biotic and industrial processes.  As a




result, clean air and clean water have become scarce  resources.




      Pure or relatively unpolluted air is no longer a free good in our




society; money outlays  must be made  to go where air is relatively




cleaner or to trap pollutants before they escape  into the air.  In the




past, those who were responsible for  pollution did not bear  the external




costs they generated to the entire society.




      With the recent enactment of the Clean Air Act of 1967 and Amend-




ments of 1970, business and industry  arfe now being required to control




the amount of pollutants that they discharge into the air.  To industry,




this  requirement means that the production costs for the same amount




of output produced prior to the legislation will be increased in propor-




tion to the required  investment in air  pollution control equipment.




Thus, certain industries and regions that have in the past enjoyed the




economic advantage of  low-cost production may face some degree of




economic decline  due to the  requirements of pollution abatement.

-------
      There would be, however, offsetting economic effects from air




quality control strategies in such regions in at least two ways.  First,




there would be increased demand for the products of the industries  that




produce pollution control equipment and low pollutant fuels leading to




increased output, employment, and income in those sectors. Second,




a variety of general economic benefits result from pollution abatement.




Some of those gains are increased labor productivity, and property




values,  reduced health expenditures,  reduced outlays on physical main-




tenance of homes and plants, and savings in agricultural production




activities.   These gains would lead to increased consumption, output,




employment and income, and would begin to offset the economic reduc-




tions resulting from the pollution control costs.




      Consequently, air pollution abatement leads to changes in eco-




nomic output, labor markets,  the availability of capital, as well as re-




distributions within the entire economy.   Further, the implementation




of air quality programs  would have a variety of other effects, such as




tax base impacts on communities or variations in land-use and  indus-




trialization in various regions.




      The Air Pollution Control Office (APCO) of  the Environmental




Protection Agency (EPA) commissioned CONSAD  Research Corporation,




under subcontract to TRW Systems, Inc. ,  for the development  and




demonstration of such an economic model as an operational analytical

-------
tool for abatement policy assessment.  This economic model initiated

in 1968 as part of the Regional Air Pollution Analysis (RAPA) program

focused on the regional economies of the various Air Quality Control

Regions (AQCRs).  As such, it will be useful for assessment of control

strategies at the  regional level, where most of the implementation plans

are prepared.  However, there is an increasing interest in air pollution

control policy decisions in the last three years at the Federal govern-

ment level.   Therefore, there has been correspondingly an increased

focus  in the model system described here toward the assessment of in-

terregional and national economic effects in addition to regional effects.

However, the model system developed by CONSAD is still  a regional

economic system with capabilities to measure, in a limited manner,

the interregional and national effects.   It is called the  APCO Regional

Economic Model  System and has been demonstrated to some degree as

an operational tool for control strategy assessment.
2.    THE APCO REGIONAL, MODEL
      SYSTEM AND THE RAPA PROGRAM
      The APCO Regional Model System was conceived as part of the

RAPA program in July 1968.  To accomplish the task of developing

timely abatement strategies requires an extensive examination of the

factors involved in the air pollution system, such as meteorology, air

-------
pollution control technology, air pollution growth trends, source emis-




sion inventories, existing regional air quality conditions, and regional




economic impact.   Figure 1 outlines the RAPA process and the role  of




the APCO Economic Model System there.




      In the first year (Phase I), CONSAD developed a Regional Econ-




ometric Model of the St.  Louis region where the RAPA was explored




first in depth.  This model describes the growth patterns of key eco-




nomic sectors  (both high emission and other industries) and estimates




the regional product, employment, capital stock and investment change,




and value-added by industry, tax receipts and regional unemployment.




These estimates are sensitive to a variety of air pollution control strat-




egies. In fact,  the economic effects of five hypothetical air quality con-




trol strategies were simulated and interpreted, using  the model for  the




St.  Louis region.




      The second year (Phase  II) witnessed an extension of the model to




31 large  metropolitan areas.  These large urban areas have a varied




industrial structure highly representative of the national industrial com-




position and comprise a significant segment of national output.  The




structure and outputs of this model are very similar to those  of the




St.  Louis model.




      In the current year (Phase III), this model system has been ex-





-------
                                       FIGURE I
                    REGIONAL AIR POLLUTION ANALYSIS PROCESS
Goals (Air
Quality or
Emission
Standards)
Abatement
Strategy
Controls
Sources
Diffusion
Receptors
Control Cost
and Benefit
(Damage)
                                  Implemen-
                                  tation Plan
  Economic
  Effect
*| (APCO Eco-

-------
embrace 100 AQCRs, at the  same time providing a broader range of





economic effects than has been possible so far.  Second, the Phase III




model provides the capabilities of capturing the economic effects of




abatement strategies in one region on all other regions  and the struc-




ture changes taking place in  the national economy.




      The Regional Model System as currently developed in Phase III




thus covers a significant portion of the AQCRs in the nation and treats




them all as  a set of interrelated regions.  Ib is a  large, complex model




which is an operational tool.  Its workability has  been demonstrated as




shown in Figure 2.  Three strategies specified by the APCO  staff and




relating to the incidence of control costs  (on the high emission indus-




tries, consumer or government), associated with the air quality stand-




ards  of the  1970 Report to the  Congress under Section 305(a) of the




Clean Air Act of 1967, were fed into the economic model and the simu-




lation program run to develop  the economic effects.




      The rest of this report will provide a brief  summary of the devel-




opment, demonstration, and assessment of the regional model system.







3.    THE MODEL SYSTEM:  AN OUTLINE







      The Regional Economic Model System has  been developed as  de-




scribed above in three phases.  At the end of Phase II,  the regional




model was structured as though AQCRs were economically independent

-------
     Inputs

Model Development
Model
Outputs
Assessment
Regional and
National
Economic
Activity


Regional
Economic
Model
System
Policy Assessment
                             1
Three
Strategies
(APCO
Specified)


Computer
Simulation
Program


Outputs
Economic
Change s


Interpretation
of Model
Output
                   FIGURE 2

THE APCO REGIONAL ECONOMIC MODEL SYSTEM:

-------
of one another.  There was no allowance for interregional effects.  If,




for example, Region A instituted air pollution control and in order to




do so imported air pollution equipment from Region B,  the model sim-




ulated the economic impact of Region A's program on Region A alone,




and not  on Region B.  Thus, the regional model gave no indication of




the increase in employment in Region B, resulting from the  increased




production of air pollution control equipment for export to Region A.




There was,  however,  a source of pessimistic bias in the  statement of




economic effects of air pollution abatement embedded in the  very struc-




ture of the model.




      Second, the models at the end of the second  year focused only on




the economic impacts  of control expenditures  and  accounted  in no way




for any  benefits which might result from air quality improvement.  This




again tended to cause unjustifiably pessimistic conclusions about the




economic effects of air pollution control.




      During the current year,  CONSAD approached the problem of




eliminating  these biases and making  some preliminary assessment of




the national impact of  air pollution control in two ways.   First, the




cross-section Regional Model was restructured to eliminate, insofar




as possible,  the pessimistic bias  induced by structural exclusion of




interregional feedback effects and benefits and to permit  preliminary




estimates, a national Input-Output (I-O) model has been introduced to

-------
capture (a) a preliminary estimate of structural changes in the national




economy, and (b) an interregional feedback scheme to the AQCRs.




      Second, the feasibility and desirability of an expressly national




model to be interconnected with the current regional model was inves-




tigated in considerable detail.




      These developments are both described next.




      The APCO Economic Model System consists of two major com-




ponents, namely, a 100 AQCR regional model and interregional feed-




back from a national I-O model,  as shown in Figure 3.  The  modules




comprising both these major components are also indicated in Figure  3.




A generalized description of the model system appears  in the text.




      The concept of "export-base" or economic base theory has tra-




ditionally been the  central guiding concept in the description of urban




economies.   In line with this concept, this  study treats  manufacturing




industries as export-oriented industries.  The growth of the  manufac-




turing industries leads to the growth of the regional economy.  This




model structure is  consistent with the familiar Keynsian-type trade




multiplier in an open economic system.   Consequently,  in Figure 3,




the manufacturing industry module feeds into the regional economy,




and growth of manufacturing industry and regional economy determine




the regional employment in the regional labor market.  By the same




token, production and  consumption activities are related to the demand

-------
                       FIGURE 3
          MAJOR COMPONENTS OF THE MODEL
I-O Model and Inter-
regional Feedback
Regional Model
1
1
1
1
1
1
1
I
1
1
1
1
1
1
1
1
1
1
1
1
1
1

National
I-O Model
1
J
f
Regional
Market
Share
Matrix

1
1
i
I
\
1
i
1
1
1
1
*
1
1
1
1
l
l
i
i
i
i










Manufac-
turing
Industries
1

Regional
Economy
Income
Consumption
Government

Regional
Labor
Market

_ — - 	 	 __.._ |
I
i
1
1
I
1
\ 1
\ i
\
| 1
	 Electricity ; |
I and Fuel i
[Demand '
V i
/ 1
/ i
/ \
r \
1
1
1
t
1

-------
for electric power and fuels.  All these relationships have been formu-




lated into a mathematical model.




     In the manufacturing module,  the production relations of the man-




ufacturing industry are presented.  Output is related to the production




factors: labor and capital.   Gross profit is given as residual between




value-added and wage bill.  Investment behavior is related to the profit




and capital stock from the previous period.  With new investment ex-




penditures and adjustment of depreciation, capital stock of present




capacity is  determined.  Finally, employment in manufacturing sectors




is derived from the level of production and wage level.




     In the regional  economy module, regional income is determined




by the  level of manufacturing production,  regional consumption expen-




ditures and local government expenditures.  On the other hand, regional




consumption is related to the regional income,  and government expen-




ditures are related to the government revenues.




      In the regional  labor market module,  employment by industry,




other than manufacturing industry,  is derived from the income  gener-




ated from non-manufacturing sectors. Regional employment,  which is




the summation of employment from manufacturing and non-manufactur-




ing industries, and unemployment rate, determines the regional labor




force in each AQCR.

-------
      Finally,  electricity and fuel demand modules provide the demand




of electricity and fuel derived from production and consumption activity




of the region.  A substitutional relationship between different types of




fuel is also included.




      A national I-O system is  introduced to serve the role of external




market for the regional economy described  in the regional model, and




also,  hopefully, to measure the structural change of the national  eco-




nomy attendant upon air pollution control in the 100 AQCRs.




      A more detailed look at the regional model which comprises 162




equations is  provided in the text.




      As a part of Phase III of the RAPA project,  CONSAD also inves-




tigated the possibilities and potentials of adding another model to the




Regional Economic Model System for the express purpose of assessing




the national economic effects of air pollution control.  CONSAD planned




the use of the Office of Business Economics (OBE) Quarterly Econo-




metric Model to make preliminary estimates of national economic




effects.  An intensive study of the structure of the OBE model, how-




ever, revealed a number of structural properties which rendered it not




an entirely desirable tool for national effects assessment.

-------
4.    TRIAL SIMULATION OF THE
      MODEL SYSTEM
      A computer simulation program of the APCO Economic Model

System was developed and a user's guide was provided in Volume II of

this report.  The Regional Model System, utilizing this simulation pro-

gram, can trace the effects of a variety of policy tools such as stand-

ards or incentives available to APCO, provided the latter are converted

into inputs constant with the model logic.

      APCO specified three alternative strategies to be tested with the

APCO Economic Model System as their basis the control costs envis-

ioned in the 1970 Report to the Congress  as required by Section 305(a)

of the Clean Air Act of 1967.  In this report,  cost estimates  were made

of controlling the emissions of selected pollutants within 100 AQCRs

during the period fiscal years  1970 through 1975.  The costs reflect

the emission reductions of particulates,  sulfur oxides, hydrocarbons

and carbon monoxides in these  100 AQCRs by 1975. *
      *Fogel, M.  E. ,  et^ aL , Comprehensive Economic Cost Study of
Air Pollution Control Costs for Selected Industries and Selected Re-
gions, Research Triangle Institute, February,  1970, Chapter 4.

-------
                                                        Emission
Pollutant                Source Category                Decrease (%)

Particulates            Solid waste                        77. 8
                        Stationary combustion             91.7
                        Industrial process                 86. 1

Sulfur  oxides            Stationary combustion             52.2
                        Industrial process                 36. 2

Hydrocarbon            Solid waste                        69.4
                        Industrial process                 57. 8

Carbon monoxide        Solid waste                        84. 7
                        Industrial process                 90. 2

The  simulation runs interpreted in this  report  reflect differing control

strategies with respect to  the incidence of  control cost  expenditures --

that  is, with respect to who finally pays the cost of controlling air pol-

lution.  In Strategy 1,  high emission industries pay.  In Strategy 2,

they are able to pass part  of the cost on to consumers as a price in-

crease.  In Strategy 3, the government  subsidizes 50 percent of the

burden the industry picks up in Strategy 2.  In  all three strategies,

electric utilities are assumed to be able to pass on the  control costs

in the form of price increase to the users

      Table 1 presents the sample summary table for one AQCR under

the "Industry Pays" strategy.   First,  a key economic indicator, change

in unemployment rate, is  analyzed in various AQCRs.  This shows that

there is a considerable difference between the  three alternatives in

terms  of incidence of  adverse effects.   The geographic distribution of

-------
                                      TABLE 1
                SAMPLE OUTPUT OF AN AQCR:
A SUMMARY TABLE
MANUFACTURING INDUSTRIES

Profit (millions)
Investment (millions)
Value Added  (millions)
Capital Stock (millions)

OTHER INDUSTRIES

Employment  (1000 S)

Regional Consumption (millions)
Total Personal Income for the Region (millions)
Total Regional Employment (1000 S)
Regional Unemployment  (percent)
Total Labor Force  (1000 S)
Government Expenditure for the Region (millions)
Government Revenue from the Region (millions)

ELECTRIC POWER DEMAND
                                                         Without
                                                         Control
        5738.90Z
         452.000
       13836.699
       10884.758
        3979. 71
                      Net
                      Change
 •33. 051
 •20. 394
  5.273
 •19.132
- 2.866
               Percent
               Change
-0.57591
-4.51190
-0.09929
-0.56227
-0.07202
32445.000
53712.000
5123.711
4.000
5337. 199
5944. 000
6416.000
-30. 164
-41.488
- 9.293
0.1741
- 9. 660
- 3.929
- 4. 165
-0.09297
-0.07724
-0. 18137
4.35349
-0. 18100
-0.06610
-0.06491
Total Electric Consumption for the Region (1000 KWS)      4878.000
Electricity Used by Manufacturing Industries (1000 KWH)    518.000
Electricity Used by Other Industries  (1000 KWH)           3205.000
Residential Consumption in the Region (1000 KWH)         1155.000
                       6. 364
                       2.736
                       2. 205
                       1.423
               -0.13047
               •0.52817
               •0.06881

-------
all 91 AQCRs show that most of the AQCRs seriously (adversely)

affected are located in the highly industrialized north-central (Michigan,

Ohio, Indiana,  Illinois) and central-east (Pennsylvania, West Virginia)

states.  AQCRs located in the west and south, in general, do not seem

to'be affected by air pollution control and some are even better off.

The  following table shows that adversely affected regions improve from

Strategies  1 to 3.

Categories of Change of                  AQCRs Included  (%)
Unemployment Rate (%)
in AQCRs	             Strategy 1  Strategy 2  Strategy 3

1. Better off (increase in
   employment)                        5.5%       7.7%        9.9%
2. Negligible (0.01% to
   1.00%)                            67.0%      69.2%       83.5%
3. Moderate (1. 01% to
   2.0%)                              19.8%      19.8%        3.3%
4. Serious (2. 01% and
   over)                               7.7%       3.3%        3.3%

   Total                             100.0%     100.0%      100.0%

      When two other economic indicators -- profits of manufacturing

industry and regional personal income -- are also analyzed.  The pat-

terns of economic change among AQCRs are  similar to that of unem-

ployment.  However, if one observes the total effects of 91  AQCRs

under the three strategies, the pattern is entirely different.

      The view of cost sharing which emerges when results for individ-

ual regions are considered is thus quite different from  that  which

-------
emerges from solely aggregative considerations.  While cost sharing




results in overall worsened economic performance,  the number of re-




gions severely affected adversely by air pollution control is reduced.




There appears to be a policy trade-off to be made.   The policy-maker




must decide whether he wishes to affect a few regions adversely to a




considerable degree while maintaining good overall economic perform-




ance, or slightly reduce overall performance so that a few regions may




be spared severe hardship.  It should be remembered that the results




are preliminary and presuppose a particular cost sharing scheme.  It




is not the only one which could be undertaken.




      These results on differential regional impact are  suggestive in




many veins,  but one in particular deserves comment.  Some regions




evidently have an economic capacity to meet more rigorous emission




and air quality standards  than do others.




      Some caveats on the structure,  data inputs,  and use of the model




system are in order at this stage.  Structurally, the APCO Model Sys-




tem  is a cross-sectional regional model; consequently,  its strength lies




in its assessment of geographical patterns  of change in  the AQCRs.  Its




estimates of aggregate changes are less reliable.  Second, the 91




AQCRs used in the  model covers the greater portion of  the economic




activity in the nation,  but not all.   Finally, the model system does not




capture the dynamic effects and macro-effects that only a national

-------
model can handle.  Consequently,  the interpretation presented below




must be viewed with a dose of caution.




      The data inputs used in demonstrating the model have limitations




that must be considered.  First, the two billion dollar estimate  of total




annual benefits from control was selected primarily because it was just




large enough to offset the direct costs of control and not as an accurate




reflection of the actual value  of damages reduced.  In this sense, the




value of benefits may be considered quite conservative.  Second, the




control cost estimates for each AQCR were drawn from the 1970 Cost




of Clean Air Report to the Congress.  As such, the control cost esti-




mates related to implementation of previous  legislation and involved a




number of assumptions regarding availability and price of various con-




trol technologies.   Furthermore, the price elasticity  of demand for the




products of high emission industries that were applied in testing strategy




alternatives 2 and 3 were based heavily on qualitative  information as




distinguished from detailed product demand studies.




      In conclusion, it must be observed that the trial simulation of




three strategies and the interpretation of results has demonstrated to





a limited degree the operational nature  and strategy assessment poten-




tial of the model system.  Surely,  what has been done here is but




scratching the  surface of the  utilization potential of the model system.




What has been  done, however, clearly points to the view of the regional

-------
economy and environmental system (quality levels) as interrelated sys-




tems,  and that control activities affect profoundly, in many dimensions,




economic activity.  However,  these economy-environment relationships




are more complex than evident in this demonstration.  This suggests




the need for further efforts in model utilization, sensitivity testing,




and model system refinement  as appropriate.  In such an effort of




further utilization of the model system,  it is hoped that the possible




extension of the model to other media -- water and solid waste -- will




receive consideration.

-------

-------
1. 0   AIR POLLUTION AND
      THE ECONOMY:
      AN OVERVIEW
1. 1   The Context


      The recent upsurge of public concern over environmental ques-

tions reflects a  recognition that man has been too cavalier in his rela-

tions with nature.  This concern has manifested itself in a spate of

legislation in the last few years on abatement of air, water, noise, and

other kinds of pollution.

      In the field of air pollution in the last few years, two parallel

trends are discernible.  On the one  hand, the Federal administration

has proposed and Congress has passed a Clean Air Act of 1967 and

Amendments to  Clean Air Act of 1970.  This legislation and adminis-

trative activity is evidence  of the increasing public pressures for

cleaner air.  On the other hand,  concerns about dire consequences of

such actions  are repeatedly and increasing frequently expressed by

business and community.

      The fears  of industry in terms of adverse economic consequences,

though expressed often extravagantly, are not baseless.  Cleaner  air

is not free and high emission industries or regions incur costs that

affect their output and profits and the regions in which they are located


-------
      In this regard, it may be useful to explore the complex chain of




interrelationships between economic activity and various abatement




policies such as setting emission standards or achieving the standards




through incentives and government expenditures on research and devel-




opment.





      The sshort-term (0-5 years) chain of events following the  promul-




gation of emission standards is fairly clear in form if not in magnitude.




Imposition of standards will result in investment in control equipment,




in process changes, and in outlays for non-capital inputs (e.g. , low




sulfur fuel, maintenance)  used in air pollution control.  As a result of




these outlays,  emission and ambient air concentrations  will be reduced.




Finally, production  and consumption activities sensitive to ambient air




concentration of pollutants will change.  This chain of interdependent




events is  represented in Figure 1. 1.




      Figure 1. 1 highlights the interdependence between the environ-




ment and  the economy.  Control outlays directly influence production




of air pollution control inputs.  In addition, implementation plans may




directly change production and consumption patterns,  as for example,




do local plans  prohibiting  the sale of non-returnable containers or




phosphate detergents.  Indirectly, control outlays influence production




and consumption in two ways: (1) by shifting aggregate  supply schedule





-------
                          FIGURE  1. 1
         SHORT-TERM EFFECTS OF STANDARDS
                          APCO
                          Designated
                          Standards
                         Implementation
Production
and
Consumption
Activities
Ambient
Air
Quality
Control Outlays
Emission

-------
changing air quality which will shift both aggregate supply and aggre-




gate demand schedules.  Moreover,  these changes  in production and




consumption will feed back on control outlays and on emissions since




level of production and consumption influence both control cost needed




to meet any given standard and also to influence the level of emissions.




The system is thus almost completely interdependent.  The medium by




which the Air Pollution Control Office (APCO) actuates a chain of events




within the system is by establishing standards and by supervising their




implementation.  Not only will the level at which standards are met be




important but also the rate at which they are applied.




      The longer-term effects (5 years +) of control standards are even




more pervasive.   Spurred on by standards,  there will be an effort to




look for cheaper  ways to meet emission control requirements.  An




accelerated private  research act development effort should, over the




longer haul,  result in significant changes in production functions, which




changes will influence both demand and production, control cost outlays,




and environment  quality.




      One greatly important longer-term effect of National Standards




which deserves preliminary discussion in this report is the effect of




differential treatment of technologies of different vintages in the setting




of performance standards.  In Section 112 of the Clean Air Amendments





-------
sources that "contribute significantly to air pollution which causes or




contributes  to the endangerment of public health or welfare. "  APCO




is  not limited in establishing such standards to pollutants fortwhich




criteria and control technology documents have been published.  It may




thus happen that older sources of some pollutants will go uncontrolled,




while their  newer counterparts will be subject to control standards.  It




may also happen that controls required of new sources will be costlier




than those required of existing sources by state and local governments.




The  net result will be to make new plants and equipment more expensive




relative to existing plants  and equipment, and, therefore, to  retard in-




vestment in new plants and equipment,  feince new investment in the




manner in which new, more productive,  technology frequently comes




in  line,  national economic growth could be slowed.




      Long-term effects of National Standards Policies are schemati-




cally presented  in Figure  1.2.




      In the longer term,  virtually every facet of modern life may be




influenced,  to a greater or lesser extent, by the  character of standards




adopted.




      The use of Incentive Policies  to implement standards differ from




other implementation methods (coercion and moral evasion) in that it




attempts to use  the economic mechanism itself to solve the pollution





-------
                  FIGURE 1.2

   LONG-TERM EFFECTS OF STANDARDS
APCO
Standa rds
Rate of
Investment
in New
Technology
                                   Research
                                   and
                                   Development
Economic and

-------
economic policies since Incentives will be effected either through tax-

ation or expenditures.  In either event, there will be a marked effect

on government finance.

      As a part of stepped-up air pollution control activity, all levels

of government will be increasing their expenditures for air pollution

control.  Some of this increased expenditure will be directly for con-

trol of emissions from governmental facilities.  The majority,  how-

ever, will be  expended in administering air  pollution control programs.

What effect such expenditures will have on the economy depends  pri-

marily on how they are financed. If air pollution control agency expen-

ditures are financed at the expense  of other  governmental expenditures,

the net effect  is likely to be small or non-existent. *  If,  however,  air

pollution control agency expenditures  are  additional expenditures,  over

and above what would otherwise  be undertaken, they would have a mul-

tiplier effect on Gross National Product (GNP).

      In summary, it appears that there are wide  and far-reaching

economic consequences  of abatement strategies expressed in whatever

form -- standards, incentives,  research and development,  and so forth.
      *No reason to think multiplier associated with APC expenditures is

-------
      Indeed,  there are signs in the Clean Air Amendments of 1970




itself and in the companion water pollution control legislation that the




President and the Congress are already beginning to worry about ad-




verse economic consequences.  Particularly with regard to the pace




at which implementation of the various standards established under




the Amendments is to proceed,  the Environmental Protection Agency





Administrator is directed to consider practicability,  a major dimen-




sion of which  is, undoubtedly,  the implementation costs  and the result-




ing economic  effects.




      Thus, it appears that as pressures for cleaning up the 
-------
      To get on with the business of cleaning up in an economically




humane fashion,  a way must be found to separate  the well-founded from




baseless fears, and to tailor policies to mitigate insofar as possible,




the undesirable consequences of air pollution control policies.




      Virtually all of the economic  policy questions  concerning air pol-




lution control which have been raised over the last four years,  and




which are continuing as a central concern, are questions of fact.  Vir-




tually everyone agrees that pollution control is, by itself, a good thing.




Virtually everyone agrees that unemployment,  high  and rising prices,




falling profits, and a falling growth rate are by themselves bad things.




The disagreement arises when the extent to which good things,  "air




pollution control strategies, " contribute to bad things,  "unemployment,




high prices,  etc. , " is considered.




      Because these disagreements concern observable facts, it is in




concept possible to think of an economic-environmental model that




would identify and estimate these relationships.  One could then manip-




ulate air pollution control strategies in the model, observe the result-




ing employment, price, output,  etc. , changes holding  all else constant,




and infer something about the direction and magnitude  of the economic




impacts  of air pollution control  strategies.




      The development and demonstration of such an Economic Model




for assessment of abatement strategies for Air Quality Control  Regions





-------
1.2   The APCO Economic
      Model System
      An economic model is a simplified abstract version of the real

world built on concepts of economic theory.  A model is more than a

hypothesis; it is a set of functional relations between the various ele-

ments of which an economy is composed.  The  crucial economic ele-

ments -- wages,  capital, investment, income,  consumption, employ-

ment and so forth -- are identified and relationships between them are

postulated.  For  the purposes of any particular policy assessment,

certa.'.n  factors are assumed to be constant and the consequences of

postulated relationships between policy and key economic indicators

worked  out for a  variety of assumptions.  The notion behind this pro-

cedure is that the many facets of economic life constitute a system and

intervention such as by air  pollution abatement strategies causes a

variety  of measurable economic effects.

      APCO commissioned  CONSAD Research Corporation, under sub-

contract to TRW  Systems, Inc. ,  for the development and demonstration

of such  an economic model  as an operational analytical tool for  abate-

ment strategy assessment.   This economic model initiated in 1968 as

part of the  Regional Air Pollution Analysis (RAPA)  program focused on

the regional economies of the various AQCRs.  As such, it will be use-

ful for assessment of control strategies at the regional level, where

-------
most of the implementation plans are prepared.  However, there is an




increasing interest in air pollution control policy decisions in the last




three years at the Federal government level.  Therefore, there has




been correspondingly an increased focus in the model system described




here toward the assessment of interregional and national economic




effects in addition to regional effects.  However,  the model system




developed by CONSAD is still a regional economic system with capabil-




ities to measure,  in a limited manner, the interregional and national




effects.  It is called the APCO Regional Economic Model System and




has been demonstrated to some degree as an operational tool for con-




trol strategy assessment.




      The rest of the report is devoted to a description of the context,




scope, and operation in the use of this  model system.

-------

-------
2. 0   THE RAPA PROGRAM AND THE
      REGIONAL ECONOMIC MODEL SYSTEM
2. 1   Purpose of the RAPA Program


      In July of 1968, the APCO initiated a systems analysis of regional

air pollution control.  It was clear,  at the outset, that this study's con-

tribution to the pollution problem's solution would be  in the integration

of contemporary air pollution control developments into a workable

analytical tool, rather than in fundamental research areas.  With this

in mind, a tool was developed -- Regional Air Pollution Analysis

(RAPA) --to demonstrate the usefulness of looking at the many facets

of the air pollution problems.  With this tool,  more precise  estimates

of the problem can be made,  and more  importantly,  estimates can be

made of specific approaches  for the solution of the problems.

      The RAPA program is  a system of mathematical models arranged

in a modular fashion and relating both engineering and economic effects

of the analysis.  Relations between the  major components of the system

are described  in Figure 2, I.

      Information on the effects of air pollution is reported in terms of

air pollutant criteria, which are a compendium of today's knowledge of

scientific findings on the range of adverse effects of specific air pollu-

tants  and combinations of pollutants on man and his environment.   The

-------
                                    FIGURE 2. 1
                    REGIONAL AIR POLLUTION ANALYSIS PROCESS
Goals (Air  '
Quality or
Emission
Standards)
Abatement
Strategy
1



Controls


Implemen-
tation Plan
,
f

Sources
Diffusion
Receptors





Control Cost
and Benefit
(Damage) 1








Economic
Effect
(APCO Eco-
nomic Mode])




-------
evidence in the criteria documents, with respect to human health,




animal health,  plant damage, material damage and visibility,  are not




necessarily the lowest levels of exposure below which there are no




adverse effects; however, they do provide quantitative  guidance in the




setting of regional air quality standards.




      Air quality standards that are developed with the guidance of




these air quality criteria are goals established for the  protection of




public health and welfare.  They provide a basis for controlling exist-




ing sources of  pollution emission and preventing future regional growth




from adding to the pollution problem.  Regional goals may reflect more




than one air quality standard,  insuring minimum air quality levels,  as




well as  higher  levels of air quality, to preclude any significant deteri-




oration of existing high air quality levels.




      On the other hand,  engineering and economic information on con-




trol of air pollution is reported  by the Federal government in terms of




handbooks on available control devices such as filters and precipitators,




and non-device control measures such as the use of low-sulfur content




fuels or alterations to basic industrial processes that inherently cause




less pollution.




      The government's role starts with setting air quality standards




which reflect goals for clean air within a specified time period.  After




the air quality  standards  are set, an  effort is made to establish

-------
implementation plans which may set forth regulatory procedures, such




as pollutant source emission  standards to achieve air quality standards.




Limiting pollutant emissions  through source emission standards, along




with other types of regulatory procedures such as zoning regulations or




fuel restrictions,  forms an abatement  strategy designed to achieve  re-




gional air quality within a specified time period.




      To accomplish the task of developing abatement strategies will




require an extensive examination of the factors involved in the air pol-




lution system,  such as meteorology, air pollution control technology,




air pollution growth trends,  source emission inventories, existing re-




gional air quality conditions,  and regional economic impact.




      The Regional Economic Model was expressly developed as a way




to respond to these requirements of information on economic impacts




of abatement strategies.







2. 2   The Role of the Economic Model in RAPA







      The recently enacted Federal air pollution abatement legislation




requires business and industry to control the amount of pollutants that




they discharge into the air.  To industry, this requirement means that




the production costs for the same amount of output produced prior to




the legislation will be  increased in  proportion to the required invest-




ments to air pollution  control equipment. Thus,  certain  industries

-------
and regions that have in the pa,st enjoyed the economic advantage of low-




cost production may face some degree of economic decline due to the




requirements of pollution abatement.




      There would be offsetting economic effects from air quality con-




trol strategies in  such regions in at least two ways.  First, there would




be increased demand for the products of the industries that produce  pol-




lution control equipment and low pollutant fuels leading to increased




output,  employment,  and income in those sectors.  Second, a variety




of general economic benefits  result from pollution abatement.  Some




of those gains are increased labor productivity, reduced health expend-




itures,  reduced outlays  on physical maintenance of homes and plants,




and savings in agricultural production activities.  These gains would




lead to increased  consumption, output,  employment and income, and




would begin to offset  the economic reductions resulting from the pollu-




tion control costs (as discussed above).




      Consequently, air pollution abatement leads to changes in eco-




nomic output, labor markets,  the availability of capital,  as well as re-




distributions within the  entire economy. Further,  the implementation




of air quality programs  would have a  variety of other effects, such as




tax base impacts on communities or variations  in land-use and indus-




trialization in various regions.

-------
      CONSAD Research Corporation has developed a regional economic

model that will provide pollution abatement policy-makers capabilities

to assess the effects of various pollution control strategies.   The

CONSAD model is intended to provide the following types of information

for public-policy analysis:

            .  Regional economic changes (e.g., output,  in-
              vestment, employment, income,  and consump-
              tion) expected to result from enforcement of
              varying abatement standards upon high-emis-
              sion industries.

            .  Regional economic effects expected to  result
              from reduction of industrial damage and growth
              in air pollution equipment industries.

              Fiscal effects of regional implementation of
              air quality control programs,  including tax
              base impacts of economic change and the
              effects of tax credits upon economic change
              and the rate  of achievement of emission
              standards in terms of the implementation plan.

      With alternative abatement strategies at hand,  not only associ-

ated direct control costs of each strategy, but also the measurement

of the  corresponding economic impacts of chain reaction throughout

the region,  will also provide useful information to the different levels

of decision-making for the  implementation plan.   The changes in major

economic indicators are the focal point in the analysis.   For  example,

given that control costs directly increase production costs, then as

profits decline, new investments may be reduced.  Therefore, demands

-------
of the products from other firms  and industries, and demands for labor




are affected,  resulting in changes in regional income and unemployment




levels.  On the other hand, those industries  (and individuals) that orig-




inally suffer damages and pay additional costs due  to the polluted air




may start to enjoy some  economic gains from the air pollution control.




Meanwhile, fuel substitutions might occur in some firms,  especially




those in the electric  power industry. The demands for pollution  control




equipment will also result in a new series of product demands within




and outside the area.  Basically,  there is a shift in economic  structure




of the region.




      Abatement strategies not only cost more to the high emission in-




dustries,  but also result in an initial decline in regional employment




and income.  Decision-making is required in the public sector to weigh




these adverse consequences.  The trade-off  between the total  costs and




benefits to the entire regional community may depend upon the informa-




tion from these measured economic impacts of alternative control




strategies.   Tax policies and techniques for  financing and distributing




air pollution control costs can be determined through a full examina-




tion of the total impacts  to the community.




      During the first year of the RAPA program,  CONS AD developed




a model of the economic  effects of air pollution control  (Regional




Econometric Model) of the St. Louis metropolitan  region based on

-------
time series data. * During this second year of the program, the aim

was to focus on the development of a model which can be applied to the

major  cities in this country.  This resulted in estimation of a cross-

sectional economic model for the  31  largest Standard Metropolitan

Statistical Areas (SMSAs) across  the country.  The roles of Keynsian

theory and regional export base theory will be central as they were  in

the St. Louis time series model.  However, the entire equation system

has been reformulated in order to integrate the cross-sectional struc-

ture  into the model.

     In the current year (Phase III), the 31 AQCR Model System has

been extended further and utilized for policy assessment.  The nature

and scope of this extension and utilization of the Regional Model is de-

scribed in the next section.


2. 3   The APCO Economic Model System  (Phase III)


     In the current year (Phase III), two  types of activities have been

carried out:
      *CONSAD Research Corporation, Structure Requirements of an
Econometric Model of St. Louis SMSA, prepared for TRW Systems,
January,  1969,  and Final Version of the CONSAD Regional Econometric
Model as  Applied to the St.  Louis SMSA for Air Pollution Control Analy-
sis,  May,  1969.  (A summary of these reports appears in Appendix A. )

-------
            .  Further model development, and

              Preliminary demonstration of model use
              and assessment of three trial strategies.

These activities are outlined in Figure 2,2.

     Model Development

     The APCO Regional  Economic Model developed in the third year

(Phase III) by CONSAD is a logical extension of the concepts and em-

pirical work of the first  two  years  into an operational tool amenable to

assessment of various control strategies. In particular,  it builds on

and extends the Phase II 31 AQCR Model in three important dimensions.

     First,  the Phase III  model is extended to cover 100 AQCRs.

These 100 AQCRs account for over 65 percent of the national output.

Thus, the APCO model will  cover a major portion of the economic
                                         i
activity in the nation.

     Second, the  Phase III model overcomes a major  drawback of the

31 AQCR  model,  which treats each AQCR as an isolated region. In

reality, economic effects attendant on pollution abatement are not

localized.  Price increases  in certain industries,  the growth of con-

trol equipment industries or demand spurred by reduced expenditures

on health, etc. ,  in one region will have feedback effects on  the  econ-

omies of other regions.  The Phase III APCO Economic Model System


-------
t\»
                Inputs	
           Model Development
*•           Model             •»
 Regional Economic Model System
      Outputs
                                                                                                Assessment
            Regional Economic
            Activity
            National Economic
            Activity
                                               Regional
                                               Model
         Interregional
         Feedback
                                            I-O Model and
                                            National Effects
                                            Assessment
           Policy Assessment
            APCO Policy Variables
             Standards
             Incentives
             Subsidies
             Fuels
              Research and
              development
              Government
              expenditures
     Computerized Simula-
     tion Program

        Program RMS
     .   Program FEE
     .   Program IOA
Outputs:  Changes in:
  Manufacturing
  activity
.  Power and fuel
  consumption
.  Income
.  Employment
.  Government
  expenditures
.  Qthe r
                                                             Interpretation
                                                             of Model
                                                             Outputs
Model. System
Assessments
                                      FIGURE 2.2

             THE APCO ECONOMIC MODEL SYSTEM:  DEVELOPMENT AND USE
                                                             Recommenda-
                                                             tions on
                                                             Further Refine-
                                                             ment and  Utili-
                                                             zation of Model

-------
share matrix to capture these interregional feedback effects and en-




compasses the 100 AQCRs as an interrelated system of regions.




      Third,  the Phase III model incorporates a regional fuel demand




submodel as a component of the Regional Economic Model System.  As




air pollution control policy is implemented, sulfur content in coal and




fuel oils will greatly affect their price in view of increased demand for




low sulfur fuels and their limited supply.  Prices of natural gas and




electricity will also tend to change.  Industries will  choose an optimal




combination of fuels and electricity which minimizes the  total cost of




energy,  to the degree  substitution is possible.  The  fuel demand model




will describe these relationships.




      Figure 2.2 shows the three modules •-- Regional Model, Inter-




regional Feedback,  and the National I-O Model --of the Regional Model




System that is described in detail in the next chapter. It should be




noted a link is needed  to a national macro model  to capture the over-




time and structure change of national effects.  Such  a link to the Office




of Business Economics (OBE) Quarterly Model was  assessed for feasi-




bility in this regard.  This effort forms the theme of Chapter 4.




      Policy Assessment




      A simulation program consisting of three modules corresponding




to the three modules of the Regional Model System was prepared.  This




program is a flexible, efficient tool  to estimate changes in the  regional

-------
and national economies when alternative abatement strategies are spe-




cified.  The control strategies have to be translated into variables con-




sistent with the model structure before the simulation can be initiated.




Such specification and use of the three control strategies specified by




APCO for input into the model is described in Chapter  5.




      Chapter 6 identifies a wide range of model outputs  resulting from




the three strategies.  It also develops an approach to organize in a pre-




liminary way these outputs for analysis.  Finally, a tentative interpre-




tation of the complex economic effects of the three strategies is pro-




vided.




      An assessment of the model  structure, data and simulation pro-




gram forms the theme of Chapter  7.




      Chapter 8 explores the utilization potential of the APCO Regional




Economic Model System and advances recommendations  for further





-------

-------
     Inputs
Mode 1 Development
       aT E c oja^S
     - i t y  x' x ' •-•
,Xa t io j\a 1 JE c on'Qjaft ic
.Activity   ' - x '
          -Model             •«.
:gional Economic Model System
                                                                Outputs
                                  XModcl
                                                ^
                                                 1
                                  ^    ,
                                 ^ nt eXr e g i o na 1
                                 Feedback
                                 	___J.__^_.^
                                 I-Q Mode>xa,rid
                                 •National" Effects
                                 Assessment
Policy Assessment
 APCO Policy Variables
   Standards
   Incentives
   Fuels
   Research and
   development
   Government
   expenditures
   Computerized Simula-
   tion Program

     Program RMS
   .  Program FBE
   .  Program IOA
Outputs:  Changes in:
.  Manufacturing
   activity
.  Power and fuel
   consumption
   Income
   Employment
   Government
   expenditures
_..  Other	
                                                                                       Interpretation
                                                                                       of Model
                                                                                       Outputs
                                                                                       Model System
                                                                                       Assessments
                                                                                       Recommenda-
                                                                                       tions on
                                                                                       Further Refine-
                                                                                       ment and Utili-
                                                                                       zation of Model

-------
3. 0   MODEL DEVELOPMENT
      AND FORMULATION
3. 1   The Model in General


      A model is essentially a representation of a real world phenom-

enon devoid of those elements deemed irrelevant or unimportant for

the  purpose at hand.  A model consists then of  variables embedded in

mathematical formulae (structural relationships), numerical constants

(parameters), solution methods (algorithms), and processes used for

establishing the values of parameters (calibrating procedures).  A

variable is a measurable quantity in which one  has some interest and

which varies from observation to observation.  Variables are to be

distinguished  from a parameter which essentially describes  the struc-

ture.  Parameters change slowly over time (in response to a structural

change or break) and in this  sense are "less  variable variables. "

      An econometric  model generally consists of a system of  equations

embodying a prior reasoning of economic theory.  This reasoning is

given empirical content by estimating and testing the  model through

statistical methods.

      In brief, an econometric model for policy simulation purposes

includes the following  logical steps as shown in Figure 3. 1.

-------
 Econor.ijc
  Theory
Test of the
  ReEulv.s
  Policy
Simulation
 Decinion
 Making
                        FIGURE 3. 1

                        ECONOMIC MODELLING FOR
                        POLICY SIMULATIONS
Step 1:  Hypothesis:  Abstract
relations between key variables
                       Step 2;  Formulation:  Mathemat-
                       ical formulation of equation sys-
                       tem includes:
                       endogeneous (output)
                       exogeneous (input)
                       variables
                       Step 3:  Estimation: Data collec-
                       tion,  statistical methods, and
                       computation
Step 4:  Evaluation: Test of eco-
nomic theory by statistical
methods,  acceptable or not
Step 5:  Simulation: Impact sim-
ulation of different strategies
Step 6:  Use:  Information from
results of policy simulations used
in decisions

-------
      From the complexity of real economic space,  economic theory




develops by simplification of the relationships  between key variables




under investigation.  For example, the relation between income and




consumption of society is isolated from other numerous factors and




presented under a set of assumptions.




      The second step includes a mathematical formulation of eco-




nomic theory into a model which consists of a  set of equations.  In this




mathematical model, endogenous variables will be determined by  the




model using  a  set of exogenous variables which are given outside of the




model.  In other words, exogenous variables provide model input  and




endogenous variables the output of the model.  A policy-oriented model




usually integrates the policy questions or variables of the structure of




the model as well.




      In the case of a mathematically formulated model,  empirical data




are corrected  and estimation of the model parameters by statistical




methods takes  place as in Step 3.





      In Step 4, the results  of statistical estimation are assessed  by




evaluation criteria from probability theory to determine whether the




results  are acceptable.  If  not, theoretical assumptions in Step 1 will




be rejected and reformulation is in order.




      After an  acceptable model is  "empirically"  obtained, the model




is ready for  policy simulation.  Policy,  by itself, is  an exogenous

-------
"shock" to the model to observe the change of endogenous variables in




the model.  It can be in the form of a change  in exogenous variables  or




an exogenous information of change in the parameters of the model.




      Outputs from the simulation provide the information of the endo-




genous variables before  and after the policy simulation for evaluation




of policy impact to the economic system under study.







