University of Southern California Dissertations and Theses

A Study Of Process And Content Of Public Policy In American States: With Reference To The Medicaid Program

(USC Thesis Other)

A Study Of Process And Content Of Public Policy In American States: With Reference To The Medicaid Program

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content A STUDY OF PROCESS AND CONTENT
OF PUBLIC POLICY IN AMERICAN STATES:
WITH REFERENCE TO THE MEDICAID PROGRAM
by
Marn Jai Cha
A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(Public Administration)
August 1970
71-7705
CHA, Marn Jai, 1938-
A STUDY OF PROCESS AND CONTENT OF PUBLIC POLICY
IN AMERICAN STATES: WITH REFERENCE TO THE MEDICAID
PROGRAM.
University of Southern California, Ph.D., 1970
Political Science, public administration
University Microfilm s, Inc., Ann A rb o r, Michigan
THIS DISSERTATION HAS BEEN MICROFILMED EXACTLY AS RECEIVED
U N IV E R S IT Y O F S O U T H E R N C A L IF O R N IA
T H E G R A D U A T E S C H O O L
U N I V E R S IT Y P A R K
LO S A N G E L E S . C A L I F O R N I A 9 0 0 0 7
'This dissertation, w ritte n by
Marn Jai Cha
under the direction of D issertation C om -
iniltee, an d a p p ro v e d by a ll its members, has
been presented to and accepted by 'The G r a d u
ate S chool, in p a rtia l f u lfillm e n t o f re q u ire
ments o f the degree of
D O C T O R O F P H I L O S O P H Y
.-•/ u-r-i Gy s -
Dean
D a te...
ACKNOWLEDGMENT
1 Lake this opportunity to thank all of those who
assisted me in completing this dissertation. 1 wish to
acknowledge my debt and gratitude to the members of the
Dissertation Committee. My special thanks are due to Dr.
William B. Storm, Chairman of the Committee for his warm
support.
1 am grateful to Professor Richard Gable, now at the
University of California, Davis, who first stimulated my
interest in national health care problems. Thanks are
also due to Dr. John Rotstan, Chairman of the Department
of Political Science at Fresno State College for providing
an environment in which a new junior member of the faculty
could finish the remainder of the dissertation.
Lastly, 1 owe much to my wife, Hack Jo, and my
parents for their continuing encouragement in bringing
this task to a successful completion.
TABLE OF CONTENTS
Pago
ACKNOWLEDGMENT ........................................ ii
LIST OF TABLES....................................... v
LIST OF FIGURES..................................... ix
LIST’ OF CHARTS......................................... x
Chapter
I. INTRODUCTION ................................. 1
Sotting of the Problem
Immediate Focus
Economic Development
Political System
Policy Content
Policy OuLputs
Statement of the Problem
Limitations of the Study
Importance of the Study
Organization of the Remainder
of the Study
II. REVIEW OF THE LITERATURE..................... 42
State Policy Studies
Critique and Summary
III. RESEARCH METHODOLOGY ......................... 69
|
Research Design i
Hypotheses and Variables ■
Selection of Sample
Data-Gathering Procedures
Statistical Techniques
Summary
iii
iv '
Chapter Page
IV. ECONOMIC DEVELOPMENT AND POLITICAL
a YD TDM CHARACTERISTICS..................... 81
Inter-Party Competition and economics
Political Participation and economic
Development
MalapportionmenL and economic
Development
Summary
V. 1NTEP.FACE OF POLICY OUTPUTS Wiiii ECONOMICS
AND POLIT]CaL SYSTLM............................ 127
Scope of Coverage
Range of Services
expenditures in Medicaid
Summary
VI. IMPACT OF POLICY OUTPUTS: MEDICAID.............. 148
Medicaid: A Brief History
Fiscal Impact
Non-Fiscal Impact
Summary and Analysis
VII. DISCUSSION AND CONCLUSIONS . ................... 196
Discussion
Conclusions
BIBLIOGRAPHY............................................ 230
LIST OF TABLES
Table
1.
2 .
3.
4.
3.
6 .
7.
8.
9.
10.
11.
12.
Page
Classification of the General Literature
on Public Policy........................... 10
Thirty-Eight States Classified by the
Degree of Inter-Party Competitiveness .... 84
Industrialization in Terms of the Number
of Population Engaged in Non-Agricultural
Sector in Thirty-Eight States ............... 85
Relationship Between Inter-Party Competi
tiveness and Industrialization in Thirty-
Eight States............................... 86
School Years Completed by Lhe Population 25
Years or Older in Thirty-Eight States .... 89
Relationship Between Inter-Party Competi
tiveness and Education in Thirty-Eight
States...................................... 90
Per Capita Income of Thirty-Eight States
1967 92
Relationship Between Per Capita Income and
Inter-Party Competitiveness 93 |
Rate of Urbanization of Thirty-Eight States j
in Terms of Percentage of Population !