3.2   The APCO  Economic Model System







      The APCO  Economic Model System consists of two major com-




ponents,  namely,  a 100 AQCR regional model and interregional feed-




back from a national I-O model,  as shown in  Figure 2.2.  The modules




comprising both these major components are  also indicated in Figure




3.2.  A generalized description of the model  system appears in this




section, and a more detailed development and estimation of the model




is provided in Appendix A  through Appendix D.




      3. 2. 1  The Regional Model




      There is  a  fundamental difference between the economy of an




AQCR and the national economy.  The former is an open economy




where growth and development are closely related to its capability to




carry on external trade with other regions.  The  latter  is rather  more




self-contained by its nature.  The concept of  "export-base" or eco-




nomic base theory has traditionally been the central guiding concept

-------
in the description of urban economies.  In line with this concept,  this




study treats manufacturing industries as export-oriented industries.




The growth of the manufacturing industries  leads to the growth of the




regional economy. This model structure is consistent with the




familiar Keynsian-type trade multiplier in an open economic  system.




Consequently, in Figure  3, 2, the manufacturing industry module feeds




into the regional economy,  and growth of manufacturing industry and




regional economy determine the regional employment in the regional




labor market.  By the same token,  production and  consumption activ-




ities are related to the demand for electric  power and fuels.  All these




relationships have been formulated into a mathematical model as




shown in Figure 3.4, using the notations defined in Figure  3.3.




      In the manufacturing module, the production  relations of the




manufacturing industry are presented.  Output is related to the pro-




duction factors:  labor and capital.  Gross profit is given as  residual




between value-added and wage bill.  Investment behavior is related to




the profit and capital stock from the previous period.  With new in-




vestment expenditures and adjustment of depreciation, capital stock




of present  capacity is determined.  Finally, employment in manufac-




turing sectors is derived from the level of production and wage level.




      In the regional economy module,  regional income is determined




by the level of manufacturing production, regional  consumption

-------
                        FIGURE 3.2
          MAJOR COMPONENTS OF THE MODEL
I-O Model and Inter-
regional Feedback
Regional Model
    National
    I-O Model
     Regional
     Market
     Share
     Matrix
1
1
1
1
1
1
1
1
1
1
1
1
»•
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1

Manufac-
turing
Industries
!
Regional
Economy
Income
Consumptior
Government

Regional
Labor
Market

i
i
1
1
r !
\ i
\ 	 , i
' i
Electricity |
and Fuel j I
Demand j '
/i
i
i
' !
!
!
i
!
1

-------
                         FIGURE 3, 3

        NOTATION OF THE REGIONAL MODEL VARIABLES


V         value-added by industry j of ith, AQCR

K..       capital stock by industry j of ith. AQCR

N..       employment by industry j of ith AQCR

!••        investment expenditure by industry j of ith AQCR

n..       capital share or gross profit by industry j  of ith AQCR

W..       average wage by industry j of ith AQCR

Y-        regional personal income of ith AQCR

C.        regional consumption expenditure of ith AQCR

G.        local government expenditure of ith AQCR

T.        local government revenue of ith AQCR

N.        employment by industries other than manufacturing
          industries of ith AQCR

N.   .-.  (*>   j\i.. -(• N. ) total regional employmcnl  of ith AQCR
        j -1

L.        regional labor force of ith AQCR

Q.        total regional consumption of electric power in ith AQCR


Q..       electric power consumed by industry j of jrh AQCR

QCJ       electricity consumed by residents  of ith AQCR

Q.       electricity consumed by industries other thnn
          manufacturing industries of ith AQCR

 ch-        price of fuel type r  of ith AQCR,  type of fuel including
           coal, coke, fuel oil and natural gas

 ^'r-•       clcmr-nd of fuel, type r by jth industry by ith AQCR

 i  •• 1,  .  . .  ,  10(1   1 L'O AQCR'r,

 j  r; 1 •  •  • •  >  I1-*     \vhichis two-digit manufacturing industries

-------
                             1*  'J

   (ID   u.(t)  =    Li(t? - Ni(t)
          1           Li(t)
   (!Z)   L.ft)  -:  f(N. (t),  u-ff) )
           i       "i       l
                                FIGURE 3.4
                 THE REGIONAL MODEL  FORMULATION
I.      Manufacturing Industry
                          a j     i - a .
   (!)    V..(l)   -  A.. K. (t-) Ktj  J(t)

   (2)    1T..(1)   -  (] -.o.) V..(t)
          IJ             .1    IJ

   (3)    I..(t)  --.  .f (ii  (i.),  K   (I:-!))                             j=l, .  .  .  ,  19
          ''"•'           ij       ij
   (4)    K  (t)   =  K..  (l-l) +I..(t) - d  K.  (t)                     j=l,.  .  .  ,  19
          i.i         i.l          ij      j   ij

   (5)    AV..(l)  - JC-.V.. (t-1),  u.(tl)                             j=l.,  .  .  ,  19
           1.1            !J         1
II.    Regional Economy
   (6 )     Y . (t )  =  f (C. (1: ) s  v Vj j (1 ) ,  G. (1. )  )
                            ,i
   Co     c.(i.j  -..  c(V.(0,  C.(i-D)
           3.        "    '       J-
   (P)     c-^t)     J'(T.(D)
III.    Regional Labor Market
          _                !
   (9)     N.(t)=  f (Y.(t) -
                           J
                  _       19
         N  i)  -  N.(t> +

-------
                               FIGURE  3. 4 (continued)
IV.   Electricity and Fuel Demand
  (13)       Q..(t)  =  f (V..(t))
               1J          XJ  •

  (H)       Q .(t)  -•   ;f(C.(iM
               r*i            •>•
  (15)       Q (t)  ::   J (Y.(t)  -  Z  V..(l) )
               1           1       .   lj
                                 J


  (16)       Q.(t)  - ŁQ..(t) +Q  .(t) +Q.(1:)
               1     j   1J        Cl       1
  (17)         Er..  ..qp.  ,  .v..
  (18)        Z..(t)  -   B  (t)

-------
expenditures and local government expenditures.  On the  other hand,





regional consumption is related to the regional income, and govern-




ment expenditures is related to the government revenues.




     In the regional labor market module, employment by industry,




other than manufacturing industry, is derived from the income gen-




erated from non-manufacturing sectors.   Regional employment,  which




is the summation of employment from manufacturing and  non-manu-




facturing  industries, and unemployment rate,  determines the regional




labor force in each AQCR.




      Finally, electricity and fuel demand modules provide  the de-




mand of electricity and fuel derived from production and consumption




activity of the region.   A substitutional relationship between different




types of fuels is also included.




     A more detailed look at the regional model which comprises  162




equations is provided in Figure 3. 5.




      3.2.2  I-O Model and Interregional Feedback




     As emphasized earlier in the model formulation, a  region's




growth is closely dependent upon its  capability to carry on external




trade with other regions.  A national input-output system is  introduced




to serve the role of  external market  for the regional economy described




in the regional model, and also, hopefully, to measure the structural




change of the national economy upon  air pollution control  to the  100




AQCRs.







-------
                            FIGURE 3.5
           REGIONAL MODEL -- A MORE DETAILED LOOK
Capital
Stock
Output
   3ZE
Employ-
ment
Electric
Power
Demand
Fuel:
Coal, oil
gas
Regional
Income
Govern -
ment Ex-
penditures
Other
Industry
Regional
Employ-
ment
Regional
Unemploy-
ment
Investment
Profit
Wage
Deter-
mination
Price of
Electrici-
ty
Price of
Fuels
Regional
Consump-
tion
Govern-
ment
Revenue
Employ-
ment in
other sec.
Regional
Labor
Force
Manufacturing
Industries
(By 2-digit
SIC)
Electricity
and Fuel
Demand
Regional Economy:
  . Income
  . Consumption
  . Government
Regional
Labor
Market

-------
      The relation between a national I-O system and a cross-sectional

regional model of Keynsian-type formulation can be better explained

from policy questions to be answered from this model system.  The re-

gional model developed in Phase II is appropriate for an isolated region

without interregional feedback scheme,  although the export activity is

explicitly treated.  When an air pollution control policy is implemented

across the nation, costs increase.  The  production will result in an up-

ward shift of the supply curve.  Whether an industry or an individual

firm may be able to pass on the increased cost per unit of output is de-

pendent upon the elasticity of demand and supply of the corresponding

products.  If price changes are obtained from exogenous information, *

the equilibrium  supply of the products will be known. **

      However,  price increases in high emission industries under air

pollution control will not only reduce  the demand of their products but

also affect the demand of those products which are  the intermediate

products in the production of the high emission industries.  For exam-

ple,  a higher price for steel products will not only  reduce the demand
      *A study of price markup which air pollution control is instituted
has been reported in D.  A. LeSourd,  e^ aL ,  Comprehensive Study of
Specified Air Pollution Sources to Assess the Economic Effects of Air
Quality Standards.  Research Triangle Institute, December,  1970.

     **See Appendix C for detailed illustration.

-------
for steel but also affect sales in transportation, coal products, and

other materials and services related to the steel industry.   Further,

such effects  originating in a given region (or AQCR) in the nation will

not be limited to the region but would affect the economic activity in

other regions (or AQCRs) and in all likelihood,  create a feedback to

the region. *

      For a limited number  of regions, the formulation of an interre-

gional I-O system is perhaps feasible.  However, a 20-sector regional

I-O system  with 100 regions, the size of the matrix will be 2000 x 2000

(with such detailed information largely non-existent).

      An alternative formulation was consequently necessary.  A na-

tional input-output  system linked to a regional market share matrix

was used  to capture the regional feedback.  It is argued that the re-

gional share of  the national market by industry (termed as the "loca-

tion quotient"**) is relatively stable.  For example,  if steel production

in the Pittsburgh AQCR is 12 percent of the nation's steel product,  then

12 percent of a  change in the national steel market will have an affect
      *These interregional feedback phenomenon were observed in a
pioneer study by Ronald E. Miller, "Interregional Feedback Effects in
Input-Output Models:  Some Preliminary Results, " Papers, Regional
Science Association, Vol.  17, 1966.

     **A location quotient is the ratio of the regional activity level in
an industry to that  in the nation in that industry.

-------
on the Pittsburgh AQCR.   This concept is particularly useful for a




cross '•sectional model which deals with the geographic distribution of




economic activity at a given period of time.





      In brief,  using exogenous information on price changes in high




emission  industries occasioned by air  pollution control,  the high emis-




sion industries in each AQCR will fall  in output to the point correspond-




ing to an upward shifting  supply curve  and the new price after control.




An aggregation of changes in regional production for each high emis-




sion industry will give the change of  demand in the nation.  By use of




the national I-O system,  the impact of changes in high emission indus-




tries on other industries  can be measured.  Finally, through the use




of regional market share matrix, the national impact can be distributed




to each AQCR as the net interregional  feedback from the other regions




under study (99  AQCRs) and the rest of the nation.







3. 3   Empirical Estimation







      A detailed account of the data used in the model and methods of




statistical estimation is provided in Appendices C,  D and E.  It is




worth noting here that the economic data used were for 1967.




      In general, the equation  in each of the modules shown in the re-




gional model component in Figure 3.2  is solved simultaneously.  How-




ever, the modules themselves are related recursively.  The model

-------
system appears to perform well on criteria of plausibility, explanatory




power,  and reliability.




      The I-O model is a 42-sector model obtained by collapsing the




1963 national I-O table.   The details of the development of this model




and the interregional market share matrix are described in Appendix D.

-------
    Inputs
Model Development
••           Model             **•
 Regional Economic Model System
      Outputs
                                                                                      Assessment
 Regional Economic
 Activity
 National Economic
 Activity
                                    Regional
                                    Model
         Interregional
         Feedback
Policy Assessment
 APCG Policy Variables
   Standards
   Incentives
   Fuels
   Research and
   development
   Government
   expenditures
     Computerized Simula-
     tion Program

     .   Program RMS
     .   Program FEE
     .   Program IOA
Outputs:  Changes in:
   Manufacturing
   activity
   Power and fuel
   consumption
   Income
   Employment
.  Government
   expenditures
_..  Other	
                                                                                      Interpretation
                                                                                      of Model
                                                                                      Outputs
Model System
Assessments
                                                              Recommenda-
                                                              tions on
                                                              Further Refine-
                                                              ment and Utili-
                                                              zation of Model

-------
4. 0   PRELIMINARY STUDY OF A MODEL
      FOR NATIONAL ECONOMIC EFFECTS
      ASSESSMENT
4. 1   The Transition from Regional to
      National Policy Analysis:  Background
      Over the past three years, there is an increasing interest in air

pollution control policy decisions at the Federal government level.  This

transfer is institutionalized under the Clean Air Amendments of 1970,

which delegates significant new authorities and responsibilities for

cleaning up the nation's air resource to APCO of EPA.  Directly or in-

directly, APCO decisions will result in billions of dollars of emission

control expenditures each year.  As air quality is changed due to these

control expenditures,  additional billions of dollars worth of changes in

production and consumption patterns can be expected to ensue.  Taking

the longer  view,  changes in technology fostered by Federal government-

sponsored  and private research and development can be expected to

make revolutionary changes in modern living patterns (e. g. ,  transpor-

tation and energy).

      Paralleling this trend, there has been a shift of emphasis in

CONSAD's efforts to develop a model for assessment of the economic

effects  of air  pollution control.   During the first year of the project,

-------
a time series econometric model of a single region -- the St. Louis




metropolitan area -- was constructed and the  economic impacts of




alternative control strategies for that region were simulated.  Over




the second year of the project,  a model like the St. Louis model was




built for 31 AQCRs using cross-section data.




      Simulations of these regional models, in the context of over-




increasing emphasis  on a Federally-coordinated national program of




air pollution control, highlighted the need for  two types of improve-




ments in the  scope and structure of these models.




      First,  the regional model was  structured as though AQCRs  were




economically independent of one another.  There was no allowance for




interregional effects.  If, for example,  Region A instituted air pollu-




tion control and in order to do so imported air pollution equipment




from  Region B,  the model simulated the economic impact of Region




A's program on Region A alone, and not on Region B.   Thus,  the  re-




gional model gave no indication of the increase in employment in




Region B resulting from the increased production of air pollution con-




trol equipment for export to Region A.  There was, therefore, a




source of pessimistic bias in the statement of economic effects of air




pollution abatement embedded in the very structure of the model.




      Second, the models at the end of the second year focused only on




the economic impacts of control expenditures  and accounted in no way

-------
for any benefits which might result from air quality improvement.




This again tended to cause unjustifiably pessimistic conclusions about




the economic effects  of air pollution control.




      During the current year,  CONSAD approached the problem of




eliminating these biases and making some preliminary assessment of




the national impact of air  pollution control in two ways.  First,  the




cross-section Regional Model was restructured to eliminate, insofar




as possible,  the  pessimistic bias induced by structural exclusion of




interregional feedback effects and benefits and to permit preliminary




estimates, a national I-O  model has been introduced to capture:  (1) a




preliminary  estimate of structural changes in the national economy,




and (2) an interregional feedback scheme  to the AQCRs.




      Second, the feasibility and desirability of an expressly national




model to be interconnected with the current regional model was  inves-




tigated in considerable detail.




      It is  the primary purpose of this chapter to present the findings




of this evaluation study.   It opens with a brief discussion of CONSAD's




attempt to  use an existing national econometric model for making pre-




liminary estimates of national effects of alternative national control




strategies.

-------
      Given the nature  and difficulties of access of existent models,




the chapter next explores  the feasibility of the development of a national




model specifically structured for air pollution policy assessment.







4. 2   The QBE Quarterly  Model







      As a part of Phase III of the RAPA project, CONSAD investigated




the possibilities and potentials of adding another model to the Regional




Economic Model System for the  express purpose of assessing the na-




tional economic effects of air pollution control.  CONSAD planned the




use of the OBE Quarterly  Econometric Model to make  preliminary esti-




mates of national economic effects.




      A brief review of the nature of concern with a national  macro




model will be a useful  prelude to the discussion of the  OBE model uti-




lization.




      The national economic  system and the environment conditions




form two interrelated dynamic systems.  Pollution is a by-product of




economic activity. In  this broad sense, growth of GNP and hence the




industrial component are  closely related to the environment  conditions




in the future.  On the other hand,  implementation of a  given  emission




(an air quality) standard,  possibly involves an over time policy planning




and stimulates a dynamic  economic  impact to the nation.  Even if the




control technology is available,  an overnight switch of entire production

-------
line, for example,  a highly efficient automobile engine, may be eco-




nomically impractical.  The question of tax policy,  choice of evalua-




tion criteria,  or optional time table  of a given standard are  all key




questions involved  in the trade-off between the growth of GNP and a




better  or worse off environmental condition of the nation.




      For example, given a national standard for a certain type of




emission, there are a set of "implementation phase. "  Each imple-




mentation phase may be thought of as depicting a schedule for achiev-




ing air quality standards over time (see Figure 4. 1).




      One factor in determining the rate of implementation of standards




will be the estimated rate at which industry can absorb the cost of air




quality control equipment installations.  A time series national model




may be designed to provide estimates of such rates  for each industrial




sector.  In this manner,  the model may provide data for facilitating




the current  process of implementation planning.




      The future process of implementation planning may be enhanced




by providing APCO with additional control measures for  mounting an




implementation strategy.  One such measure mentioned in the  past at




the Federal level is the  provision of tax credits to industries that in-




vest in costly pollution control equipment.  One effect of the proposed




measure would be  to accelerate the rate at which industry can  absorb

-------
                        FIGURE 4. 1

              EFFECT OF TAX CREDIT STRATEGIES
                UPON IMPLEMENTATION PLAN
Emission
  Rate
Present
esent.
Level
                      Gradual Implementation, No Tax Credits

                            .Low Tax Credit Policy

                                   High Tax Credit Policy

                                              Desired
                                                  Standard

                                                      Time

-------
pollution control costs, and this,  in turn,  would increase the rate at




which desired standards can be implemented.




      Although acceleration of the feasible implementation schedule




will occur under a tax-credit strategy, the effectiveness of this mea-




sure will depend upon the growth  rate of industrial composition of the




nation.  The impacts upon the feasible rate of implementation that may




be expected of a proposed tax credit strategy may be estimated with a




national model.




      On the other hand, the regional  model developed in the last chap-




ter is a cross-sectional model, although some lagged variables are




included.  It is argued that economic impact of air pollution control




strategies can be described in two facets,  namely, spatial impact of




resource allocation and over  time impact related to  the growth of the




economy.   The regional model is  particularly useful as demonstrated




in Chapter 6 when a region-by-region differential  of control impacts




are of concern.   It will be shown  that  while the aggregate net change




on all 91 AQCRs under study  is an increase of unemployment rate of




0. 5 percent (3. 5 to 4 percent),  the unemployment  rates in some  of the




AQCRs with concentrations of high emission industries will be 2 per-




cent or more.  Some other AQCRs are shown  to be even "better  off"




in terms of unemployment effects under the same  air quality standards.

-------
      But,  the limitations  of the regional model is also clear; a cross -




sectional model needs different sets of economic data over time to pro-




vide a preliminary picture of dynamic impact of the air pollution con-




trol.  Thus,  it requires projections of large sets of data,  and restrict-




ing thereby the accuracy of the model.





      Thus, it appears that a cross-sectional regional model linked




with a time series national model will be an ideal system for the study




of economic effects.




      It was with this objective that CONSAD began the exploration of




the link-up and use of an extant national model to the Regional Model.




The candidate for this experiment was the OBE Quarterly Model.   This




model was to provide a "quick look" at the national effects  over time.




The nation model  will cover the rest of the nation other than the AQCR




and hopefully a preliminary over time economic impact of air pollution




control will emerge.




      An intensive study of the structure of the OBE model, however,




revealed a number of structural properties which rendered it not an




entirely desirable tool for national effects assessment.  In particular,




the model is  structured primarily for the purpose of making short-




term (roughly eight quarters) forecasts of the effect of conventional




government fiscal and monetary strategies.  Both the temporal and

-------
industrial aggregation of the model are inappropriate to simulations of




the national economic effects of air pollution control.




      In contrast to the high degree of time disaggregation of the OBE




mode..,  it exhibits a high degree of industrial aggregation.  No differ-




entiation is made,  for example, of investment  by industrial sector,




GNP by sector or employment by sector.  Since  a principal concern of




APCO must be the effect of its  policies on particular industries,  the




OBE model must be judged less than satisfactory for APCO strategy




analysis purposes.




      Meeting with OBE personnel generally confirmed CONSAD's find-




ings.  Indeed,  OBE professionals had  such serious reservations about




the appropriateness of the model for air pollution control strategy sim-




ulations that access to the model was effectively refused.  It would




seem, therefore,  that if an effective vehicle is to be sought for national




economic effects assessment, APCO  must look to some other national




model,  or perhaps consider building one expressly for its purposes.




      Given the results of the OBE model applicability to air pollution




control strategy assessment, it appears appropriate to inquire whether




other models are  to be explored or whether a special model needs to




be developed for APCO use.  Before this query is addressed,  it will




be useful to identify the  characteristics  of what CONSAD believes to




be a desirable  national model.

-------
4. 3   Desired Attributes in a
      National Model

      Ideally, a national model and the current Regional Model would

be moulded into one large interdependent system of models.  In rough

terms,  the  system of models would be composed of three submodels,

each of which would pertain to a particular geographic area:

            .  the nation,

            .  specific AQCRs, and

            .  the rest of the  United States not in AQCRs
              specifically included in the  model.

CONSAD stresses the importance  of retaining specific regions within

the model since the economic effects of air pollution control have been

found to be  strikingly different from region to region.  A solely national

picture would thus be misleading in important ways.

      Submodels (1) and (2) of the  ideal model structure would have

stochastic  equations.  Submodel (3) would be made up wholly of identi-

ties.  Each of the three  submodels would  be sectored conformably to

facilitate interconnection.  The submodels would then be interconnected

into one large model as  indicated in the following simple Keynesian

model.

            Submodel 1

            YN  = CN + IN


-------
where:      N stands for nation, and

            Y,  C and I are income consumption and investment,
               respectively.

            Submodel 2
            GI = ai + bj YI + ej

where:      "1" stands for region i.

            Submodel 3

            Yo = YN - Yj

            C0 = CN - Cj

where:      "0" stands for the nation that is outside the AQCRs.

      Submodel (2) would consist of the CONSAD's  regional model for

all AQCRs currently has or will develop, further refined to incorpor-

ate interregional flows of air pollution control factors.  Submodel (1)

will be estimated according to the sectoring and specification neces-

sary for air pollution control study.

      The output of this  integrated model would be  a comprehensive

and logically consistent  forecast of the  economic and gross environ-

mental effects of air pollution control programs for each of the specif-

ically included AQCRs,  for the nation,  and for all other areas of the
                                           *
United States not in the specifically included AQCRs.

      Submodels (1) and (2) should be sectored at least into three broad

categories, namely, sectors  affected by pollution,  sectors affecting

-------
pollution, and environment sectors which provide gross environmental




indicators related to economic variables.  Consideration should be




given to  the sectors affected by the level of pollutants.  For example,




reduction of agriculture productivity, health and productivity losses,




extra household maintenance,  cleaning services  are highly related to




the level of  pollutants in ambient air.  These effects can be reflected




in the production function of agriculture.




      For sectors affecting pollution, emission of pollutants from a




given source depends upon the level of output. Air pollution abatement




will lead to  cost increases in the high emission industries and be re-




flected in price increases and in changes  in interindustrial structure.




Such high emission industries include steam-electric  power, iron and




steel mills, petroleum refining,  sulfate pulp  mills,  hydraulic cement




manufacturers, gray-iron foundries, etc.  For these  industries, de-




mands for low sulfur fuel and control devices will depend both on out-




put level and emission standards.  Environmental sectors, since pollu-




tion and  economic activity are interrelated,  should also contain a set of




equations predicting  emission from production and level of control activ-




ity.  These  sectors would also include supply and demand relations of




control device and low sulfur fuels.




      Besides the sector specification,  environmental variables should




be explicitly integrated into the  related equations. The following exam-




ples will provide some ideas  on this approach.








-------
      Sectors Affected by Pollution





      The production function of the agriculture sector can be specified




as follows:





            X = f  (K,  L, P)




where X,  K and L are output, capital and labor,  respectively, and P is




measurement of pollution concentration such that:





               < 0
Reduction of emission increases output level, such treatment is quite




similar to the technical change introduced in production function.




      In the consumption functions, benefits from air pollution control




(B) can be introduced as follows:




            C = f (Y, B)




            B = f (P)
Reductions in air pollution levels (P) will save expenditures on health,




household maintenance, cleaning service,  etc.  Types of benefits and




consumption expenditures can be further divided if such data are avail-





able.




      Sectors Affecting Pollution




      Production functions of the high emission industries can explicitly




include low sulfur fuel (S) and control devices (D) as additional factors

-------
of production:




            X = f (K,  L, S, D)




From these production functions can be derived projections of emis-




sions and projections  of the demand for inputs for air pollution control.




      In sum, the ideal model is a comprehensive model, fairly com-




plete not only in its coverage of national economic phenomena, but also




regional economic phenomena and gross environmental phenomena.







4. 4   Data Availability and Reliability







      Apart from the theoretical and structural advantages of a national




model in dealing with  the dynamic nature of the economic impacts of




air pollution control,  a major profit of the national modelling is the




ability to capitalize on the rich and reliable economic data and the  con-




trol input data available at the national level.




      It often happens that data on a two-digit SIC industry in a given




AQCR is not disclosed in order to protect the financial conditions of




firms when less than four operate in that AQCR in that industry.




Therefore,  there are  limits to specifying industrial detail in the  re-




gional model.  At the  national level, all  economic data are available




at four-digit SIC detail for any given industry.  In other words, a much




more detailed industry study is possible for any specific industry of




interest, if a national model was  linked to the  APCO Economic Model




System.








-------
      The same situation is true for the availability of the control input

data as well.  The regional control cost data available in the 305(a)

Report of 1970 is the only source of control cost by region for 100

AQCRs and even that is  a set of aggregate control costs by region.

Only for a limited number of AQCRs control costs  at the two-digit de-

tail are available.  However, data on control inputs are available for

the nation in greater specificity.  Utilization of such information will

strengthen the understanding of the economic impact of air pollution

control strategies.
4. 5   Alternative Approaches to
      National Effects Assessment
      If it is granted that evaluation of the economics of alternative

national strategies is desirable and that an economic model is a useful

tool for such evaluation,  it remains to consider the different  kinds of

models which might fulfill this need.  CONSAD has identified three

broad modelling alternatives to capture in varying degrees the alter-

natives described in the previous section.

            1.   One alternative would be to expand the geographic
                scope of the current Regional Model System to in-
                clude the remaining AQCRs in the nation.  In  the
                process,  the model structure may be refined  as
                appropriate.

            2.   An existing national economic model (e. g. , Klein-
                Goldberger) may be modified and adopted for  pur-

-------
            3.  A riew model could be built expressly for the
               purpose of assessing national effects  of Con-
               gressionally mandated responsibilities of
               APCO.

      For reasons which are best left to the last chapter,  CONSAD

believes an effort which takes,  in part, all three approaches is advis

able.

      CONSAD also believes such a national model is feasible on the

basis of preliminary  work on formulation of such a model.

-------

-------
                                      MiuK-1
                                       Outputs
                            Assessment
Model Development
Regional Economic Model System
 Regional Economic
I Activity
 National  Economic
 Activity
                                    Rc-gLonal
                                    Model
        Interregional
        Feedback
                                 I-O Model and
                                 National Effects
                                 Assessment
Policy Assessment
/ IhcenfeiyXs
   F^erls
 .  Re sea-re h an
  /developme
   x-expendirfure'S
     .,, R-fqgrarn. IOA
Outputs: Changes in:
.   Manufacturing
   activity
.   Power and fuel
   consumption
.   Income
   Employment
   Government
   expenditures
.   Other
                                                              Interpretation
                                                              of Model
                                                              Outputs
                                                                                        Model System
                                                                                        Assessments
                                                              Recommenda-
                                                              tions on
                                                              Further Refine-
                                                              ment and  Utili-
                                                              zation of Model

-------
5. 0   SIMULATING THE IMPACT OF AIR
      POLLUTION CONTROL STRATEGIES:
      A SIMULATION OF THREE STRATEGIES
5. 1   Introduction


      The raison d'etre of any economic model is to facilitate accurate

and comprehensive prediction of the consequences of changes in partic-

ular variables and parameters under the control of policy-makers.

The  adequacy of predictions so made depends not only on the adequacy

of the model itself, but also on the proficiency of the model-user in

translating the changes in policy variables he wishes to investigate in-

to appropriate changes in variables and parameters of the model.  The

computerized economic simulation model does not supplant the  expert,

but rather,  follows through difficult but routine complex logic and cal-

culations, thus freeing the expert for the more creative  and demanding

task of using the model intelligently.

      This chapter of the report is intended to  explore the simulation

potential of the Regional  Economic Model System in the following order.

            .  First,  it describes the range of  air pollution control
              policies for which the model is developed and how
              these policies need to be translated into appropriate
              variables of the model before  they are introduced in-
              to the  simulation.  An outline  of the way the Regional
              Economic  Model System transforms  control policy
              inputs  into various model outputs of interest to policy
              assessment is also provided.

-------
              Second, it describes the implications  of the three
              strategies prescribed by APCO staff for the excer-
              cise of the model.

              Third, the chapter outlines the manner in which
              the three strategies are prepared as inputs for the
              simulation run and the nature of the outputs of the
              simulation.
5. 2   The Scope and Nature of Simulation
      of the Regional Model
      5.2.1  Nature of Simulation in General

      An economic model purports to describe the relationships be-

tween a set of variables  of interest in some context, most though not

all of which are economic variables.  Variables are typically classi-

fied dichotomously as either exogenous or endogenous,  paralleling the

engineer's classification of variables as either input variables or out-

put variables.  Exogenous variables  are variables not determined by

the model, but variables whose values  influence the values taken by

variables determined by the model.   The latter type of  variable is

called an endogenous variable.  For  instance, in a model of the wheat

market,  the average annual price of  wheat and annual output would be

endogenous variables, while weather indicators,  since  the weather is

not significantly (if at all) influenced by wheat prices  and output, are

exogenous variables.

-------
      Model variables are related to one another in a system of simul-




taneous equations which system actually comprises the model.  The




parameters (or coefficients) which appear in these equations are called




the structural parameters of the model.  They are hopefully stable con-




stants which characterize relatively invariant relationships among




model variables.  The model is either linear or non-linear depending




upon whether all its equations are linear or not.




      These concepts may perhaps be clarified by consideration of an




actual representation of a model.  Ley Y be an nxl vector,  the com-




ponents of which are values  of variables endogenous to some model and




let Z  be an rxl vector  of values of variables exogenous or lagged endo-




genous to some model.  Then the notion that Y and Z are linearly re-




lated  can  be represented as:




(1)         A Y +  B Z = 0




where A and B are respectively nxn and nxr matrices of structural




parameters.




      A simulation is really nothing more nor less than solving of the




model for the endogenous variables given values of the structural




parameters and the exogenous variables.   The process is particularly




simple where the model is linear,  as is that outlined above in equation




(1).  In this common case, the model solution may be obtained explicitly




as:

-------
(2)          Y =  -A'1 B Z





Where the model is  non-linear, more complicated techniques for solv-




ing systems of non-linear equations (e.g., modified Newton) must be




invoked.




      Simulation of the effect of a .policy involves (1)  solving the model




for no change in policy,  (2) translation of the proposed policy into




changes in exogenous variables and/or structural parameters,  and




(3)  solving the model for these  changed  values.  Comparison of the




solution values with and without the policy demonstrates the effect of




the policy.  Steps 1  and  3 are computerized and are thus trivially sim-




ple.  Step 2, however,  demands thorough familiarity with the model,




patience,  and sometimes considerable ingenuity.  Because Step 2 is




essentially an endeavor  of skillful and critical decision in the absence




of all but the most general guidelines, it is  best mastered by example




and practice.  And that is the purpose of the following sections of this




chapter.




      5.2.2  APCO  Strategies  and the Regional Model




      The purpose of this section is to review the range of policies at




the disposal of APCO and how they should be transferred for simula-




tion through the model system.




      There is a variety of tools available to APCO to meet its Con-




gressionally mandated responsibilities to improve air quality.  Under

-------
current and foreseeable law,  APCO has, as indicated earlier,  policy

or advisory responsibilities in the following areas:

            1.  Setting of standards  (air quality, emission)

               a.   For stationary sources
               b.   For mobile sources

            2.  Incentive policies

               a.   Environmental subsidies
               b.   Environmental degradation taxes
               c.   Accelerated depreciation
               d.   Low  interest loans

            3.  Government expenditures

            4.  Research and  development

            5.  Fuel regulation

      The Regional Economic Model System can trace a significant por-

tion of these effects,  say when standards set by APCO are used in a

simulation of the model.  The standards assumed in the 305(a) Report

to the Congress are an illustration of such  standards.

      However, before the 305(a)  standards can be used in the  simula-

tion, they need to  be translated into model  variables or model  inputs.

In other words, the 305(a) standards must be translated into model in-

puts  reflecting these standards, suitable to the model logic, before

their effects can be simulated.  The 305(a) standards are also  express -

able  in terms of the control costs incident to the high emission indus-

tries in the AQCR and are entered into the model as control costs.

-------
These control costs are distinguishable into annualized investment

costs and operating costs and are fed into the appropriate component

of the model (see Figure 5. 1).

      Figure 5.  1 provides an illustrative translation of the various

APC  policies into appropriate inputs into the model.   The model inputs

into which APC  policies are translatable are of two kinds:

              Changes in exogenous variables of the model,
              e.g., control costs,  damage functions, etc.
              These are represented by the rectangles in
              Figure  5. 1.

            .  Changes in structural-parameter (changes
              in behavior patterns  in the economy), e. g. ,
              modified consumption patterns,  modified tax
              structure and changes in the production pro-
              cess; represented by the diamonds in Figure
              5. 1

The first step, therefore, in simulation is to  identify the model inputs

in which form the control policies can be expressed.

      5. 2. 3  Regional Model System and Control Inputs

      The scope  of this section is to illustrate how the various  control

inputs into the model relate  to various components of the model.   Such

an understanding is vital to effective use of the model.

      Figure 5.2 identifies the four major  modules of the  Regional

Model and identifies their relationships to  five types  of control policy

model inputs.

-------
                          FIGURE 5. 1
            APC POLICIES AND MODEL, INPUTS
     APC Policies
  Model Inputs
Standa rds
 Air Quality Standards
 Emission standards
   a. stationary sources
   b. vehicles
Control Costs
 Damage
 Estimates
                                             of consump
                                             tion patterns
Incentives
subs idies
taxes
accelerated
depreciation



Emission
Reductions



Control Costs
Damage
Estimates



Government
Expenditures
                                                  od-
                                             ification of
                                             tax struc-
                                                ture
Research and
Development
 Control Costs
                                                  re-
                                               duction
                                               process
                                               change
Fuel Regulation
Control Costs
  Damage
  Estimates
       Input Changes
       Parameter Changes

-------
                        FIGURE 5.2

         REGIONAL MODEL AND CONTROL INPUTS
Regional Model
   Manufacturing
   Industry
   Power and
   Fuel Demand
   Regional Income
   Determination
   Regional Labor
   Market
                                       Control Policy
                                       (control input)
   Control
cost in high
emission in-
   dustry
                                        Interregional
                                        feedback effect
                                         Fuel price
                                           change
                                           Property
                                         value bene-
                                         fit in AQCR
  Local tax
    policy

-------
      The control cost to the high emission industries associated with




a given emission or air quality standard will lead to a cost increase in




those industries and power industries in particular.  These cost in-




creases will enter into the manufacturing industry module (in the in-




vestment and profit components) and the fuel demand module (price of




electricity equations)  and  will cause changes in the output of manufac-




turing industries.  Changes  in the  production will affect the fuel de-




mand of the industries and affect the regional income and regional




labor market.




      Interregional feedback effect, from the benefits to be realized




from cleaner air is brought  to the  regional model from a national I-O




model. This feedback effect can also feed in price changes of products,




increased demand for  control equipment,  etc. , through the link it pro-




vides between the regional model and the national I-O system.  Fuel




price changes will affect the fuel substitution in the regional model and




relate to the other parts of the model.  If property value benefits by




AQCR can be estimated, net increase of disposable income, will aug-




ment additional local consumption expenditures.   An increase  of the




productivity attendant on pollution abatement can lead to an increase




in the profit.  Finally, if any local taxes are introduced to aid the  im-




plementation of air pollution abatement, its impacts on the regional




economy can be traced.

-------
      Figure 5. 3 provides a more detailed look at the way control in-


puts course through the model.


      A simulation program consisting of the following three major


modules has been prepared:


            .   RMS  (Regional Model),


            .   IOA  (Input-Output Model),  and


            .   FEE  (Interregional Feedback).


A detailed description of the  simulation program  appears in Volume


II.  It will be used in the next section to simulate  the effects of three


strategies specified by APCO staff.
 5. 3   Simulating Three Strategies:
      The Definition of Strategies
      APCO specified three strategies to be simulated by the model.


All three of the simulations to be discussed here have as their basis


the control costs envisioned in the  1970 Report to Congress as required


by Section 305(a) of the  Clean Air Act of 1967.  The simulations re-


flect differing policies and assumptions with respect to the incidence


of control cost expenditures -- that is, with respect to who finally pays


the cost of controlling air pollution.  Before beginning a detailed de-


scription and analysis of the strategies studied here,  some general
                                            i

comments on incidence  and its  determinants are in order.

-------
                            FIGURE 5. 3
      REGIONAL MODEL AND CONTROL POLICIES:
                  MORE DETAILED LOOK
Capital
Stock
                      Investment
                            ontrol
                         cost in high
                      emission industries
                        operation cost
                        investment cost
                        subsidies
                        price increase
                       Wage
                       Determina-
                       tion
                                                 In te r -
                                               regional feed-
                                               back:
                                               . change in value
                                                  added
Electric
Power
Demand
Price of
Electricity
                      Price of
                      Fuels
coal,  gas,
                                                Fuel Policy:

                                                .  price changes
Regional
Income
                      Regional
                      Consump
                      tion
                                                 increase in
                                                 disposable
                                                 income
Govern-
ment
Expenditure
Govern-
ment
Revenue
                                                 increase in
                                                  roductivity
                      Employ-
                      ment in
                      other sec
Other
Industry
                                                   subsidies
Regional
Employment
                      Regional
                      Labor
                      Force
Regional
Unemploy-
ment
     REGIONAL MODEL
                                            CONTROL INPUT

-------
      The diagram presented in Figure 5.4 is a supply schedule (a




schedule relating market price to quantities which suppliers will wish




to produce and sell per unit time) and a demand schedule (a schedule




relating market price to quantities which consumers will wish to pur-




chase per unit time).  That price at which  the quantity consumers  wish




to purchase just equals the quantities suppliers wish to produce  and




sell is called  the equilibrium price,  and the corresponding quantity is




called the equilibrium quantity.  The effect of shifts  in either schedule,




i.e. ,  shifts in willingness to  sell or buy, causes equilibrium price and




quantity to change.  For example,  in Figure 5. 5(a) below,  consumers'




disposable income increases,  causing  the demand schedule to shift out-




ward  to the right.  In Figure  5. 5(b), the supply schedule is shifted up-




ward  reflecting an increase in costs of production.




      In simple graphical terms,  the effect of imposing  air pollution




control on an  industry is rather like Figure 5. 5(b).   Air pollution  con-




trol adds to the cost  per unit  of output,  causing the supply curve to




shift upward by the amount of the additional cost of air pollution con-




trol.  This is  depicted in Figure 5.6.  Equilibrium price is then in-




creased from po  to pi and equilibrium decreased from qo to qj.  The




total air pollution control expenditure is (pj - a) times qj. Of this




total amount (pj - po) times qj is paid by purchasers in the form of




higher prices  for each unit purchased.  The amount paid by producers

-------
    FIGURE 5.4
SUPPLY SCHEDULE
                            Supply
                            Schedule
                              Demand
                              Schedule
                          Units Purchased
                          Per Unit Time

-------
                            FIGURE 5. 5
 Price
P'


P
(a)
Shift due to increase in
disposable income
                         q'
                                         Quantity
 Price
p1

p
(b)
Shift due to increase in
production costs
                                         Quantity

-------
                                   FIGURE 5.6
                  AIR POLLUTION CONTROL COST PER UNIT OUTPUT
       Price
                                                            Additional Air
                                                            Pollution Control
                                                            Cost/Unit Output
Paid by
purchaser
                                      qi
                                                             D
Quality/
Unit Time

-------
in the form of lower revenues net of control costs is given by (p  - a)




times qj.  The difference in prices paid by consumers,  (pi  - po) is





called  the incidence of control costs on consumers,  and the  comple-




mentary price difference (po  ~ a) is called the incidence of control




costs on producers.  The sum of the incidences thus equals  the total




per unit cost of control.




     Incidence for a single industry is not, of course, the  whole story




of the economic impact of air pollution control.   Not only is one inter-




ested in changes in prices and output,  but also changes in wages, prof-




its, employment, consumption,  investment, use of raw materials and




other artifacts of production.  And this interest extends  to those quan-




tities not primarily for the case where control is applied to  a single




industry, but where most, if not all,  industries are controlled in




greater or lesser degree.  There are, therefore, important interin-




dustrial effects.   This is where  the APCO  Economic Model System




makes its contribution.   Using the results  of incidence calculations




and other input information, the APCO Economic Model System fore-




casts the consequences of alternative control policies for a wide range




of economic variables, taking into account interindustrial effects.




      The three strategies selected by APCO for simulation and analysis




in this report differ from one another  in the kind and degree of govern-




mental intervention to shift the incidence of air  pollution control costs.

-------
A general description of the three strategies follow.




      Strategy 1:  Industry Pays




      There are some who contend that industry should bear the full




cost of controlling air pollution; that is, that the incidence of control




costs on consumers or government should be zero.  The argument




buttressing this policy position is  a moral one (as  are  those which lie




behind other policy positions) and  is thus beyond the purview of scien-




tific enquiry.  The concern here is solely with discovery of the prob-




able effects of such a policy if it were  pursued.




      The most plausible specific  policy which the government could




use to insure that industry alone bears the incidence of air pollution




control costs is to freeze  prices at their level prior to instituting air




pollution control.  This  effectively produces a kinked demand curve,




perfectly elastic, at quantities less than that which was obtained  before




air pollution control.  This is diagrammed in  Figure 5. 7.  Since the




effect air pollution control is to shift the supply curve  upward, the new




equilibrium price and quantity will fall along the perfectly elastic por-




tion of the  demand curve.   Remembering that where demand is per-




fectly elastic, the full burden of air pollution control costs falls on the




supplier,  it is demonstrated that price freezes coupled with air pollu-




tion control will achieve the distribution of incidence desired under




Strategy 1.  This approach is diagrammed in Figure 5. 8.

-------
                          FIGURE 5.7
              PRICE FREEZE DEMAND SCHEDULE
                                         Pre -abatement
                                         Equilibrium
                                         Post-Price Freeze
                                         Demand Schedule
Pm

-------
                                    FIGURE 5.8

                 DEMAND AND SUPPLY RELATIONS UNDER
                             STRATEGY 1
Paid by
producer
                                                     S1 (post-abatement)
                                                            S (pre-abatement)
                                                      D (post-price freeze)

-------
      It is assumed that Strategy 1 differentiates between public util-




ities and other industries, allowing the former to pass along in toto




the additional cost of control so as not to jeopardize the already sup-




posedly minimal rate of return in this  sector.