Living in Urban Areas 96 !
Relationship Between Inter-Party Competi
tiveness and Rate of Urbanization.......... 97
Correlation Coefficient Between Party
Competition and Economic Development .... 100
Average Percentage of Voter Turnout in
Gubernatorial and Neatorial Elections in
Non-Presidential Years (1952-65) 102
v _________________
vi
age
103
106
109
112
113
116
117
119
121
123
125
130
132
Relationship Between Voter Turnout and
Industrialization .................
Relationship Between Voter Participation
and Per Capita Income .....................
Relationship Between Voter Turnout and
Education .................................
Relationship Between Political Partici
pation and Rate of Urbanization ..........
Correlation Coefficient Between Political
Participation and Economic Development . .
Schubert-Press Ranks of Apportionment in
Thirty-Eight States .......................
Relationship Between Industrialization and
Apportionment in Thirty-Eight States . . .
Relationship Between Education and Apportion
ment in Thirty-Eight States ...............
Relationship Between Per Capita Income and
Apportionment in Thirty-Eight States . . .
Relationship Between Urbanization and
Apportionment in Thirty-Eight States . . .
Relationship Between Economic Development
and Political System Characteristics . . .
Medicaid Coverage Groupings .................
Number of Persons Covered by Medicaid
Plans in Thirty-Seven States ............
Relationship Between Political System and
Scope of Coverage in Medicaid States . . .
Vll
Page
134
136
137
138
141
142
146
155
156
157
159
161
163
Correlation Coefficient Between Economic
Development and Scope of Coverage Under
Medicaid .................................
Types of Services Provided under Medicaid . .
Number of Services Provided under
Medicaid by States .......................
Correlation Coefficient Between Number of
Services Provided and Political and
Economic Variables .......................
Costs of Medicaid as Adopted by Thirty-
Seven States .............................
Correlation Coefficient Between Three
Measures of Title 19 States' Expenditure
and Political and Economic Variables . . .
Correlation Coefficients Between Seven
Measures of Medicaid Policy Outcomes and
Political and Economic Variables ........
Personal Health Care Expenditures by
Source of Funds (FY 1960-67) ............
Per Capita Personal Health Care Expenditures
in Constant 1967 Dollars (FY 1960-67) . . .
Percent of Physicians Reporting Fee
Increases (FY 1966-67) ...................
Medical Care Price Changes on Selected Items
and Periods (June, 1965-Dec., 1967) . . . .
Some Factors Underlying the Rise of Medical
Care Prices Derived From Criticisms of
Title 19 Program Operations ...............
Federal Expenditures for Title 19 Programs
Actual and Estimated (FY 1966-69) ........
Vlll
Table Page
40. Comparison of Recent Rates of Per Capita
Medical Vendor Payments with Pre-Title 19
Experience by Scope of Coverage ............. 165
41. Vendor Payments by the States (FY 196 5
and 1967).................................... 167
42. Some Criticisms of Medicaid Program
Operations as Reported by Twenty-Six
States........................................ 176
43. Experiences of Medicaid States as of
Summer, 1968 179
44. Medicaid Problems Other Than Fiscal Policy
as Reporled by State Governors and
Legislative Leaders, Summer, 1968 .......... 181
43. State Reports on Measures to Deal with
Certain Operating Problems in Medicaid . . . 191
46. Relationship Among Urbanization, Industri
alization, Income, and Education .......... 203
LIST OF FIGURES
Figure Page
1. Three Basic Cells of the M o d e l ............... 12
2. Schematic Diagram of the Model:
Input-Output Analysis ....................... 13
3. Full Version of the Conceptual Model
Depicting Policy Process and Content .... 31
4. Schematic Classification of the Approaches
and Variables in Policy Studies ............ 36
ix
LIST OF CHARTS
Chart
1.
Page
Per Capita Public Assistance Medical
Payments 1965 and 1967 170
CHAPTER I
INTRODUCTION
Setting of the Problem
As the scope of public activities has expanded at an
unprecedented rate to meet increasingly complex social
problems, the study of public policy has become an important
area of inquiry among the social and political scientists.
The purpose of this introductory chapter is twofold: (a) to
review and assess the issues, concepts, and ideas involved
in various approaches to the study of public policy and (b)
to propose a conceptual scheme for studying a public policy
based on this critical review of the literature.
The literature in the field reveals four different
types of approaches to the study of public policy, viz.,
(1) actor, (2) problem, (3) process, and (4) content
approaches. There also emerges to be four major analytical
units or levels to which these approaches are applied: (1)
general purpose, (2) national, (3) state-local-municipal,
and (4) international problems. j
The actor approach is primarily concerned with the
analysis of cognitive, perceptual, and motivational process |
through which a decision-maker arrives at his choice of
alternatives. The range of thoughts in this actor-oriented
approach is diverse; from an effort to build a generalized
2
abstract model of a decision-maker to particular case
description of the situations in which an actor makes a
difficult decision.
Herbert Simon^ attempts to build a model of policy
maker or decision-maker in terms of variations on the theme
of rationality of "economic man." Those who followed
2
Simon's lead subsequently attempt to build a generalized
abstract construct of a man, the decision-maker, in a
milieu of complex organizations.
There are others who are interested in probing into
perceptual and motivational aspects of behavior of a single
3
policy-maker or a defined group. Lucian Pye, for example,
has attempted to explain the behavior of a class of policy
makers in Burma in this fashion.
Robert Lane^ has also applied this motivational
analysis to fifteen "common men" in New Haven to find the
^"Herbert Simon (ed.), Models of Man (New York: John
Wiley, 1957), pp. 25-27.
2
See Richard Cyert, et al., A Behavioral Theory of
Firm (Englewood Cliffs, N. J.: Prentice-Hal1, 1963); M. B.
Smith, "Opinion, Personality, and Political Behavior,"
American Political Science Review, LII (March, 1958), 10.
3
Lucian Pye, Politics, Personality and Nation
Building (New Haven: Yale University Press, 1962).
Robert E. Lane, Political Ideology (New York: Free
Press, 1962), pp. 131-36.
sources of their political ideology. Lasswell intensively
scrutinized the life histories of different types of
political actors and administrators to draw a profile of
classes or role incumbents.
Using the concept of "office" or "position" as a
point of departure, Richard Neustadt^ studies strategic
pattern of decision-making behaviors of the recent Presi
dents. Thomas Flynn^ concentrates on the office of
Governor Freeman in describing Minnesota's budget-making
g
process. John Keeley studies Robert Moses, the renowned
Commissioner of the Park Department of the City of New
York, in analyzing the dynamic nature of public policy
making process.
The problem approach is usually concerned with some
particular problem and the policies that are appropriate to :
I
dealing with it. There is, usually quite explicitly, a !
general notion that there is a problem, and that it is !
I
^Harold Lasswell, Psychopathology and Politics (New j
York: Viking, 1960). I
^Richard E. Neustadt, Presidential Power (New York:
John Wiley, 1963). j
^Thomas Flynn, Governor Freeman and the Minnesota
Budget, #60 (New York: Bobbs-Merril, 195? ).
g
John Keeley, Moses on the Green, #45 (New York:
Bobbs-Merril, 1953). Also, see other Inter-University Case
Programs publication; this program was once caught with
this idea of descriptive case studies probing into the
situational factors surrounding an individual personality.
possible to take steps to solve it.
9
Vincent Ostrom, for example, expends considerable
effort in solving the local water resource problems. James
Davis and Kenneth Dolbeare^ show their expertise in study
ing national selective service system.
Hitch and McKean^ apply their particular analytic
skill, cost-effectiveness analysis, to the problem of
allocation of military resources and weaponry systems.
12
Yehezkel Dror proposes his "normative" model for evalu
ating any general public policies.
They usually evaluate the past or present policies
with their particular skills or analytic tool and come up
with exhortation for new directions or policies. Style is
descriptive and prescriptive ranging from highly quantita
tive analysis to mere verbal exposition.
9
Vincent Ostrom, "Water Resources Development," in
Austin Ranney (ed.), Political Science and Public Policy
(Chicago: Markham, 1968), pp. 123-150.
10
James Davis and Kenneth Dolbeare, "Selective Service!
and Military Manpower," in Ranney, op. cit., pp. 83-122. !
^^Charles Hitch and Roland McKean, The Economics of
Defense in the Nuclear Age (Cambridge: Harvard University
Press, 1961).
12
Yehezkel Dror, Public Policymaking Re-examined
(San Francisco: Chandler Publishing Company, 1968). Dror
is openly for a normative prescriptive approach to the
policy studies. He comments on how the past policy analysts
neglected in assessing the policy consequences on the
pretext that they are only concerned with value-free
l"explanation."__________________________________________________
A bulk of political science literature on policy
studies can be characterized as process-oriented. Unlike
those who apply the issue-centered approach, process-
oriented scholars aspire to be "scientific" disclaiming the
role of prescription in policy studies.