      Strategy 2:  Hands Off




      Under Strategy 2,  the  incidence of control costs falls where it




may,  with no  attempt by the government to alter the distribution




achieved by the market.  In  other words, Strategy 2 is simply an air




pollution control policy with no accompanying economic policy.  In




diagrammatic terms,  the consequences of this policy for incidence




and changes in price and output may be seen by referring to Figure




5.6.




      Strategy 3:  Cost Sharing




      It is frequently argued that if government shares some of the




cost of control which would normally fall on industry,  undesirable eco-




nomic effects of control would be much attenuated.  In particular, the




net result of cost sharing, it is argued,  would be  to reduce the amount




by which prices will rise and output will fall.  The reasoning behind




this argument is presented diagrammatically in Figure 5. 9.  As shown




earlier,  the immediate effect of air pollution control is to shift the




supply curve upward by the amount of the additional cost of air pollu-




tion control.  Producers are willing to produce and sell only if they

-------
                           FIGURE 5.9
  DEMAND AND SUPPLY RELATIONS UNDER STRATEGY 3
P '
Pi
Po
                                                xSj(abatement without
                                                   cost sharing)
                                                        ^(abatement
                                                           with cost
                                                           sharing)
                                               Cost borne by  .,$ (pre-
                                               governmentx^  abatement)
                                                 incidence on buyers
                                              Incidence on sellers

-------
are able to cover all costs of so doing.  Hence,  without cost sharing,





the supply curve shifts from S to S1.  If the government were then to




undertake to reimburse industry for a portion of the control cost, the




supply curve would shift downward to some position intermediate be-




tween S and S',  say S".  The  effect  of cost sharing is simply to reduce





the additional cost of control  as viewed  by industry.  It is clear, that





as argued by proponents of this policy,  price increases and output falls




are attenuated.




      Because elasticities of  demand, at least in the short run, are




typically low for the services of public utilities, these industries





should have little difficulty in passing along cost increases with little





reduction in output.   It has, therefore,  been supposed here that no




cost sharing is to be done with utilities.




      For purposes of the simulation, it is assumed that government




is to bear 50 percent of all industrial process and combustion control




costs,  which would otherwise fall on industry, financed by an increase




in Federal personal income taxes, and no portion  of other control costs




as indicated above.







5.4   Preparing Inputs for Simulation







      The difficult task remains of translating each of these three




strategies into values of exogenous  variables and parameters in the

-------
APCO Economic Model System so that simulation can be executed.




The process by which this is done is described in the text and figures




of the following pages.




      The raw data on which simulations with the APCO Economic




Model System rest are (1) control costs by AQCR (and by 2-digit SIC




if available) for industrial processes and combustion, (2) control  costs




for steam electric power generation,  and (3) benefits which will occur




because of air pollution control outlays.  Data on items (1) and  (2),




for the simulations reported here, were taken from the 1970 Report




to the Congress required by Section 305(a) of the Air Quality Act of




1967.  For  purposes  of benefit determination as required under item




(3), the conservative estimate is that air  pollution  causes $10 billion




worth of damage annually.  These data, however, form only the basis




for simulation.   The  model user must still exercise considerable judg-




ment in deciding precisely how they shall be used in the simulation.




      To fully understand the simulation process for this model, how-




ever, one must delve yet deeper into its inner workings.  Appendices C,




D  and Volume II provides such a detailed review of these details.




      With this  rather detailed background in mind, it is time to dis-




cuss conversion of raw data into model inputs.   Benefits are discussed




first since the treatment of benefits is precisely the same under the

-------
three strategies.  A conservative total damage figure of $10 billion*

is assumed,  and it is assumed that roughly a 20  percent reduction in

annual damages would occur if the control costs  programmed in the

1970 Report to the Congress under Section 305(a) of the Air Quality

Act of 1967 were incurred.   An annual damage reduction of $2 billion

is thus implied.  This annual damage  reduction is treated as an in-

crease in consumers' disposable income, which  treatment implies an

increase in consumption, assuming a  marginal propensity to consume

0. 8, of $1. 6  billion.  This increase in consumption was treated as an

increase in final demand in the Interregional  Feedback Submodel, dis-

tributed among the various  sectors in that model according to the his-

torical distribution of consumption in  each of the feedback models'

sectors.

      All the raw cost of control information  was taken from the 1970

305(a) Report.   These costs reflect the following emission reductions

of particulates,  sulfur oxides,  hydro carbons and carbon monoxides  in

the 100 AQCRs:
      *U. S.  Department of Health,  Education and Welfare, National
Air Pollution Congrol Administration, Costs and Economic Impacts  of
Air Pollution Control,  Fiscal Years 1970-1974. January,  1969,  p. iv.

-------
                                                      Emission
Pollutant               Source Category                Decrease (%)

Particulates           Solid waste                        77. 8
                       Stationary combustion             91.7
                       Industrial process                 86. 1

Sulfur oxides           Stationary combustion             52.2
                       Industrial process                 36.2

Hydrocarbon           Solid waste                        69.4
                       Industrial process                 57. 8

Carbon monoxide       Solid waste                        84. 7
                       Industrial process                 90.2

      The only difference between the three strategies is in the treat-

ment of control cost, which treatment reflects, as noted above,  differ-

ing policies with respect to who shall bear  the cost of control. Under

Strategy 1  (Industry Pays),  the full burden  of industrial control costs

and increased costs of electric power used by the manufacturing  and

non-manufacturing sector falls upon  industry.  Consumers pay only

the increased cost of the electric power they use.

      For each AQCR,  control costs can be identified in three cate-

gories, namely,  investment expenditures on control equipment,  annual

operation cost,,  and  control cost of electric power industry. *  By
      '''Aggregation of the five-year average control cost over 91 AQCRs
under the present study is $1, 333 million.  Unfortunately,  control costs
by AQCR are not available on 2-digit SIC breakdown,  although the pres-
ent model can use such data.  A separate simulation run, based on 2-
digit SIC detail control costs for St. Louis,  Cincinnati, and Washington,
D. C.  AQCR, has been conducted.

-------
notation:

            A Ci =  A Cn + A Ci2                      i=l ..... 91

where      A C^   is the total control cost of manufacturing industry
                   in ith AQCR,

            A Cji  is the investment expenditure of control in i"1 AQCR,

            A C^2  is the annual operational cost of control in i*n AQCR.

It has been assumed that electric power industry will pass on the entire

control cost to the users by a price increase of:

            A  -  -  °i
             qi     Qi
where      Aqi    price increase of electricity in  i^1 AQCR,

            Oi      control  cost of electric power industry in ith AQCR,

            Qi      total regional consumption of electricity in i"1 AQCR.

      Under Strategy 1 (Industry Pays), annual operation cost A C^ has
been reduced from the gross profit (Ł flj;) of the manufacturing indus-

try,  and the investment expenditure AC^j from the gross investment

(Ł I^j) in each AQCR.  On the other hand, control cost to the electric

power industry becomes a price increase of electricity and passed on

to three types of users:

           Control cost passed to
           manufacturing industry        Acii'? Q{\
                                              j    J
           Control cost passed to
           the AQCR residents           Aqi'Qci

           Control cost passed to             _
           the other industries           Aqi' Qj

-------
where QJJ is electricity used by j'k industry in i"1 AQCR.  Therefore,

Aqi ? Qij becomes further reduction to the gross profit from manufac-

turing industry in each AQCR.  Qc^ is the electricity consumed by res-

idence of i"1 AQCR, hence, Acjj Qc^  is a reduction on the regional dis-

posable income  (Y-).   Finally, Aqi Qj is passed to the other industries.

      The benefits from control (i.e. , $1. 6 billion additional consump-

tion as the total benefit in the nation) was distributed to corresponding

sectors of the nation based on the average propensity to consumers by

sectors. *  By use of the national I-O model (Program IOA) direct and

indirect increase  of production  (value of shipment) by sector were es-

timated and then,  by use  of regional market share matrix (Program

FEE), the impact of benefit at the  national level is distributed  to each

AQCR.

      Under Strategy 2 (Hands Off), a part of the control cost is passed

on as price  increase.  A prior estimation of price increase by industry

was reported by the 305(a) Report  of 1971.  In this study, a weighted

average  price increase of 0.4 percent for manufacturing industries was
      *See Appendix D for percentage distributions among consumption
items and regional market share matrix.

-------
derived.* Therefore, under Strategy 2,  the control cost to the manu-

facturing industries becomes:

            AC*  =
                 =  r •     jj
                       j   J
r is percentage price increase and Ł Vjj is total production (value-

added) of manufacturing industries.  Hence, APj is the part of control

cost passed on to the consumer by price increase which reduces ACj

and AC2 proportionally, although the price increase in electricity re-

mains the same as before.  On the other hand, the price increase of

the manufacturing product in the nation will result in an inflationary

effect on disposable income, which will be reduced by the same amount.

It is estimated that $624 million out of $1,333 million control cost has

been passed on the consumer by the price increase.   This reduction of

the disposable income in the nation is distributed to  each AQCR in the

same manner as benefit.
      *In a recent study of the price increase among high emission in-
dustries, it is shown that the percentage price increase among differ-
ent products depends upon the price elasticity of the corresponding
product.  An 0.4 percent "average" price increase for manufacturing
industries was derived from the weighted average  of the percentage
price increase by two- to three-digit SIC industries weighted by the
corresponding volume of sales by each industry.  Data used in this
calculation is obtained from LeSourd, D. A. , et al. ,  Comprehensive
Study of Specified Air Pollution Sources to Assess the Economic
Effects  of Air Quality Standards,  Research Triangle Institute,
December,  1970.

-------
      Under Strategy 3 (Cost Sharing),  it has been assumed that the


Federal government will subsidize 50 percent of the remaining control


cost which industry cannot pass on by price increase.  Therefore, the


control cost of the manufacturing industries  become:

               **
           ACi   =  AC^ - APj - AFi


           AFL   =  J(ACi - APi)


      The government subsidy of A Fi is 50 percent of the remaining


control cost to the manufacturing industries.  This again is propor-


tionally reduced from ACji and AC^-  On the other hand,  it is


assumed that the Federal government will raise the same amount of


tax from the public.   An estimated amount of $355 million reduction


of disposable income in the nation was  assumed.  The impact on each


AQCR is also  treated through the I-O model and regional market share


matrix as before (see Table 5. 1).


      The results of the  simulation of the three strategies are pre-


sented and interpreted in the next chapter.

-------
                              TABLE 5. 1

             INCIDENCE OF CONTROL COSTS UNDER THE
                    THREE STRATEGIES (in millions)*
Strategy
1
2
3
Cost Incurred
Manufactur-
ing Industry
$1333
$ 709
$ 354
Consumers

$ 624
$ 629
By
Government


$ 355
Total
$1333
$1333
$1333
*In Strategy 1, high emission industries are assumed to absorb the
entire  cost of control.  In Strategy 2, high emission industries will
pass on part of the control costs as a 0.4  percent price increase to the
consumer.  In Strategy 3, it is  assumed that the Federal government
will raise, by a special tax to subsidize the high emission industries
by an amount equal to 50  percent of the control costs not covered by
the price increase in Strategy 2.  In all three  strategies,  it is further
assumed that electric utilities will pass on their control costs to the
consumers as a price increase.  For a detailed description of the three
strategies,  see page 94 or Chapter  5.

-------

-------
            	Inputs	
            Model Development
•»           Model             ^
 Regional Economic Model System
Outputs
                                                                                                  Assessment
             Regional Economic
             Activity
             National Economic
             Activity
                                                Regional
                                                Model
         Interregional
         Feedback
                                             I-O Model and
                                             National Effects
                                             Assessment
vO
tsJ
            Policy Assessment
             APCO Policy Variables
               Standards
               Incentives
               Fuels
               Research and
               development
               Government
               expenditures
     Computerized Simula-
     tion Program.

     .   Program RMS
     .   Program FEE
     .   Program IOA
                       Model System
                       Assessments
                                                              Recommenda-
                                                              tions on
                                                              Further Refine-
                                                              ment and Utili-
                                                              zation of Model

-------
6. 0   INTERPRETATION OF THE EFFECTS
      OF THREE TRIAL STRATEGIES
      This chapter is intended to provide a brief interpretation of the

preliminary results of the simulation of the three strategies specified

by APCO.  It opens with a brief recapitulation of the assumptions un-

derlying the three strategies.  It proceeds  to a brief discussion of the

framework of analysis of the model output through a series of questions.

Next, it identifies the geographic patterns of incidence of a key eco-

nomic indicator unemployment rate under the three alternatives.

Finally, it explores the patterns of change  of two other key indicators --

profits  and personal income -- under the three strategies on the 91

AQCRs, and assesses the potentials of the  interpretations attempted

here.


6. 1   The Three Strategies


      As described in the previous chapter, the three alternative strat-

egies tested with the APCO Economic Model System have as their basis

the control costs envisioned in the 1970 Report to Congress as required

by Section 305(a) of the Clean Air Act of 1967. * These three strate-

gies are summarized in Figure 6. 1.
      *Hereafter, it will be referred to as the 305(a) Report.

-------
                            FIGURE 6. 1
               THE THREE STRATEGIES AT A GLANCE
APC Strategy 1

1.  High emission industries will absorb the entire control cost as
    costs increase in the production without price increase.

2.  Electric utilities  (SIC 4911) will pass over the  control cost to the
    users as  price increases to industries and consumers.

3.  The benefits from air pollution control will result in a $1.6 billion
    increase  of consumption in the nation based on average propensity
    for consumption of goods and services,  generating an increase in
    production of corresponding industries in each AQCR.

APC Strategy 2

1.  High emission industries will increase the price of products as part
    of control cost increase.  It is estimated that $624 million out of a
    total control cost of $1, 333 million will be passed on to the con-
    sumer as the result of an  average 0.4 percent  increase of the  man-
    ufacturing product price.

2.  Because of the price increase in the products of the high  emission
    industries, consumers in  the nation will reduce consumption to
    $623 million.

APC Strategy 3

1.  Besides price increases assumed in the APC Strategy 2,  high  emis-
    sion  industries will receive 50 percent of Federal government sub-
    sidies for the part of control cost which is not  passed on  by price
    increases of the  products.

2.  Federal government will raise an amount of tax equal to the subsidy
    to the high emission industries; therefore, the national disposable
    income will be reduced by $355 million,  in addition to the price in-
    crease  on the products of high emission industries.

-------
6. 2   An Approach to Assessment of
      Efi'ects of the Three Strategies
      For each strategy simulated,  the APCO Economic Model System

gives a varied and detailed output describing the economic situation in

the region with and without the control strategy.  For example,  Table

6. 1 presents the summary table for the New York AQCR under the

"Industry Pays" strategy.  Evidently, the model produces  so great an

output that one needs a framework to analyze these  outputs and compare

the three  strategies, otherwise it may be difficult to "see the forest for

the trees. "

      The approach  taken here is  to pose questions  on which the model

impinges.  The questions to be considered are the following:

            .  What  effect does air pollution control have on the
              nation's  regions?

              How is this effect modified by government strate-
              gies for  modifying the incidence of control expen-
              ditures ?

            .  How are geographical effects modified  by govern-
              ment  strategies for modifying the incidence  of
              control expenditures?

            .  Are there any geographical patterns  of economic
              effects of air pollution control?

            .  If so,  what explanations can be advanced for these
              patterns ?

            .  What  changes in the structure of economic activity
              occur within regions?

-------
TOTAL NET EFFECT OF  ALL CONTROL STRATEGIES PURSUED  IN  "HIS RUN

AOCR   1   NEK  YORK,  N.Y.


    MANUFACTURING  INDUSTRIES

      PROFIT  (MILLIONS)
      INVESTMENT  (MILLIONS)
      VALUE ADDED  (MILLIONS)
      CAPITAL STOCK  (MILLIONS)
      EMPLOYMENT  ( 1000 S)

    OTHER INDUSTRIES

      EMPLOYMENT  (1000 S)
    REGIONAL CONSUMPTION (MILLIONS)
    TOTAL PERSONAL  INCOME ?OR THE. REGION  (MILLIONS)
    TOTAL REGIONAL  EMPLOYMENT (1000 S)
    REGIONAL UNEMPLOYMENT (PERCENT)
    TOTAL LABOR  FORCE (1000 S)
    GOVERNMENT EXPENDITURE FOR THE REGION  (MILLIONS)
    GOVERNMENT REVENUE FROM THE REGION (MILLIONS)
NO
    ELECTRIC POWER  DEMAND

      TOTAL ELECTRIC  CONSUMPTION FOR THE REGION  (1000   KWS)
      ELECTRICITY USED BY MANUFACTURING  INDUSTRIES  (1000  KWH)
      ELECTRICITY USED BY OTHER INDUSTRIES  (1000 KWH)
      RESIDENTIAL CONSUMPTION IN THE REGION  (1000   KWH/
                                                                           WITHOUT
                                                                           CONTROL
                                                                           5738.002
                                                                            452.000
                                                                          13836.699
                                                                          10884.758
                                                                           1144.000
                                                                            3979.71
 NET
CHANGE
-33.051
-?0.394
  5.273
-19.132
 -6.432
 -2.866
 PERCENT
 CHANGE
•0.57591
•4.51190
-0.09929
-0.17577
•0.56227
-0.07202
32445.000
537 12.000
5123.711
4.000
5337.199
5944.000
6416.000
-30.164
-41.4PB
-9.293
0.1741
-9.660
-3.929
-4.165
-0.09297
-0.07724
-0.18137
4.35349
-0. 1 a 100
-0.06610
-0.06491
4878.000
518.00
3205.00
1155.000
-6.364
-2.736
-2.205
-1.423
-0.13047
-0.52817
-0.068S1
-0.12322
                                                    TABLE 6.. 1


-------
              How are these changes moderated by incidence
              strategies ?

      The issues raised by these questions are addressed in several

steps.  First, a key economic indicator -- change in unemployment

rate --is analyzed in the various AQCRs and patterns of change under

the alternative strategies are interpreted.  Then the consistency of two

other key indicators -- profits in manufacturing industries, regional

personal income -- with the patterns evidenced in unemployment rate

changes is explored.   Next, an interpretation of the  net effects aggre-

gated over 91 AQCRs as evidenced by these three indicators is

attempted.  Finally, the potentials of the interpretation of  results

displayed here are assessed.
6. 3   Changes in Unemployment Rates
      Under the Three Strategies
      A key economic indicator,  change in regional unemployment rate

attendant on the three alternative air pollution control strategies, is

analyzed here.

      Table 6. 2 shows the change of unemployment rate in each AQCR

from the simulation outputs of the three alternative APC strategies.

      If one compares the change of unemployment rate among differ-

ent AQCRs, it is obvious that the air quality standard to be imple-

mented in the 305(a) Report will  have considerably different economic

-------
                                TABLE 6.2
   CHANGE OF UNEMPLOYMENT RATE FROM SIMULATION OF THREE
                            APC STRATEGIES
                                     Change of Unemployment Rate (%)
AQCR                                Strategy   Strategy   Strategy
Code    AQCR                            I          2           3

   1    New York,  N. Y.                0.17       0.13      0.07
   2    Chicago,  111.                    1.38       1.15      0.69
   3    Los Angeles,  Calif.            -0.003      0..02      0.03
   4    Philadelphia,  Pa.               0.58       0.46      0.29
   5    Detroit, Mich.                  1.21       1.07      0.62
   6    San Francisco,  Calif.          -0.03       0.07      0.13
   7    Boston, Mass.                  0.28       0.24      0.26
   8    Pittsburgh, Pa.                 2.06       1.57      1.03
   9    St. Louis, Mo.                  1.74       1.53      0.91
  10    Washington, D. C.               0.22       0.13      0.06
  11    Cleveland,  Ohio                 0.98       0.60      0.40
  12    Baltimore,  Md.                 1.00       0.81      0.47
  14    Minneapolis-St. Paul, Minn.     0.53       0.47      0.15
  15    Houston,  Texas                 0.49       0.58      0.39
  16    Buffalo, N. Y.                   1.40       1.08      0.69
  17    Milwaukee, Wis.               .1.39       1.20      0.86
  18    Cincinnati,  Ohio                1.54       1.14      0.58
  19    Louisville,  Ky.                 1.58       1.39      0.96
  20    Dallas, Texas                   0.13       0.08      0.02
  21    Seattle-Everett, Wash.          0.09       0.03      0.06
  22    Kansas City, Mo.                0.25       0.22      0.16
  23    San Diego,  Calif.               -0.03     -0.02     -0.01
  24    Atlanta, Ga.                    0. 08       0. 04      0. 06
  25    Indianapolis, Ind.                0.78       0.67      0.31
  26    Miami, Fla.                    0.03     -0.01       0.0
  27    Denver,  Colo.                   0.51       0.38      0.20
  28    New Orleans,  La.               0. 52       0. 12      0. 08
  29    Portland, Ore.                  0.40       0.15      0.08
  30    Providence-Pawtucket,  R.I.     0.23       0.18      0.14
  31    Phoenix,  Ariz.                  0.29       0.17      0.09
  32    Tampa, Fla.                    0.47       0.29      0.15
  33    Columbus, Ohio                 0.07       0.04      0.02
  34    San Antonio, Texas              0. 12       0. 04      0. 02
  35    Dayton, Ohio                    0.78       0.63       0.45
  36    Birmingham, Ala.               1.95       1.51       1.10
  37    Toledo, Ohio                    0.91       0.78       0.44
  38    Steubenville-Weirton,
         Ohio/W. Va.                   8.89       7.37       3.92
  39    Chattanooga, Tenn.              0.43       0.45       0.13

-------
                               TABLE 6. 2 (continued)
                                    Change .of Unemployment Rate (%)
AQCR
Code    AQCR

  40    Memphis,  Tenn.
  41    Salt Lake City,  Utah
  42    Oklahoma City,  Okla.
  43    Omaha,  Neb.
  44    Honolulu, Hawaii
  45    Beaumont-Port  Arthur-
          Orange, Texas
  46    Charlotte, N. C.
  47    Portland, Maine
  48    Albuquerque,  N. M.
  49    Lawrence-Haverhill/
          Lowell, Mass.
  50    El Paso, Texas
  51    Las Vegas, Nev.
  52    Fargo-Moorhead, N.D. ,
          Minn.
  53    Boise, Idaho
  54    Billings, Montana
  55    Sioux City,  Iowa
  61    Allentown-Bethelehem-
          Easton, Pa. ,  N. J.
  63    Bakersfield, Calif.
  64    Davenport-Rock Island-
          Moline, Iowa,  111.
  66    Grand Rapids/Muskegon-
          Muskegon Hts. ,  Mich.
  67    Greensboro, N. C.
  68    Harrisburg, Pa.
  69    Jacksonville,  Fla.
  70    Knoxville, Tenn.
  71    Nashville, Tenn.
  72    Peoria,  111.
  73    Richmond,  Va.
  74    Rochester,  N. Y.
  75    Saginaw/Bay City, Mich.
  76    Scranton-Wilkes Barre-
          Hazelton,  Pa.
  77    Syracuse,  N.  Y.
  78    Tulsa, Okla.
Strategy
1
-0.03
0. 72
0.31
0.69
0. 18
0. 77
0.50
0.07
0.02
0.87
0.35
0. 16
0. 57
0.29
0.30
0.02
2.03
0.22
1.22
1.61
0.08
1. 14
0.25
1.04
0.08
1.80
0.45
-0.63
5.23
5.27
0.33
0.61
Strategy
2
-0.04
0.38
0. 18
0.51
0. 12
0. 18
0.35
-0. 74
0. 08
0. 79
0. 11
0. 13
0.25
0. 11
0.05
-0.01
1.56
0.05
0.95
1.45
0.05
0.69
0. 10
0.50
0.09
1.54
0.35
0.22
4.67
4.49
0.30
0.33
Strategy
3
-0. 02
0. 19
0. 12
0.22
-0.05
0.09
0. 16
0.20
-0.02
0.39
0.09
0. 07
0. 12
0.07
0.03
-0. 06
1.24
0.02
0.61
0.69
0. 02
0.31
0.02
0.20
0.05
0.81
0.20
-0. 17
2.35
2.25
0.23
0. 17

-------
                               TABLE 6.2  (continued)


                                    Change of Unemployment Rate (%)
AQCR                              Strategy   Strategy   Strategy
Code    AQCR                          1         2          3
  80    Youngstown-Warren, O.        1.69       1.13       0.61
  81    Albany-Schenectady-
          Troy, N. Y.                  0.28       0.22       0. 15
  82    Binghamton,  N. Y.              0.84       0.65       0.32
  83    Charleston, S. C.              0.51       0.24       0.12
  84    Charleston, W. Va.             1.73       1.48       0.74
  85    Des Moines, Iowa              0.79       0.59       0.30
  86    Fresno, Calif.                 0.07       0.01       0.01
  87    Fort Wayne, Ind.               0.91       0.73       0.36
  88    Jackson, Miss.                 0.01     -0.28       0.05
  89    Johnstown, Pa.                 2.07       1.25       0.62
  90    Lancaster, Pa.                 0.77      0.62       0.34
  91     Mobile, Ala.                    0.64      0.43       0.20
  92     Norfolk-Portsmouth/
         Newport  News-Hampton, Va.   0.42      0.21       0.11
  93     Raleigh/Durham,  N.  C.         0.77      0.6l       0.30
  94     heading, Pa.                    1.52      1.23   .    0.75
  95     Rockford,  111.                   1.47      1.34       0.70
  96     Sacramento, Calif.              0.02      0.01       0.0
 97     South Bend, Ind.                2.05      1.66       0.86
 98     Utica-Rome, N.  Y.              0.42      0.35       0.25
 99     Wichita, Kan.                   0.01      -0.07     -0.03
100     York,  Pa.                       1.08      0.92      0.48

-------
effects on each AQCR.  Some AQCRs,  for example, Steubenville-




Weirton, Pittsburgh, etc. , will possibly expect a 2 percent or more




increase in the regional unemployment.  However,  for others, the im-




pacts are negligible; some AQCRs are even better off with a decrease




in employment rate as a result of air pollution control.  This could




mean that higher air quality standards for those AQCRs  with negligible




effects may be "practical. "  The 91 AQCRs under this study can be




classified into four major categories, namjely, better off (decrease in




unemployment rate), negligible (increases  under one percent), mod-




erate (increases between one to two percent), and serious (increases




more than two percent).




      The geographic distribution of all 91  AQCRs under this  classifi-




cation is shown in Figure 6.2.  Under APC Strategy 1 (Industry Pays),




most of the AQCRs seriously adversely affected are located in the




heavily industrial north-central (Michigan,  Ohio, Indiana, Illinois) and




central-east (Pennsylvania,  West Virginia) states.   AQCRs located in




the west and south,  in general, do not seem to be affected by air pollu-




tion control and some are even better off.  Therefore, air pollution




control under Strategy  1 may conceivably lead to a locational redistri-




bution of the economic activity of the nation, by increased growth in




the new metropolitan areas and under the greater economic pressure




on the older heavy industrial areas.

-------
                            FIGURE 6. I
      GEOGRAPHIC DISTRIBUTION OF ECONOMIC EFFECTS UNDER
APC STRATEGY 1 (MEASURED  BY CHANGE OF UNEMPLOYMENT RATE)
                                          O   Better Off
                                              Negligible

                                          3   Moderate


-------
      The geographic distribution of the outputs from Strategy 2 and




Strategy 3 is provided in Figures  6.3 and 6.4, respectively.  By com-




paring the outputs for  alternative  strategies,  the economic effects of




different strategies seem to be clear.




      Under Strategy 2, the high emission industries, to some degree,




would be able to pass on a portion of control costs to the public as price




increase in the products.   Therefore, the  AQCRs with heavy concentra-




tions of high emission industries will be in a better condition than under




Strategy 1.




      Under Strategy 3, the cost pressures on the high emission indus-




tries are further released by 50 percent of government subsidies.




Therefore, those AQCRs  with such industries  are improved further




than in Strategy 2.  However, price increases and additional taxes (for




pollution subsidies) will have an adverse economic effect on those




AQCRs  with large populations (and,  hence, regional income and con-




sumption).




      The number of AQCRs in each category changes  considerably




under different strategies.  The percentage distribution between the




four categories  is provided below:

-------
                         FIGURE 6. 3
   GEOGRAPHIC DISTRIBUTION OF ECONOMIC EFFECTS UNDER
APC STRATEGY 2 (MEASURED BY CHANGE OF UNEMPLOYMENT RATE)
                                       O  Better Off
                                       ©  Negligible

                                       3  Moderate


-------
                          FIGURE 6.4
    GEOGRAPHIC DISTRIBUTION OF ECONOMIC EFFECTS UNDER
APC STRATEGY 3 MEASURED BY CHANGE OF UNEMPLOYMENT RATE)
                                        O   Better Off
                                        ź   Negligible

                                        3   Moderate


-------
Categories of Change "of                AQCRs Included (%)
Unemployment Rate (%)
in AQCRs                        Strategy 1  Strategy 2  Strategy 3

1.  Better off (increase
    in employment)                   5. 5%       7. 7%         9. 9%
2.  Negligible (0.01% to
    1.00%)                          67.0%      69.2%        83.5%
3.  Moderate (1. 01% to
    2.0%)                           19.8%      19.8%         3.3%
4.  Serious  (2. 01% and
    over)                            7.7%       3.3%         3.3%

    Total                          100.0%     100.0%       100.0%

      Under Strategy 1, 27. 5 percent of AQCRs will have a moderate or

serious economic effect with an increase of more than one percent of

unemployment rate.  This figure is reduced to 23 percent under Strate-

gy 2 and further down to 6. 6 percent under Strategy 3.  In other words,

regional difference of economic impacts among AQCRs will be much

greater in Strategy 1 and Strategy 2 than Strategy 3, although all three

alternative strategies were  under "same" emission standard specified

in the 305(a) Report of 1970.


6.4   Changes in Profits and Personal Income


      Unemployment rate,  though important,  is but one indicator.   Con-

sequently, this section explores the patterns of change in two other key

indicators.  The first is the decline in profits of manufacturing indus-

tries in the  AQCRs --a key indicator of the high emission industries.

The second  indicator is the  decline in regional personal income.

-------
      The changes by region, in the unemployment rate, and percent-




age changes in personal income and profits under each policy,  are pre-




sented in Tables 6. 3 through 6. 5.  For each column,  the median and




quartiles have been identified.  The figure beside each number indi-




cates the quartile in which the region falls where (1) indicates least




negatively affected  quartile,  and (4) indicates most negatively affected




quartile.  In addition,  a rank column has been placed by each result




column ranking the AQCRs from least negatively affected to most neg-




atively affected.




      These tables, singularly and severally,  tell an interesting story.




First, for any given strategy,  there  appears to be rough consistency as




between the quartile rankings of any  given AQCR for each of the three




indicators of economic activity selected for analysis.  That is,  if un-




employment increases  a relatively great amount in some region,  per-




sonal income and profits decrease by a relatively great amount.   There




also appears to be  a rough consistency between the ranking of an AQCR




for the changes arrayed for  any given policy.  It would, indeed, be in-




teresting and perhaps  informative exercise to look at the rank correla-




tion coefficients  between these various measures.




      For all indicators studied here, in fact, median values tend to




decrease as incidence becomes less  geographically localized.  The

-------
                                              TABLE 6. 3

                   CHANGES IN UNEMPLOYMENT RATE, PROFITS AND PERSONAL INCOME

                               IN THE AQCRs UNDER STRATEGY 1
o
00
AQCR
1
2
3
4
5
6
7
8
9
10
11
12
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
Increase in
Unemploy-
ment Rate
0. 1741
1.3850
-0. 0034
0. 5851
1.2074
-0. 0380
0.2775
2.0567
1.7359
0.2174
0.9812
1. 0027
0.5293
0.4904
1.4004
1.3947
1.5448
1.5802
0. 1312
0. 0946
0.2510
-0. 0303
0. 0763
0. 7753
0. 0291
0. 5057
0. 5179
0.3992
0.2309
0.2901
Rank
22
72
7
50
70
2
29
86
82
24
65
66
48
43
74
73
77
78
20
18
28
3
15
58
11
45
47
37
26
31
Q
1
4
1
3
4 '
1
2
4
4
2
3
3
3
2
4
4
4
4
1
1
2
1
1
3
1
2
3
2
2
2
% Change
in Profit
-0.5759
-1.8681
-0.0177
-0.7452
-1.5538
-0.0297
-0.3103
-4.3805
-2. 1337
-2.9775
-1.8266
-1.2875
-0.6567
-0.2687
-2. 1073
-1. 1307
-1.4598
-1.0973
-0. 1315
-0.2603
-0.3093
0.0887
-0.4437
-0.7649
-0.2047
-1.4104
-0.8355
-0.7502
-0.3380
-0.2044
Rank
43
80
10
50
77
11
28
89
83
86
79
70
48
25
82
68
75
67
16
24
27
6
33
53
20
73
55
51
29
19
Q
2
4
1
3
4
1
2
4
4
4
4
4
3
2
4
3
4
3
1
2
2
1
2
3
1
4
3
3
2
1
% Change
in Personal
Income
-0.0772
-0.8325
0. 0303
-0.3880
-0.6638
0. 0847
-0. 1329
-1..8569
-1.0087
-0. 1485
-0.8609
-0.7273
-0.2745
-0.3593
-1.3009
-0.8505
-0.8754
-0.9246
-0. 1060
-0.0631
-0. 1560
0.0301
0.0080
-0.4213
-0.0478
-0.3106
-0.6643
-0.4447
-0. 1783
-0.2474
Rank
17
67
5
47
61
2
22
85
74
23
69
65
32
41
82
68
70
71
20
16
24
6
7
49
15
35
62
50
25
30
Q
1
3
1
3
3
1
1
4
4
1
3
3
2
2
4
3
4
4
1
1
2
1
1
3
1
2
3
3
2

-------
                           TABLE 6.3 (continued)
CHANGES IN UNEMPLOYMENT RATS,  PROFITS AND PERSONAL INCOME
            IN THE AQCRs UNDER STRATEGY 1
AQCR
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
61
63
64
66
67
68
Increase in
Unemploy-
ment Rate
0.4742
0.0681
0.1195
0.7827
1.9490
0.9080
8.8929
0.4265
-0.0284
0.7163
0.3130
0.6940
. 0. 1757
0.7718
0.4986
0.0726
0.0199
0.8728
0.3488
0. 1643
0.5717
0.2932
0.3017
0.0194
2. 0333
0.2176
1.2150
1.6144
0.0766
1. 1409
Rank
42
12
19
59
84
64
91
40
4
54
34
53
23
57
44
14
10
62
36
21
49
32
33
9
85
25
71
79
16
69
Q
2
1
1
3
4
3
4
2
1
3
2
3
1
3
2
1
1
3
2
1
3
2
2
1
4
2
4
4
1
3
% Change
in Profit
-1.4185
-0.0554
-0.5954
-0. 7164
-3.9978
-1. 0426
-6.7709
-0.2110
0,0983
-0. 8865
-1. 0664
-0.4881
-0.5232
-0.0497
-0.5148
-0.4591
-0.3007
-1.0052
-0. 7613
-0.5383
0.4038
-0.5470
0.0691
0. 1336
-2.2695
-0.5119
-0.5209
-1.5296
-0. 0064
-1. 7794
R.ank
74
14
44
49
88
63
91
21
4
57
65
35
39
13
37
34
26
61
52
41
1
42
7
3
84
36
38
76
9
78
Q
4
1
2
3
4
3
4
1
1
3
3
2
2
1
2
2
2
3
3
2
1
2
1
1
4
2
2
4
1
4
% Change
in Personal
Income
-0.3768
-0.0438
-0. 1233
-0.5055
-2. 1093
-0.5222
-12.0438
-0,3667
0.0673
-0.6629
-0.3216
-0.3869
-0. 1013
-0.9856
-0.3625
-0.3381
-0.0248
-1. 1710
-0.4044
-0. 1807
-0.5441
-0.2882
-0.3472
0. 0679
-1.6683
-0. 1870
-1.0585
-0.9517
-0. 0281
-1. 0803
Rank
44
14
21
52
88
53
91
43
4
60
36
46
19
73
42
37
10
79
48
26
54
33
38
3
84
27
75
72
12
77
Q
2
1
1
3
4
3
4
2
1
3
2
2
1
4
2
2
1
4
3
2
3
2
2
1
4
2
4
4
1

-------
                           TABLE 6.3 (continued)
CHANGES IN UNEMPLOYMENT RATE, PROFITS AND PERSONAL INCOME
            IN THE AQCRs UNDER STRATEGY 1
AQCR
69
70
71
72
73
74
75
76
77
78
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
Increase in
Unemploy-
ment Rate
0.2502
1. 0365
0.0766
1. 8022
0.4490
-0.5312
5.2258
5.2672
0.3257
0.6129
1.6879
0.2815
0. 8404
0.5143
1.7346
0.7975
0.0681
0.9059
-0.0058
2.0708
0.7706
0.6360
0.4220
0.7677
1.5157
1.4734
0. 0178
2.0593
0.4194
-0.0138
1.0832
Rank
27
67
17
83
41
1
89
90
35
51
80
30
61
46
81
60
13
63
6
88
56
52
39
55
76
75
8
87
38
5
68
Q
2
3
1
4
2
1
4
4
2
3
4
2
3
2
4
3
1
3
1
4
3
3
2
3
4
4
1
4
2
1
3
% Change
in Profit
-0.6084
-0.9269
0.0971
-0.9415
-0.2514
0.0042
-3.3248
-4.6192
-0. 1973
-1.0661
-2.0096
-0.3918
-0.5369
-0.9075
-1.3058
-1.079-6'
-0. 1798
-0.8417
-0. 1189
-2.9529
-0.6337
-0.8280
-1. 1985
-0.4098
-1.0319
-0.3757
-0.0404
-1.3659
-0.2200
0. 1466
-0.6439
Rank
45
59
5
60
23
8
87
90
18
64
81
31
40
58
71
66
17
56
15
85
46
54
69
32
62
30
12
72
22
2
47
Q
2
3
1
3
1
1
4
4
1
3
4
2
2
3
4
3
1
3
1
4
2
3
3
2
3
2
1
4
1
1
3
% Change
in Personal
Income
-0.2712
-1. 1647
-0.0430
-1.0764
-0. 3089
0.3802
-2.6546
-3.6113
-0.2157
-0.5510
-1.9618
-0.2080
-0. 5480
-0.5832
-1.2865
-0.6645
-0.0894
-0.6359
-0. 0016
-2.0432
-0.6940
-0.4740
-0.3776
-0.3589
-1. 1798
-0. 7762
-0. 0227
-1.3341
-0.3563
-0.0251
-0.6172
Rank
31
78
13
76
34
1
89
90
29
56
86
28
55
57
81
63
18
59
8
87
64
51
45
40
80
66
9
83
39
11
58
Q
2
4
1
4
2
1
4
4
2
3
4
2
3
3
4
3
1
3
1
4
3
3
2
2
4
3
1
4
2
1

-------
                          TABLE 6.4
CHANGES IN UNEMPLOYMENT RATE, PROFITS AND PERSONAL INCOME
            IN THE AQCRs UNDER STRATEGY 2
AQCR
1
2
3
4
5
6
7
8
9
10
11
12
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
Increase in
Unemploy-
ment Rate
o. 1290
1. 1541
0.0233
0.4594
1.0678
0. 0749
0.2380
1.5702
1.5329
0. 1257
0.5983
0.8060
0.4729
0.5811
1.0833
1. 1975
1.1433
1.3923
0. 0804
0.0279
0.2169
-0.0219
0. 0360
0.6666
-0.0148
0.3834
0. 1200
0. 1466
0. 1848
0. 1698
Rank
28
75
10
51
71
18
39
87
84
27
57
68
52
55
72
76
74
80
20
11
36
5
13
62
6
48
26
30
33
3]
Q
2
4
1
3
4 '
1
2
4
4
2
3
3
3
3
4
4
4
4
1
1
2
1
1
3
1
3
2
2
2
2
% Change
in Profit
-0.2086
-0.9241
0.0055
-0.3901
-0.9805
-0.0153
-0. 1558
-2.2961
-1.2837
-0.9358
-0.8366
-0. 7327
-0.3125
-0. 1317
-1.2047
-0.7096
-0.6951
-0.5938
-0.0512
-0.0218
-0. 1541
0.0669
-0. 1453
-0.4332
-0.0160
-0.5753
-0. 1794
-0. 1833
-0. 1732
-0. 1138
Rank
38
77
8
53
79
12
30
88
84
78
74
70
49
25
83
69
68
62
19
16
29
5
28
57
13
61
34
35
32
22
Q
2
4
1
3
4
1
2
4
4
4
4
4
3
2
4
3
3
3
1
1
2
1
2
3
1
3
2
2
2
1
% Change
in Personal
Income
-0.0298
-0.5672
-0.0014
-0.2394
-0.5200
-0.0245
-0.0910
-1.2675
-0.7640
-0.0559
-0.4117
-0.4924
-0.2003
-0.4418
-0.8931
-0.6022
-0.4295
-0.6888
-0.0455
0.0003
-0. 1130
0.0217
0. 0611
-0.2929
-0. 0019
-0. 1643
-0. 1502
-0. 1404
-0. 1192
-0. 1029
Rank
17
71
11
53
70
16
28
87
78
23
65
69
48
67
82
72
66
74
22
10
30
8
7
55
12
41
38
37
31
29
Q
1
4
1
3
4
1
2
4
4
1
3
3
3
3
4
4
3
4
1
1
2
1
1
3
1
2
2
2
2

-------
                            TABLE 6.4 (continued)
CHANGES IN UNEMPLOYMENT RATE,  PROFITS AND PERSONAL INCOME
            IN THE AQCRs UNDER STRATEGY 2
AQCR
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
61

63
64
66
67
68
Increase in
Unemploy-
ment Rate
0.2898
0.0360
0. 0446
0.6327
1. 5072
0. 7756
7.3687
0.4463
-0.0363
0.3819
0. 1753
0. 5088
0. 1153
0. 7802
0.3502
-0. 7406
0.0776
0. 7851
0. 1112
0. 1301
0.2502
0. 1138
0. 0492
-0.0132
1. 5642

0.0539
0.9514
1.4548
0. 0497
0.6891
Rank
41
12
14
60
83
65
91
50
4
47
32
54
25
66
45
1
19
67
23
29
40
24
15
7
86

17
70
81
16
63
Q
2
1
1
3
4
3
4
3
1
3
2
3
2
3
2
1
1
3
1
2
2
2
1
1
4

1
4
4
1
3
% Change
in Profit
-0.6123
-0.0198
-0.2052
-0.4063
-2. 1350
-0.5692
-3.0258
-0.0829
0.0735
-0.2759
-0.6231
-0.2200
-0. 1168
-0.0113
-0.2317
-0.2198
-0.0204
-0.8491
-0. 1367
-0. 1122
-0.3681
-0. 1389
0. 1058
0.2249
-1.4259

-0. 1168
-0.3859
-1. 1449
0.0026
-1.0432
Rank
63
14
37
56
87
60
90
20
4
44
64
41
23
11
43
40
15
75
26
21
50
27
3
1
85

24
52
82
9
80
Q
3
1
2
3
4
3
4
1
1
2
3
2
1
1
2
2
1
4
2
1
3
2
1
1
4

2
3
4
1
4
% Change
in Personal
Income
-0. 1557
-0.0020
-0.0346
-0.3308
-1.4813
-0.3728
-8.0880
-0.3922
0.0720
-0.2312
-0. 1271
-0. 1661
-0.0407
-0.2292
-0. 1631
0. 5802
-0.0898
-0.9060
-0. 1265
-0. 1276
-0. 1697
-0.0795
-0.0420
0.0985
-1.0557

-0.0455
-0.7013
-0. 7463
0.0078
0.4522
Rank
39
13
18
57
88
58
91
62
6
52
34
42
19
51
40
1
27
83
33
36
43
26
20
4
85

21
75
76
9
2
Q
2
1
1
3
4
3
4
3
1
3
2
2
1
3
2
1
2
4
2
2
2
2
1
1
4

1
4
4
1

-------
                                                TABLE 6.4 (continued)
                    CHANGES IN UNEMPLOYMENT RATE, PROFITS AND PERSONAL INCOME
                                IN THE AQCRs UNDER STRATEGY 2
uo
AQCR
69
70
71
72
73
74
75
76
77
78
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
Increase in
Unemploy-
ment Rate
0.0958
0. 5045
0. 0934
1.5374
0.3534
0.2158
4.6720
4.4876
0.3003
0.3339
1. 1264
0.2170
0.6460
0.2376
1.4824
0.5910
0.0135
0. 7348
-0.2814
1.2532
0.6178
0.4294
0.2096
0.6087
1.2275
1.3387
0. 0070
1.6596
0. 3480
-0. 0697
0.9217
Rank
22
53
21
85
46
35
90
89
42
43
73
37
61
38
82
56
9
64
2
78
59
49
34
58
77
79
8
88
44
3
69
Q
1
3
1
4
2
2
4
4
2
2
4
2 .
3
2
4
3
1
3
1
4
3
3
2
3
4
4
1
4
2
1
3
% Change
in Profit
-0.2110
-0.3738
0. 0431
-0. 5421
-0. 1750
0. 0434
-2.6109
-4. 0735
-0. 1684
-0.6621
-1.0612
-0.2268
-0.3116
-0.3905
-0.8828
-0.8235
-0.0410
-0.6403
-0.0271
-1. 7245
-0.5345
-0.2828
-0.6494
-0.3062
-0.8172
-0.2946
-0.0092
-0.7454
-0. 1939
0. 1070
-0.4029
Rank
39
51
7
59
33
6
89
91
31
67
81
42
48
54
76
73
18
65
17
86
58
45
66
47
72
46
10
71
36
2
55
Q
2
3
1
3
2
1
4
4
2
3
4
2
3
3
4
4
1
3
1
4
3
2
3
3
4
2
1
4
2
1
3
% Change
in Personal
Income
-0.0780
-0.3851
-0.0676
-0. 7500
-0. 1866
-0.3886
-2.0691
-2.4319
-0. 1837
-0.2035
-1.2202
-0. 1276
-0.2961
-0. 1877
-0.9238
-0.3730
-0.0179
-0.4115
0.4219
-0.8606
-0.4570
-0. 1936
-0. 1251
-0.2161
-0. 7900
-0.6081
-0. 0083
-0. 8302
-0.2638
0. 0949
-0.4024
Rank
25
60
24
77
45
61
89
90
44
49
86
35
56
46
84
59
15
64
3
81
68
47
32
50
79
73
14
80
54
5
63
Q
2
3
2
4
2
3
4
4
2
3
4
2
3
2
4
3
1
3
1
4
3
3
2
3
4
4
1
4
3
1

-------
                           TABLE 6. 5
CHANGES IN UNEMPLOYMENT RATLJ, PROFITS AND PERSONAL INCOME
            IN THE AQCRs UNDER STRATEGY 3
AQCR
1
Z
3
4
5
6
7
8
9
10
11
12
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
Increase in
Unemploy-
ment Rate
0.0714
0.6910
0.0349
0.2918
0.6206
0. 1270
0.2562
1.0321
0.9080
0.0629
0.3981
0.4717
0. 1463
0.3858
0.6931
0.8577
0.5819
0.9592
0.0176
0.0557
0. 1579
-0.0069
0. 0623
0.3100
0. 0022
0.2009
0. 0796
0. 1430
0. 1381
0.0889
Rank
25
75
18
55
73
35
54
86
84
23
65
68
39
63
76
82
70
85
12
21
42
7
22
59
9
47
27
38
37
29
Q
2
4
1
3
4
2
3
4
4
1
3
3
2
3
4
4
4
4
1
1
2
1
1
3
1
3
2
2
2
2
% Change
in Profit
-0. 1128
-0.5141
-0.0199
-0.2475
-0.5788
-0.0568
-0.0969
-1.4975
-0.7246
-0.4698
-0. 5404
-0.4738
-0. 1349
-0. 1054
-0. 7274
-0.4764
-0.3673
-0.3142
-0.0409
-0.0295
-0. 1054
0.0350
-0. 1046
-0.2271
-0.0245
-0.2936
-0. 1221
-0. 1386
-0. 1347
-0.0850
Rank
34
76
14
54
81
21
27
89
82
73
79
74
39
32
83
75
67
60
19
18
33
4
30
53 -
16
57
35
41
38
25
Q
2
4
1
3
4
1
2
4
4
4
4
4
2
2
4
4
3
3
1
1
2
1
2
3
1
3
2
2
2
2
% Change
in Personal
Income
-0.0225
-0.3900
-0. 0219
-0. 1871
-0.3500
-0.0891
-0. 1632
-0.9354
-0.4859
-0.0273
-0.2978
-0.3930
-0. 0578
-0.3050
-0.6356
-0.4053
-0.2069
-0.5045
0. 0022
-0.0216
-0. 1144
0.0075
0.0188
-0. 1429
-0.0063
-0. 0911
-0. 1010
-0. 1374
-0. 1201
-0.0714
Rank
22
72
21
57
69
37
54
86
80
23
66
73
27
67
84
74
61
81
12
19-
45
10
8
51
13
38
43
50
48
30
Q
1
4
1
3
3
2
3
4
4
1
3
4
2
3
4
4
3
4
1
1
2
1
1
3
1
2
2
3
3

-------
                            TABLE 6. 5 (continued)
CHANGES IN UNEMPLOYMENT RATE, PROFITS AND PERSONAL INCOME
            IN THE AQCRs UNDER STRATEGY 3
AQCR
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
61
63
64
66
67
68
Increase in
Unemploy-
ment Rate
0. 1466
0. 0234
0.0176
0.4475
1. 1020
0.4393
3.9276
0. 1342
-0.0157
0. 1946
0. 1174
0.2215
-0.0503
0.0913
0. 1678
0.2042
-0.0202
0.3947
0.0863
0.0724
0. 1155
0.0661
0.0251
-0. 0596
1.2386
0. 0235
0.6115
0.6943
0. 0177
0.3068
Rank
40
15
11
67
87
66
91
36
6
46
34
51
3
30
43
50
5
64
28
26
32
24
17
2
88
16
72
77
13
58
Q
2
1
I
3
4
3
4
2
1
2
2
3 .
1
2
2
3
1
3
2
2
2
2
1
1
4
1
4
4
1
3
% Change
in Profit
-0. 3040
-0.0272
-0. 1048
-0.2188
-1.4771
-0.3317
-1.6558
-0.0411
0.0339
-0. 1391
-0.3409
-0.0991
0.0312
-0.0057
-0. 1294
-0.3884
0.0197
-0.4246
-0.0651
-0. 1334
-0.3159
-0.0733
0. 0445
0. 1714
-0.9607
-0.0584
-0.2913
-0.5651
0.0026
-0. 5159
Rank
58
17
31
52
88
64
90
20
5
42
66
28
6
11
36
68
8
71
23
37
61
24
3
1
86
22
56
80
9
77
Q
3
1
2
3
4
3
4
1
1
2
3
2
1
1
2
3
1
4
1
2
3
2
1
1
4
1
3
4
1
4
% Change
in Personal
Income
-0.0799
-0.0137
-0.0093
-0.2365
-1.2348
-0.2453
-4.5999
-0..1512
0.0281
-0. 1182
-0.0803
-0.0592
0.0194
-0.1159
-0.0836
0.2040
0.0201
-0.4589
-0. 0979
-0. 0302
-0.0776
-0.0464
-0. 0218
0.0824
-0.9283
-0.0204
-0.5241
. -0.3578
0.0085
-0. 1893
Rank
32
16
15
64
90
65
91
52
5
47
34
28
7
46
35
2
6
79
42
24
31
25
20
3
85
17
82
70
9
59
Q
2
1
1
3
4
3
4
3
1
3
2
2
1
2
2
1
1
4
2
2
2
2
1
1
4
1
4
4
1

-------
                           TABLE 6.5 (continued)
CHANGES IN UNEMPLOYMENT RATE,  PROFITS AND PERSONAL INCOME
            IN THE AQCRs UNDER STRATEGY 3
AQCR
69
70
71
72
73
74
75
76
77
78
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
Increase in
Unemploy-
ment Rate
0.0180
0.2036
0. 0473
0.8108
0. 1897
-0. 1650
2.3466
2.2496
0.2262
0. 1709
0.6056
0. 1548
0.3215
0.1169
0. 7427
0.2981
0.0051
0.3630
0.0526
0.6218
0.3399
0.2039
0. 1078
0.2955
0.7491
0.6990
-0. 0035
0.8587
0.2542
-0.0287
0.4812
Rank
14
48
19
81
45
1
90
89
52
44
71
41
60
33
79
57
10
62
20
74
61
49
31
56
80
78
8
83
53
4
69
Q
1
3
1
4
2
1
4
4
3
2
4
2
3
2
4
3
1
3
1
4
3
3
2
3
4
4
1
4
3
1
3
% Change
in Profit
-0.0925
-0. 1702
0. 0246
-0.3190
-0. 1001
-0.0088
-1.3073
-2.0409
-0. 1377
-0.3387
-0.7389
-0. 1544
-0.1590
-0.1931
-0.4437
-0.4128
-0.0205
-0.3195
-0.0136
-0.8602
-0.2903
-0. 1421
-0.3080
-0.1542
-0.5197
-0. 1678
-0.0046
-0.3905
-0. 1595
0.0511
-0.2178
Rank
26
49
7
62
29
12
87
91
40
65
84
45
46
50
72
70
15
63
13
85
55
43
59
44
78
48
10
69
47
2
51
Q
2
3
1
3
2
1
4
4
2
3
4
2
2
3
4
4
1
3
1
4
3
2
3
2
4
3
1
3
3
1
3
% Change
in Personal
Income
-0.0206
-0. 1742
-0.0530
-0.4343
-0. 1212
0.2035
-1.0372
-1.2092
-0. 1780
-0.0963
-0.9520
-0. 1144
-0. 1584
-0.0947
-0.4543
-0. 1877
-0.0069
-0.2023
-0.0800
-0.4271
-0.3757
-0.0852
-0.0700
-0.0977
-0.5721
-0.3477
0. 0034
-0.4403
-0.2361
0.0396
-0.2279
Rank
18
55
26
76
49
1
88
89
56
40
87
44
53
39
78
58
14
60
33
75
71
36
29
41
83
68
11
77
63
4
62
Q
1
3
2
4
3
1
4
4
3
2
4
2
3
2
4
3
1
3
2
4
4
2
2
2
4
3
1
4
3
1

-------
median for each indicator under each of the three policies is presented

below in Table 6.6.


                             TABLE 6.6

             Median Values of the Distributions of the Changes
             in Unemployment Rate, Percentage Changes in
             Profits and Personal Income in the Three Strategies*

Increase in Unemployment Rate
Percentage Decrease in Profits
of Manufacturing Industries
Percentage Decrease in
Personal Income
Strategy 1
(Industry
Pays)
0.5455
0.6667
0.4286
Strategy 2
(Hands Off)
0.3654
0.2885
0.2035
Strategy 3
(Cost
Sharing)
0.2102
0.2000
0. 1557
*In Strategy  1, high emission industries are assumed to absorb the
entire cost of control.  In Strategy 2, high emission industries will
pass on part of the control costs as a 0. 4  percent price increase to the
consumer.  In Strategy 3, it is  assumed that the  Federal government
will raise, by a special tax to subsidize the high  emission industries
by an amount equal to 50  percent of the control costs not covered by
the price  increase in Strategy 2.  In all three  strategies, it is further
assumed that electric utilities will pass on their  control costs to the
consumers as a price increase.
      The trend in the three indicators presented in this table tend to

be consistent with the analysis of changes in unemployment rate pre-

sented in the previous section.  The profits of manufacturing industries

show the highest sensitivity among the three variables.

-------
      A variety of other economic indicators that reflect the impact of

alternative strategies are printed out in the simulation as  shown in

Table 6. 1.
6. 5   Total Net Effects of Three Alternative
      Strategies for 91 AQCRs
      This section presents some evidence on the effects of the three

air pollution control strategies totalled for 91 AQCRs.  A number of

caveats are in order.

      First,  the APCO Model System,  as it now stands, is a cross-

sectional regional model; consequently, its  strength lies  in its assess-

ment of geographical patterns  of change in the AQCRs.   Its estimates

of aggregate changes are less  reliable.  Second,  the 91 AQCRs used in

the model cover the greater portion of the economic activity in the

nation, but not all.  Finally, the model system does not capture the

dynamic  effects and macro effects that only a national model can handle.

Consequently, the interpretation presented below must  be viewed with

a dose of caution.

      Some  evidence on the overall impact of air pollution control and

the effect of incidence policy thereon is presented in Table 6. 7, which

-------
                             TABLE 6.7
             NET EFFECTS OF THREE ALTERNATIVE
              STRATEGIES ON 91 AQCRS*
Changes
in Key
Indicators

Profits in
Manufacturing
Industries
Regional
Personal
Income
Employment**
Strategy 1
No. %
-$ 345 -0.47
(million)
-$ 841 -0.21
(million)
-146,000 -0.43
Strategy 2
No. %
-$ 396-0.54
(million)
-$ 1, 123 -0. 28
(million)
-171, 000-0.48
Strategy 3
No. l/0
-$ 429 -0.59
(million)
-$ 1,279 -0. 32
(million)
-185,000 -0.50
      *For a detailed description of the three strategies, see page 94
or Chapter 5.                           :  .

     **This indicates an increment of base unemployment rate of 3. 45
to 3. 88,  3. 94 and 3. 95,  respectively.

-------
presents totals for the 91 AQCRs in the simulation. * The table reflects

a decline in economic activity regardless of the incidence policy pursued.

Under the "Hands Off" strategy, which is in many ways the most plaus-

ible Federal posture, the unemployment rate for the 91 regions in-

creases from 3.45 percent to 3.93 percent, personal income declines

by 0. 28 percent and profits fall by 0. 54 percent.

      Somewhat more surprising at first blush are the results  as one

moves from a strategy which localizes the incidence of control (Strate-

gy 1 -- Industry Pays) costs to one which spreads the incidence over the

whole of the nation (Strategy 3 -- Cost Sharing).  From an aggregative

point  of view, Table 6. 7 shows that the  total effect of control is mini-

mized if steps are taken to see that the  sources incurring air pollution

control costs bear their  full incidence.  This can be seen by comparing

the percentage changes in these three tables.  Note that those  in the

declines  for Strategy 1 are uniformly smaller than those for Strategy 2,

which are themselves uniformly smaller than those for Strategy 3.  One

might tentatively conclude, therefore,  that from a purely aggregative

standpoint, policies which spread the incidence  of control costs tend to
      *The 100 AQCRs for which data was corrected in this study, only
91 AQCRs have been included in the model estimation and simulation.
For the remaining nine AQCRs, some of the economic data was not
available.  These AQCRs are: Hartford,  Connecticut; Cheyenne,
Wyoming; Anchorage, Alaska; Burlington, Vermont; San Juan, Puerto
Rico; Virgin Islands; Anderson, Indiana; Flint, Michigan;  and
Worcester,  Massachusetts.

-------
increase the aggregative negative economic effects of air pollution con-




trol.  If one were willing to extrapolate these findings to the nation,




what this implies is that cost sharing financed by increases in the per-




sonal income tax, could actually make unemployment worse rather than




better.  This result is not unreasonable since an increase in personal




income taxes for the purpose of financing subsidies (which are not gov-




ernment purchases of goods  and  services) reduces final demand by the




amount of the marginal propensity to consume times the tax increase.




      The view of cost sharing which emerges when results for individ-




ual regions  are considered is thus quite different from that which




emerges from solely aggregative considerations.  While cost sharing




results in overall worsened economic performance,  the number of re-




gions severely affected adversely by air pollution control is reduced.




There appears to be a policy trade-off to be made.  The policy-maker




must decide whether he wishes to have a few  regions experience  con-




siderable  adverse effects while maintaining good overall economic per-




formance,  or slightly reduce overall performance so that a few  regions




may be spared severe hardship.   It should be  remembered that the re-




sults  that  support this  finding presuppose a particular cost sharing




scheme.   It is not the only one which could be undertaken.




      These results on differential regional impact are suggestive in




many veins, but one in particular deserves comment.  Some regions

-------
evidently have an economic capacity to meet more rigorous emission




and air quality standards than do others.







6. 6   Concluding Comments







      In conclusion, it must be observed that the trial simulation of




three strategies and the interpretation of results has demonstrated to




a limited degree the operational nature and strategy assessment poten-




tial of the model system.  Surely,  what has been done here is but




scratching the surface of the utilization potential of the model system.




What has been done, however, clearly points to the view of the regional




economy and environmental system (quality levels) as interrelated sys-




tems and that control  activities affect profoundly in many dimensions




economic activity.  However,  these economy-environment relationships




are more complex than evident in this demonstration.  This suggests




the need for further efforts in model utilization,  sensitivity testing,




model system refinement as appropriate, and more up to date informa-




tion.




      The fundamental question of incidence  of economic burden of con-




trol policies  and the caveats expressed in this  interpretation do cer-




tainly warrant further exploration of the  model utilization.

-------

-------
CsJ
               Inputs
          Model Development
•••           Model             JB»
 Regional Economic Model System
      Outputs
                                                                                                Assessment
           Regional Economic
           Activity
           National Economic
           Activity
                                              Regional
                                              Model
         Interregional
         Feedback
                                           I-O Model and
                                           National Effects
                                           Assessment
          Policy Assessment
           APCO Policy Variables
             Standards
             Incentives
             Fuels
             Research and
             development
             Government
             expenditures
     Computerized Simula-
     tion Program

        Program RMS
        Program FEE
     .   Program IOA
Outputs:  Changes in:
.  Manufacturing
  activity
  Power and fuel
  consumption
  Income
  Employment
  Government
  expenditures
.  Other
                                                              Interpretation
                                                              of Model
                                                              Outputs
                                                                                                              /
                                                              Recommenda-
                                                              tions on
                                                              Further Refine-
                                                              ment and Utili-
                                                              zation of Model

-------
7.0   AN ASSESSMENT OF THE APCO
      REGIONAL ECONOMIC MODEL SYSTEM:
      PROMISES AND PITFALLS
      The objective of the effort described in previous chapters was to

develop and demonstrate the use of an operational analytical tool for

assessing the complex and far-reaching economic effects of air pollu-

tion abatement strategies.

      First in St.  Louis, then in 31 AQCRs and now in 100 AQCRs,

such an operational model amenable to some policy assessment was

developed.  In other words, the elements  of the economic system

were defined; the air pollution control tools identified in this context;

the equations set up; the  relations  established which determine the

values of the elements; that is, the economic system described; effects

of certain alternative strategies estimated, and the properties of this

"solution"  investigated and assessed.   The logical question at this

stage is: How good or useful is the Regional Economic Model System?

      An unambiguous answer to this'question is not possible at this

stage.   It is true that the use of the Regional Model System as a tool

to assess these  strategies has been demonstrated; this  is no mean

feat considering the complexity of  the economic system modelled.

Further, the simulations showed considerable sensitivity of the model

system and the complex and profound interrelationships between

-------
control policies and various facets of the regional economies.  How-

ever, this demonstration has  explored only a limited portion of the

utilization potential of the model systems.  Judgments  on the utiliza-

tion potential of the model system have to be necessarily tentative.

      The model system, however,  can be assessed from a set of

morŁ general criteria.  The three parts of the Regional Model Sys-

tem -- Model Structure,  Data and Simulation --  can be evaluated by

the following five criteria.

            1. Model Scope:  What sorts of control policies
              are  covered by the model structure?  Which
              are  excluded?  How does this policy  coverage
              of the model affect its relevance as an ongoing
              policy assessment tool?

            2. Model Structure:  Given the policy scope,  how
              efficiently is the model structured?  How
              comprehensible is the model  structure?  Is
              the model system a "black box" or a "white
              box"?* How extravagant is the model in
              terms of data requirements?

            3. Model Sensitivity and Accuracy: How readily
              manipulable is the model?  How sensitive is
              it to policy inputs?  How accurately is the
              model system representing economic reality?

            4. Data Limitations: What are the quality and
              currency of the data  used in the model?   What
              additional data would improve model scope and
              performance?
      *A black box model is one whose structure is not clear but de-
fined in terms  of inputs or outputs and can be contrasted with "white
box" models whose structure  is "transparent. "

-------
            5. Efficiency of Simulation Program:  How efficiently
              is the simulation program assembled?  Is the
              program extravagant of processing time?  How
              user oriented is the program?

      Model Scope

      The range of policies to which the Regional Model System is

designed for was described in Chapter 5 and is quite large. However,

there are two major limitations in scope.  First, the Regional Model

System in its  present stage of development does not fully incorporate

the interdependency between the economic system and environmental

quality.  Since air pollution is a by-product of economic activity, the

level of total emissions is directly related to the level of economic

growth (say GNP), geographic distribution of the growth and the indus-

trial composition of the economy.  The present model system provides

only an estimate of economic effects,  with given air pollution levels

(before and after the control is instituted).

      A feasible approach for addressing this limitation would  be to

build into the  model system relationships  between unit levels of eco-

nomic activity by industrial sector and emission levels  and control

costs.  In this manner, as economic activity increases,  emission

levels and control costs  will also  increase and the effects of such

emission and  cost increases on economic  activity can be computed.

Thus, a more interdependent economic environmental regional model

system can be developed.

-------
      The second limitation of the model system is its cross-sectional




and static structure.  In its current stage, the model system traces




the cross-sectional reallocation of economic activity among the ACQRs




attendant on control strategies pursued in the 305(a) Report.  Thus, it




is possible to identify which AQCRs are affected adversely by the con-




trol policies.  But the overall effects of control on different industrial




regions and their varying growth rates  in the future are not captured




now.  A  dynamic model, if developed, would assess, for instance, a




"practical timetable" for control of stationary emissions, the minimal




economic dislocations, demand for control devices, etc. , over time




and be a realistic aid to important issues likely to confront policy-




makers in the next several years.




      Model Structure




      The Regional Economic Model System must be viewed as a well




structured regional model with elaborate  sectoring of high emission




industries and careful integration of regional economic concepts.   The




sectoring is elaborate enough to be sensitive to a wide range  of high




emission industries and electric power industry --no mean achieve-




ment.  Given its complexity,  the operationability of the  model is




worthy of note.  Furthermore,  the equations all conform to a priori




expectations of theory.

-------
      As is to be expected, this model's structure provides room for




improvement in many different ways.





      First, the treatment of interregional effects in the model leaves




something to be desired.  The use of the locational quotient or regional




market share matrix assumes a fixed regional distribution of economic




activity -- over time an unrealistic assumption.




      Second, the nonavailability of a national macro model limits the




estimation of national effects to the structural  changes obtainable from




the I-O model.  The latter represents only a partial statement of




national effects.   This drawback can be addressed only by one of the




alternatives proposed at the end of Chapter 4.




      Third, there is  one  important kind of variable not explicitly




treated in the model --  price variables.  The model itself thus makes




no forecasts of changes  in prices  occasioned by air pollution control,




but rather requires such information as input information. Since one




of the chief concerns about air pollution control is that it will raise




prices, it would be desirable to modify the model so that price effects




of air pollution control are predicted as well.




      While the model gives generally good coverage to important vari-




ables of regional economic activity,  it is deficient in geographical




scope for national economic effects assessment.  Although the model




is large and covers all major urban AQCRs,  only about 65 percent of

-------
GNP is accounted for by output from these regions.  There is thus a




large portion of economic activity and impacts  of air pollution thereon




left unaccounted for by the model.




      Fourth^ the method by which the model treats  air pollution con-




trol costs and benefits is restrictive. Such information now is treated




as input information into the model.  That is, the model user must




translate his policies into control costs  and control benefits,  feed




these values into the economic model system, and the  computer does




the rest.  Or to put it more technically, control costs  and benefits are




treated as exogenous variables in the system (see Chapter 5).  In fact,




control costs and benefits ought to be endogenous in  the system since




control costs depend in major measure  on the level of  economic activ-




ity, which indicators are themselves the key endogenous variables of




the model.  In like fashion, damages reductions (or  benefits) depend




upon changes in the level of economic activity and control activity.




Treating these variables as exogenous is thus likely to result in faulty




predictions  of the net effect of air pollution control of  as yet unknown




extent.




      The remedy for this particular defect of structure is the same




as proposed earlier to incorporate interdependencies between the eco-




nomy and the environment.

-------
      Model Sensitivity and Accuracy

      The model system appears to be sensitive to the range of strat-

egies tested.  The differences in key economic indicators among the

various AQCRs  are encouragingly significant.  The same assertion,

however,  cannot be made about effects totalled over the AQCRs.  In

any event,  further sensitivity testing is warranted before any assess-

ment in this regard can be made.

      As yet,  no thorough test of model accuracy has  been conducted

since the data on which model estimation was based are the only  data

CONSAD has assembled.  For a true model test, it is necessary to

see whether the model accurately predicts  data that has not been used

in estimating the model.  Insofar as accuracy of the model in explain-

ing sample data itself is concerned, however, the following points

should be made:

              By themselves, individual equations  of the model
              appear to satisfactorily explain model  data.  The
              percentage of variation accounted for in the equa-
              tions are, on the .average, about 90 percent.

              Comparisons of actual with predicted values for
              a few AQCRs suggest that the model does  reason-
              ably well in predicting values for industrial cities,
              but not so well for non-industrial AQCRs like
              Washington,  D. C.

More testing will be needed, however, to adequately assess the model

system1 s accuracy and to pinpoint the source of any inaccuracies for

elimination.

-------
      Data Limitations





      This model system uses a large mass of data for 1967 (see





Appendix E).  The parameters of the model reflect 1967 situation and




the continued use of the model calls for an updating of these data or the




use of forecast data from other sources such as the  OBE data for 50




AQCRs.




      Five year average data on control costs have been used from the




305(a) Report of 1970.  Year by year data for  1971-75 would have been




more realistic.




      Benefit data used in the model is very much in the nature of




"informed judgment". Alternative levels  of the benefit estimates or




better estimates would improve the outputs from the model.




      Simulation




      As yet, little experience has been had with the use of the  model




system.   It can be said with some confidence that the model simulation




program works, which,  for a model of the size and complexity  of this




one, is no small achievement.  Also,  it is efficient in terms of com-




puter processing time.  Furthermore, the model and its simulation




program appear to be sufficiently flexible to permit  investigation of a




wide range of policy alternatives.




      There  are, however,  two drawbacks in the manner in which the




simulation program is assembled.  First, the  model,  including its

-------
simulation program,  is very large. As a consequence, it is difficult

to understand and,  therefore, less than simple to manipulate by the

general user for policy evaluation purposes other than those expressly

provided for in the  simulation program.

      Second, there is a non-linear component for which there is no

"exact solution. "  Therefore, an interative algorithm  --an extended

Newton's method*-- has been applied.  The accuracy of this method is

dependent upon the  number of iterations involved in the computer runs

and process of convergence of the system. This aspect calls  for a

further testing of this algorithm for errors due to  aborted convergence,

      Third, at the present stage,  only a limited number of strategies

has been tested.  It is more plausible that additional strategies and

hence nature of  inputs,  be applied to improve the use potential of this

simulation program.
      *See M. K.  Evans and L. R. Klein, The Wharton Econometric
Forecasting Model,  University of Pennsylvania, 1967, Chapter IV.

-------
     Inputs
            Modol
      Outputs
                                                                                       Assessment
Model Development
Regional Economic Model System
 Regional Economic
 Activity-
 National Economic
 Activity
                                    Regional
                                    Model
        Interregional
        Feedback
                                 I-O Model and
                                 National Effects
                                 Assessment
Policy Assessment
 APCO Policy Variables
   Standards
   Incentives
   Fur-Is
   Research and
   development
   Government
   expenditures
     Computerized Simula-
     tion Program

     .  Program RMS
     .  Program FEE
     .  Program IOA
Outputs:  Changes in:
   Manufacturing
   activity
   Power and fuel
   consumption
   Income
   Employment
   Government
   expenditures
.   Other	  	
                                                              Interpretation
                                                              of Model
                                                              Outputs
Model System
Assessments
                                                                   j/'    /
                                                                irVcie r Re ij^ie •
                                                               S  •'    "^  —S
                                                              rfient and .LTtili-
                                                              .,zatiari of Model

-------
8. 0   REFINEMENT AND UTILIZATION OF
      THE APCO ECONOMIC MODEL SYSTEM:
      RECOMMENDATIONS
      This report covers a great deal of material in fairly minute

detail.  Much of the material concerns what has been done over the

past three years.  It was clear at the outset that  CONSAD's contribu-

tion to the Regional  Air  Pollution Analysis (RAPA) would be to develop

and demonstrate the use of a workable analytical tool, to assess the

economic effects of  various abatement strategies in the AQCRs.

      The model  system has been developed and demonstrated in a

limited manner for three strategies.

      In describing the model system and the uses to which it has been

put to date, CONSAD has attempted to alert the potential user to  defi-

ciencies in the model system, as well as to indicate its achievements.

This chapter is different in character since it deals not with what is

past, but  rather the potential of the model system for future assess-

ment of economic effects of air pollution control  strategies.  In a

nutshell,  the key question which this chapter addresses itself is

"Where, if anywhere,  should APCO go from here?"

      In the process of describing the model system as it stands and

in evaluating the  potential of a program of model refinement and utili-

zation,  a  number  of ideas have been expressed.  It has been suggested,

-------
for example,  that (1) the model be employed, if possible, for national

economic effects assessment, (2) that the model be applied to assess-

ing alternative timing of implementation policies,  (3) that the model be

used to predict the effects of different strategies for setting secondary

ambient air quality standards and performance standards for new

sources,  (4) that the model be applied to the problem of forecasting

pollution emissions,  and (5) that the effect of ambient air quality im-

provements on the productivity of labor and capital be investigated.

There is evidently no dearth  of interesting and important questions

which could be posed for the  model system,  some  of which the model

is now fully capable of evaluating and others of which the model could

be refined to handle.  Indeed, what is needed is an orderly plan of

attack.

      Based upon  its own evaluation of the  issues and possibilities

raised in  this report,  CONS AD has identified the following four recom-

mendations for maximum utilization of the economic model system.

      Recommendation 1:  Further Utilization of the
                          APCO Economic Model System

      CONSAD recommends that an in-depth utilization of the APCO

Economic Model System developed in the current year be carried out.

The further utilization of the  model system will serve three purposes.

First,  it will  follow-up on the preliminary analyses of the three

-------
strategies  and their variations to assess in-depth the economic effects

of various  strategies.  Second, it will incorporate data that has become

available since the 1967 economic data used in the model development.

In this manner, the model system will be more honed in.  Third, the

utilization  of the model system with new data will identify potential

areas for model refinement,  sensitivity analysis, and further applica-

tions  of the model.  Among other possible applications  could  be poten-

tial opportunity to extend the model to assess the effects of water qual-

ity and solid waste control strategies.

      Recommendation 2:  Sensitivity Analysis and Refine-
                          ment of the Model System

      With the experience gained from the in-depth utilization of the

model system, CONSAD recommends  a sensitivity analysis of the

model system.  The model system  should be refined and modified

accordingly, to remove any  deficiencies insofar as possible.   As these

modifications are carried out, the effort  should be directed to develop

a broad adaptable operational tool.  Therefore, a generalized computer

simulation program should be developed for the refined model system.

      Recommendation 3:  Expanded National Model Study

      The model system, in its current form,  is largely "comparative

static" and "regional. "  CONSAD recommends that "dynamic" and

"national"  aspects of the  economic  impacts of air pollution control

-------
neeids to be built in.  Ideally, this can be accomplished by links between

the cross-sectional regional model and a time-series national model,

with interaction between economic and environment systems to be inte-

grated in the model system.

      There are a number of ways, however, in which one might go

about attaining the  ideal.   CONSAD has identified three alternatives

which it believes to be worthy of further study:

              One  might try to make do with the Economic Model
              System as it is currently structured (or as refined
              if Recommendation 2 is pursued), since the model
              encompasses approximately 65 percent of the na-
              tion's economic activity.  This percentage could
              be increased by increasing  the number of regions
              included in the model.

           .  An existing  national economic model (e. g. , Klein-
              Goldberger,  Michigan) might be modified and
              adopted for  purposes of strategy  assessment.

              A new model could be built  expressly for the pur-
              pose of assessing  national effects of national air
              pollution control strategies.

-------
                   APPENDIX A

A MODEL TO ASSESS THE ECONOMIC EFFECTS OF
 AIR POLLUTION ABATEMENT IN ST. LOUIS AQCR

-------
Appendix A:  A Model to Assess the Economic
             Effects of Air Pollution  Abatement
             in the St. Louis AQCR (Phase I)
      The purpose of this appendix is to describe  an econometric model

which simulates the basic economic  structure of an Air Quality Control

Region (AQCR) for testing the impacts of air pollution control policy in

that AQCR.   The model will be specified for the St. Louis Standard

Metropolitan Statistical Area (SMSA).   The model will be presented in

its  abstract form  in order to illustrate  the theoretical background and

implications of the entire equation system.

      A comparison of the development of regional models  with several

advanced models at the national level* reveals that the existing econo-

metric models at  a regional level are still few -- their  scope much

more limited and  the in results less sophisticated.  The relative back-

wardness of the regional model stem from three reasons.

      First, at the national level,  good time series data have been

available for all major economic variables  and other variables,  such
      *For example, Klein, L. R. ,  "A Postwar Quarterly Model, "
Models of Income Determination, Studies in Income and Wealth, Vol.
28, Princeton, New Jersey, Princeton University Press,  1964; Klein,
L. R. ,  and Goldberger, A.  S. , An Econometric Model of the  United
States.  1929-1952, Amsterdam, North-Holland, 1955; Suits, Daniel B. ,
"Forecasting and Analysis with an Econometric Model, " American
Economic Review, Vol. 52, March, 1962, pp. 104-132; and Dusenberry,
J. S. , Fromm, G. , Klein,  L. R. ,  and Kay, E. (eds. ),  The Brookings
Quarterly Econometric Model of the United States,  1965.

-------
as capital stock are even published as the result of theoretical inquiry.





But at the regional level -- either state or SMSA -- data are unrelia-




ble and unavailable, especially in a continuous time series form.




      Second, econometricians began the development of their models




at the national level.  From time to time, different equations and theo-




ries have been tested with the same base of data in a well-defined econ-




omy, say the United States.   By repetition of tests and the accumula-




tion of experience, models calibrated to good  theoretical structures




are now in sight.  On the other hand,  regional models always deal with




different geographic units.   Beyond the  uniqueness of data, suggested




by location theory,  geographic and cultural environment and socio-




economic structures may differ from region to region.




      Third, theoretically speaking, macro-economic models are usu-




ally based on well-established economic theory in their formulation.




However, at the regional level, economic base theory, location theory,




gravity concepts,  migration theory,  etc. (especially the concept of




"distance" emphasized by regional scientists for some time), are much




harder  to integrate into an overall hierarchical system.

-------
      In spite of such shortcomings, some notable regional models

have been recently estimated.  For example,  Bell  estimated a Mass-

achusetts model.  An Alaskan model by Tuck,  a Puerto Rico model

by Lakshmanan and Lo, a Northeast Corridor model by Crow, a

Michigan model by Suits,  a Philadelphia SMSA model by Glickman,

and a Hyogen-ken model in Japan by Kaneko can be indicated. *  Klein**

recently suggested a general strategy in building regional econometric

models to link with national models.
      *See Frederick W.  Bell, "An Econometric Forecasting Model
for a Region, " Journal of Regional Science. Vol. 7, No. 2, 1967;
B. H.  Tuck, An Aggregate Income Model of a Semi-Autonomous
Alaskan Economy,  Anchorage, Federal Field Development Committee
for Development Planning, 1967; T. R. Lakshmanan and Fu-chen Lo,
A Regional Growth Model of Puerto Rico:  An Analysis of Municipio
Growth Patterns and Public Investments.  Pittsburgh,  Pa. , CONSAD
Research Corporation, September, 1970; Robert Crow, "Econometric
Model of the Northeast Corridor, " (mimeograph) MATHEMATICA,
October,  1967; Daniel B.  Suits, "Econometric Model of Michigan, "
(mimeograph), Research  Seminar in Quantitative Economics of the
University of Michigan; Norman J. Glickman,  "An Annual Econometric
Model of the Philadelphia SMSA,  1949-1966, " (mimeograph), Ph. D.
Dissertation,  Department of Economics,  University of Pennsylvania,
November,  1968; Yukio Kaneka,  "An Econometric  Approach to Annual
Forecast  on Regional Economy by Local Government, " Paper  and Pro-
ceedings, The Second Far East Conference of Regional Science Associ-
ation,  1965,  University of Tokyo Press,  Tokyo,  1967;  pp.  119-144;

     **Klein, L. R. ,  "Econometric Analysis of the Tax Cut of 1964, ''
(mimeograph), Department of Economics, University  of Pennsylvania,
1967.

-------
      For the present study,  the following characteristics in this model

may be different from national models:

              national influence upon local economy,
            .  migration patterns,
              interregional commodity flows,
            .  interindustrial structure, and
            .  function of local government.

The details are presented formally below.

-------
Notation



*X    =  gross regional output



*C    =  regional consumption



*I     =  regional investment



*E    =  regional export



*M    =  regional import



*G    =  local government expenditures


   F
  G    =  federal  government expenditures



  p^.   =  local consumption price index
   O


*Y    =  regional capital stock



  r     =  national interest rate



  II     ~  total rcgioucu  profit or property income



 • T    =  local direct taxes



*T    =  local indirect  taxes
   Łj

   F
*T    =  taxes  paid to federal government by the region



*T    =  transfer payments



  D    =  depreciation on regional capital stock



  Q.    =  regional gross output by industry i         (i = 1, .  .  .  , n)


   N
  Q.   =  national gross output by industry i



  p.    =  regional price of output i


   N
  p.     -  national price of output i

-------
 11.    =  profit level of regional industry i



*E.    =  exports  by regional industry i



 GNP =  gross national product




 O.    =  domestic final demand for regional output i



 a..    =  input-output coefficient of local economy




 W    =  wage bill



 P    =  regional population




 U    =  regional unemployment



*N    =  regional total employment



*N.    =  employment by regional industry i



 w.    =  wage rate of  regional industry i



*L    =  regional labor force


   M
*P    =  regional net migration


   p
 N    =  regional female employment


   N
 w    =  national wage rate



 t     =  time period

-------
      The Model Framework


      The gross regional product can be first identified as an identity:


(1)         X =  C+I + E-M + GL +  GF


Three of the  variables on the right-hand side of this identity -- con-


sumption, investment and export -- always have been emphasized in


other regional growth models.  These three variables determine the


regional multiplier effect on regional growth through the Keynesian


type multiplier-acceleration principle or regional export base theory.


The  government sector plays an important role in the regional econ-


omy and is, therefore,  included in the  regional model.  It is classified


into  local and Federal components.


      The consumption and investment  function can be formulated as:

(2)         C-   =  _
(3)         I  =  f3 (X,  Kt_lf  r,  n)


where the consumption function has been presented in real term as func-


tion of disposable income.


      The regional disposable income identity is:


(4)         Y =  X - T^ - T^ - TF + TR - D,

                                          T    T    -r-»
that is,  gross regional product  after tax (Tj ,  To ,  T ) and depreciation


(D) of existing capital stock plus transfer payments (T  ).  The invest-


 ment function may be disaggregated into manufactures  and non-

-------
manufactures or even according to the sectoral structure.  However,

the general form of equation (3)  should do.

      It has been indicated that for a regional model, the export sector

usually plays an important role, especially for a relatively small re-

gional unit such as the St.  Louis SMSA.  Trade with the  "rest-of-the-

world" is the key to economic  growth.  Regional export may be strongly

related with metropolitan's interindustrial structure.  Thomas sen, *

Bell,  Glickman, and  Klein** link exports also with Gross National

Product (GNP).   However, it is  important that the export sector  be

disaggregated,  and that sectoral exports should be taken as a function

of national sectoral output and of comparative sectoral prices between

the region and the "rest-of-the-world. "  The reason is that the inter-

industrial  structure of a large metropolitan area may  be quite different

from  that at the national level.   As  suggested by location theory,  re-

gional resource endowments and other socio-economic characteristics

determine the industrial structure of a local economic unit.  Therefore,

the use of  a GNP trend is a less desirable indicator  of external market

demand than national outputs by industry. In general, regional sectoral

output (Qi) is a  function of local interindustrial demand (Qi,  j=l, . . . , n),
      *See H. Thomassen,  "A Growth Model for a State, " Southern
Economic Journal, No. 24,  1957,  pp. 123-139-
     **Frederick W.  Bell,  Norman J. Glickman, and L.  R.  Klein,
op. cit.

-------
national sectoral output (Q  ), comparative price (	), profit (17;),  and
                                                pN

regional output (x).


(5)         Qt =f5 (Qif...,Qn,  Q^, ~, Hi,  X)        i = 1	n
                                    PiN


      However,  such a formulation may face a serious multicollinearity


problem upon estimation.  A possible solution to this problem is to


separate (5) into two  equations,  namely,  export demand  and domestic


demand.  Noting that export data is unavailable,  residual exports of


the i   industry  (Ej) can be defined as:

           A        n
(6)         Ei = QJ - 2 ajj QJ - Oi                     i = 1, . . . , n
where O^ is domestic final demand of industry i.   Equation (6) may be

obtained from an output-input table,  a technique that has been applied

in the  recent Brookings' model.* For these export-oriented  industries,

the residual export function will be:



<7>        *i  =   /7 (Q*  -V n.)     1= i,  .  .  .  , k, k+i, .   . .  m

                           pi


where Ihe  first k  industries arc high emission sources.

      However, since there is  some industry (i - k+1,  .  . .  , n) which

has a small role in  export, we can aggregate into one equation:
      *Dusenberry, J. S. ,  e_t aL , op. cit.

-------
           n      ~
(8)         Z      Ei = ER
(9)         ER  =  fg (GNP,
      Imports may be divided into consumption goods (M^) and indus

                                   *
trial demand (Mj).

(10)        M  =  Mc + Mj
(11)        MC  = fn (Y,    )
                          PC

(12)        MI =  f 12 (x, ^)


      Equation (12) can be disaggregated according to export sectors,

or equations  (10),  (11) and (12) can be aggregated into one equation:

(10')       M  =  f (X, ~).


The ability to estimate the import function is highly dependent upon

availability of data and relations with other equations.

      The government sector plays an important role in the regional

economy, and must therefore be included in the regional model.

      In most of the national models, government expenditures have

been treated  as exogenous. *  Klein has strongly suggested to make
      *The Brookings Model has allowed for endogeneous treatment.
See Dusenberry,  J. S. ,  eŁ aL ,  op. cit. ; and Ando,  A. ,  Brown,  E.G. ,
and Adams, Jr., E. W. , "Government Revenue and Expenditures. "

-------
them endogeneous at the regional level.* After a small modification of




his proposal,  we have:




(13)       T^   =   fn (W,  H)






(14)       T2   =   f14 (x)




(15)       TF   =  i
                   315 (w,   n)




(16)       TR   -  f   (U,  P)





(17)       GL   =  f   (T^+ T^ ,  P)
                    *l    J.     d





where direct local taxation (Tj ) and Federal tax (T^) are functions of




wage bill and profit; indirect local tax (Tjf) is a function of regional




output,  while transfer payments (T  ) is a function of unemployment



and population. Finally, local government expenditures  (GL|) is a func-




tion of local taxation and population.



      Through the multiplier effect, the mix of expenditures from pri-




vate and public sectors (effective demand) determines gross regional



product; by the same  token,  level of gross regional product determines
      "Klein, L.  R. ,  op.  cit^

-------
the regional employment levels.  From our last experience in the Eco-




nomic Development Administration (EDA) labor force model, * the




labor market  behavior can be given by the following equations:









                                 \A    TT

           L  =  f!9  (ZNi' P» P  •  * .  . . .)




(20)       PM=  -f    fpM    W   ,  N  •
                 J*>/\L *^   i  "••••""••  f  ••" '-• 1    i






(21)       U  =  L - 2N.

                      i  1





      Employment is  a function of output,  wage rate  and capi.tal stock;




and labor force  is determined by employment, population (P),  migra-
tion (p) ancj female worker.  While regional migrations are deter-




mined by past figures (P.  ,),  wage difference (iTTivf) and employment-
                        t"" 1                    VV

            N

labor force (j^t-l'  Finally,  the differences between labor force and




employment determines regional unemployment.
      *CONSAD Research Corporation,  A Study of the Effects of Public

Investment.  Pittsburgh,  Pa. ,  1968.

-------
      The St.  Louis Model




      Except for the income-determination block,  the model is recur-




sive in nature.  GNP is expressed in terms of national  income by sec-




tors (two-digit SIC in manufacturing industries) and,  as external mar-




ket demand,  determines the level of exports from the St. Louis SMSA.




The exports then determine the output levels of St. Louis manufactur-




ing sectors and,  hence,  their import demands.  By the production




functions,  once output level is given, capital share (profit)  can be




obtained (since the Cobb-Douglas production function is  assumed).




Investments are a function  of profit and previous capital stocks.   After




adjustment for depreciation, capital stock is given by the following




accounting relationship:




           Kfc =  Kt_1 + It - dKt




where I and K are investment and capital, respectively, and ci is depre-




ciation rate.   Given output  level and capital stock, the employment level




also can be obtained through the production function.




      On the  other hand, given exports, imports,  investment, and gov-




ernment expenditures (the latter is a function of tax revenue), regional




income  and consumption expenditures are determined simultaneously.




As to income determination, two types of theories are emphasized in




the Keynesian type of model -- the multiplier-acceleration principle




which is based on the propensity to consume,  and the economic-base

-------
      Finally, non-manufacturing employment level is a function of




non-manufacturing income.  Labor force is then determined by total




employment, national unemployment rate,  and migration.   The differ-




ence between labor force and total employment determines local un-




employment.  Because  the electric-power  industry has been consid-




ered a major pollutant in the St. Louis SMSA, electric-power demand




functions have also been estimated.




      The disaggregation of manufacturing industries into  two-digit




SIC levels requires further explanation.  There were three major rea-




sons for disaggregation:  first, inter industrial structur at the regional




level may be quite different from that at the national level. Second,




the present  model is  designed to measure the economic impacts of air




pollution control,  and it is more realistic to observe the impact for




each major  manufacturing industry at the  two-digit SIC level  than to




do so for the manufacturing industry as a  whole.  Third, measurement




of exports,  as emphasized by economic-base theory, requires sectoral




disaggregation.




      However,  in the St. Louis economy,  because certain two-digit




manufacturing industries either are too small in scale or lack suffi-




cient data, only 11 two-digit SIC industries were so isolated;  and the




manufacturing industries other than these  11 have been aggregated as




"other manufacturing industry.."  The final 12 groups are as follows:

-------
            .  SIC 20 - Food and kindred products,

            .  SIC 26 - Paper and allied products,

              SIC 27 - Printing and publishing,

              SIC 28 - Chemicals  and alliec products,

            .  SIC 29 - Petroleum  and coal  products,

            .  SIC 32 - Stone,  clay and glass products,

            .  SIC 33 - Primary metal industries,

              SIC 34 - Fabricated metal products,

              SIC 35 - Machinery, except electrical

            .  SIC 36 - Electrical machinery,

            .  SIC 37 - Transportation equipment,  and

            .  Other manufacturing -  Including tobacco products
                       (SIC 21), textile mill products (SIC 22), apparel
                       and related products (SIC 23), lumber  and wood
                       products (SIC  24), furniture and fixtures (SIC
                       25),  rubber and plastics (SIC 30),  leather
                       products (SIC  31), instruments  and related
                       products (SIC  38), and miscellaneous .-manufac-
                       turing (SIC 39).

      The Variables of the St.  Louis Model

Y     = national income by sector i ($ million) at time t


E   ,  = export by industry i ($ million) at  time t


V     = value added by industry i  ($ million) at time t

-------
M    = import by industry i ($ million) at time t
  it


N    = employment by industry i (1000 employees) at time t



K.    = cap:.tal stock of industry i (million dollars) at time t



I.    = investment by industry i ($ million) at time t



n.    = capital share of value  added or profit by industry i ($ million) »t time t



Q    = electric power demand by industry i (1000 kilowatt hours) at time t




For  the above variables:


i     = 20,  26,  27, 28, 29, 32, 33,  34, 35, 36,  37, and other manufactures.

        (These SIC numbers identify the 12 manufacturing sectors of the model)


P    = industrial commodity  price index (1957-59 = 1.00)



T    = local tax revenue ($ million) at time t



G    = local government expenditure  ($ million) at time t



C    = consumption ($ million) at time t



Y*   (= C + 21. + Z E. - SM. + G)  =  regional expenditure ($ million) at time t



Y    =  regional income ($ million)  at time t



N    =  employment by industries  other than manufacturing industries

         (100 employees) at time t


Nfc   (= Z N. + ~N)  = total employment



L    =  total labor force


  N
u    =  national unemployment rate



U    =  unemployment

-------
u.    = local unemployment rate

Q    = electric power demand for residential use at time t (1000 kilowatt hrs)

Q    = electric power demand for non-manufacturing industries at time t
  *      (1000 kilowatt hrs. )

t     = time  period (1954 = 0, 1955 = 1, .  .  . ,  1966 = 12)

w.    = wage per man-year by industry i (1000 dollars)

-------
Estimation of the Model
The
(1)
general form
Ei = aj + bj
of the export
YNit
function will be:



The results of estimation are:
Equation
Number
(1.1)
(1.2)
(1.3)
(1.4)
(1.5)
(1.6)
(1.7)
(1.8)
(1.9)
(1.10)
Industry
Code
SIC 20
SIC 26
SIC 27
SIC 28
SIC 29
SIC 32
SIC 33
SIC 34
SIC 35
SIC 36
a.
i
244. 585
(45. 371)
27.688
(4. 163)
10. 998
(7. 128)
- 4.462
(22. 090)
-344.071
(86. 110)
21. 522
(10. 528)
-144. 539
(83.865)
55.472
(12.206)
93. 853
(7.792)
141. 997
(18.689)
b. R2
1
. 0284 . 832
(.000366)
.00663 .830
(.000861)
.0152 .947
(.00104)
.0664 .986
(. 00230)
.124 .789
(.0183)
.0151 .791
(. 00222)
.0435 .750
(.00716)
.0155 .915
(. 00136)
.0105 .965
(. 000574)
.00539 .435
(.00169)
Durbin-Watson
Statistic
1. 713
1. 697
. 534
2. 295
2.55
.772
2.218
2. 064
2. 105
1.079

-------
(continued)
Equation Industry
Number Code
(1. 11) SIC 37
(1.12) Other
Manu.
All Manufacturing Ind.
a.
i
-468.752
(88. 511)
230. 314
(19. 383)
-176. 337
(181. 595)
2 Durbin-Watson
i Statistic
.13Q .940
(.00945)
0.05181 .747
(.00053)
29.527 . 975
(1.351)
1. 812
2. 517
1.287
      Manufacturing (or base) industries export supply function:
(2)
            V4  =
This is a set of identities expressing the proportional relationships of




exports to value-added in corresponding manufacturing industries; aj's




have been obtained from the St. Louis  input-output table. *
Equation
Number
(2.1)
(2.2)
(2.3)
(2.4)
(2.5)
(2.6)

#Liu, Ben-chieh,
Output Study of the St.
Industry
Code
SIC 20
SIC 26
SIC 27
SIC 28
SIC 29
SIC 32
Interindustrial Structure
a.
i
.639
1. 143
1.143
. 553
. 553
1. 320
Analysis: An Input-
Louis Region, 1967, St. Louis Regional Indus -
trial Development Corporation, December, 1968.

-------
(continued)
Equation
Number
(2.7)
(2.8)
(2.9)
(2.10)
(2.11)
(2.12)
Industry
Code
SIC 33
SIC 34
SIC 35
SIC 36
SIC 37
Other
Manu.
a
i
.688
.972
.781
.708
. 809
.949
      When the model is to be used to simulate  the impact of air pollu-




tion control costs upon the St. Louis  economy,  value-added can be de-




rived in terms of both (a)  export demands and (b) the ratio of the level




of capital stock in the presence  of air pollution  control costs to the




level of capital stock that  would have occurred without such control




costs:
      Import demand function:





(3)          Mit - aj Vit





The  levels of manufacturing industry output determine the  imports of





raw  materials  and intermediate goods from other regions.  Import co-





efficients (a^) also were obtained from the St.  Louis input-output table





as was done in the previous formulae.

-------
Equation
Number
(3.1)
(3.2)
(3.3)
(3.4)
(3,5)
(3.6)
(3.7)
(3.8)
(3.9)
(3.10)
(3.11)
(3.12)
Industry
Code
SIC 20
SIC 26
SIC 27
SIC 28
SIC 29
SIC 32
SIC 33
SIC 34
SIC 35
SIC 36
SIC 37
Other
Manu.
a.
i
.661
.285
.285
,907
. 907
.329
.605
.391
. 342
. 548
1.567
.501
     Production function with price index adjustment factor:
               =  A.
where P. has been calculated as the geometric mean of the annual




St.  Louis labor  share.

-------
Equation
Number
(4.1)
(4.2)
(4.3)
(4.4)
(4.5)
(4.6)
(4.7)
(4.8)
(4.9)
(4.10)
(4.11)
(4.12)
Industrial
Code
SIC 20
SIC 26
SIC 27
SIC 28
SIC 29
SIC 32
SIC 33
SIC 34
SIC 35
SIC 36
SIC 37
Other
Manu.
All Manufacturing
Industry
A.
i
3.257
(. 02073)
2.766
(.0177)
4.637
(. 0282)
2. 044
(. 0174)
1.439
(.121)
2. 605
(.0188)
3. 090
(.0658)
4.473
(.0278)
4.259
(.0278)
2.234
(. 0422)
4.241
(. 0469)
4.269
(.0368)
3. 554
(.0217)
\
. 0285
(. 00293)
.0225
(. 00250)
. 0274
(.00399)
.0467
(.00247)
.0309
(.01790)
. 0251
(. 00266)
. 0247
(.009305)
. 0284
(.00274)
.0213
(. 00393)
-.0156
(. 00596)
. 04904
(. 00663)
. 0186
(. 00623)
.0360
(. 00366)
\
.450
. 581
.617
.319
.316
.440
.580
. 558
. 577
. 586
. 501
.848
.528
(1-P.)
.550
. 419
. 383
.681
. 684
. 560
.420
.412
.423
.414
.499
. 152
.472
R2
.711
. 821
. 914
. 974
. 159
. 908
. 841
. 906
.856
. 328
. 957
. 560
.969
D-W
Stat.
1.629
.955
.847
2. 120
. 777
1.0818
1.735
1. 178
.829
1.228
2. 338
2.478
.942

-------
      Capital share (profit) relation:

(5)          nit = (i -TI) vit

Given the Cobb-Douglas production function,  capital share or profit in

the  broad sense can be derived and the coefficients can be obtained as

in the preceding table for the production function.   Altogether, there

are 12 equations (5. 1) through (5. 12), comparable to those in 4. 1.

through 4. 12.

      In addition,  to consider the effect of air pollution control costs

(C)  upon industry investment,  a modification of the equation(s) is re-

quired:

(5')         n-t = nt -  ci

      Investment function:

      The investment functions presented here differ more from indus-

try  to industry than do  the other  equations.  This is so, first, because

investment functions differ from model  to model because  there are dif-

ferent theories behind each model (usually involving a lag structure*)

and, second, because investment behavior differs from industry to in-

dustry and tends to be non-linear. Therefore, the results have been

presented individually.
      *For example, see Shirley Almon, "The Distributed Lag Between
Capital Appropriations and Expenditures, " Econometrica, 1965,  pp.
178-196.

-------
(6.1)


(6.2)

where
(6.3)


where
(6.4)

(6.5)
(6.6)

(6.7)

(6.8)


(6.9)




(6.10)

where

(6.11)


(6.12)


I- = 73. 865 + . 261FI- - . 421 K?
"
(2.031) (3.27) (2.38)
I_, = 1.250 + 3. 861 6 + . 0648IT+ . 321 t
Zbt t t
(2.768) (.811) (.104) (.0763)
6=1, when t = 7; otherwise 6 = 0.
I2?t = 11.672 + .0983II27t - .235K2 + 7.9276

(14.910) (.321) (.562) (3,911)
6 = 1, when t = 8, 9, 10; otherwise 6 = Q
I2gt = 10.222 + .114n-.0394K28t_1
(.0849) (1.328) (.086)
I29 = 51.014 .29in29t-.236K29t_1
(1.04) (1.78) (1.24)
I32t = 174. 52+ .0486 II - 1.424 K32t_j
(2.268) (.417) (2.348)
I33t= 146.13 + .347n33t-.76733t_1 + 9.788t
(1.22) (2.84) (1.33) (1.755)
I34t = -3.950 + .l64II34t
(2.794) (.0359)

I = 29.697 + . 0829FI - . 369K ,, + 5. 301 6
.3 3 1 J C* C .3 3 v

(26.549) (.132) (.236) (2.165)
+ 1. 317t where 6=1, when t = 4, 5,
(. 889) 6, 7; otherwise 6 = 0
I = -.553 + .0750FI + 4. 497 6 + . 169T
3ot t t
(1.838) (.0311) (1.0527) (.0790)
6 = 1, when t = 8; otherwise 6=0

I,_ = 66. 549 + . I66n__ - . 4tf6
j ( t J 1 1 J ( C— i
(2.05) (3.40) (2.01)
I = -. 0919 + . 0635R . + .685t
other other
(11.111) (.256) (.342)
S I. = 618. 82 - .424 ZK. ,+.307Sn.
R =
D-W =

R2 ,
D-W =

R2 .

"

R2 .
D-W =
R2 .
D-W =
R2 =
D-W =
R2 ,
D-W =
R2 =
D-W =
2
R =




R2 =
D-W =

2
R =
D-W =

R2 ,
D-W =
R2 =
.443
2. 131

. 844
2. 131

.637

'

.440
2. 013
.239
1.62
. 353
2.615
.685
2. 425
.622
1. 854

. 709
2? An
. i*\j\j


. 709
1. 783


. 589
1. 911

. 500
2. 89
. 906
               i    -         i                  = 2  89
(457. 33)  (.275)         (.007)

-------
      Capital stock identity:

(7)         Kit^K^t.j +Iit-diKit

where dj is depreciation rate.
Equation
Number
(7-1)
(7.2)
(7.3)
(7.4)
(7.5)
(7.6)
(7.7)
(7.8)
(7.9)
(7.10)
(7.H)
(7.12)
Industry
Code
SIC 20
SIC 26
SIC 27
SIC 28
SIC 29
SIC 32
SIC 33
SIC 34
SIC 35
SIC 36
SIC 37
Other
Manu.
Depreciation
rate d.
i
.0674
. 0534
. 0686
.06)69
.0614
. 0646
. 0560
. 0700
.0722
.0738
.0784
.0751
      Government expenditure equation:

(8)          Gt = 27. 517 + 0.951  Tfc
                (15.978) (0.0529)

      Regional expenditures:
(9)          YŁc = Ct + 2 Iit + 2 Eit - S Mit + G
                      i       i      i
IT = 0.973
D-W d = 2. 173

-------
     Consumption function:



(10)          Ct = 287.524  +  . 592YŁ                   R2 = . 990
                  (467.215)    (.0797)                   D-W d = . 882



(10) has been estimated by two-stage least squared method.






     Relation between regional expenditure Y* and regional incon-e Y:




(11)         Yt =  -3667.206  +  1.565^*               R2 = . 989


                   (310.85)     (.053)                   D-W d = 2. 06





     Employment by industries other than manufacturing industries:



(12)         l*t=  384.798 +  . 0821 (Yt - S V.^           R2 = . 793


                   (37.88)     (.0146)     {   l           D-W d = 2. 19





     Total employment:



(13)          N =  S N.+ N.
              t    .   it   t
     Labor force function:



                = 68. 01 + .           .                         .

                  (63.74)  (.0539)   (3. 427) m          D-W d= 1.18
(14)          L =  68. 01 + .9384 N+ 5.366 u             R2 =  .995
     Regional unemployment:



(15)          US  L-N
     Regional unemployment rate:




              t
(16)          u, -  Uf  X 100
     Electric power demand functions for manufacturing industries:


(17)   Q.  =  a. + b. V. + c. t
        it    i    i  it   i

-------
Equation
Number
(17.1)
(17.2)


(17.3)
(17.4)
(17.5)
(17.6)
(17.7)
(17.8)
(17.9)
(17.10)
(17.11)
(17.12)
Industry
Code
SIC 20
SIC 26


SIC 27
SIC 28
SIC 29
SIC 32
SIC 33
SIC 34
SIC 35
SIC 36
SIC 37
Other
Manu.
All Manufacturing
Industry
a. b. c.
i i i
-39. 302
(71.82)
. 5970
(.2054)
8. 171
(1.974)
-6. 706 + 44. 12d + . 2978V_,i.
tot
(12.46) (1.95) .204)
+ 5.758t
(.449)
where d = 1 if t < 6; d = 0
if t > 6
-11. 361
(1.586)
79. 910
(.919)
120. 806
(10.103)
160. 527
(.837)
-65.049
(.873)
-58.124
(1. 814)
24.652
(2.955)
19.996
(.692)
24.206
(1.10)
-33. 100
(287.8)
-633. 620
(193.74)
.4063
(.7791)
.3252
(.0495)
.2672
(.1935)
1.0533
(.0558)
1. 3238
(.3553)
.8176
(.4963)
. 1070
(.2430)
.7681
(.4305)
.4236
(.3925)
.2178
(. 0966)
1. 1356
(.861)

38.992
(2.781)
21. 031
(15.450)
7.424
(1. 036)
2. 505
(.296)


-1.450
(1.750)
8.095
(1.180)
1.665
(8. 516)
13.410
(13.693
R2
.937
. 984


. 857
. 914
. 975
. 450
.787
.703
.329
. 623
.972
. 555
.993
f .
D-W
d
1. 875
2.41


1.259
. 906
1. 944
.806
1. 823
.521
1. 475
2.296
1.925
1.443

-------
      Electric power demand function for residential users:


                -1548.94 + . 9485


                (201.75)   (.538)
(18)     C-   = -1548. 94 + .9485 C                       R2  =  .972
          ct                      t


                                                        D-Wd = . 566
      Electric power demand function of the industries other than man-



ufacturing industries:



(19)      a  = 479. 98  + .4552 (Yt - SV.J                R2 =  .895
          t                     t   .   iv
                                   i

              (129.36)  (.0516)                          D-W d  =  .1.81





      Wage function:  Collective bargaining power should also be in-



cluded in the model for wage determination;




(20)      w.^  = w.  a.
          it      10 i



where w. is wage  (per man-year) at 1954 by industry i.
Equation
Number
(20.1)
(20.2)
(20.3)
(20. 4)
(20. 5)
(20.6)
(20.7)
(20.8)
(20.9)
Industry
Code
SIC 20
SIC 26
SIC 27
SIC 28
SIC 29
SIC 32
SIC 33
SIC 34
SIC 35
w.
4. 505
3.918
4.744
4.641
5. 181
4.062
4.509
4.211
4.410
a.
i
1.0413
1.0407
1.0343
1.0422
1.0409
1.0461
1.0474
1.0430
1.0419
R2
.996
.991
.990
.998
.993
.986
. 988
.995
.998

-------
(continued)
Equation   Industry      w.
Number	Code	($10$))	i	R

(20.10)     SIC 36       4.305           1.0323        .994

(20.11)     SIC 37       4.662           1.0519        .995

(20.12)     Other        3.283           1.0562        .986
           Manu.

-------
                  FIGURE A. 1

      ACTUAL VERSUS PREDICTED REGIONAL
EXPORTS BY MAJOR TWO-DIGIT MANUFACTURING
         INDUSTRIES IN ST. LOUIS SMS A

-------
                                      Figure  A.  1
         ST. IjOUIS SMSA
         Export by Industry 20

                                                                                                ST. LOUIS SMSA
                                                                                                Export by Industry 28
   1954  '55   'it,   '57    '5'8   '59   '60   '6'l   '62   '63   '64   '65    '66
	1	-(—•	1	(	1	1	r	1	I	1	1	1	1
 1954  '55   '56   '57   '58   '59   '60   '61   '61   '63   '64   '65  '66
         ST.  LOUIS SMSA
         Export by Industry 26
Predicted/
                                                                                       (million)
                                                                                      400 T
                                                                                                   ST. LOUIS SMSA
                                                                                                   Export for Industry 29
  1954  '55   '56   '57   '58   '59   '60   '61    '62   '63   '64    '65   '66
                                                                                             1954   '55   '56  '57   '58   '59   '60   '61   '62   '63   '64  '65   '66
      ST. LOUIS SMSA
      Export for Industry 27
 1954  '55   '56   '^7   '58   '59    '60   '61    '62   '63   '64    '65   '66
                                                                                                   ST.  LOUIS SMSA
                                                                                                   Export for Industry 32
                                                                                                Predicted
 	T	1	1	1	1	1	1	1	
   1954  '55   '56   '57   '58   '59    '60   '61  '62   '61    '*>4   'dS   '66

-------
                             Figure  A.  1 (continued)
                                                                                         ST. LOUIS SMSA
                                                                                         Export by Industry 36
                                                                                 1954  '55  '56  '51
                                                                                                                    61   '62  '63  '64   '65   '66
     ST.  LOUIS SMSA
     Export by Industry 34
 (milUi

1300-


1200


1100-


1000-


 900-


 800-

 700-


 600-


 500 -


 400-


 300-
ST. LOUIS SMSA
Export by Industry 37
                                                                                Actual
1954  '55   '56   '57   '58   '59
                                                 '64   '65   '66
                                                                                	1	1	1	1	1	1	1	1	1	1	1	'	1
                                                                                 1954  '55   '56   '57   '58  '59   '60   '61  '62  '63   '64   '65   '66
       ST. LOUIS SMSA
       Export by Industry 35
                                                                          (million)

                                                                          425 '
                                                                                   ST. LOUIS SMSA
                                                                                   Export by Other Industry
                                                                                 1954  '55   '56   '57   '5«  '59  'bO   '41

-------
  APPENDIX B

31 AQCR MODEL

-------
Appendix B:  SlAQCRModel


      Introduction

      In the first year of model development, CONSAD developed a

Regional Econometric Model of the St.  Louis metropolitan region,

based on the time series data.  In the second year program plan,

the aim has been to focus on the development of a cross-sectional re-

gional econometric model which includes 31 AQCRs across the country.

In this model, the roles of Keynsian theory and regional export base

theory will remain as the basic framework as before.  However, the

entire equation system has been reformulated in order to integrate the

cross-sectional structure into the model.

      At the regional level, cross-sectional models are not uncommon

for the  single equation model (as are some partial equilibrium-type

models), partially due to the fact that time series data are usually un-

available in continuous time  series form.  However, an economy-wide

cross-section model,  at this stage,  is more a conceptual constrict

than a well-developed operational model. * This is partially due to the
      *Some Keynsian type cross-sectional models have been sug-
gested in Carl F.  Christ,  Econometric Models and Methods, New York,
1966, and a cross-sectional  income model has been estimated in
J. M. Mattila and W.  R.  Thompson, "Toward an Econometric Model
of Urban Economic Development, " (Appendix),  Issues in Urban Eco-
nomics,  edited by H. S. Perloff and L. Wingo, Jr. ,  Resources  for
the Future, 1968.

-------
fact that growth models at the national level always deal with time

series data, for a country, unless international comparisons of the

growth patterns are the focus.

      A Cross-Section Regional Econometric Model

      There is a fundamental difference between the economy of metro-

politan regions and the national economy.  The former is based on an

open economy where growth  and development is closely related to its

capability to carry on external trade with other regions.  The latter is

rather more self-contained by  its ;aature.  In spite of the Keynsian

macro-economic theory,  the concept of "export-base"  or economic

base theory, has been the core of the analytical frameworks of urban

economies since its first appearance in 1928. *  However, the measure-

ment of the economic base multipliers originally based on calculation

of the ratio between export-oriented (or basic) employment and local-

oriented (or service) employment has been changed by using the  con-

cept of value-added instead of employment. **
      *Haig,  Robert M. , Major Economic Factors in Metropolitan
Growth and Arrangement, Vol. I, Regional Survey of New York and
Environs, New York, 1928.  See also Thompson,  Wilbur R. , A Pref-
ace to Urban Economics,  Resources for the Future,  1965.
     -••-'•As an early example, see Leven, Charles  L. ,  "Measuring the
Economic Base, "  Papers and Proceedings, Vol. 2, The Regional
Science Association,  1956.

-------
      This later development now takes into account different produc-

tion structures among local industries,  so that factor intensity of cap-

ital and labor may contribute different weight to the multiplier.   Since

manufacturing industries are more capital-intensive, the role of man-

ufacturing industries in regional growth becomes decisive in that they

usually dominate the value of exports greater than 80 percent of  the

total value-added. *  It is therefore quite safe to treat manufacturing

industry as an export-oriented industry in the regional growth model

that in some of the recent successful regional econometric models ex-

plicitly or implicitly embodied a causal relationship of manufacturing

industry leads the overall regional growth. **

      The change in the measurement from employment  to value-added

in economic base theory not only has improved the  applicability of the

regional multiplier in recent regional growth analysis, but it also

seems to be consistent with the familiar Keynsian-type trade multiplier

in the open economic system.  However, economic base theory is only

a partial phenomenon in that it explains the demand side of the urban
      :;Tn a case study of five midwest metropolitan areas, Charles C.
Leven reported that manufacturing industry dominates 80 to  96 percent
of export measured by value-added,  while it only counts 45 to 71 per-
cent if it were  measured by employment.
     **A typical and successful model of this nature is  Frederick W.
Bell, op. cit.

-------
growth without proper inclusion of the supply side. *  The  relations

between labor and capital markets, export-oriented industry and other

local service industry, are also important.  Details are given in the

formulation of the entire model.

      Since the main interest in this model is focused on air pollution

control,  the  sectoral structure must be based on criteria  that reflect

the major air pollution industries being considered in the  present

study. In total,  13 manufacturing industries have been chosen as

shown in Table B. 1. **

      The variables of the model developed in this  section can be spe-

cified and defined as follows:

           •YI   = regional income of i*  SMSA,

            Ci   = regional consumption expenditure  of i"1 SMSA,

           Ij;   = investment expenditure  by industry j of i^ SMSA,

            Fiji  = capital share or gross profit by industry j of
                  ith SMSA,

           K^   = capital stock by industry j of i^1 SMSA,
      '"Thompson, W. R. , op. cit. ,  pp.  27-60.
     #*See CONS AD Research Corporation, Progress Report on Sec-
toral Structure and Data Collection,  prepared for TRW Systems, Inc. ,
September 20,  1969.  This classification has been based on the infor-
mation available in R. L..  Duprey, Compilation of Air Pollution Emis-
sion Factors. Washington, D. C. , Department  of Health, Education and
Welfare,  Public Health Service, 1968.

-------
                            TABLE B..1
           CLASSES OF MANUFACTURING INDUSTRIES
Industry Title

Canning and grain mill products

Fabrics, yarn and thread mills

Paper and allied products

Chemicals  and selected chemical
  products

Petroleum  refining and related
  industries

Rubber  and plastics products

Glass and glass products

Stone and clay products

Primary iron and  steel,  copper
  and aluminum manufacturing
Other non-ferrous metals
 manufacturing

Motor vehicles and equipment

Aircraft and other transportation
 equipment

Other manufacturing industries
Related SIC Codes

203,204

221, 222, 223, 224, 226, 228

26


281,286,287,289


29

30

321-323

324-329

331,332,3391,3399,3331,
3351,3362,2819,3334,3352,
3361

3332, 3333, 3339,3341,3356,
3357,3369, 3392

371


372-379

20-39, except included in
above classification

-------
           Xjj  =  value-added by industry j of it  SMSA,

           Nij  =  employment by industry j of i**1 SMSA,

           Wij  =  average wage by industry j of i*h SMSA,

           Ui   =  unemployment rate of i^ SMSA,

           Ui   =  unemployment of i*n SMSA,

           Ni   =  employment by industries other  than manufacturing
                  industries of i*h SMSA,

           NT  =  (S Ntj + Nj) total regional employment of ith SMSA,

           Li   =  regional labor force of i*h SMSA,

           Pi   =  population of ith SMSA,

           Qji  =  electric power demand by industry of i*h SMSA,

           Mt   =  migration of ith SMSA,

           Gj   =  government expenditure of i*^1 SMSA,

           Tj   =  government revenue of i"1 SMSA.

      Integration of the Keynesian system and economic  base theory

can be best explained by the  income determination block of this model:

(1)         Yi  = f (Ci,  S Xjj, Gi)


(2)         ^  = f (Yi,  Ci (t-1))

(3)         Gi  = f(Ti)

This model is a cross-sectional regional growth model.   Thus, the

national influence upon a local economy can be measured by the re-

gional market share of the output in each manufacturing  industry.

-------
Because manufacturing is also regarded as export-oriented, the value-

added,  S Xi4, also reflects the role of economic base theory of local
        j    J
demand, namely,  local consumption and government expenditures.  In

this model, therefore, government expenditure is a simple  function of

government revenue.

      The  sectoral structure of the export-oriented industries has been

based on  13 sectors.

(4)          Xij (t) = f (Yi (t), Yi (t-1), Xij (t-1))        j = l	13

(5)          lij (t) = f (Ilij  (t), Kij (t-1))                 j = l	13

(6)          Kij (t) = KIJ (t-1) +Iij (t)  - dj KIJ (t)         j=l	13

(7)          Xij (t) = AIJ Nij (t) K^"* (t)                  j = l	13

(8)          Wij(t) = f (14,  M^ Wij (t-1))                j = l	13

(9)          Qij (t) = f (Xij (t))                          j = l,	13

Equation (4) is the demand function for the products in terms of the

value-added by industry j  in region i, where Yi (t) and Yi (t-1) includes

both local and external market demand of present and previous years,

and Xij (t-1) measures market share from the previous period. Equa-

tion (5) is  a typical investment function and equations (6) and (7) are

familiar production structures developed in the St.  Louis Model.  Since

output level (value-added) and capital stock have been determined by

the three previous equations  ((4), (5), and (6)), equation (7) becomes

an implicit demand function for labor requirements by industry j (Njj).

-------
Wages (equation 8) are a function of the local unemployment rate (Ui),


migration (Mj) and lagged wages (Wij (t-1)).  This  relationship not only


includes labor productivity but it also shows, in part, the union bar-


gaining power based on the last period wage  level.   Finally, equation


(9), electric power demand by industry,  is expressed in terms of a


single function, current industrial output levels.


      The equations for  total industrial employment, other than for


manufacturing industries, and therefore a functional description for


the labor market in the regional economic model, is as follows:


(10)        Ni  = f (Yi  - S Xij)


            N?  =  ? Nii  + Ni
                   J   J

(12)        Li  = f(ui,  Pi,  2 Wy  Nij/2 Nij)


(13)        Ui  =  ^  - N?


                   Lj  - N?
(14)        u-   = 	;	 x 100
             1        Lt

Equation (10) is a labor  demand function for the industries other than


manufacturing industry.  Equation (11) is derived from equations (7)


and (10).  Equation (12)  represents  the supply side of the labor market.


The labor force (Li) is expressed as a function of regional unemploy-


ment  (ui), the weighted average wage rate (S W^ Nii/Z N:;) and size
                                          J    J    J  J    J

of the population (Pi). Unemployment in equation (13) is  the result of


the difference between the size of the labor force and total regional

-------
employment.  The amount of unemployment is expressed as a rate of

unemployment in equation (14).

      AQCRs Included in the Model

      The main interest of the Phase II Regional Econometric Model

was on the ability of the model to be applied to the major cities across

the country.  In total, the 31 largest SMS As have been chosen for the

present model.  The  size of the city,  in general, is reported to be con-

sistent with the  degree of the severity of the air  pollution problem. *

In the following  Table B. 2, both rank in size of population and in sev-

erity  of air pollution have been included for the cities under study.

      Model Estimation

      The model developed in Section 2. 2. 2 has been estimated on the

data from 31 largest SMS As:

      Regional income equation:
                                           13
(1)          Yi  =  64. 78 + 0. 702  Ct + 0.481 2 X^ + 3.02 Gi
                                          J = 1                R2 = 0.994
                  (194.4)  (0. 137)    (0. 100)       (0.591)

      Regional consumption function:

(2)          CA   =  15. 11 + 0.012  Yi + 1.061 Ci,..!             R2 = 0. 999

                  (40.9)   (0.030)    (0.048)
      *Middleton, John T. ,  "Air Pollution: How 65 Metropolitan Areas
Rank in Severity, " Nation's  Cities. Vol. 5, No. 8, August,  1967,  pp.
8-11/

-------
                          TABLE B.2

                                               Rank of     Rank of
Code               SMSA                      City*       Severity**

  1     New York, New York                      1          1
  2     Chicago, Illinois                          22
  3     Los Angeles/Long Beach,  California       3          4
  4     Philadelphia,  Pa. -  New Jersey            4          3
  5     Detroit,  Michigan                         5          9
  6     San Francisco/Oakland,  California         6         35
  7     Pittsburgh, Pennsylvania                  8          6
  8     St. Louis,  Missouri - Illinois              9         10
  9     Washington, D. C. - Maryland - Va.       10         18
 10     Cleveland, Ohio                          11          5
 11     Baltimore, Maryland                     12         13
 12     Newark, New Jersey                     13          8
 13     Minneapolis/St. Paul, Minnesota          14         32
 14     Houston, Texas                          15         43
 15     Buffalo,  New  York                       16         30
 16     Milwaukee, Wisconsin                    17         20
 17     Cincinnati, Ohio - Indiana - Kentucky      18         19
 18     ' Paterson/Clifton/Passaic, New Jersey    19         21
 ?9     Pall?.?. Tex?.?                           20         57
 20     Seattle/Everett,  Washington              21         36
 21     Kansas City,  Kansas                     22         25
 22     San Diego, California                     23         61
 23     Atlanta,  Georgia                         24         51
 24     Indianapolis,  Indiana                     25         14
 25     Miami, Florida                          26         63
 26     Denver,  Colorado                        27         27
 27     New Orleans, Louisiana                  28         59
 28     Portland, Oregon - Washington            29         49
 29     Columbus, Ohio                          33         46
 30     Rochester, New York                     34         53
 31     Dayton, Ohio                             35         26
      *Rank of city is based on the size of the population reported in
Current Population Reports:  Technical Studies, Series P-23, No. 23,
October, 1967.  Boston SMSA ranked 7th, Providence-Pawtucket-
Warwick SMSA ranked 30th, San Bernadino-Riverside-Ontario SMSA
ranked 31st and  Tampa-St. Petersburg SMSA have been excluded from
present study because of changes in definition of SMSA or lack in
certain type of data.
     **See  John  T.  Middleton,  ibid.


-------
      Demand function of export-oriented industry:




(3)          Xijt=  aj + bj  Yit + Cj Y^.j +dj X^^      j = l,2	13




      The results of estimation are:
Equation Industry aj b;
Number Code
(3.1) 1
(3.2) 2
(3.3) 3
(3.4) 4
(3.5) 5
(3.6) 6
(3.7) 7
(3.8) 8
(3.9) 9
(3.10) 10
(3.11) 11
-441.9
(532.0)
224. 3
(242.3)
-529.3
(1138.7)
-97.52
(927.9)
-449. 9
(927.8)
-695.9
(946.7)
-57.36
(117.7)
-1745.2
(1089.7)
-6063. 1
(5835. 1)
-412.4
(710.5)
4680. 7
(6835.9)
. 103
(.65)
.116
(.218)
.672
(1.38)
.252
(1.11)
.313
(1.08)
.635
(1.20)
1.899
(.717)
.738
(1.248)
3. 783
(6.07)
9.380
(2.59)
1.309
(. 814)
Cj
-. 155
(.698)
-. 160
(.232)
-.942
(1.47)
-.220
(1.18)
-.234
(1.15)
-.864
(1.29)
-2.028
(.767)
-.890
(1.34)
-3.580
(7.04)
10. 127
(2.77)

dj
1. 135
(.009)
1.085
(.006)
1. 122
(.013)
1.090
(.005)
1.097
(.006)
1. 135
(.012)
1. 061
(.003)
1. 131
(.013)
1.095
(.008)
1. 121
(.011)
1.047
(.006)
*2
.998
.999
.998
.999
.999
.998
.999
.997
.998
.999
.999

-------
(continued)
Equation
Number
(3.12)
(3.13)
Industry
Code
12
13
AJ bj
-3165.2 60.785
(3635.3) (23.32)
23296.6 7.337
(29916.3) (41.2)
Production functions:
(4) Xij = Aj N"J KJp
Equation
Number
(4.1)
(4.2)
(4.3)
(4.4)
(4.5)
(4.6)
(4.7)
(4.8)
(4.9)
Industry
Code
1
2
3
4
5
6
7
8
9
AJ
4.505
(1.002)
2.6375
(.659)
3.1312
(.418)
4. 708
(.919)
2.924
(.811)
4. 000
(.366)
3.548
(.531)
3.579
(.525)
3.296
(.423)
CJ dj
-66.56 1.159
(24.89) (.008)
-26.331 1.154
(42.7) (.030)


•j X-J
5217 .4783
5208 .4792
5315 .4685
5269 .4731
5293 .4707
5215 .4785
5085 .4915
5269 .4731
5301 .4699
R2
.999
.997
, 13
R2
.974
.993
.993
.973
.983
.997
.986
.995
.998

-------
(continued)
Equation Industry
Number Code
(4. 10) 10
(4.11) 11
(4. 12) 12
(4.13) 13
J J J
3.559 .5292 .4708
(.525)
4.517 .52152 .4748
(1.101)
6.204 .5231 .4769
(.358)
4.656 .5269 .4731
(.358)
*2
.997
.998
.996
.998
      Capital share  (or gross profit) relation:




      Given the Cobb-Douglas production function,  capital share or




gross profit can be derived as follows:




(5)         ny  = (1 -«j) Xjj                           j = l,...,13




Capital shore  coefficients,  cj, can be obtained as in the preceding




table for the production function.  Therefore, there are 13 equations




(5. 1) through (5. 13).




      Investment function:
(6)
= a
                    .
                                               13

-------
Equation
Number
(o. 1)
(6.2)
(6.3)
(6.4)
(6.5)
(6.6)
(6.7)
(6.8)
(6.9)
(6. 10)
(6.11)
(6.12)
(6.13)
Industry
Code
1
2
3
4
5
6
7
8
9
10
11
12
13
aj
-420. 6
(326.1)
-40.24
(183.7)
-903.6
(949.3)
-123. 1
(824. 1)
-624.5
(1668.6)
-782.2
(504.4)
87. 10
(108.2)
-2046.9
(1078.2)
-2923.3
(2378)
-62.79
(414. 1)
5832.6
(2637.3)
1459.8
(997.9)
(11708.3)
(11516)
bJ
. 168
(.067)
.4707
(.116)
.6033
(.099)
-.067
(. 047)
-.4419
(.133)
.4573
(.0942)
.0686
(. 147)
.0686
(.126)
. 6747
(.0595)
.6177
(.118)
.2157
(.0398)
.2174
(. 044)
.3586
(.048)
Cj
.0788
(.0516)
-.0809
(.0526)
-0.0852
(.0411)
.2032
(. 0304)
.2792
(. 034)
-. 1403
(.0682)
. 1004
(.079)
. 1004
(.0589)
-.0979
(.023)
-. 1568
(.062)
.0034
(.025)
-. 1571
(.078)
-. 1348
(.0412)
R2
.963
.986
.963
.971
.974
.967
.989
.989
.992
.982
.986
.979
.964

-------
     Capital stock identity:
(7)
where
Equation
Number
(7.1)
(7.2)
(7.3)
(7.-4)
(7.5)
(7.6)
(7.7)
(7.8)
(7.9)
(7.10)
(7. 11)
(7. 12)
(7.13)
Kijt = Kijt-l + Iijt = dj Kijt
d; is depreciation rate in industry j.
Industry
Code
1
2
3
4
5
6
7
8
9
10
11
12
13
j = l 	 13
Depreciation Rate

-------
Equation
Numbe r
(8.1)
(8.2)
(8.3)
(8.4)
(8.5)
(8.6)
(8.7)
(8.8)
(8.9)
(8.10)
(8.11)
(8. 12}
(8.13)
Industry
Co3e
1
2
3
4
5
6
7
8
9
10
11
12
13
aj
.:388
(.402)
-.658
(.458)
-2. 103
(.482)
.6105
(.432)
.0244
(.454)
.4938
(.399)
.0817
(.171)
. 1974
(.571)
.2051
(.522)
.2925
(.120)
-.2676
(1.34)
.2202
(.416)
. 8552
(.90)
bJ
.985
(.031)
1. 009
(.063)
1.2906
(.066)
.9961
(.032)
.9782
(.023)
.9593
(.059)
1.0179
(.018)
1.048
(.055)
1.0629
(.085)
.8941
(.018)
.9721
(. 064)
.9936
(.029)
1. 0008
(.078)
CJ
-.0516
(.0969)
.2109
(. 106)
. 1455 .
(.112)
-. 1556
(.146)
.0098
(.135)
-.0909
(. 076)
-.0214
(. 044)
-.1399
(.127)
-. 1701
(.175)
. 0419
(.025)
.2279
(.490)
.0906
(.157)
-.2850
(.278)
dJ
-.0183
(.022)
-.0069
(.0195)
.0384
(. 0263)
-.0135
(.0298)
.0473
(. 028)
.0025
(.018)
.0062
(.010)
-.0275
(.032)
-.0233
(.039)
.0206
(.006)
.0057
(. 064)
-.0050
(.0195)
-.0443
(.035)
R2
.981
.949
.944
.979
.987
.911
.996
.934
.917
.998
.938
.984
.847

-------
      Employment by industries other than manufacturing industries:

                                       1 1
(9)         Ni   =  121. 73 + 0. 1262 (Yi - S X^)         R2 = 0.968


                  (31.96) (0.004)


      Total employment:

           _~    _   13
(10)        N^  =  Ni + 2 N^
                      j = l


      Labor force supply function:

                                                13        13
(11)        1^   =  312. 735 + 0.4662 Pi - 39.749 (S W^ Nij/2 Nij)


                 (149.9)   (0.005)     (18.62)


                  - 31.319 ut                          R2 = 0.995
                   (17. 1)


      Regional employment


(12)        ^   =  Lj_ - NiT


      Regional unemployment rate:
(13)
j.
Li
Electric power
(14)
Equation
Number
(14. 1)
(14.2)
(14.3)'
Qijt = »j
Industry
Code
1
2
3
x 100
demand function by
+ bj Vijt
aj
4.6152
(2.81028)
29. 0536
(11.39258)
2.49808
(. 00002362)

industry:

bJ
. 79804
(.042681)
.205259
(.27310)
-.0049799
(.0031956)



R2
.933
.799
.999

-------
Equation
Number
(14.4)
(14.5)
(14.6)
(14.7)
(14.8)
(14.9)
(14. 10)
(14. 11)
(14. 12)
(14.13)
Industry
Code
4
5
6
7
8
9
10
11
12
13
aj

-22. 75882
(32. 82634)
-.0051536
(.0052080)
-0.0054016
(.0017908)
-5.84172
(4. 16915)
0. 028437
(3.85147)
2. 01174
(253.2238)
-5.53004
(2. 19407)
-11.33485
(10.35392)
-0.005127
(.007177)
-.004861
(.0020049)
bJ
11.98117
(.20547)
3. 76464
(.00003890)
1.50925
(.00001555)
1.83096
(. 14389)
2.38909
(.033961)
5.15529
(.39807)
1. 76572
(. 05378)
. 75855
(.010675)
. 75402
(.00001392)
. 10026
(.00000058)
R2
.991
.999
.999
.915
.994
.879
.994
.996
.999
.999
      Government expenditure function:




(15)        G^  =  55.2434 + 0.92961T




                 (10. 1457) (0. 007116)
R2 = 0.998

-------
           APPENDIX C

THE APCO REGIONAL ECONOMIC

-------
      The APCO Regional Economic Model developed in three years




(Phase HI) by CONSAD is a logical extension of the  concepts and em-




pirical work of the first  two years into an operational tool amenable




to assessment of various control strategies.  In particular, it builds




on and extends in Phase II 31 AQCR model (Appendix B) in three im-




portant dimensions.




      First,  the Phase III model is extended to cover 100 AQCR's.




These 100 AQCR's account  for over 65% of the nation. Thus,  the APCO




model will cover a major portion of the economic activity in the nation.




      Second, the  Phase in model overcomes a major drawvack of the




31 AQCR  model, which treats each AQCR as an isolated region.  In




reality, economic effects are attendant on pollution abatement.  Stra-




tegies such as price increases in certain industries,  the growth of




control equipment industries or demand spurred by reduced expendi-




tures on health, etc. ,  in one region will have  feedback affects on the




economies of other regions.  The Phase m APCO Economic Model




system uses a national input-output system linked to a regional market




share matrix to capture  these interregional feedback effects and en-




compasses the 100 AQCR's as an interrelated system of regions.




      Third, the Phase III model incorporates a regional fuel demand




submodel as a component of the Regional Economic Model  system.   As

-------
air pollution control policy is implemented, sulfur content in coal and




fuel oils will greatly affect their price in view of increased demand for




low sulfur fuels and their limited supply.  Prices of natural gas and




electricity will also tend to change.  Industries will choose an optimal




combination of fuels and electricity which minimizes the total cost of




energy to the degree substitution  is possible.  The fuel demand model




will describe these relationships.




      This appendix is addressed in particular to the description of




the incorporation of the  interregional feedbacks and regional fuel de-




mand model into the APCO Regional Economic Model system.  The




rest of the model system represents an extension of the 31 AQCR




model to the 100 AQCR's.

-------
1.   I-O System and
    Interregional Feedback
      As emphasized earlier in the model formulation, a region's

growth is closely dependent  upon  its  capability to carry on external

trade with other regions.  A national input-output system is  introduced

to serve the role of external market  for the regional economy described

in the regional model, and also, hopefully, to measure the structural

change of the national economy upon  air pollution control to the 100

AQCR's.

      The input-output system has been one of the most popular ap-

proaches used in economic structure analysis in the past several

decades.  With sufficient knowledge  in input-output coefficients and

details of the final  demand,  levels of interindustrial demand can be

derived by its inverse matrix.  However,  one major weakness of the

input-output system is that its operational feasibility is solely depen-

dent upon exogeneous  information regarding final demand.  Therefore,

a combined structure  of the  Keynsian macro-economic system and in-

put-output table has been introduced  in some recent econometric model

development.* A modification of Klein's  formulation will be sum-
      *  For example,  Dusenberry,  J.  S. ,  Fromm,  G. ,  Klein,  L. R. ,
and Kuh,  E.  (eds. ), The Brookings Quarterly Econometric Model of
the United States, 1965, Chapter 17.

-------
marized as follows. *  The input-output system will be

                  (I - A) X = Y

where:

      X is a vector of industry outputs

      Y is a vector of final demand by sector

      A is a matrix of technical coefficient

      I is a unit matrix


      In this model, aggregation of sectoral demand (the elements of

Y) is equal to GNP.

                  n
                    Y.  =  GNP
      This approach can be viewed as a model system when those sec-

tor demands are explained in terms of Engel curves or its analogues

are related to the Keynsian type macro-economic system.

      The relation between a national input-output system and a cross -

sectional regional model of Keynsian type formulation can be better ex-

plained from policy questions to be answered from this model system.

The regional model developed in Phase II is appropriate for  an isolated
      **  Klein, Lawrence R. ,  "What Kind of Macro-Econometric
Model for Developing Economies?",  Econometric Annual of the Indian
Economic Journal, Volume 13,  No. 3, 1965, also reprinted in A.
Zellner (ed. ),  Readings in Economic Statistics and Econometrics,  1968.

-------
region without interregional feedback scheme, although the export

activity is explicitly treated.  When an air pollution control policy is

implemented across the nation,  costs of production increases.   This

will result in an upward shift of the supply curve.  Whether an indus-

try or an individual firm may be able to  pass on the increased cost per

unit of output is dependent upon  the elasticity of demand and supply of

the corresponding products.  If  price changes are obtained from exo-

geneous  information, * the equilibrium supply of the products will be

known as shown in the following  figure.

      However, price increases in high emission industries under

air pollution control will not only reduce the demand of their products

but also  effect the demand of those products  which  are the intermediate

products  in the production of the high emission  industries.   For ex-

ample, a higher price for steel  products will not only reduce the de-

mand for steel but also affect sales in transportation,  coal products,

and other materials and services related to the steel  industry.

Further,  such effects originating in a given  region (or AQCR) in the

nation will not be limited to the  region but would affect the  economic
      * A study of price markup which air pollution control is insti-
tuted has  been reported in LeSourd, D. A.  etc. , Comprehensive Study
of Specified Air Pollution Sources to Assess the Economic Effects of
Air Quality Standards, Research Triangle Institute,  December, 1970.

-------
Price
        (with abatement)  S,
                                S   (pre-abatement)
                                                       Quantity

-------
activity in other regions (AQCR's) and in all likelihood, create a

feedback to the region. *

      For a limited number of regions, the formulation of an inter-

regional I-O system is perhaps feasible.   However,  a  20  sector

regional I-O system with 100 regions, the size of the matrix will be

2000 x 2000 (with  such detailed information largely nonexistent).

      An alternative formulation was consequently necessary.  A

national input-output system linked to a regional market share matrix
/

was used  to capture the regional feedback.  It is argued that the re-

gional share of the national market by industry (termed as the  "loca-

tion quotient") is  relatively stable.  For example,  if steel production

in Pittsburgh AQCR is 12% of the nation's steel product, a change in

the national steel  market will have a 12% effect on the  Pittsburgh

AQCR.  This concept is particularly useful for a cross-sectional model

which deals with the geographic distribution of economic activity at

a given period of time.

      In brief, using exogeneous information  on price  changes  in high

emission  industries occasioned by air pollution control, the high

emission  industries in each air quality control region  (AQCR) will fall
      *  These inte rregional feedback phenomenon were observed in a
pioneer study by Ronald E.  Miller,  "Interregional Feedback Effects in
Input-Output Models:  Some Preliminary Results, " Paper; Regional
Science Association, Vol.  17,  1966.

-------
in output to the point corresponding to an upward shifting supply curve

and the new price after control.  An aggregation of changes in regional

production for each high emission industry will give the change of

demand in the nation.  By use of the national input-output system,

the impact of changes  in high emission industries on other industries

can be measured. Finally, through the use of regional market share

matrix, the national impact can be distributed to each region  (AQCR)

as the net interregional feedback from the other regions under study

(99 AQCR's) and the rest of the nation.

      Mathematically, high emission industry in each region will have

an exogeneous determined price equation and supply curve as follows.

                 P.  = P. (exogeneous  price determination
                  •^      ^  of industry j)

                 Q..  =  f(P.,  factor prices, etc.) (regional supply
                                                   curve of high emis-
                                                   sion industry j
                                                   in region i)

Price increase after air pollution control,  AP.,will determine change

of production A  Q.. with other factors of the regional supply curve of
                  J
industry j in region i.

      Aggregation of AQ.., over regions (say 100 AQCR's) under con-

trol will give the "national" change  of the demand (supply = demand

at equilibrium in each region).
                          m
                 AY.  =        AQ..            '         j =  1, . .  .  n
                    J

-------
Change of final demand by high emission sectors,  AY., will affect


the structure change of the national economy since


                  (i - A) AX  = AY


then


                 AX =  (I - A)"1 AY


A X is a vector of change in gross product by sector in the nation as


the results of air pollution  control are implemented,  say in 100 AQCRs.


A X is then distributed to each region with the regional market share


matrix B.  B is an nxm matrix.


                  B  =  [b.. ]                               i=j,  .  .  . m
                                                          j = 1,  .  .  . n


By  definition,
where
      b.. is the  regional market share of jth industry product
         in the  ith region


      X  is the output of the jth industry in the ith region
       U

      X.  is  the output of jth  industry in the nation


      Appendix  D provides the empirical development of the  input-


output model system and the interregional feedback formulated here

-------
2  Regional Fuel Demand Model







      It has been observed that the burning of coal, fuel oil and natural




gas to produce power and heat is onS of the most important sources of




particulates,  sulfur oxides,  and nitrogen emission to the air.  Coal,




coke, fuel oil, natural gas and electricity are also the most important




energy sources available to the manufacturing industries in the nation.




Demand for energy increases as the manufacturing output increases.




However, each type of manufacturing industry differs from the other




in the production process; therefore, the type of fuels and combination





01 different type of fuel and electric power also  differ from industry




to industry.




      On the other hand, it is true that there are substitutional rela-




tions among the different types of fuel and/or electricity to produce




the energy (power and heat) necessary for any given level of product




of an industry.  Hence, industry may choose an optimal combination




of fuels and electricity which minimizes the total cost of energy.  This





is to say, the prices of fuels and electricity also affect the demand




for each type  of fuel or amount of electricity in  the production process




of each  type of manufacturing industry.  Therefore, if the price of




electricity or any type of fuel changes,  then the  demand for the fuels




and electricity changes according to a new optimal combination which




minimizes the total cost of the energy.

-------
      As the  air pollution control policy is implemented,  sulfur con-



tent in coal and fuel oils will greatly affect the price because of in-



creased demand for low sulfur fuels and their limited supply.  Prices



of natural gas and electricity (partly by the increase in production cost)



tend to change  because of changes in demand and supply relations.



      Demand for energy,  and hence fuels, like the demand for labor



or capital is  an induced demand from moderation.  Therefore, an



appropriate way to incorporate an energy demand model into the regional



model would  be to reformulate the production functions in the regional



model.



      A production function describes the maximum output obtainable



from every possible combination of inputs.  Some of the inputs are



substitutable for  one another while others are non-substitutable and



are proportional  to the output.  A general production relation can be



conceived of  various type of inputs, with substitutional relations among



a group of inputs categoried into  a number of sub-groups.  Between



any pair of sub-groups of inputs there is no substitutional relation.



Further assume that inputs for a  given industry have  been classified into



3 groups of inputs.  Since there is no substitutional relations among



those 3 inputs, the productions function can be given as



      X = min (- X,, -X_, -X,)
                a1  1  a2  2  a^  3'



Each group input is proportionate to the output.   However,  within each

-------
group of inputs there are substitutional relations

       X.  =  f(xu, x.2	x.n)                  1.1. 2, 3

 f has all properties of a new-classical production function.  Therefore,

       X =  min  i  fl(Xl.  . . x  ). ±  f-2 (x     ....  x  ),
                 L  1               1    Z       1            ^
       In the Phase II model,  the production functions treated the rela-

  tion between value added as output, and labor and capital as inputs in

  the form of a Cobb-Douglas type function.  Consequently,  interme-

  diari products included in the production were omitted as explicit fac-

  tors of production.

       Assume that X is gross product (or value of shipments) of a given

  industry, V is its value-added,  Z is energy requirement in this pro-

  duction, M is  another intermediate good. *

       X = min
lin  - V, - Z, -  M
   Lai    a2    a3   J
  Thus we can derive the relation between value-added and total energy as
                       *2
                   Z = —  V
                       al

  i.e. , energy demand is proportionate to the value-added.
       *  Although M can be an aggregation of other intermediate goods,

-------
      Suppose there are five types of fuels E.  (i = l, .  .  . , 5), the tech-



nical relations between fuels as inputs and total energy produced can be



described as an energy production function which is  similar to the



relation between labor and capital to the value-added.



                   Z =  f(Er ....  E5)



This relation provides the substitutional relation between each pair of



E. with a given level of energy (Z).



      As shown in Figure 1, with a given level of energy, there are



various combinations of fuel which produce the same level of energy.



For example, in Figure 1 combination of E, and E^  gives the same


                          it       "
energy as combination of E  and E^ .



      Given the price of each type of fuel i(i=l, .  . .  , 5), total cost of



energy, C,  is the  sum of the cost of all types  of fuel which is product



of price and quantity of fuel



                   C =  C,  + C  +C + C + C
                        12345



                     =  
-------
                         FIGURE 1
        FUEL SUBSTITUTION: ISO-ENERGY LINE
(fuel type 1)
      M
    E
      1
                                                iso-energy line Z


                                                     E  (fuel type 2)

-------
      Given the cost function C and energy production function E,  the


optional behavior for an industry will be:

                                 5
           Minimize cost   C = S  q. E*
                               i= 1


           Subject  to energy production function  Z = -f (Ej, .  . . , Eg)


The solution gives the optional condition.


                     M
                     9E
            1L       	[-               (i * j), i, j = 1	5
            —   =    !Ł
           'qj       8E.



That is to say, price ratio  of the types  of fuel equals to the  ratio of


marginal (or additional unit of) productivity of energy.  If price of fuel


q.and total energy   Z(which is proportion to  the output)  are  given then


as optimal combination of fuels E. (i=l  .  . . , 5) can be divided.  This


relation can better be explained in the following figure

-------
                    FIGURE 2
DETERMINATION OF OPTIMUM FUEL COMBINATIONS

-------
3.  The Regional Model



      With interregional relations and fuel demand submodel being


formulated as indicated above,  the rest of the Phase III model is quite


similar to that of Phase n 3 1 AQCR regional model.


Notation:


V        value-added by industry j of ith AQCR
  ij
K. .       capital stock by industry j of ith AQCR


N. .       employment by industry j of ith AQCR


L •        investment  expenditure by industry j of ith AQCR


n..       capital share or  gross profit by industry j of ith AQCR


W. .       average wage by industry j of ith AQCR


Y-        regional personal income of ith AQCR


C.        regional consumption expenditure of ith AQCR


G.        local government expenditure of ith AQCR


T.        local government revenue of ith AQCR


N.        employment by industries other than manufacturing
          industries of ith  AQCR
N
                 _
             i' + N- ^  total regional employment of ith AQCR
             ^
L.        regional labor force of ith AQCR

Q.        total regional consumption of electric power in ith AQCR

-------
Q..       electric power consumed by industry j of ith AQCR

QCJ       electricity consumed by residents of ith AQCR

Q.        electricity consumed by industries other than
          manufacturing industries of ith AQCR

qr        price of fuel type r of ith AQCR, type of fuel including
          coal, coke, fuel oil and natural gas

E  ..       demand of fuel,  type  r by jth industry by ith AQCR

i = 1, .  .  .  , 100   100 AQCR's

j = 1, .... 19    which is two-digit manufacturing industries

      The sector structure of the manufacturing industries included

two digit SIC industries as follows.

MANUFACTURING INDUSTRY CODE

SIC       Industry

20        Food and kindred products
22        Textile mill products
23        Apparel and related products
24        Lumber and wood products
25        Furniture and fixtures
26        Paper and allied products
27        Printing and publication
28        Chemicals and allied products
29        Petroleum and coal products
30        Rubber and plastics products
31        Leather and leather products
32        Stone,  clay, and glass products
33        Primary metal industries
34        Fabricated metal products
35        Machinery, except electrical
36        Electrical machinery
37        Transport equipment
38        Instruments and related products
39        Miscellaneous products

-------
 For manufacturing industry in each AQCR, the following equations



 provide a submodel of manufacturing sectors.
(1)    V  (t)  =  A.. N..Q(t) Ky  QJ(t)                          j-1	19




(2)    n..(t)  =  (1 -a.) V..(t)                                j=l	19
       *-J           J    J



(3)    I..(t)  =  f(FI  (t), K   (t-1))                           j=l	19

       1J          ij      ij



(4)    K  (t)  =  K.. (t-D  +I..(t)  - d  K..(t)                   j=l	19
       ij        iJ         iJ      j  ij



(5)    W..(t)  = f(W..  (t-1), u.(t))
 Equation  (1) is a typical Cobb-Douglas  production function.  Equa-


 tion (2) states the  profit as capital share from value-added.  Equa-


 tion (3) relates investment to the current profit and existing capital


 stock.  Equation (4) is  capital identity which defines capital stock


 of time t  as capital stock of previous period plus new  investment


 M>>nus depreciation,  dj is the depreciation  rate by industry (j  =  1,



 .  •  ,  19).

-------
      Integration of Keynsian system and economic base theory can be

treated in a regional income determination as before (Phase II).

(6)    Y.(t)  = f (C.(t). rVjjCt), Gj(t))
                      j

(7)    C.(t)  = f(Y.(t), C.(t-l))

(8)    G.(t)  = f(T.(t))


In Equation (6),  manufacturing industries are  treated as export activity

while regional consumption C. and local government expenditure also

are included in  income determination.  Equation (7) is a typical con-

sumption function, and Equation (8) merely relates local government

expenditure to local government revenue.

      Industries other than manufacturing industries have been treated

as residual of regional economic activity.  Therefore, the employment

by industries other than manufacturing industries is related to the non-

manufacturing income in Equation (9).
      _              19
(9)    N.(t)=  f (Y.(t)- Ł>..(t)


The labor market in each AQCR can be given by
              __      19
(10)   N (t)  = N(t) + y)N..(t) )
        i        !     *rr  ij

(ID   u.(t,  .    Li(t> - Ni(t)
       1         L.(t)

(12)   L.(t)  =f(N.(t), ujft))

-------
      Electric power demands are vital to the air pollution control



problem, it gives
(13)       Q..(t)  - f(V..(t))                                j = l, .  .  . ,  19
(14)      Q .(t) =  f(C.(t))
                       !
            Cl




(15)       Q.(t) =  f (Y.(t) - Z  V.,.(t))





(16)       Q.(t) = ZQ..(t) +Q .(t) +Q.(t)
            1     j  IJ       Cl       1




      Electricity consumed by  each economic sector is related to the



level of production of manufacturing industries, consumption expendi-



ture of residents, and income of other industrial sectors respectively.



Equation (16) gives the total regional demand of electricity.



      Fuel demand model developed in the previous  section can be



given as
(18)      Z..(t)  =  B  (t) E^1-",  Ef2j,  . . .  E^.5.J            j=l .....  19
            ij        ij    lij   2tj         5i3



Equation (17) gives the relation that the cost of fuel in the production



is proportionate  to the value-added by industry in each AQCR.  And



Equation (19) is an energy production function for a given industry.

-------
4.  Statistical Estimations
This section provides the results of statistical estimation of 162
equations in the region component of the Model System. *
(1) Production
V.. = A.
ij J
Equation
Number SIC
(1. 1) 20

(1.2) 22

(1.3) 23

(1.4) 24

(1.5) 25

(1.6) 26

(1.7) 27

(1.8) 28

(1.9) 29

functions
Ni°iKij"
*i
3. 6408
(.2714)
3. 5518
(.5138)
7.0735
(1.0456)
3. 8879
(.6094)
5.4057
(.3146)
3.3285
(. 0830)
5. 6188
(.4513)
2.4928
(. 1923)
1. 2408
(. 1455)

j = l.. . . , 19
•j '-«,
.3846 .6154 .997
(1.3488)
.5614 .4386 .995
(1.1609)
.5696 .4304 .998
(1. 1444)
.5780 .4220 .988
(1. 1726)
.5569 .4431 .997
(1.3215)
.5153 .4847 .994
(1. 1829)
. 5306 .4694 . 995
(1. 1687)
.2890 .7110 .994
(1.5465)
.2499 .7501 .994
(1.6557)
     * This count of equations does not include the I-O Model and
the market share matrix of the Regional Model System.

-------
(1.
(1.
(1.
(1.
(1.
(1.
(1.
(1.
(1.
(1.
10)
11)
12)
13)
14)
15)
16)
17)
18)
19)
30
31
32
33
34
35
36
37
38
39
4. 1808
(.4282)
6. 2163
(.7523)
3.4098
(. 3146)
3. 5046
(.3079)
4. 9152
(. 3368)
5. 2706
(.4248)
5. 9796
(.5209)
6. 2431
(.6891)
5. 7923
(.6361)
5.098
(.4871)
. 5030
(1.1932)
. 5423
(1. 1617)
.4649
(1. 2411)
. 5156
(1. 1858)
. 5192
(1. 1801)
. 5257
(1. 1753)
. 5356
(1. 1665)
. 5726
(1. 1365)
. 5013
(1. 1999)
. 5271
(1. 1718)
.4970
.4577
. 5351
.4844
.4808
.4743
.4644
.4274
.4987
.4729
. 995
. 995
. 988
. 998
. 998
. 995
. 998
. 998
. 998
. 994
          Given the Cobb-Douglas production function, capital share


or gross profit  can be derived as  follows.


(2)        n..  = (1 -a) V..                                j = l,. ...  19
            ij         J    iJ

          Capital share coefficients,  (1 -a.),  can be obtained as in the


preceding table for the production function.   Therefore, there are  19


equations (2. 1)  through  (2. 19).

-------
(3) Investment function
          I..  = a  + b n
           *J     J    J  ij
=l.. .  .  ,  19
Equation
Number SIC
(3.1)
(3.2)
(3.3)
(3.4)
(3.5)
(3.6)
(3.7)
(3.8)
(3.9)
(3. 10)
(3. 11)
20
22
23
24
25
26
27
28
29
30
31
a
J
2.
(1
0.
(0
0.
(0
0.
(0
0.
(0
2.
(1
2.
(0
1.
(2
-1.
(2
3.
(3
0.
(0

1030
.047)
1677
. 1721)
8537
.6129)
4099
.4073)
6597
.4689)
8315
.7716)
6523
.6559)
6595
.0899)
1015
. 1012)
6158
.6618)
0728
.0671)
b.
J
0.
(0
0.
(0
0.
(0
0.
(0
0.
(0
0.
(0
0.
(0
0.
(0
0.
(0
0.
(0
0.
(0

084
.0050)
1068
. 0044)
0265
.0035)
1111
.0382)
0725
. 0174)
1502
. 0348)
0585
. 0027)
1042
. 0093)
2979
. 0205)
2204
. 0784)
0342
.0028)
R2
0.799
0.934
0.478
0. 119
0.222
0.229
0.869
0. 670
0.853
0. 117
0.778

-------
(3. 12)
(3. 13)
(3. 14)
(3. 15)
(3. 16)
(3. 17)
(3. 18)
(3, 19)
(4) Capital

32 1.0897
(0.8574)
33 -0. 5863
(1.6544)
34 -0.0969
(0.4430)
35 5. 1995
(6.5116)
36 1,8848
(1. 1608)
37 1. 1828
(1.2960)
38 -0.0229
(0. 1575)
39 0. 1637
(0.3164)
Stock Identity
K = K. . , i I
ijt ijt-1 ijt
where d. is depreciation rate in
Equation
Numbe r
(4. 1)
(4.2)
(4,3)
(4.4)
SIC
20
22
23
24
0.1564 0.506
(0.0190)
0.2715 0.933
(0.0095)
0.1260 0.967
(0.0029)
0.1343 0.216
(0. 0320)
0. 1116 0.844
(0.0063)
0.1122 0.922
(0. 0043)
0.1076 0.992
(0. 0012)
0.0715 0.917
(0. 0028)

<*. K j=l,
J ijt
industry j.
Depreciation Rate
d.
J
.06738
.05248
.09060
. 09149
                               . ,  19

-------
(4.5)
(4.6)
(4.7)
(4.8)
(4.9)
(4. 10)
(4. 11)
(4. 12)
(4. 13)
(4. 14)
(4. 15)
(4. 16)
(4. 17)
(4. 18)
(4. 19)
(5) Regional
Y
i
(6) Regional
C
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
Income Equation
13,
= .42836 7 V.. + . 94812
Ł1 1J
(.0564) (.0661
Consumption Function
= 137.99 + .6133 Y. + .012
.07366
.05343
.06859
.06690
.06139
.06182
.07549
. 06457
.05598
.06995
.07216
.07379
. 07840
.06714
.07321
C. + 2.8304 G. R2 = . 997
i i
1) (.30932)
14 C , R2 = .992
(50.09)   (.0057)    (.0012)

-------
(7)  Local Government Expenditure



         G.  = 23.4227 + . 94215T                        R2 = . 998


               (4. 108)   (.0048)



(8)  Employment by industries other than

    manufacturing industries


         -                            19
         N.  = 56. 3560 + . 10323  (Y. - J^ V..)            R  = . 978
                                   - V* V  )
                                  1   FI  ij
               (9.523)     (.0016)



(9)  Total employment
        N  = N  +      N
          i      i     Z-f   ij
(10)  Labor force supply function



         L.  =  -13.958 + 1.0392 N. + 361.374 U.          R2 = . 999
          i                       11


               (2.080)  (.0009)     (55.965)



(11)  Regional Unemployment Rate



                  L. -N .

           U.  =   — - - — x 100

                     L.
                       i
(12)  Electric power demand

     function by industry



        Q . =  a + b V..                                 j = l, .... 19

-------
Equation
Number
(12.
(12.
(12.
(12.
(12.
(12.
(12.
(12.
(12.
(12.
(12.
(12.
(12.
1)
2)
3)
4)
5)
6)
7)
8)
9)
10)
11)
13)
14)
SIC
20
22
23
24
25
26
27
28
29
30
31
32
33
a
J
1.4987
(.6969)
1.4681
(.7354)
-.01340
(.0931)
. 025681
(. 1584)
.05072
(.07771)
1.9802
(1.6883)
1.0508
(. 3265)
12.0593
(5.492)
2. 00296
(2. 5082)
-. 3307
(.3239)
.07498
(. 0530)
3. 1259
(1.3261)
8. 8908
(6. 1280)
b.
J
. 006517
(.00024)
. 009954
(.00109)
.002932
(.000030)
. 009854
(.00065)
.004665
(.00015)
.01484
(.002:.2)
. 002547
(.00010)
.0120
(.0022)
. 0325
(.01533)
.01533
(.00045)
.00308
(.00012)
. 01581
(.0018)
. 02730
(.0021)
R2
.918
. 667
. 994
. 809
. 943
.457
.913
.329
. 7971
.958
. 939
. 549
. 749

-------
(12.
(12.
(12.
(12.
(12.
(12.
14)
15)
16)
17)
18)
19)
34
35
36
37
38
39
. 2557
(. 5316)
. 6298
(. 6848)
2.08744
(1. 1652)
-.2536
(1.3973)
. 6778
(.3588)
.7320
(.7793)
.0074
(.00025)
. 005378
(.00023)
. 00547
(.00038)
.007513
(.00026)
. 002845
(.00024)
.006066
(.00075)
. 938
. 902
. 796
. 939
. 728
. 540
      Electricity demand by residentials
      in each AQCR

(13)        Q .  =  .04718C
             C1    (.00216)                            R2 = .880

      Electricity demand by other industries in each AQCR

(14)        Q~:   =  . 07947 (YŁ = Z VH)                R2 = . 846
                  (.00423)      j   J
      Regional demand of electricity
                   19            —
(15)        Qi   =  Z  Qij + Qci + Qi

-------
     Energy demand function
(16)
= aJ
j  = 1,..., 19
Equation
Number
(16.1)
(16.2)
(16.3)
(16.4)
(16.5)
(16.6)
(16.7)
(16.8)
(16.9)
(16. 10)
(16. 11)
(16. 12)
(16. 13)
(16. 14)
(16. 15)
(16. 16)
(16. 17)
(16. 18)
(16. 19)
Industry
SIC
20
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
aj
2. 13525
3. 04152
0. 74246
2. 57780
1.33572
3. 97124
0.61895
2. 79554
9.43424
2. 83225
1. 15811
6. 14368
10. 68980
2. 10111
1. 34915
1. 19740
1. 35564
1. 00903
1.44677

-------
      Energy production function




M71         7    R  TT71J F72J F"1^
(17)         Z.j - Bj Enj E2ij E3ij  4i
j  =  1	19
Equation SIC B- "^ "^ y3 y4
Number Code
(17.
(17.
(17.
(17.
(17.
(17.
(17.
(17.
(17.
(17.
(17.
(17.
1) 20
2) 22
3) 23
4) 24
5) 25
6) 26
7) 27
8) 28
9) 29
10) 30
11) 31
12) 32
17.304
19.563
(5.246)
21.914
(7.570)
19.061
(3.693)
23.206
(4. 134)
20.941
(4.732)
19.616
(1.963)
12. 312
(4. 514)
3. 063
(. 700)
14. 935
(1.617)
17.965
(12.313)
9. 586
(2.476)
>0993
(.0597)
. 1133
(.0523)
.0
.0
. 0882
(.0237)
.2150
(.1233)
.0
. 1564
(.0798)
. 0
. 0
. 1191
(.0318)
. 1382
(. 0785)
. 0
.0
.0
.0
.0
. 0
.0
. 0
. 0
.0
. 0
. 0
. 1107
(.0687)
.1544
(.0786)
. 0934
(.0572)
. 1707
(.0337)
.0873
(.0282)
. 1318
(.0961)
. 0670
(.0206)
.0587
(.0618)
.0620
(.0594)
. 1030
(.0307)
. 1535
(.0387)
. 0818
(. 0455)
.2250
(. 0908)
. 1070
(.0429)
.0822
(.0359)
. 1110
(. 0323)
. 1243
(. 0280)
. 1378
(.0995)
. 1325
(. 0352)
.2067
(. 1109)
. 5371
(. 1387)
. 1236
(.0347)
. 1105
(.0199)
.3780
(.1195)
ys
. 5651
(. 0820)
.6254
(. 1204)
. 8244
(.0817)
. 7183
(.0493)
. 7002
(. 0618)
. 5154
(. 1390)
. 8006
(. 0503)
. 5782
(. 1352)
.4009
(. 1308)
. 7734
(.0436)
.6169
(. 0601)
.4020
(.0857)
R2
.984
.976
.998
.970
.995
.930
.982
. 920
. 959
.989
.941
. 857

-------
(17.13)  33


(17.14)  34


(17.15)  35


(17. 16)  36


(17.17)  37


(17.18)  38


(17.19)  39
19.482
(4.450)

17. 059
(1.719)

19.981
(2. 734)

16.827
(2.912)

17. 572
(5.738)

18.982
(8.577)

20. 846
(4.008)
 .0465
(.0195)

 .0352
(.0173)

 .0688
(.0363)

 .0493
(.0142)

 .0905
(.0321)

 . 1141
(.0591)

 . 0608
(.0182)
 . 1838    .1062
(.0852)  (.0546)
 . 0
 .0
 .0
 .0
 . 1108
(.0588)

 . 1087
(.0663)

 .0800
(.0391)

 .0717
(.0406)

 . 1159
(.0226)

 . 1289
(.0436)
 .2293
(.0701)

 .2338
(. 0762)

 . 1656
(.0597)

 . 1567
(.0436)

 . 1345
(.0394)

 . 1199
(.0197)

 . 1437
(.0661)
 .4342
(.0982)

 .6202
(.0691)

 .6569
(.0687)

 . 7139
(.0556)

 . 7032
(.0536)

 .6500
(.0593)

 .6666
(.0980)
.983


.993


.987


.955


.990


.969


.985

-------
         APPENDIX D

 INPUT-OUTPUT MODEL SYSTEM

-------
1.  I-O Model







      An input-output table of a nation Shows the total output and the




interindustry transactions in the national economy.   The national eco-




nomy is broken down into a large number of industries or sectors (up




to more than 380 sectors in 1963 OBE Input-Output table) and repre-




sented in a matrix form.  Each row provides the distribution of domes-




tic outputs of a given industry to all other industries as intermediate




products, and to the final demand for consumption.   Each column  repre-




sents the purchase of all inputs including intermediate products from all




other industries and contribution of labor capital.  The transactional




patterns provided in an  input-output table are assumed to be  relatively




stable regardless of level of output in each  sector.   The percentage




distribution (or ratios) of inputs in a given column to one unit of out-




put is called input-output coefficients which describe the present stage




of production technology in a given industry.  In this manner, the  input-




output table displays the pattern of the present stage of production




technology of the nation. The  input-output  table is particularly useful




when addressing the following  type of policy.  "If the final demand for




goods and services changes, what would be the  new  gross domestic  pro-




duction in each industry?"  Such information can be  obtained through




the use of so-called "inverse coefficient" matrix.  The inverse

-------
matrix describes the direct and indirect input requirements of all




sectors to deliver one unit of final demand.




      As national air pollution control policies are implemented, the




resulting  cost increases in high emission industries would induce a




price increase in the products of those industries.  This increase in




price may result in a reduction in  the demands for the products  of




those industries.  On the other hand,  air pollution abatement would




bring about  cleaner air in the nation as a whole,  leading in turn  to a




reduction in health expenditures,  increase in property values, etc.




These benefits can be viewed as additional money available to the




United States residents for consumer expenditures which would create




additional demand for a variety of  goods and services.




      Thus, the implementation of air pollution abatement  policies




in the nation will cause reduced demand for certain industries but the




resulting  benefits of cleaner air may stimulate the demand for products




of the same or other industries. Such changes in demand and resulting




outputs  of various industries  attendant upon an air pollution control




policy can be ideally captured by the use of the national I-O model




when it is properly sectored for the purpose at hand.  Such a use of




the model is explained in detail in  the  following section.

-------
2.  Use of the I-O Model


      In order to provide an analytical tool for the study of national air

pollution policies on the U. S. economic structure,  CONSAD has de-

veloped a computer program entitled Program IOA  (Input-Output Analy-

sis).  This program accepts as input a hundred sector input-output table

and aggregates to any desired number of sectors.  It also calculates the

corresponding input-output coefficient table and inverse matrix.  This

program also takes changes of final demand by sectors and costs and

benefits of national air pollution control policies to estimate the changes

in production of each industry in the national economy.

      To illustrate the use of this model,  a 42-sector analysis  has

been conducted in this appendix.  First, the total control costs by in-

dustry were  transformed into the changes in  the final demand in the

nation.*

      Second, the effects of benefit are  estimated to be $l.-i>  billion of

additional consumption expenditures to  the national economy.   This

additional consumption expenditure is then distributed among 42-sectors
      * In this example, we use total control cost by industry provided
in Comprehensive Economic Cost Study of Air Pollution Control Cost
for Selected Industries and Selected Regions (February, 1970) by
Research Triangle Institute to APCO. Control costs by sector are
then transformed into changes of demands by the use of coefficients
estimated in the  regional model.

-------
to the consumption pattern of the nation; in other words,  proportionally




increase the sector consumption as before (this is average propensity




to consume by sector).  Sectoring  of the I-O model, coefficient and




inverse matrices are given in the following tables.

-------
Input-Output Coefficient Matrix and Inverse Matrix
Sectoring
SECTOR
SECTOR
SECTOR
SECTOR
SECTOR
SECTOR
SECTOR
SECTOR
SECTOR
SECTOR
SECTOR
SECTOR
SECTOR
SECTOR
SECTOR
SECTOR
SECTOR
SECTOR
SECTOR
SECTOR
SECTOR
SECTOR
SECTOR
SECTOR
SECTOR
SECTOR
SECTOR
SECTOR
SECTOR
SECTOR
SECTOR
SECTOR
SECTOR
SECTOR
SECTOR
SECTOR
SECTOR
of the Input- Output Model
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
33
34
35
36
37
AGRICULTURAL
IRON MINES
NON-FERROUS MINES
COAL MINES
CRUDE PETRO-NATURAL GAS
STONE AND CLAY MINING
CHEMICAL MINING
CONSTRUCTION
ORDINANCE (SIC 19)
FOOD PRODUCTS (SIC 20)
TOBACCO PRODUCTS (SIC 21)
TEXTILE PRODUCTS (SIC 22, 23)
LUMBER-WOODPRODUCTS (SIC 24)
FURNITURE-FIXTURES (SIC 25)
PAPER PRODUCTS (SIC 26)
PRINTING- PUBLISHING (SIC 27)
CHEMICAL PRODUCTS (SIC 28)
PETROLEUM REFINING (SIC 29)
RUBBER-PLASTIC PROD. (SIC 30)
LEATHER PRODUCTS (SIC 31)
STONE, CLAY.AND GLASS (SIC 32)
IRON AND STEEL MANU. (SIC 33)
COPPER MANUFACTURING (SIC 33)
ALUMINUM MANUFACTURE (SIC 33)
OTHER NON-FERROUS MET (SIC 33)
FABRICATED METAL PROD (SIC 34)
MACHINERY, EX. ELECTRIC (SIC 35)
ELECTIRCAL MACHINERY (SIC 36)
MOTOR VEHICLES (SIC 371)
OTHER TRANSPORTATION (SIC 37)
INSTRUMENTS (SIC 38)
MISCELLANEUS MANUFAC. (SIC 39)
TRANSPORTATION- WAREHOUSING
OTHER SERVICES
MEDICAL, EDUC. & NON-PROFIT
ELECTRIC UTILITIES (SIC 4911
GAS UTILITIES (SIC 492)

-------
Input-Output Coefficient Matrix and Inverse Matrix

Sectoring of the Input-Output Model (continued)
SECTOR  38             WHOLESALE & RETAIL
SECTOR  39             FINANCE & SERVICES
SECTOR  40             IMPORT OF  GOODS AND SERVICES
SECTOR  41             GOVERNMENT ENTERPRISES
SECTOR  42             MISCELLANEOUS BUSINESS

-------
Input/Output Coefficient Matrix:

ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
SOW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
SOW
ROW
ROW
ROW
ROW
ROW
ROW
ROW

1
2
3
4
5
6
7
8
9
10
11
12
13
—r*r-
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42

0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
•— 'o.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
COL I
3063585
0
0
0001381
0
0012640
0005380
0123653
0
0669180
0
0023912
0019046
0
0008181
0002341
0225669
0174438
0032423
0000921
0005985
0
0
0
0000381
0024070
0042577
0005380
0011825
0004538
0
0000802
0163981
0641467
0030594
0027280
0003420
0337471
0096294
0203269
0001973
0009497
COL 2
0.0
0.0665909
0.0311749
0.0044598
0.0
0.0
0.0
0.0005684
0.0
0.0
0.0
0.0001312
0.0058590
0.0
0.0000437
0.0000437
0.0127235
0.0093131
0.0005247
0.0
0.0007433
0.0170959
0.0000874
0.0000874
0.0017489
0.0021425
0.0247038
0.0021862
0.0005247
0.0022299
0.0002186
0.0
0.0997771
0.0764724
0.0007433
0.0139915
0.0008745
0.0194570
0.0053343
0.2990252
0.0007433
0.0044598
COL 3
0.0
0.0088175
0.1529400
0.0009241
0.0
0.0000385
0.0008086
0.0008086
0.0
0.0
0.0
0.0013092
0.0010011
0.0
0.0003080
0.0002310
0.0294945
0.0056217
0.0023873
0.0
0.0047746
0.0360403
0.0001540
0.0001540
0.0047746
0.0016942
0.0250279
0.0036194
C. 0007316
0.0
0.0003465
0.0
0.0316507
0.0338070
0.0007316
0.0144777
0.0054676
0.0244H9
C. 0114743
0.1587927
0.0009241
0.0038505
COL 4
0.0
0.0
0.0
0.1821136
0.0
0.0004641
0.0000273
0.0008463
0.0
0.0
0.0
0.0006825
0.0074254
0.0
0.0026753
0.0002457
0.0174170
0.0108651
0.0078076
0.0
0.0023477
0.0084082
0.0004095
0.0000819
0.0064426
0.0118752
0.0501215
0.0032213
0.0021566
0.0042041
0.0000546
0.0012285
0.0073435
0.0278999
0.0009555
0.0227403
0.0000546
0.0328684
0.0096640
0.0010101
0.0012012
0.0040949
COL 5
0.0
0.0
0.0
0.0000179
0.0263655
0.0
0.0
0.0003882
0.0
0.0
0.0
0.0002090
0.0005793
0.0
0.0004658
0.0000717
0.0047476
0.0046998
0.0032069
0.0000060
0.0004121
0.0002747
0.0000119
0.0000119
0.0007465
0.0061330
0.0144219
0.0038100
0.0008779
0.0
0.0000597
0.0000358
0.0254876
0.1597037
0.0008480
0.0047416
0.0016482
0.0124870
0.0097878
0.0822316
0.0004419
0.0063958
COL 6
0.0
0.0002724
0.0005837
0.0017123
0.0
0.0070050
0.0003892
0.0012064
0.0
0.0
0.0
0.0000389
0.0000389
0.0
0.0096124
0.0002724
0.0118306
0.0272416
0.0187189
0.0
0.0649907
0.0141267
0.0000778
0.0000389
0.0011286
0.0007783
0.0706336
0.0019458
0.0039306
0.0001557
0.0002335
0.0003113
0.0158780
0.0365816
0.0008951
0.0215598
0.0017902
0.0603597
0.0106242
0.0592700
0.0015956
0.0049424
COL 7
0.0
0.0006233
0.0005194
0.0008311
0.0018699
0.0151673
0.0639934
0.0005194
0.0
0.0001039
0.0
0.0003117
0.0004155
0.0
0.0055059
0.0
0.0312695
0.0088303
0.0042593
0.0
0.0005194
0.0177644
0.0002078
0.0002078
0.0031166
0.0017661
0.0348016
0.0043632
0.0021816
0.0003117
0.0002078
0.0002078
0.0597342
0.0196343
0.0008311
0.0208810
0.0203615
0.0243286
0.0058176
0.0891337
0.0010389
0.0082070
COL 8
0.0028276
0.0
0.0
0.0000010
0.0000010
0.0100778
0.0
0.0001299
0.0000459
0.0001563
0.0
0.0000840
0.0507102
0.0064024
0.0052489
0.0001319
0.0215386
0.0181455
0.0054081
0.0000088
0.0707602
0.0333012
0.0052821
0.0003741
0.0171297
0.1041405
0.0193126
0.0252892
0.0000225
0.0000332
0.0031978
0.0018919
0.0313585
0.0506789
0.0009396
0.0020706
0.0000088
0.0879836
0.0081888
0.0
0.0002393
0.0038629
COL 9
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0015735
0.0184039
0.0
0.0
0.0007553
0.0011959
0.0000126
0.0047206
0.0017875
0.0034491
0.0021526
0.0236154
0.0000378
0.0036631
0.0134819
0.0073641
0.0142372
0.0287765
0.0190962
0.1096930
0.0693732
0.0041667
0.2121102
0.0238419
0.0025176
0.0104985
0.0201662
0.0009567
0.0029204
0.0006294
0.0288772
0.0063067
0.0018882
0.0009693

-------

ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW

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
33
34
35
36
37
38
39
40
41
42

0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
COL 10
.3086993
.0
.0
.0006669
.0
.0000480
.0001243
.0033984
.0
.1710971
.0000075
.0025205
.0013931
.0
.0186234
.0018951
.0063447
.0038100
.0019045
.0000057
.0084461
.0000245
.0000207
.0004380
.0000876
.0271165
.0002618
.0004728
.0
.0
.0000019
.0004606
.0382386
.0599993
.0009306
.0034389
.0013328
.0348505
.0050128
.0352716
.0009155
.0054838
COL 11
0.1955482
0.0
0.0
0.0002436
0.0
0.0
0.0
0.0000530
0.0
0.0065455
0.1907821
0.0002012
0.0016523
0.0
0.0240106
0.0019594
0.0205896
0.0004448
0.0015146
0.0000424
0.0000106
0.0
0.0000212
0.0011650
0.0000212
0.0030609
0.0001165
0.0001059
0.0
0.0
0.0
0.0011333
0.0121801
0.0469408
0.0009850
0.0006355
0.0001589
0.0141818
0.0020547
0.0033998
0.0021500
0.0018535
COL 12
0.0435766
0.0
0.0
0.0006306
0.0
0.0000020
0.0000315
0.0004927
0.0
0.0009242
0.0
0.4189151
0.0000532
0.0006602
0.0095339
0.0009045
0.0549413
0.0012415
0.0052872
0.0019687
0.0010503
0.0002168
0.0000138
0.0000296
C. 0000769
0.0012829
0.0025126
0.0001360
0.0000315
0.0000512
0.0003015
0.0103143
0.0158340
0.0214346
0.0010168
0.0053010
0.0004178
0.0380115
0.0060695
0.0197497
0.0014287
0.0057405
COL 13
0.1142030
0.0
0.0000227
0.0001744
0.0
0.0000152
0.0000076
0.0015015
0.0
0.0000076
0.0
0.0014863
0.2677022
0.0024418
0.0089254
0.0028892
0.0158337
0.0075225
0.0049215
0.0000607
0.0042163
0.0019868
0.0000152
0.0007962
0.0001896
0.0100326
0.0034883
0.0011754
0.0000455
0.0007432
0.0000152
0.0013043
0.0435503
0.0215514
0.0008797
0.0048381
0.0002806
0.0393720
0.0054144
0.0624325
0.0009782
0.0065670
COL 14
0.0
0.0
0.0
0.0004359
0.0
0.0
0.0
0.0004480
0.0
0.0046615
0.0
0.0508838
0. 1031579
0.0308868
0.0210554
0.0005085
0.0189728
0.0019857
0.0295187
0.0016830
0.0232953
0.0466269
0.0000363
0.0087297
0.0008233
0.0635898
0.0091777
0.0028332
0.0007991
0.0005933
0.0019978
0.0074099
0.0196388
0.0331994
0.0010170
0.0048310
0.0003753
0.0553566
0.0050368
0.0001332
0.0008839
0.0096014
COL 15
0.0
0.0
0.0
0.0057325
0.0
0.0017786
0.0011649
0.0038579
0.0000919
0.0049017
0.0000084
0.0078911
0.0500813
0.0001169
0.2826012
0.0081040
0.0340194
0.0095278
0.0128095
0.0001253
0.0044424
0.0002129
0.0000084
0.0007766
0.0003006
0.0111394
0.0051229
0.0013945
0.0
0.0
0.0003298
0.0007933
0.0362698
0.0236816
0.0009645
0.0087261
0.0039664
0.0378856
0.0060331
0.0646193
0.0020208
0.0063003
COL 16
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0034071
0.0004181
0.0
0.0
0.0012735
0.0000481
0.0003844
0.1801769
0.1155335
0.0137917
0.0008458
0.0010716
0.0000481
0.0000144
0.0
0.0000144
0.0009995
0.0002547
0.0021144
0.0036810
0.0009323
0.0001538
0.0013359
0.0026190
0.0025854
0.0145510
0.0935868
0.0010668
0.0040270
0.0
0.0249357
0.0094428
0.0024700
0.0069631
0.0249212
COL 17
0.0016479
0.0023061
0.0018100
0.0041014
0.0010806
0.0009578
0.0158431
0.0015472
0.0000049
0.0176605
0.0
0.0022300
0.0019082
0.0000049
0.0300801
0.0025222
0.2598557
0.0323150
0.0096272
0.0000147
0.0100987
0.0033523
0.0004224
0.0020556
0.0047719
0.0185078
0.0078049
0.0009382
0.0000123
0.0000393
0.0017633
0.0010241
0.0302422
0.0706838
0.0009603
0.0085368
0.0057616
0.0281178
0.0088217
0.0174149
0.0011902
0.0118547
COL 18
0.0
0.0002255
0.0
0.0006063
0.5085889
0.0032976
0.0000665
0.0014824
0.0
0.0006913
0.0
0.0002736
0.0001664
0.0
0.0051867
0.0000370
0.0343695
0.0720884
0.0003734
0.0000037
0.0022662
0.000022?
0.0000333
0.0000333
0.0000739
0.0184398
0.0002773
0.0004695
0.0000148
0.0
0.0000776
0.0003586
0.0531975
0.0327836
0.000^464
0.0048946
0.0101145
0.0107023
0.0066395
0.0317337
0.0022625

-------

ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
"OW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW

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
33
34
35
36
37
38
39
40
41
42

0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
COL 19
0
0
0
0016051
0
0005813
0012234
0009631
0
0001996
0
0870933
0017353
0002256
0114788
0028892
1869843
0024207
0325971
0019695
0108281
0021951
0000087
0001822
0022385
0153658
0054488
0036961
0000174
0027851
0016051
0040258
0222375
0402409
0010498
0088586
0010238
0327967
0058912
0372823
0011540
0095180
COL 20
0.0003601
0.0
0.0
0.0006220
0.0
0.0
0.0002128
0.0001146
0.0000164
0.0007038
0.0
0.0362211
0.0091494
0.0002619
0.0182006
0.0044683
0.0208194
0.0011785
0.0528668
0.2661997
0.0042883
0.0
0.0000491
0.0000491
0.0003437
0.0077909
0.0006220
0.0014240
0.0000491
0.0000818
0.0019805
0.0030771
0.0153526
0.0389708
0.0011948
0.0037645
0.0002619
0.0337496
0.0075454
0.0135686
0.0031589
0.0073326
COL 21
0.0003353
0.0007803
0.0008656
0.0071993
0.0
0.0457013
0.0022677
0.0004450
0.0
0.0007010
0.0
0.0022921
0.0074005
0.0005303
0.0413610
0.0016825
0.0383313
0.0090525
0.0079735
0.0000975
0.1080201
0.0022921
0.0001219
0.0010790
0.0010302
0.0127527
0.0034869
0.0045232
0.0001463
0.0000975
0.0006218
0.0018715
0.0504988
0.0295592
0.0010180
0.0137951
0.0162213
0.0329973
0.0095889
0.0141182
0.0023957
0.0096865
COL 22
0.0
0.0524958
0.0002456
0.0218467
0.0
0.0025895
0.0003335
0.0059491
0.0000515
0.0003548
0.0
0.0009127
0.0011704
0.0000728
0.0030837
0.0012098
0.0115070
0.0067951
0.0027259
0.0000030
0.0147242
0.2052591
0.0016161
0.0019285
0.0112524
0.0253610
0.0207096
0.0042208
0.0016738
0.0007944
0.0002820
0.0004730
0.0477566
0.0185174
0.0009248
0.0117890
0.0091207
0.0330657
0.0067951
0.0473745
0.0014130
0.0043390
COL 23
0.0
0.0
0.1268308
0.0012342
0.0
0.0010696
0.0
0.0003497
0.0
0.0002674
0.0
0.0005348
0.0001234
0.0
0.0027563
0.0006788
0.0062531
0.0040727
0.0013781
0.0
0.0032088
0.0080015
0.2751155
0.0185330
0.0908755
0.0182039
0.0123828
0.0030648
0.0007611
0.0
O.OOOC617
0.0006994
0.0215361
0.0151185
0.0007611
0.0074667
0.0049778
0.0389996
0.0060268
0.1343592
0.0007611
0.0045047
COL 24
0.0
0.0
0.0339620
0.0016575
0.0
0.0002706
0.0000507
0.0005243
0.0
0.0
0.0
0.0012008
0.0003383
0.0
0.0024186
0.0007949
0.0160338
0.0054123
0.0013362
0.0
0.0042960
0.0070698
0.0211924
0.2833830
0.0931586
0.0173531
0.0144778
0.0096068
0.0035180
0.0000677
0.0003214
0.0005412
0.0182833
0.0187907
0.0011839
0.0259112
0.0111797
0.0248288
0.0072727
0.0691248
0.0008457
0.0054968
COL 25
0.0
0.0065186
0.1005513
0.0008878
0.0
0.0000851
0.0003527
0.0001824
0.0003040
0.0
0.0
0.0043538
0.0021161
0.0000122
0.0042565
0.0007419
0.0294552
0.0024201
0.0015080
0.0
0.0053997
0.0165275
0.0985325
0.0393303
0.1307119
0.0216475
0.0156275
0.0232285
0.0
0.0011918
0.0006446
0.0018607
0.0200300
0.0135722
0.0006324
0.0079171
0.0033323
0.0458853
0.0061902
0.1633776
0.0006081
0.0040741
COL 26
0.0
0.0
0.0001762
0.0002715
0.0
0.0001155
0.0000116
0.0006614
0.0002224
0.0000116
0.0
0.0019465
0.0054324
0.0017993
0.0091580
0.0013632
0.0102642
0.0048346
0.0050310
0.0002310
0.0077949
0.2379037
0.0149803
0.0350379
0.0137096
0.0694693
0.0403952
0.0134525
0.0076389
0.0044967
0.0052794
0.0021025
0.0166179
0.0233066
0.0009444
0.0052851
0.0012765
0.0350523
0.0071624
0.0055335
0.001 1726
0.0093631
COL 27
0.0001125
0.0000792
0.0
0.0003084
0.0
0.0003667
0.0
0.0010918
0.0005000
0.0000458
0.0
0.0012522
0.0024315
0.0003438
0.0037837
0.0006938
0.0036003
0.0040045
0.0099384
0.0004417
0.0064173
0.0916253
0.0073757
0.0097905
0.0122574
0.0416623
0.1472430
0.0447334
0.0147222
0.0085987
0.0032399
0.0020981
0.0112573
0.0304029
0.0008813
0.0043004
0.0004084
0.0399621
0.0064756
0.0103281
0.0011793

-------

ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW

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
33
34
35
36
37
38
39
40
41
42

0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
COL 28
0
0005931
0003239
0002691
0
0000100
0
0010142
0106875
0
0
0016272
0022925
0098877
0128754
0009195
0149386
0018614
0155267
0003763
0164263
0406296
0076325
0102415
0353718
0445717
0285715
1659720
0032693
0049364
0115921
0017194
0131246
0397724
0010466
0046398
0004136
0398247
0041514
0061499
0026488
0187686
COL 29
0.0
0.0
0.0
0.0006382
0.0
0.0000041
0.0
0.0028133
0.0002442
0.0000021
0.0
0.0109125
0.0005520
0.0002319
0.0043584
0.0004412
0.0069973
0.0018776
0.0250671
0.0002873
0.0122258
0.0836145
0.0032298
0.0052121
0.0025014
0.0626678
0.0296923
0.0222128
0.2960653
0.0011676
0.0043194
0.0007244
0.0184228
0.0343195
0.0009131
0.0033283
0.0006320
0.0293537
0.0037141
0.0329221
0.0017073
0.0041717
COL 30
0.0
0.0
0.0
0.0003402
0.0
0.0000187
0.0
0.0013671
0.0389611
0.0
0.0
0.0019507
0.0085425
0.0043602
0.0012984
0.0006430
0.0070101
0.0028714
0.0067292
0.0001405
0.0063234
0.0572197
0.0021286
0.0142979
0.0102248
0.0429187
0.0539175
0.0394886
0.0087673
0.1609098
0.0131774
0.0018227
0.0114389
0.0159240
0.0009239
0.0043009
0.0006710
0.0272132
0.0035456
0.0057741
0.0010549
0.0064233
COL 31
0.0007061
0.0
0.0006012
0.0004962
0.0
0.0000191
0.0
0.0002004
0.0192092
0.0019658
0.0
0.0091322
0.0007920
0.0028246
0.0202779
0.0004008
0.0218715
0.0016318
0.0092754
0.0013550
0.0133023
0.0148291
0.0064031
0.0093517
0.0221101
0.0306220
0.0310801
0.0561197
0.0094376
0.0119568
0.0617689
0.0040365
0.0116228
0.0420826
0.0009256
0.0030918
0.0002863
0.0470639
0.0051625
0.0232170
0.0014696
0.0189038
COL 32
0.0020368
0.0
0.0
0.0001419
0.0
0.0001013
0.0
0.0027968
0.0000912
0.0013984
0.0001115
0.0260324
0.0203071
0.0011755
0.0578609
0.0046106
0.0379693
0.0027866
0.0320718
0.0131834
0.0076607
0.0273801
0.0078533
0.0105487
0.0343315
0.0307139
0.0076709
0.0142372
0.0010741
0.0026752
0.0012971
0.0578103
0.0158788
0.0448904
0.0010133
0.0045802
0.0
0.0657243
0.0083599
0.0528450
0.0019659
0.0132948
COL 33
0.0010322
0.0
0.0
0.0008328
0.0
0.0000369
0.0000240
0.0358942
0.0
0.0029398
0.0
0.0013055
0.0007867
0.0
0.0012206
0.0019722
0.0025280
0.0429113
0.0068620
0.0001163
0.0002844
0.0011320
0.0000979
0.0002604
0.0012446
0.0018411
0.0044078
0.0042693
0.0027311
0.0136390
0.0008107
0.0014090
0.0628583
0.0762683
0.0009048
0.0032500
0.0003305
0.0289916
0.0201520
0.0319166
0.0252726
0.0054087
COL 34
0.0177620
0.0000429
0.0000386
0.0004623
0.0008692
0.0000530
0.0000111
0.0500873
0.0014637
0.0006002
0.0000159
0.0044165
0.0002512
0.0008437
0.0467155
0,0015721
0.0052424
0.0069461
0.0009150
0.0001919
0.0017428
0.0001644
0.0000265
0.0000371
0.0004835
0.0016594
0.0098879
0.0067870
0.0094641
0.0026149
0.0052284
0.0049811
0.0056411
0.1005907
0.0013629
0.0055176
0.0010274
0.0224413
0.0232382
0.0015774
0.0199816
0.0080969
COL 35
0.0004350
0.0
0.0
0.0000099
0.0
0.0
0.0
0.0305973
0.0
0.0079045
0.0000074
0.0048149
0.0001681
0.0038756
0.0141060
0.0002101
0.0249791
0.0027090
0.0
0.0001285
0.0003164
0.0
0.0
0.0
0.0
0.0009244
0.0001409
0.0009615
0.0000840
0.0009442
0.0140566
0.0012902
0.0054254
0.1178879
0.0133126
0.0125711
0.0045158
0.0189753
0.0118864
0.0002175
0.0013273
0.0237877
COL 36
0.0
0.0
0.0002641
0.0286448
0.0000393
0.0000225
0.0000056
0.0334714
0.0000225
0.0000393
0.0000112
0.0001854
0.0002697
0.0000056
0.0012080
0.0002641
0.0011743
0.0162778
0.0004551
0.0000056
0.0013710
0.0020677
0.0002023
0.0005675
0.0000225
0.0078046
0.0006686
0.0008653
0.0000730
0.0001011
0.0000056
0.0003034
0.0337636
0.0202503
0.0007810
0.0606216
0.0277907
0.0152102
0.0056183
0.0012530
0.1388021

-------

ROW 1
ROW 2
KOW 3
ROW 4
ROW 5
ROW 6
ROW 7
ROW 8
ROW 9
ROW 10
ROW 11
ROW 12
ROW 13
ROW 14
ROW 15
ROW 16
ROW 17
ROW 18
ROW 19
ROW 20
ROW 21
ROW 22
ROW 23
ROW 24
ROW 25
ROW 26
ROW 27
ROW 28
ROW 29
ROW 30
ROW 31
ROW 32
ROW 33
ROW 34
ROW 35
ROW 36
ROW 37
ROW 38
ROW 39
ROW 40
ROW 41
ROW 42
COL 37
0.0
0.0
0.0
0.0005441
0.1902156
0.0
0.0
0.0233889
0.0
0.0
0.0
0.0
0.0
0.0
0.0001360
0.0001280
0.0000160
0.0048650
0.0001280
0.0000080
0.0004161
0.0004401
0.0
0.0
0.0
0.0028326
0.0001680
0.0002721
0.0
0.0000160
0.0
0.0001440
0.0010722
0.0105782
0.0002721
0.0000160
0.3569863
0.0044329
0.0020004
0.0027926
0.0170996
0.0013363
COL 38
0.0015527
0.0
0.0
0.0000498
0.0000027
0.0000375
0.0
0.0081364
0.0000942
0.0062929
0.0000321
0.0013842
0.0016749
0.0002716
0.0083254
0.0022516
0.0025915
0.0077787
0.0023684
0.0002321
0.0025185
0.0000792
0.0000212
0.0000416
0.0001003
0.0022701
0.0025977
0.0019097
0.0023547
0.0008416
0.0006580
0.0012449
0.0041873
0.1273597
0.0010013
0.0150497
0.0025499
0.0162250
0.0165696
0.0003044
0.0144081
0.0189940
COL 39
0.0
0.0
0.0
0.0002397
0.0
0.0
0.0
0.0044963
0.0
0.0
0.0
0.0020231
0.0
0.0
0.0047865
0.0134762
0.0004332
0.0034910
0.0018486
0.0000421
0.0
0.0
0.0
0.0
0.0
0.0
0.0001746
0.0000736
0.0
0.0003680
0.0
C. 0008517
0.0088811
0.1348231
0.0052765
0.0032261
0.0007928
C.C096529
0.2022780
0.0028097
0.0151229
0.0188242
COL 40 COL 41
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0111706
0.0000336
0.0
0.0025558
0.0004505
0.0000069
0.0
0.0223111
0.0
0.0053317
0.0
0.0001239
0.0000012
0.0005246
0.0009705
0.0004192
0.0008882
0.0001123
0.0
0.0000151
0.0002652
0.0000440
0.0
0.0
0.0
0.0004922
0.0000359
0.0000359
0.0004354
0.0
0.0
0.0000058
0.0151381
0.0061342
0.0000046
0.0067075
0.0012739
0.0019536
0.0009276
0.0033074
0.0002223
0.0018622
COL 42
0.0125010
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0005118
0.2218103
0.0161653
0.0007438
0.0002047
0.0425865
0.0874385
0.0011600
0.0035551
0.0013647
0.0
0.0028455
0.0005391
0.0002934
0.0
0.0
0.0
0.0009144
0.0007506
0.0068783
0.0
0.0
0.0023269
0.0289938
0.2994363
0.1188616
0.0041147
0.0
0.0
0.0406213
0.0010167
0.1043680
0.0

-------
                     THE LEONTIEF INVERSE MATRIX OF THE INPUT/OUTPUT TABLE ( 42 SECTORS  )
DETERMINANTS. 0012276925

ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW

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
33
34
35
36
37
38
39
40
41
42

1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
COL 1
.5044756
.0007345
.0007447
.0017042
.0199930
.0028480
.0019132
.0300775
.0004494
.1258757
.0001632
.0105866
.0087251
.0008705
.0220194
.0022061
.0569692
.0364245
.0079898
.0004072
.0070416
.0096064
.0013008
.0017623
.0025341
.0161414
.0138565
.0055938
.0056236
.0025892
.0015779
.0018834
.0450953
.1511621
.0055629
.0092303
.0038912
.0715984
.0287046
.0464898
.0078832
.0077860
COL 2
0.0083713
1.0738344
0.0406267
0.0081650
0.0121228
0.0006251
0.0005629
0.0136298
0.0006007
0.0049675
0.0002013
0.0032771
0.0116229
0.0008535
0.0137504
0.0013887
0.0276031
0.0211718
0.0036557
0.0001981
0.0046322
0.0353789
0.0019402
0.0021781
0.0053378
0.0106710
0.0386597
0.0087203
0.0044029
0.0061959
0.0017166
0.0015978
0.1287200
0.1256626
0.0015820
0.0205304
0.0046768
0.0389340
0.0179324
0.3412825
0.0107569
0.0098421
COL 3
0.0059906
0.0150318
1.1829691
0.0046175
0.0106211
0.0009801
0.0020400
0.0093633
0.0003812
0.0050108
0.0001932
0.0055348
0.0043449
0.0007533
0.0125054
0.0016070
0.0552313
0.0149342
0.0056759
0.0001820
0.0102453
0.0644230
0.0027596
0.0026665
0.0099489
0.0109119
0.0411521
0.0102392
0.0039457
0.0020551
0.0015772
0.0013052
0.0562844
0.0764168
0.0015807
0.0225635
0.0137255
0.0446320
0.0246478
0.2058599
0.0089575
0.0094748
COL 4
0.0076513
0.0021121
0.0025958
1.2253637
0.0114803
0.0012816
0.0007828
0.0081456
0.0006578
0.0050696
0.0002134
0.0056044
0.0154670
0.0009043
0.0162194
0.0016255
0.0392294
0.0203316
0.0129469
0.0002647
0.0071311
0.0332212
0.0043599
0.0040981
0.0132567
0.0258527
0.0782850
0.0122211
0.0072367
0.0081806
0.0014766
0.0030129
0.0251693
0.0690820
0.0018958
0.0337646
0.0038237
0.0569393
0.0219854
0.0160336
0.0099369
0.0104448
COL 5
0.0086419
0.0005819
0.0006685
0.0010188
1.0335855
0.0004306
0.0002795
0.0131649
0.0005377
0.0045486
0.0002203
0.0038056
0.0037002
0.0009132
0.0180575
0.0012825
0.0134146
0.0105010
0.0053227
0.0002117
0.0032348
0.0088774
0.0012484
0.0015353
0.0029012
0.0123561
0.0225795
0.0092167
0.0050045
0.0018622
0.0017222
0.0018303
0. 0380905
0.2021729
0.0015677
0.0081753
0.0042537
0.0259726
0.0215468
0.0925642
0.0076138
0.0107470
COL 6
0.0067015
0.0025293
0.0020880
0.0054629
0.0211999
1.0111170
0.0012801
0.0097217
0.0004459
0.0056718
0.0002431
0.0063580
0.0048789
0.0009509
0.0304192
0.0019549
0.0340308
0.0371648
0.0235299
0.0003107
0.0777666
0.0355744
0.0025022
0.0030334
0.0052027
0.0130378
0.0899910
0.0106030
0.0094971
0.0025507
0.0016956
0.0020862
0.0382642
0.0819654
0.0017892
0.0291223
0.0085313
0.0814304
0.0229149
0.0767699
0.0106914
0.0118742
COL 7
0.0060722
0.0029827
0.0022526
0.0038142
0.0196298
0.0169745
1.0693502
0.0096666
0.0003991
0.0063491
0.0002765
0.0038106
0.0037355
0.0009342
0.0193404
0.0012334
0.0536638
0.0194659
0.0080865
0.0002122
0.0054265
0.0362080
0.0024927
0.0026739
0.0071203
0.0109666
0.0505215
0.0111588
0.0061187
0.0028916
0.0013690
0.0016092
0.0839990
0.0583800
0.0015870
0.0279521
0.0373156
0.0433877
0.0157917
0.1118456
0.0103176
0.0136514
COL 8
0.0252139
0.0056614
0.0067590
0.0046237
0.0181908
0.0148954
0.0012303
1.0118361
0.0010066
0.0089489
0.0002958
0.0069759
0.0767707
0.0084283
0.0324601
0.0026589
0.0520707
0.0318599
0.0119609
0.0004302
0.0878886
0.0914516
0.0155797
0.0110019
0.0293244
0.1277483
0.0402830
0.0399607
0.0053223
0.0036265
0.0060491
0.0045870
0.0662074
0.1158310
0.0023001
0.0120068
0.0070573
0.1233608
0.0250815
0.0346334
0.0093855
0.0143888
COL 9
0.0112770
0.0054838
0.0121006
0.0039565
0.0076194
0.0013408
0.0007501
0.0106925
1.0314569
0.0118704
0.0006780
0.0122457
0.0107600
0.0043829
0.0288629
0.0042943
0.0327976
0.0120020
0.0342861
0.0007956
0.0154592
0.0852984
0.0246626
0.0373659
0.0548241
0.0583279
0.1656926
0.1156542
0.0165225
0.2653965
0.0333077
0.0064593
0.047225?
0.080R800
0.0025415
0.0129247
0.0063227
0.0737397
0.0211921
0.0423067
0.0086183

-------

ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW

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
33
34
35
36
37
38
39
40
41
42
COL 10
0.5678055
0.0012012
0.0010727
0.0030720
0.0145968
0.0021423
0.0013139
0.0245993
0.0005081
1.2583876
0.0002991
0.0123989
0.0106423
0.0012003
0.0531135
0.0048173
0.0405034
0.0252257
0.0078510
0.0003716
0.0171186
0.0183495
0.0021265
0.0040551
0.0034628
0.0464972
0.0115589
0.0063834
0.0047204
0.0028653
0.0013351
0.0028001
0.0786241
0.1643877
0.0038560
0.0116158
0.0062454
0.0832513
0.0264005
0.0714816
0.0104300
0.0140039
COL 11
0.3735555
0.0005210
0.0006638
0.0017312
0.0080619
0.0010433
0.0012373
0.0134555
0.0003217
0.0440034
1.2359066
0.0056276
0.0094225
0.0006501
0.0570177
0.0043915
0.0544195
0.0140859
0.0057880
0.0003080
0.0037438
0.0062321
0.0009704
0.0033061
0.0019489
0.0118828
0.0066916
0.0034211
0.0029032
0.0015276
0.0011683
0.0028552
0.0356039
0.1184679
0.0030104
0.0056048
0.0027196
0.0446639
0.0153997
0.0240891
0.0082021
0.0071643
COL 12
0.1224919
0.0008790
0.0010536
0.0034480
0.0091070
0.0009409
0.0026521
0.0113538
0.0003351
0.0207081
0.0003440
1.7276945
0.0062111
0.0022537
0.0433977
0.0038274
0.1430452
0.0151457
0.0142837
0.0052224
0.0066463
0.0084072
0.0014175
0.0020664
0.0033223
0.0117046
0.0115080
0.0041908
0.0025994
0.0018757
O.C018934
0.0206145
0.0502056
0.0944411
0.0029888
0.0153499
O.OC48199
0.0884924
0.0241046
0.0520657
0.0103108
0.0167666
COL 13
0.2410139
0.0009759
0.0009116
0.0019852
0.0134804
0.0012218
0.0010415
0.0145099
0.0004008
0.0257287
0.0003000
0.0089355
1.3711281
0.0044672
0.0326338
0.0061027
0.0469749
0.0241988
0.0107704
0.0004222
0.0104100
0.0143353
0.0014609
0.0036336
0.0028808
0.0229468
0.0124981
0.0060140
0.0029870
0.0035766
0,0012631
0.0035693
0.0835018
0.0887179
0.0027046
0.0121179
0.0035917
0.0784793
0.0211984
0.1042699
0.0089831
0.0147091
COL 14
0.0426248
0.0054904
0.0028374
0.0050096
0.0086112
0.0022629
0.0012883
0.0108434
0.0005386
0.0172709
0.0003907
0.1005805
0.1524943
1.0337019
0.0530885
0.0033572
0.0598559
0.0139275
0.0375090
0.0031777
0.0340861
0.0926486
0.0043896
0.0188757
0.0083581
0.0840557
0.0242213
0.0105696
0.0050088
0.0034878
O.C040552
0.0115274
0.0566008
0.0925626
0.0024205
0.0141696
0.0060369
0.0931891
0.0197854
0.0364417
0.0090926
0.0191565
COL 15
0.0291225
0.0010015
0.0010899
0.0117860
0.0158130
0.0035623
0.0032330
0.0153606
0.0005023
0.0177633
0.0003342
0.0251526
0.0993707
0.0015036
1.4142389
0.0146730
0.0815269
0.0253792
0.0223761
0.0005949
0.0118492
0.0130837
0.0016914
0.0040315
0.0035545
0.0261674
0.0159329
0.0067524
0.0026097
0.0024007
0.0018288
0.0032480
0.0767000
0.0840295
0.0023467
0.0186365
0.0121719
0.0761216
0.0210801
0.1116329
0.0114692
0.0157212
COL 16
0.0165404
0.0006236
0.0008421
0.0033282
0.0069672
0.0011256
0.0011640
0.0168315
0.0011056
0.0144462
0.0007150
0.0106139
0.0234353
0.0025283
0.3052018
1.1347790
0.0436987
0.0114344
0.0077394
0.0005265
0.0051664
0.0082794
0.0013360
0.0035655
0.0029783
0.0131360
0.0125630
0.0063740
0.0034555
0.0038885
0.0048481
0.0059102
0.0499144
0.1595440
0.0024009
0.0115576
0.0040589
0.0565127
0.0251282
0.0354122
0.0161989
0.0353075
COL 17
0.0284836
0.0048957
0.0054359
0.0093752
0.0347586
0.0032484
0.0234345
0.0157274
0.0005684
0.0392487
0.0004684
0.0119092
0.0117136
0.0015488
0.0781650
0.0061029
1.3715668
0.0567939
0.0175367
0.0004058
0.0202205
0.0233657
0.0039842
0.0074500
0.0114449
0.0375579
0.0225131
0.0077802
0.0039460
0.0026972
0.0042854
0.0039166
0.0715456
0.1546938
0.0024730
0.0189643
0.0172948
0.0630493
0.0274112
0.0522261
0.0142367
0.0230704
COL 18
0.0095524
0.0013018
0.0010204
0.0025527
0.5740113
0.0043219
0.0012074
0.0159502
0.0005239
0.0070663
0.0002327
0.0045483
0.0048168
0.0009781
0.0268452
0.0017449
0.0620990
1.0905399
0.0054260
0.0002378
0.0071430
0.0147172
0.0018366
0.0027409
0.0035089
0.0324068
0.0173410
0.0081464
0.0044692
0.0027070
0.0018878
0.0023175
0.0905425
0.1726167
0.0023639
0.0126992
0.0210899
0.0364505
0.0266943
0.0936819
0.0119678

-------

ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
SOW
ROW
ROW
ROM
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
«*OW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW

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
33
34
35
36
37
38
39
40
41
42

0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
COL 19
.0223425
.0018206
.0021206
.0054482
.0119825
.0022313
.0063147
.0112317
.0005979
.0146502
.0003914
.1602871
.0087157
.0016471
.0461808
.0059069
.2843417
.0193402
.0404663
.0035615
.0192489
.0179386
.0024810
.0039157
.0072796
.0299066
.0165031
.0096333
.0029216
.0054794
.0037044
.0082145
.0540078
.1057435
.0023431
.0176117
.0074336
.0649900
.0206498
.0626673
.0100804
.0192578
COL 20
0.0188650
0.0007525
0.0008882
0.0029621
0.0061575
0.0009407
0.0017779
0.0090769
0.0004377
0.0101055
0.0003425
0.0998992
0.0229461
0.0015841
0.0548501
0.0090156
0.0736083
0.0101131
0.0782984
1.3635445
0.0109548
0.0092054
0.0014145
0.0021095
0.0031438
0.0193668
0.0073412
0.0061564
0.0024434
0.0019849
0.0042721
0.0074706
0.0427476
0.1000878
0.0026144
0.0111067
0.0035549
0.0686827
0.0226032
0.0374972
0.0111618
0.0168159
COL 21
0.0117189
0.0021264
0.0024613
0.0125403
0.0184544
0.0525192
0.0041398
0.0109037
0.0004157
0.0095061
0.0003487
0.0099237
0.0191134
0.0017976
0.0807824
0.0042670
0.0732394
0.0229300
0.0143141
0.0004610
1.1289406
0.0161931
0.0019882
0.0041799
0.0045692
0.0239958
0.0161076
0.0104448
0.0030260
0.0024800
0.0020480
0.0040274
0.0816974
0.0810924
0.0020833
O.C228921
0.0322899
0.0601478
0.0237265
0.0392945
0.0121352
0.0171637
COL 22
0.0069106
0.0724767
0.0072330
0.0363077
0.0144572
0.0049137
0.0011315
0.0170422
0.0005840
0.0060425
0.0002380
0.0056051
0.0070194
0.0011199
0.0187835
0.0031578
0.0325852
0.0193431
0.0072952
0.0002454
0.0255263
1.2825737
0.0078258
0.0082618
0.0216064
0.0445675
0.0427287
0.0133558
0.0063829
0.0043449
0.0018449
0.0022379
0.0882140
0.0708023
0.0020442
0.0225186
0.0217308
0.0634040
0.0207494
0.0998092
0.0107982
0.0116729
COL 23
0.0069436
0.0060835
0.2310888
0.0053929
0.0119682
0.0024824
0.0009755
C. 0095584
0.0005228
0.0067267
0.0002843
0.0062757
0.0045347
0.0010876
C.0196051
0.0027647
0.0370188
0.0159998
0.0059920
0.0002576
0.0115076
0.0477176
1.4039736
0.0482170
0.1565312
0.0407012
0.0382534
0.0163322
0.0050835
0.0025497
0.0016900
0.0029182
0.0607491
0.0736406
0.0021774
0.0220322
0.0178954
0.0855585
0.0240640
0.2679546
0.0099087
0.0139411
COL 24
0.0071526
0.0044265
0.0858477
0.0067497
0.0154365
0.0014815
0.0011487
0.0105084
0.0006529
0.0067172
0.0003045
0.0077217
0.0048972
0.0012509
0.0198830
0.0029242
0.0501109
0.0176709
0.0059872
0.0002655
0.0128257
0.0412353
0.0651284
1.4095707
0.1621291
0.0411164
0.0382958
0.0268053
0.0102499
0.0026423
0.0021979
0.0027863
0.0534888
0.0739221
0.0027077
0.0465884
0.0310906
0.0648118
0.0241994
0.1600282
0.0131533
0.0149775
COL 25
0.0079437
0.0132201
0.1685048
0.0050512
0.0101900
0.0014108
0.0018321
0.0089249
0.0011897
0.0069181
0.0002737
0.0131164
0.0077463
0.0013483
0.0228430
0.0028258
0.0671172
0.0140047
0.0065505
0.0003142
0.0139338
0.0562724
0.1647955
0.0730444
1.1810884
0.0427868
0.0393138
0.0406493
0.0039569
0.0042903
0.0026876
0.0041769
0.0553068
0.0701323
0.0019299
0.0212299
0.0139658
0.0860575
0.0224541
0.2641056
C. 0093544
0.0134038
COL 26
0.0093724
0.0198867
0.0124152
0.0113559
0.0109919
0.0023325
0.0008780
0.0112885
0.0012852
0.0078740
0.0003791
0.0093818
0.0137383
0.0035144
0.0308720
0.0037845
0.0355194
0.0165541
0.0110457
0.0007111
0.0199716
0.3447180
0.0313340
0.0594881
0.0349929
1.0977659
0.0696896
0.0286980
0.0165357
0.0090868
0.0079658
0.0045906
0.0587681
0.0778509
0.0023112
0.0175351
0.0114080
0.0730790
0.0225024
0.0569022
0.0095501
0.0186893
COL 27
0.0091822
0.0099960
0.0079037
0.0063035
0.0081723
0.0019377
0.0005899
0.0101789
0.0022987
0.0085746
0.0004608
0.0089806
0.0088206
0.0026108
0.0233383
0.0026879
0.0240458
0.0130641
0.0172239
0.0011168
0.0165672
0.1687656
0.0197381
0.0241722
0.0289604
0.0698200
1.1897917
0.0711396
0.0284347
0.0151104
0.0066631
0.0048545
0.0434669
0.0827346
0.0021178
0.0132300
0.0063565
0.0748756
0.0199640
0.0451855
0.0084348

-------

ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW
ROW

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
33
34
35
36
37
38
39
40
41
42
COL 28
0.0118712
0.0069692
0.0128505
0.0048314
0.0075485
0.0020218
0.0010295
0.0110062
0.0143199
0.0116801
0.0006256
0.0124838
0.0118823
0.0142745
0.0427327
0.0033030
0.0465654
0.0117642
0.0247532
0.0011388
0.0293977
0.0978632
0.0246436
0.0273250
0.0602185
0.0740098
0.0565677
1.2116385
0.0104115
0.0135401
0.0172976
0.0051057
0.0486079
0.0999770
0.0024338
0.0141620
0.0064219
0.0789868
0.0180740
0.0462436
0.0108564
0.0309877
COL 29
0.0103308
0.0114705
0.0054321
0.0081049
0.0086241
0.0022683
0.0009581
0.0148686
0.0015542
0.0071698
0.0003123
0.0368958
0.0071903
0.0021218
0.0287439
0.0028422
0.0413302
0.0132551
0.0424011
0.0010996
0.0289198
0.1979286
0.0135014
0.0199943
0.0164358
0.1141499
0.0679085
0.0488436
1.4262447
0.0059424
0.0092721
0.0038334
0.0593245
0.1030271
0.0025381
0.0142746
0.0079583
0.0755836
0.0195192
0.0799911
0.0106569
0.0153269
COL 30
C. 0096340
0.0078093
0.0073294
0.0053812
0.0075493
0.0014598
0.0006669
0.0094258
0.0493466
0.0071350
0.0003516
0.0106795
0.0194556
0.0074275
0.0188515
0.0026335
0.0297957
0.0118349
0.0154526
0.0006952
0.0169832
0.1312855
0.0131483
0.0343002
0.0295789
0.0762557
0.0980608
0.0734903
C. 0206787
1.2076855
0.0207544
0.0046945
0.0420057
0.0652112
0.0022208
0.0133691
0.0065022
0.0641669
0.0159144
0.0380248
0.0078571
0.0173552
COL 31
0.0139856
0.0036166
0.0095093
0.0037570
0.0066544
0.0016105
0.0010748
0.0093025
0.0230140
0.0139610
0.0005926
0.0240882
0.0085552
0.0057446
0.0509659
0.0025657
0.0516336
0.0105709
0.0169226
0.0025267
0.0230843
0.0540121
0.0187482
0.0228705
0.0395160
0.0531676
0.0561446
0.0834436
0.0190050
0.0237532
1.0693836
0.0074282
0.0416379
0.0968558
0.0021462
0.0109393
0.0050270
0.0802597
0.0180179
0.0558505
0.0086569
0.0293155
COL 32
0.0239006
0.0041050
0.0105488
0.0041996
0.0090076
0.0016340
0.0017762
0.0137326
0.0009319
0.0138532
0.0006252
0.0600355
0.0405369
0.0031494
0.1094448
0.0085861
0.0866556
0.0147706
0.0419946
0.0196524
0.0163883
0.0603122
0.0210783
0.0230793
0.0510627
0.0498473
0.0222847
0.0258373
0.0052907
0.0059933
0.0035927
1.0645638
0.0503054
0.1091086
0.0024083
0.0142195
0.0055749
0.1037990
0.0244275
0.0969812
0.0107275
0.0235706
COL 33
0.0112794
0.0008098
0.0010287
0.0022533
0.0291162
0.0011094
0.0003556
0.0472284
0.0011263
0.0087762
0.0002166
0.0062544
0.0066447
0.0011930
0.0162858
0.0037231
0.0152599
0.0543040
0.0099049
0.0004016
0.0062373
0.0123745
0.0019249
0.0024014
0.0046662
0.0135373
0.0126793
0.0110376
0.0068965
0.0187440
0.0025382
0.0031554
1.0827579
0.1228515
0.0017699
0.0074882
0.0031075
0.0473876
0.0348267
0.0461963
0.0325729
0.0105861
COL 34
0.0374613
0.0009534
0.0011333
0.0023955
0.0096402
0.0014592
0.0005416
0.0611715
0.0022721
0.0092724
0.0003058
0.0125521
0.0116514
0.0023593
0.0827535
0.0039283
0.0213339
0.0148298
0.0050735
0.0006282
0.0096743
0.0138802
0.0021851
0.0024204
0.0045747
0.0159729
0.0198740
0.0149932
0.0166907
0.0052455
0.0074052
0.0072818
0.0248553
1.1472855
0.0023945
0.0104681
0.0042569
0.0456011
0.0386974
0.0176855
0.0269010
0.0139812
COL 35
0.0173555
0.0006186
0.0007785
0.0016430
0.0072554
0.0009418
0.0008351
0.0422655
0.0007911
0.0207360
0.0005976
0.0124764
0.0075534
0.0059718
0.0401551
0.0017999
0.0439604
0.0098973
0.0029095
0.0005314
0.0062436
0.0082014
0.0014085
0.0016641
0.0029014
0.0113198
0.0067151
0.0067137
0.0032155
0.0028140
0.0167602
0.0038264
0.0262571
0.1635222
1.0143499
0.0172519
0.0096433
0.0383261
0.0241195
0.0132968
0.0092014
C. 0291467
COL 36
0.0072243
0.0007666
0.0010585
0.0385862
0.0223180
0.0009753
0.0002068
0.0453490
0.0002534
0=0043872
0.0001677
0.0020366
0.0052479
0.0008938
0.0093450
0.0011494
0.0090272
0.0245145
0.0024688
0.0001303
0.0066329
0.0123091
0.0016563
0.0023064
0.0026761
0.0176733
0.0075332
0.0047587
0.0017974
0.0016680
0.0008163
0.0013030
0.0509868
0.0501166
0.0013505
1.0687637
0.0478263
0.0298846
0.0135649
0.0110276
0.1524147

-------

ROW 1
ROW 2
ROW 3
ROW 4
ROW 5
ROW 6
ROW 7
ROW 8
ROW 9
ROW 10
ROW 11
ROW 12
ROW 13
ROW 14
ROW 1.5
ROW 16
ROW 17
ROW 18
ROW 19
ROW 20
ROW 21
ROW 22
ROW 23
ROW 24
ROW 25
ROW 26
ROW 27
ROW 28
ROW 29
ROW 30
ROW 31
ROW 32
ROW 33
ROW 34
ROW 35
ROW 36
ROW 37
ROW 38
ROW 39
ROW 40
ROW 41
ROW 42
COL 37
0.0053363
0.0005657
0.0005597
0.0017805
0.3112421
0.0008016
0.0001655
0.0429335
0.0002654
0.0029575
0.0001353
0.0018788
0.0044356
0.0007896
0.0095150
0.0009407
0.0074405
0.0132548
0.0025522
0.0001354
0.0053988
0.0090410
0.0011864
0.0012510
0.0023201
0.0140719
0.0095424
0.0053351
0.0021920
0.0009695
0.0009673
0.0012399
0.0185812
0.0878640
0.0011100
0.0037179
1.5571680
0.0214603
0.0126469
0.0351312
0.0302036
0.0066253
COL 38
0.0162305
0.0004118
0.0004600
0.0014736
0.0082723
0.0006817
0.0002678
0.0194045
0.0005706
0.0161312
0.0005065
0.0058003
0.0065064
0.0017274
0.0293644
0.0039113
0.0113272
0.0131940
0.0044131
0.0005820
0.0060291
0.0061440
0.0009020
0.0011526
0.0017821
0.0086291
0.0078348
0.0060056
0.0061928
0.0023090
0.0021095
0.0033114
0.0203835
0.1656057
0.0017799
0.0190590
0.0061704
1.0300436
0.0286646
0.0103655
0.0221482
0.0230862
COL 39
0.0121505
0.0002912
0.0003483
0.0013299
0.0058625
0.0004804
0.0002193
0.0190405
0.0005057
0.0095735
0.0005637
0.0079926
0.0043403
0.0016880
0.0330240
0.0202463
0.0088747
0.0096522
C. 0042070
0.0003717
0.0029604
0.0042250
0.0006665
0.0008062
0.0013981
0.0053038
0.0050987
0.0038771
0.0033364
0.0020476
0.0017204
0.0035271
0.0282086
0.2099456
0.0074163
0.0074697
0.0031252
0.0263918
1.2627697
0.0133718
0.0258499
0.0279068
COL 40
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
1.0000000
0.0
0.0
COL 41
0.0213432
0.0002380
0.0002256
0.0036311
0.0024369
0.0004493
0.0000986
0.0246993
0.0000735
0.0092151
0.0000581
0.0009269
0.0023983
0.0008866
0.0041787
0.0007442
0.0040851
0.0027291
0.0007614
0.0000638
0.0027684
0.0031686
0.0004801
0.0004413
0.0009217
0.0045794
0.0019226
0.0014779
0.0011191
0.0005144
0.0002931
0.0003533
0.0204232
0.0162152
0.0002292
0.0080133
0.0027026
0.0079115
0.0033399
0.0066084
1.0024595
0.0028643
COL 42
0.1648893
0.0012300
0.0014076
0.0032523
0.0169084
0.0015946
0.0010171
0.0309133
0.0015405
0.2889343
0.0202470
0.0166590
0.0232192
0.0454040
0.1595598
0.0060204
0.0364149
0.0300317
0.0110452
0.0049730
0.0112361
0.0185895
0.0026250
0.0042418
0.0056629
0.0263813
0.0141588
0.0174380
0.0062742
0.0079004
0.0053546
0-0344236
0.3585163
0.2379844
0.0064732
0.0100850
0.0048984
0.0969115
0.0279144
0.1528694
0.0184115
1.0126419

-------
Distribution by Average Consumption by Sector

SECTOR- 1
SECTOR- 2
SECTOR- 3
SEC TQ"- 4
SECTOR- 5
SECTOR- 6
SECTOR- 7
SECTOR- 8
SECTOR- 9
SECTQR-IO
SSCTTR-11
SECTOR-12
SECTOR-13
SSCTC;?-14
SEC TOP- 15
SECTOR-16
SECTOR-17
SECTOR-IB
SEC TOR- IQ
SECTOR-20
SECTOR-?!
SECTHR-22
SECTOR-23
SECTQR-24
SECTnR-25
SECTQP-26
SECTOR-27
SECTOR-28
SECTOR-29
SECTOR-30
SECTOR-31
SECTOR-32
SECTOR-33
SECTOR-34
SECTOP.-35
SECTQR-36
SECTOR-37
SECTOR-38
SECTOR-39
SECTOR-40
SECTOR-41
SECTOR-42
PERCENT
0.011847
0.0
0.0
0.000704
0.0
0.000052
0.000004
0.0
C.OOC595
0. 154956
0.014550
0.0479BO
0.000510
O.OOH813
0.002897
0.008361
0,014029
0.0??549
0. 004912
C. 00394 7
0.001244
C. 000059
C.O
C.OOC020
0.000017
0.002557
C. 00 1701
0.016406
0.037152
0.003175
C. 003129
0.009590
0.026651
0.225173
0.080286
0.015634
0.009600
0. 199479
0.049824
0.013115
0.003497
0.000015
                                SECTOR  DEFINITION
                              AGRICULTURAL
                              IRON  MINES
                              NON-FERROUS MINES
                              COAL  MINES
                              CRUDE  PETRO.-NATURAL  GAS
                              STONE  AND CLAY  MINING
                              CHEMICAL  MINING
                              CONSTRUCTION'
                              ORDINANCE (SIC  IS)
                              FOOD  PRODUCTS  (SIC  20)
                              TOBACCO PRODUCTSISIC  21)
                              TEXTILE PRODUCTS  (SIC 22,?31
                              LUMBER-VOHD PRODUCTS!SIC 24)
                              FURNITURE-FIXTURES  (SIC  25)
                              PAPER  PRODUCTS  (SIC  26)
                              PRINTING-PUBLISHING  (sic. 27)
                              CHEMICAL  PRODUCTS  (SIC  28)
                              PETROLEUM REFINING  (SIC  29)
                              RUBBER-PLASTIC  PROD.(SIC 30)
                              LEATHER PRODUCTS  (SIC 31)
                              STONE,CLAY AND  GLASS(SIC 32)
                              IRON  AND  STEFL  MANU.'SIC 33)
                              COPPER MANUFACTURING;STC 3?)
                              ALUMINUM  MAM;FACTURE< sic 3?)
                              OTHER  NGN-FERROUS  ME"f(SIC33)
                              FABRICATED METAL  PROD'SIC34)
                              MACHINERY, EX.ELECT*IC(STC35)
                              ELECTRICAL MACHINERY(SIC 36)
                              MOTOR  VEHICLES  (SIC  371)
                              OTHER  TRANSPORTATIONS SIC 37)
                              INSTRUMENTS  (SIC  3B)
                              MISCELLANEUS  MANUFAC.(S1C 39)
                              TRANSPORTATION-WAREHDUSING
                              OTHER  SERVICES
                              MEDICAL*  EDUC.  &  NGN-PROFIT
                              ELECTRIC  UTILITIES  (SIC  4911
                              GAS UTILITIES  (SIC  49?)
                              WHOLESALE Ł RETAIL  TRADE
                              FINANCE & SERVICES
                              IMPORT OF GOODS AND  SERVICES
                              GOVERNMENT ENTERPRISES
                              MISCELLANEOUS  BUSINESS

-------
3.  Interregional Feedback
    Regional Market Shares
    by AQCR by Industry
      As described in Appendix C, changes in the final demand of the

nation (by sectors) will have a chain reaction on the national economy.

      The next question is how will such structural changes affect

each AQCR.  It is argued that the regional share of the national mar-

ket by each industry is stable. *  Therefore, any changes of production

at national level will have a proportional effect on each region of its

market shares.

      By definition,  regional  market share by industry by AQCR is:

                 ->..   -
      b.. is the regional market share of jth industry product
       IJ from ith AQCR

      X.. is the output (value-added) of jth industry in ith AQCR
       N
      X. is the output (value-added) of jth industry in the nation


A regional market share matrix

                  B =   [b..]
                        1 iJJ

has been estimated for 100 AQCR's and two-digit manufacturing indus

tries  (SIC 20 to SIC39).
      * Sometimes this is called "locational quotient" of the industry.

-------
      CONSAD has developed a computer program,  namely, Program




FEE (Feedback Effects), which uses  B matrix to distribute the change




obtained from the I-O model.

-------
                               B = (bŁj)  REGIONAL MARKET SHARE COEFFICIENT MATRIX
po
o

AQCR
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
33
34
35
36
37
38
39
40
<»1
S 1C
20
.0624
.0832
.0464
.0387
.0158
.0356
.0201
.0 131
.0237
.0060
.0091
.0180
.0043
.0161
.0133
.0121
.0 189
.0205
.0153
.0104
.0083
.0130
.0034
.0093
.0089
.0046
.0108
.0093
.0079
.0037
.0038
.0
.0062
.0
.0044
.0037
.0040
.0009
.0024
.0069
.0033
SIC
22
.0359
.0
.0062
.0259
.0020
.0006
.0053
.0004
.0018
.0
.0028
.0012
.0019
.0008
.0003
.0016
.0012
.0009
.0004
.0004
.0002
-.0001
.0
.0033
.0
.0005
.0001
.0
.0023
.0152
.0
.0
.0
.0
.0009
.0008
.0
.0
.0074
.0008
.0
SIC
22
.17«0
.0161
.02(5
.02C,9
.0070
.0042
.01 17
.OOC9
.0061
.OOU
.0076
. 00 «0
.OOC6
.0026
.OOC8
.0011
.00 14
.0026
.0012
.0061
.0013
.0040
.OOC8
.0062
.OOC4
.0027
. OOC6
.0051
.0020
.0018
.OOC8
.0
.OOC3
.0
.OOC4
.OOC5
.OOC2
.OOC2
.OOC3
.OOC8
.OOC3
SIC
24
.0099
.0091
.0134
.0051
.0040
.0035
.0035
.0015
.0022
.0008
.0012
.0020
.0008
.0054
.0034
.0014
.0014
.0015
.0051
.0012
.0125
.0011
.0006
.0012
.0027
.0009
.0007
.0008
.0119
.0
.0010
.0
.0006
.0
.0012
.0008
.0010
.0
.0009
.0053
.0
SIC
25
.0478
.0468
.0624
.0166
.0070
.0131
.0080
.0026
.0079
.0011
.0115
.0066
.0014
.0041
.0037
.0043
.0031
.0097
.0073
.0057
.0021
.0044
.0015
.0063
.0021
.0056
.0024
.0007
.0049
.0
.0015
.0
.0031
.0
.0024
.0023
.0024
.0
.0007
.0042
.0014
SIC
26
.C440
.C504
.0283
.0464
.0091
.0160
.0171
.0053
.0113
.0022
.0094
.0113
.0029
.0102
.0074
.0143
.0069
.0120
.0027
.0059
.0133
.0076
.0001
.0090
.0050
.0020
.0017
.0036
.0184
.0038
.0009
.0
.0030
.C
.C089
.0012
.C056
.0007
. C022
.0102
.0
SIC
27
.2470
.1135
.0442
.0469
.0205
.0233
.0304
.0089
.0157
.0211
.0195
.0104
.0035
.0187
.0057
.0085
.0109
.0149
.0060
.0097
.0052
.0136
.0036
.0076
.0069
.0049
.0066
.0031
.0039
.0040
.0033
.0
.0062
.0
.0120
.0024
.0022
.0
.0009
.0031
.0022
SIC
28
.1077
.1244
.0606
.1054
.0440
.0288
.0177
.0143
.0510
.0022
.0299
.0328
.0019
.0195
.0811
.0294
.0075
.0486
.0308
.0117
.0023
.0205
.0013
.0090
.0270
.0015
.0053
.0039
.0048
.0046
.0017
.0
.0065
.0
.0064
.0028
.0049
.0
.0147
.0174
.0011
SIC
29
.0109
.0654
.1304
.1396
.0341
.1437
.0100
.0055
.0883
.0042
.0251
.0078
.0
.0251
.2215
.0135
.0032
.0
.0032
.0035
.0032
.0328
.0016
.0011
.0064
.0005
.0080
.0141
.0048
.0
.0006
.0
.0
.0
.0010
.0016
.0388
.0
.0
.0036
.0
SIC
30
.0323
.0631
.0612
.0415
.0286
.0069
.0420
.0050
.0067
.0004
.0137
.0092
.0003
.0
.0042
.0126
.0033
.0122
.0009
.0030
.0008
.0026
.0004
.0014
.0114
.0015
.0148
.0005
.0014
.0205
.0005
.0
.0049
.0
.0271
.0005
.0024
.0
.0013
.0090

-------
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
7C
71
72
73
74
75
76
77
78
7V
80
81
82
83
84
S5
66
67
88
.0
.0112
.0059
.0
.0035
.0
.0010
.0
.0016
.0
.0
.OC05
.0009
.0036
.0
.0
.0005
.0
.0
.3043
.0
.0
.0029
.0
.0
.0016
.0
.0
.0
.0001
.0098
.0028
.0107
.0
.0
.0053
.0
.0
.0015
.0042
.OOOtt
.0
.0
.0
.0
.0
.u
.0
.0
.0
.0
.0044
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0060
.0
.0
.0
.0
.0
.0092
.0
.0
.0
.0014
.0
.0
.0007
.0
.0
.0005
.0
.0
.0
.0041
.0001
.0
.0
.0
.0
.0
.0
.0
.OOC3
.0
.0
.0011
.0
.0
.0
.0029
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0073
.0
.0
.OOC4
.0
.0
.0016
.0
.0
.0
.0017
.OOC1
.0014
.0045
.0
.0
.OOC3
.0
.0
.OOC2
.0027
.OOC2
.0
.0
.0
.0
.0
.c
.c
.0
.0
.0
.0
.0
.0007
.0
.c
.0
.0
.0
.0
.0
.c
.0
.c
.0
.0
.0002
.0
.0
.0013
.0
.0
.0012
.0
.0
.0
.COlO
.0007
.0017
.0006
.C
.0
.0006
.C
.0
.0003
.0007
.0005
.C
.0
.C
.0
.0
.0
.0
.0022
.0
.0
.0016
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0009
.0
.0
.0010
.0
.0
.0143
.0
.0
.0
.0031
.0006
.0028
.0027
.0
.0
.0006
.0
.0
.0043
.0006
.0022
.0
.0
.0
.0
.0
.0
.0
. C009
.0
.0
.0018
.0
.0
.0
.C
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0046
.0
.C
.0008
.0
.0
.0023
.0
.0
.0
.0009
.C014
. C056
.0038
.0
.0
.0060
.C
.0
.0005
.0105
.0007
.0
.C
.0
.0
.C
.0
.0
.0027
.0017
.0
.0025
.0
.0008
.0
.0008
.0007
.0
.0005
.0
.0
.0
.0
.0
.0
.0
.0035
.0
.0
.0014
.0
.0
.0012
.0
.0
.0
.0065
.0022
.0035
.0068
.0
.0
.0039
.0
.0
.0013
.0051
.0015
.0
.0
.0
.0
.0
.0
.0
.0020
.0
.0
.0037
.0
.0
.0
.0
.0026
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0051
.0
.0
.0003
.0
.0
.0060
.0
.0
.0
.0127
.0016
.0161
.0053
.0
.0
.0116
.0
.0
.0009
.0102
.0008
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0152
.0
.0
.0
.0
.0
.0
.0019
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0016
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0010
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0011
.0
.0
.0040
.0
.0
.0019
.0
.0
.0
.0033
.0
.0005
.0057
.0
.0
.0020
.0
.0
.0031
.0009
.0045
.0
.0
.0
.0
.0

-------
39
90
91
92
93
94
95
96
97
98
99
100
.0
.0026
.0013
.0
.0
.0029
.0028
.0
.0013
.0019
.0030
.0035
.0
.0016
.0
.0
.0
.0049
.0001
.0
.0
.0010
.0
.0029
.0
.0023
.0
.0
.0
.0017
.OOCO
.0
.OOC5
.ooo>
.OOC2 .
.00 18
.0
.0005
.0022
.0
.0
.C004
.0002
.0
.C
.0006
.0001
.0007
.0
.0007
.0
.0
.0
.0015
.0021
.0
.0
.0020
.0005
.0087
.0
.0007
.0142
.0
.0
.C025
.0005
.C
.C
.C022
.C004
.0066
.0
.0019
.0008
.0
.0
.0012
.0010
.0
.0011
.0010
.0016
.0030
.0
.0032
.0071
.0
.0
.0026
.0015
.0
.0
.0026
.0015
.0004
.0
.0
.0
.0
.0
• w
.0
.0
.0
.0
.0
.0351
.0027
.0
.0003
.0
o
• v/
.0016
.0005
n
. \j
n
. \j
.0005
.0004

-------

AQCR
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
33
34
35
36
37
38
39
40
41
S 1C
31
.0616
.0182
.0094
.0085
.0004
.0031
.0418
.0003
.0 139
.0
.0002
.0042
.0002
.0011
.0002
.0
.0 160
.0034
.0
.0005
.0005
.0016
.0002
.0046
.0002
.0026
.0059
.0
.0003
.0039
.0002
.0
.0041
.0
.0
.0
.0002
.0
.0016
.0
.0
SIC
32
.0231
.0403
.0430
.0268
.0263
.0162
.0060
.0306
.0217
.0055
.0095
.0135
.0022
.0201
.0118
.0149
.0051
.0060
.0064
.0081
.0069
.0093
.0022
.0072
.0046
.0040
.0055
.0091
.0045
.0043
.0043
.0
.0091
.0
.0052
.0058
.0276
.0030
.0047
.0020
.0024
SIC
33
.0170
.0655
.0251
.0430
.0671
.0140
.0036
.1185
.0244
.0
.0458
.0448
.0008
.0022
.0118
.0378
.0163
.0043
.0023
.0050
.0022
.0073
.0001
.0019
.0040
.0009
.0010
.0036
.0049
.0095
.0024
.0
.0023
.0
.0028
.0428
.0053
.0112
.0014
.0005
.0
SIC
34
.0466
.0952
.0553
.0411
.0468
.0226
.0230
.02 Ł4
.01 Ł0
.0021
.0329
.01C4
.0043
.00<7
.0123
.OOS8
.0152
.0124
.00*5
.0056
.0051
.0073
.OOC9
.0029
.00t6
.0051
.0039
.0041
.0048
.0062
.0018
.0
.0123
.0
.0027
.OOfO
.00*3
.0022
.0054
.00 18
.0019
SIC
35
.0283
.0828
.0484
.0349
.0732
.0099
.0202
.0148
.0129
.0008
.0376
.0082
.0139
.0269
.0144
.0096
.0241
.0121
.0082
.0079
.0041
.0059
.0021
.0020
.0100
.0007
.0033
.0007
.0040
.0065
.0034
.0
.0069
.0
.0252
.0012
.0056
.0
.0013
.0029
.0020
SIC
36
.0634
.0927
.0638
.0390
.0049
.0127
.0334
.0148
.0085
.0057
.0195
.0174
.0028
.0129
.0018
.0089
.0290
.0088
.0068
.0148
.0016
.0060
.0030
.0019
.0108
.0008
.0017
.0001
.0034
.0043
.0055
.0
.0106
.0
.0167
.0006
.0034
.0
.0005
.0017
.0008
SIC
37
.0466
.0309
. 1118
.0231
.1788
.0112
.0140
.0051
.0474
.0021
.0393
.0208
.0363
.0076
.0012
.0267
.0109
.0252
.0036
.0115
.0636
.0222
.0091
.0290
.0234
.0010
.0143
.0063
.0030
.0014
.0048
.0
.0100
.0
.0064
.0050
.0139
.0
.0005
.0008
.0
SIC
38
.0936
.1268
.0643
.0510
.0085
.0061
.0677
.0123
.0119
.0028
.0148
.0019
.0048
.0335
.0019
.0087
.0123
.0048
.0004
.0028
.0008
.0022
.0008
.0004
.0018
.0010
.0026
.0012
.0035
.0149
.0059
.0
.0060
.0
.0006
.0004
.0020
.0
.0
.0008
.0
SIC
39
.1678
.0678
.0482
.0164
.0104
.0096
.0134
.0059
.0087
.0010
.0102
.0037
.0022
.0087
.0019
.0048
.0057
.0102
.0018
.0023
.0020
.0020
.0013
.0061
.0035
.0024
.0029
.0012
.0023
.0553
.0007
.0
.0032
.0
.0019
.0009
.0017
.0
.0005
.0024

-------
d
•
IN)
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0023
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0049
.0
.0002
.0007
.0
.0
.0011
.0
.0
.0009
.0
.0151
.0
.0
.0
.0
.0
.0
.0
.0015
.0025
.0
.0013
.0
.0022
.0
.0011
.0033
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0106
.0
.0
.0022
.0
.0
.0015
.0
.0
.0
.0134
.0007
.0020
.0033
.0
.0
.0031
.0
.0
.0033
.0126
.0005
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0010
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0179
.0
.0
.0080
.0
.0
.0001
.0
.0
.0
.0001
.0044
.0019
.0010
.0
.0
.0047
.0
.0
.0407
.0034
.0004
.0
.0
.0
.0
.0
.0
.0
.0014
.0
.0
.0017
.0
.0
.0
.OOC3
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0040
.0
.0
.0022
.0
.0
.0011
.0
.0
.0
.0029
.0015
.0017
.0045
.0
.0
.0041
.0
.0
.0075
.00 16
.OOC7
.0
.0
.0
.0
.0
.0
.0
.0025
.0
.0
.0013
.0
.0
.c
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0039
.0
.0
.0167
.0
.0
.0005
.0
.0
.0
.0006
.0155
.0005
.0122
.0
.0
.0091
.0
.0
.0038
.0052
.C080
.0
.0
.0
.0
.0
.C
.0
.0
.0
.0
.0005
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0057
.0
.0
.0001
.0
.0
.0011
.0
.0
.0
.0011
.0001
.0
.0124
.0
.0
.0144
.0
.0
.0040
.0047
.0078
.0
.0
.0
.0
.0
.0
.0
.0004
.0
.0
.C
.0
.0
.0
.0
.0
.0
.0
.c
.0
.0
.0
.0
.0
.0
.0034
.0
.0
.0003
.0
.0
.0005
.0
.0
.0
.0021
.0001
.0002
.0004
.0
.0
.0022
.0
.0
.C083
.0021
.0016
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0006
.0
.0
.0
.0
.0
.0004
.0
.0
.0
.0032
.0004
.0013
.2883
.0
.0
.0013
.0
.0
.0004
.0013
.0126
.0
.0
.0
.0
.0
.0
.0
.0031
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0062
.0
.0
.0011
.0
.0
.000$
.0
.0
.0
.0012
.0004
.0010
.0067
.0
.0
.0032
.0
.0
.0003
.0025
.0002
.0
.0
.0
.0
.0

-------
89
90
91
92
93
94
95
96
97
98
99
100
.0
.0053
.0
.0
.0
.0033
.0
.0
.0
.0043
.0
.0069
.0
.0037
.0017
.0
.0
.0015
.0025
.0
.0
.0006
.0007
.0025
.0
.0020
.0
.0
.0
.0070
.0015
.0
.0005
.0050
.0002
.0013
.0
.0051
.0
.0
.0
.0021
.01C9
.0
.00 11
.0018
.0026
.0032
.0
.0037
.0
.0
.c
.0028
.0116
.C
.0050
.0051
.0007
.0050
.0
.0060
.0
.0
.0
.0030
.0002
.0
.0005
.0047
.0
.0014
.0
.0004
.0
.0
.0
.0026
.0029
.0
.C006
.C020
.0183
.0005
.0
.0087
.0
.0
.0
.0036
.0138
.0
.0
.0009
.0013
.0032
.0
.0094
.0
.0
.0
.0015
.0016
.0
.0
.0061
.0005
.0012
to

-------
      APPENDIX E





-------
Appendix E







      This appendix provides a listing of sources of data, and types of




data specified by input card format of the model.

-------
                         LIST OF DATA
Sources

Annual Survey of Manufacturers, 1954-1966
Census of Manufacturers,  1963,  1967
County-City Data Book,  1967,  1969
Individual Income Tax Returns, 1968
Current Population Reports
County Business  Patterns, 1967, 1969
Employment and  Earnings Statistics for States and Areas, 1968, 1969
Manpower Report of the  President,  1969
Census of Government,  1962,  1967
Consumer Expenditures  and Income, 1960-1961, for SMS A
Sales Management Magazine,  1956,  1969
Survey of Current Business (monthly publication)
Statistical Abstracts of the United States,  1967, 1968, 1969
Costs and Economic Impact of Air Pollution  Control, Fiscal Years
      1970-1974 (NAPCA,  January,  1969)
Comprehensive Economic Cost Study of Air Pollution Control Costs
      for Selected Industries and Selected Regions (RTI, February,
      1970)
The  Fuel of Fifty Cities  (Ernst and Ernst, November, 1968)
Manufacturer's Report of Air Pollution Control Equipment Sales,
      1966,  1967, 1968 (Industrial Gas  Cleaning Institute Co. )
Economic Projections for Air Quality Control Regions (OBE Depart-
      ment of Commerce,  June,  1970)

-------
LIST OF THE 100 AIR QUALITY
CONTROL REGIONS (AQCRs)
Code         AQCR

   1          New York, New York
   2          Chicago, Illinois
   3          Los Angeles, California
   4          Philadelphia, Pa.
   5          Detroit, Michigan
   6          San Francisco, California
   7          Boston, Massachusetts
   8          Pittsburgh,  Pa.
   9          St.  Louis, Missouri
  10          Washington, D. C.
  11          Cleveland, Ohio
  12          Baltimore, Maryland
  13*         Hartford-New Haven,  Connecticut
  14          Minneapolis-St. Paul, Minnesota
  15          Houston, Texas
  16          Buffalo, New York
  17          Milwaukee,  Wisconsin
  18          Cincinnati, Ohio
  19          Louisville, Kentucky
  20          Dallas, Texas
  21          Seattle-Everett, Washington
  22          Kansas City, Missouri
  23          San Diego, California
  24          Atlanta, Georgia
  25          Indianapolis, Indiana
  26          Miami, Florida
  27          Denver, Colorado
  28          New Orleans,  Louisiana
  29          Portland,  Oregon
  30          Providence-Pawtucket, Rhode Island
  31          Phoenix, Arizona
  32          Tampa, Florida
  33          Columbus, Ohio
  34          San Antonio, Texas
  35          Dayton, Ohio
  36          Birmingham, Alabama
  37          Toledo, Ohio

-------
100 AQCRs (continued)
Code         AQCR

 38          Steubenville-Weirton, Ohio/Wheeling,  West Virginia
 39          Chattanooga,  Tennessee
 40          Memphis,  Tennessee
 41          Salt Lake City,  Utah
 42          Oklahoma City, Oklahoma
 43          Omaha,  Nebraska
 44          Honolulu, Hawaii
 45          Beaumont-Port Arthur-Orange,  Texas
 46          Charlotte, North Carolina
 47          Portland, Maine
 48          Albuquerque,  New Mexico
 49          Lawrence-Haverhill/Lowell,  Massachusetts
 50          El Paso, Texas
 51          Las Vegas, Nevada
 52          Fargo-Moorhead, North Dakota/Minnesota
 53          Boise, Idaho
 54          Billings, Montana
 55          Sioux City,  South Dakota
 56*         Cheyenne, Wyoming
 57*         Anchorage, Alaska
 58*         Burlington, Vermont
 59*         San Juan, Puerto Rico
 60*         Virgin Islands
 61          Allentown-Bethlehem-Easton, Pa., New Jersey
 62*         Anderson-Muncie, Indiana
 63          Bakersfield, California
 64          Davenport-Rock Island-Moline, Iowa, Illinois
 65*         Flint,  Michigan
 66          Grand Rapids/Muskegon-Muskegon Hts. , Michigan
 67          Greensboro, North  Carolina
 68          Harrisburg, Pa.
 69          Jacksonville,  Florida
 70          Knoxville, Tennessee
 71          Nashville, Tennessee
 72          Peoria,  Illinois
 73          Richmond,  Virginia
 74          Rochester,  New York
 75          Saginaw/Bay City, Michigan

-------
100 AQCRs (continued)
Code         AQCR

 76          Scranton/Wilkes Barre-Hazelton, Pa.
 77          Syracuse,  New  York
 78          Tulsa,  Oklahoma
 79*         Worcester, Massachusetts
 80          Youngstown-Warren, Ohio
 81          Albany-Schenectady-Troy,  New York
 82          Binghamton, New York
 83          Charleston, South Carolina
 84          Charleston, West Virginia
 85          Des Moines, Iowa
 86          Fresno, California
 87          Fort Wayne, Indiana
 88          Jackson, Mississippi
 89          Johnstown, Pa.
 90          Lancaster, Pa.
 91          Mobile, Alabama
 92          Norfolk-Portsmouth/Newport News-Hampton,  Va.
 93          Raleigh/Durham, North Carolina
 94          Reading, Pa.
 95          Rockford,  Illinois
 96          Sacramento, California
 97          South Bend, Indiana
 98          Utica-Rome, New York
 99          Wichita, Kansas
 100          York, Pa.
*Data not available for this AQCR.

-------
CARD TYPE IDENTIFICATION
AND VARIABLE LIST
      The general format for all cards is 7F10.0, II,  12, 13, and 212.

The first seven fields contain specific data depending upon the codes

in columns 71 through 80.   These codes are as follows:


Column             Contains

71                  1 -- Indicates the card contains figures for fuel
                        and electric energy consumption

                    Blank  -- Anything else

72-73               Last two digits of year of the data

74-76               From  001 to 100 --  Indicates card contains data
                        for a specific AQCR (Air Quality Control
                        Region)

                    Blank  -- Indicates card contains totals for the
                        United States

77-78               00 -- Indicates the card contains  all industry
                        totals

                    From  20 to 39  -- Indicates the card  contains
                        data for a  particular industry (2-digit SIC)

                    40 -- Indicates card contains Air Quality Control
                        costs

                    41 -- Indicates card contains Unemployment
                        figures and Sulfur Maximum Restriction
                        costs

                    42 -- Indicates card contains Local Government
                        figures and Population figures

79-80               Number of cards within a specific AQCR or SIC
                        and year for a particular set of data

-------
DESCRIPTION OF CARD TYPES
Card
Type

1
Identifying
Codes Col.

  71
          77-78
          71
          77-78
          77&7S
          77&78
          71
          72&73

          71
          7Z&73
          74-76
          79-80

          71
          72&73
          74-76
          79-80
Contains

Blank


00



Blank


20 to 39
77&7S
79&80
77&7S
79&80
40
01
40
02
              41
              42
               1
               67

               1
               62 or 58
               000
               01

               1
               62 or 58
               000
               02
Data Sets Using This Card Type

1963 General Statistics -- All Industries
Total for AQCR's

1967 General Statistics -- All Industries
Total for AQCR's  1954-1967 Time Series
Data -- All Industries  Total for AQCR's

1963 General Statistics --by SIC for
AQCR's

1967 General Statistics --by SIC for
AQCR's  1954-1967 Time Series Data by
SIC for AQCR's

1971-1975 Air Quality  Control Costs
for AQCR's (Card 1)

1971-1975 Air Quality  Control Costs for
AQCR's  (Card 2)

1967 Unemployment &  Work  Force
figures for AQCR's

1967 AQCR's Local Government and
Private Income

1967 Total for U.S. of Fuel and Electric
Energy Consumption by SIC

1962 and 1958 Total for U. S. of Fuel
and Electric Energy Consumption
by SIC (Card 1)
              1962 and 1958 Total for U. S. of Fuel
              and Electric Energy Consumption
              by SIC (Card 2)

-------
DESCRIPTION OF CARD TYPES (continued)
Card
Type

10
11
Identifying
Codes Col.

  71
  72&73
  74-76
  79-80

  71
  7Z&73
  74-76
  79-80
Contains

1
62
001 to  100
01

1
62
001 to  100
01
Data Sets Using This Card Type

1962 AQCR's Fuel and Electric Energy
Consumption by SIC (Card 1)
1962 AQCR's Fuel and Electric Energy
Consumption by SIC (Card 2)

-------
VARIABLE LIST


General Statistics All Industries Total for AQCR

Card
Type     Field      Description

  1          1         Number of Total Employees for AQCR (in 1, 000)

  1          2         All Employees Wages for  AQCR (in $100, 000)

  1          3         Value added  by manufacture for AQCR (in $100, 000)

  1          4         Value of shipment for AQCR (in $100, 000)

  1          5         New Capital  Expenditures for AQCR (in $100, 000)

  1          6-7      No Data


General Statistics for SIC in AQCR

Card
Type     Field      Description

  2          1         Number of Total Employees for SIC in AQCR
                     (in 1, 000)

  2          2         All Employees Wages for  SIC in AQCR (in  $100, 000)

  2          3         Value added  by manufacture for SIC in AQCR
                     (in $100, 000)

  2          4         Value of shipment for SIC in AQCR (in $100, 000)

  2          5         New Capital  Expenditures for SIC in AQCR
                     (in $100, 000

  2          6-7      No Data

-------
VARIABLE LIST (continued)


Air Quality Control Costs  1971-1975 (Card 1)

Card
Type      Field      Description
                                         s
  ?          1         Investment by Industrial Process in AQCR
                     (in $100, 000)

  3          2         Annual Cost  by Industrial Process in AQCR
                     (in $100, 000)

  3          3         Investment by Stationary Combustion in AQCR
                     (in $100, 000)

  3          4         Annual Cost  by Stationary  Combustion in AQCR
                     (in $100, 000)

  3          5         Investment by Solid Waste in AQCR (in $100, 000)

  3          6         Annual Cost  by Solid Waste in AQCR (in $100, 000)

  3          7         No Data


Air Quality Control Costs  1971-1975 (Card 2)

Card
Type      Field      Description

  4          1         Total Investment - Lower  Limit in AQCR (in $100, 000)

  4          2         Total Annual Cost  - Lower Limit in AQCR (in $100, 000)

  4          3         Total Investment - Expected in AQCR (in $100, 000)

  4          4         Total Annual Cost  - Expected in AQCR (in $100, 000)

  4          5         Total Investment - Upper Limit in AQCR (in $100, 000)

  4          6         Total Annual Cost  - Upper Limit in AQCR (in $100, 000)

  4          7         No Data

-------
VARIABLE LIST (continued)


1967 Unemployment and Work Force Figures for AQCR

Card
Type      Field      Description

  5         1         Work Force for March 1967 in AQCR (in 100)

  5         2         Unemployment Rate in AQCR (in . 001)

  5         3         1.0 Percent Sulfur Maximum Fuel Restriction
                     Annual Regional Cost to Steam-Electric Power
                     Generation Combustion Sources:  Low for 1974
                     (in $100, 000)

  5         4-5-      No Data
           6-7


1967 AQCR's Local Government Income and Expenditures

Card
Type      Field      Description

  6         1         Population of AQCR's on July 1, 1968 (in  1, 000)

  6         2         Per  Capita Personal Income  1968 (in dollars) for
                     AQCR

  6         3         AQCR Local Government  Total 1967 General
                     Revenue (in $1, 000, 000)

  6         4         AQCR Local Government  Total 1967 Direct
                     General Expenditures (in  $1, 000, 000)

  6         5-6-7    No Data

-------
VARIABLE LIST (continued)
Total for U. S. of Fuel and Electric Energy Consumed by SIC's -1967

Card
Type      Field      Description

  7          1         Total Cost of Purchased Fuels and Electric Energy
                     by SIC (in $1,000,000)

  7          2         Total Cost of Purchased Fuels by SIC's  (in $1, 000, 000)

  7          3         Quantity of Electric Energy Purchased by SIC's
                     (million kw/hrs. )

  7          4         Cost of Electric Energy Purchased by SIC's
                     (in $1, 000, 000)

  7          5-6-7    No Data
Total for U. S. of Fuel and Electric Energy Consumption by SIC's -
1958 and 1962 - Card 1

Card
Type      Field      Description

  8         1         Total cost of purchased fuels and electric energy
                     by SIC in U. S.  ($1, 000)

  8         2         Total cost of purchased fuels by SIC in U. S.
                     ($1,000)

  8         3         Quantity of bituminous coal,  lignite and anthracite
                     purchased by SIC in U.  S.  (1, 000 short ton)

  8         4         Cost of bituminous coal, lignite and anthracite
                     purchased by SIC in U.  S.  ($1,000)

  8         5         (Only Industry 33) Quantity of coke and breeze
                     purchased by SIC in U.  S.  (1, 000 short ton)

-------
VARIABLE LIST (continued)
Total for U. S. of Fuel and Electric Energy Consumption by SIC's -
1958 and 1962 - Card 1 (continued)

Card
Type     Field      Description

  8         6          (Only Industry 33) Cost of coke and breeze purchased
                     by SIC in U.  S.  ($1, 000)

  8         7         No Data
Total for U.S.  of Fuel and Electric Energy Consumption by SIC's -
1958 and 1962 - Card 2

Card
Type      Field      Description

  9          1         Quantity of Fuel Oil (distillate and residual) pur-
                     chased by SIC in U.  S.  (1, 000 barrels of 42 gal. )

  9          2         Cost of Fuel Oil purchased by SIC in U.  S.  ($1,000)

  9          3         Quantity of Gas  - natural, manufactured, still blast
                     furnace, and coke oven - purchased by SIC in U. S.
                     (million cubic foot)

  9          4         Cost of Gas purchased by SIC in U.  S. •($!, 000)

  9          5         Quantity of Electric Energy purchased by SIC in
                     U.  S.  (million kw/hrs. )

  9          6         Cost of Electric Energy purchased ($1, 000)

  9          7         No Data

-------
VARIABLE LIST (continued)
1962 Fuel and Electric Energy Consumption by SIC in AQCR  - Card 1

Card
Type      Field       Description

  10         1          Total cost of purchased fuels and electric energy
                      by SIC in AQCR ($1, 000)

  10         2          Total cost of purchased fuels by SIC in AQCR ($1, 000)

  10         3          Quantity of bituminous coal, lignite and anthracite
                      purchased by SIC in AQCR  (1, 000 short tons)

  10         4          Cost of bituminous coal, lignite and anthracite pur-
                      chased  by SIC in AQCR ($1, 000)

  10         5          (Only Industry 33) Quantity of coke and breeze pur-
                      chased  by SIC in AQCR (1, 000 short tons)

  10         6          (Only Industry 33) Cost of coke  and breeze purchased
                      by SIC in AQCR ($1, 000)

  10         7          No Data
1962 Fuel and Electric Energy Consumption by SIC in AQCR - Card 2

Card
Type      Field      Description

  11         1         Quantity of fuel oil (distillate  and residual) purchased
                     by SIC in AQCR (1, 000 barrels  of 42 gal. )

  11         2         Cost of fuel oil purchased by SIC in AQCR ($1, 000)

  11         3         Quantity of gas (natural, manufactured',  still blast
                     furnace, and coke oven) (million cubic feet)

  11         4         Cost of gas purchased by SIC  in AQCR ($1, 000)

-------
VARIABLE LIST (continued)
1962 Fuel and Electric Energy Consumption by SIC in AQCR - Card 2
(continued)
Card
Type

  11
  11

  11
Field
 6

 7
Description

Quantity of electric energy purchased by SIC in
AQCR (million kw/hrs. )

Cost of  electric energy purchased by SIC in AQCR

No Data

-------
Source:  1963 Census of Manufacturers, Vol. 1, Summary and Subject
         Statistics, Chapter 7, Fuels and Electric Energy Consumption,
         Table 6,  Total Consumption of Heat and Major Industry Groups
         Table 10,  Consumption According to SMSAs
Card 1

Column      Contains

1-10         Total cost of purchased fuels and electric energy ($1, 000)

11-20        Total cost of purchased fuels ($1, 000)

21-30        Quantity of bituminous coal, lignite and anthracite pur-
             chased (1,000 short ton)

31-40        Cost of 11-20 ($1, 000)

41-50        (Only Industry 33) Quantity of coke and breeze purchased
             (1, 000 short ton)

51-60        (Only Industry 33) Cost of 41-50 ($1, 000)

71-80        Codes
             72-73  Last two digits of year

             74-76  AQCR (Air Quality Control Region)
                    Blank for total of all AQCR's

             77-78  Industry code  - SIC

             79-80  Card number 01 to 02 for this industry
Card 2

Column

1-10


11-20
Contains

Quantity of fuel oil (distillate and residual) purchased
(1, 000 barrels of 42 gal. )

Cost of fuel oil ($1, 000)

-------
Card 2 (continued)

Column      Contains

21-30        Quantity of gas (natural,  manufactured,  still blast furnace,
             and coke oven (million cubic feet)

31-40        Cost of  gas ($1, 000)

41-50        Quantity electric energy  (million kw/hrs. ) purchased

51-60        Cost of  electric energy ($1, 000)

71-80        Codes
             Same of card 1

-Fuel and Electric Energey Consumed 1962 + 1958 totals for United
 States According to  SIC.

-Fuel and Electric Energy Consumed  1962 for AQCR's by SIC's.

-------
Base  Year Air Quality Control Costs
Annual and Investment Costs by
Metropolitan Area, Fiscal Years 1971-1975
Card
1
1
1
1
1
1




Card
2
2
2
2
2
2
Column
1-10
11-20
21-30
31-40
41-50
51-60
72-73
74-76
77-78
79-80
Column
1-10
11-20
21-30
31-40
41-50
51-60
Contains
Investment by industrial process (in $100, 000)
Annual cost by industrial process (in $100, 000)
Investment by stationary combustion (in $100, 000)
Annual cost by stationary combustion (in $100, 000)
Investment by solid waste (in $100, 000)
Annual cost by solid waste (in $100, 000)
Year - last two digits
Air Quality Control Region
Identifies card as "Air Quality Control Cost"
card (contains 40)
Card numbers 01 or 02 of "Air Quality Control
Costs" cards
Contains
Total investment - lower limit (in $100, 000)
Total annual cost - lower limit ($100, 000)
Total investment - expected (in $100, 000)
Total annual cost - expected (in $100, 000)
Total investment - upper limit (in $100, 000)
Total annual cost - upper limit (in $100, 000)

-------
Base  Year Air Quality Control Costs
Annual and Investment Costs by
Metropolitan Area, Fiscal Years 1971-1975  (continued)
Card     Column      Contains

          72-73        Year - last two digits

          74-76        Air Quality Control Region

          77-78        Identifies card as "Air Quality Control Cost"
                       card

          79-80        Card 01 or 02 of "Air Quality  Control Cost"
                       cards
*Air Quality Control Costs  1971-1975 for AQCR's.
Source:  Comprehensive Economic Cost Study of Air Pollution Costs
         for Selected Industries and Selected  Regions, prepared for
         NAPCA, Appendix D.

-------
Area Trends in Employment and Unemployment
June,  1968
U. S.  Department of Labor
Manpower Administration

Pages 51-54:  Work Force,  Unemployment, and Employment in 150
              Major Labor  Areas
Column      Contains

1-10         Work force for March 1967 in hundreds (100)

11-20        Unemployment rate in tenths  of one percent (. 001)

21-30        1974 Low Exhibit V. 17 (1.0% Sulfur Maximum Restric
             tion: Annual Regional Costs to Steam-Electric Power
             Generation Combustion Sources, Fiscal 1970-1974;
             Costs and  Economic Impacts  of Air Pollution Control,
             Fiscal  1970-1974,  Ernst and  Ernst)

71-80        Codes
             72-73      Last two digits of the year

             74-76      AQCR code

             77-78      Card identification (blank - for these cards)

             79-80      41
Source:  1967 Unemployment and Work Force figures.

-------
Statistical Abstract of the United States
Section 33:  Metropolitan Area Statistics
Table  1:  SMSA's with 250, 000 population or more
Table 2:  SMSA's with population below 250, 000

AQCR Income and Expenditures for 1967
Column      Contains

1-10         Population (1968 July 1) Total (in 1,000)

11-20        Personal income,  1968 (per capita total) (in dollars)

21-30        Local governments, 1967 (total general revenue)
             (in million dollars)

31-40        Local governments, 1967 (direct general expenditures,
             total) (in million dollars)

71-80        Codes

             72-72     Year "67"

             74-76     AQCR code

             77-78     Identifies the card 42

             79-80     Card no. 1 in all cards



*AQCR's Income and Expenditures 1967.

-------
1967 Census of Manufacturers
Summary Series  - Fuels and Electric Energy Used
All AQCR Total
Column      Contains

1-10         Total cost of purchased fuels and electric energy
             (million dollars  - $1,000,000)

11-20        Total cost of purchased fuels (million dollars)

21-30        Quantity of electric energy purchased (million kw/hrs. )

31-40        Cost of electric  energy (million dollars)

41-50

71-80        Codes

             72-73     Last two digits of year

             74-76     AQCR (Air Quality Control Region)
                       Blank -  total for all regions

             77-78     Industry Code - SIC

             79-80     Card number within industry
                       (always  01  for this data)
-Fuel and Electric Energy Consumed,  1967 -- Totals for the United
 States According to SIC.

-------