They are concerned with accurate description and
analysis of the institutional process through which a
decision or policy outputs are produced. Thus, the elec
torate, interest group, political party, and legislature
are the important parameters for this group of scholars.
13
Arthur Bentley's The Process of Government and
14
David Truman s Governmental Process are prototype of this
process-oriented approach. David Easton^ later elaborated
the process notion of policy-making with systems concept.
Robert Dahl and C. E. Lindblom's Politics, Economics, and
16
Welfare is another good example of process-oriented
I
analysis of domestic policies. Their notion of incremen-
talism is typically process-conscious view of American
policy-making.
—
Arthur Bentley, The Process of Government (Chicago:
University of Chicago Press, 1908).
"^David Truman, The Governmental Process (New York:
Knopf, 1951). i
^David Easton, A Systems Analysis of Political Life
(New York: John Wiley, 1965).
^Robert Dahl, et al., Politics, Economics, and
Welfare (New York: Harper, 1953).
6
V. 0. Key^7 took an early lead in applying this
process approach to the analysis of policy outcomes at state
level in his Southern Politics. Subsequently, Duane
18 19 20
Lockard, Thomas Dawson, and Thomas Dye followed suit.
21 22
Cutwright and James Rosenau employ process-notion in
their respective analysis of foreign policy-making.
They all take the present policies as the "givens"
or the outcomes of the system. They attempt to explain and
analyze how these outputs came to be as they are. They take
the policy outputs as the "dependent variable" to be
explained by analyzing the interaction of the interest
group, political party, legislature, and electorate within
a political system. They consider the political actors as
17V. 0. Key, Southern Politics in State and Nation
(New York: Knopf, 1951) .
1 R
Duane Lockard, New England State Politics
(Princeton: Princeton University Press, 1959).
^Richard Dawson, et al., "The Politics of Welfare"
in Herbert Jacob and Kenneth Vines (ed.), Politics in the
American States (Boston: Little, Brown, 1965), pp. 371-410.1
20
Thomas Dye, Politics, Economics, and the Public
(Chicago: Rand McNally, 1966;.
21
Phillips Cutwright, Political Structure, Economic
Development, and National Security Programs," American i
Journal of Sociology, LXX (1965), 537-48.
22
James Rosenau, "Pre-Theories and Theories of
Foreign Policy" in Barry Farrell (ed.), Approaches to
Comparative and International Politics (Evanston, 111.:
Northwestern University Press, 1966), pp. 45-51.
7
merely instruments of the system in contrast with actor-
oriented scholars who consider the actors as crucial
variables.
The content approach is largely a consequence of
reaction to the process-orientation which dominated so much
the policy inquiries in political science. The reactions
23
are of two kinds of motive, pragmatic and scientific.
Content-oriented scholars claim that preoccupation
of process-oriented people with value-neutral concern with
"explanation" of the system resulted in serious neglect in
making the "findings" bear upon the practical solution of
the problem. They say that merely explaining how policies
are made is not enough; study of the effect of implemen
tation of policies is equally important and vital to
respond to the increasing demands made upon the political
scientists as a professional. That is, policy recommending
and advising task.
This, they contend, calls for concentration on the
content of public policies in terms of investigating their
I
intended effects on the target group. They mean to respond j
2^+
seriously to Lasswell's call for concern with who gets
I
' . i ■ ■ — I
^See Austin Ranney, "The Study of Policy Content:
A Framework for Choice" in Ranney, op. cit., pp. 3-22.
0 /
Harold Lasswell and D. Leraer (eds.), The Policy
Sciences (Stanford, Calif.: Stanford University Press,
1951), pp. 3-5.
what and how for a practical reason.
Those who are motivated to revolt against the
process-oriented approach for scientific reason argue that
the past process explanation of the workings of the system
fall short of full satisfaction. The nature of public
policies is too complex and varying in their content to
make a sweeping generalization that all of them are pro
cessed through the same institutional mechanism.
They suspect that different types of policies may
go through different types of institutional process. Find
ing these differences and similarities in terms of the
content of the policies would shed a new light on how the
political system is really working.
25
To do this, this group represented by Lowi,
26 27 28
Froman, Wildavsky, and Wayne suggests taking the
policy outcomes as the "independent" rather than "dependent"
variable to "observe the differential impact of different
25
Theodore Lowi, "American Business, Public Policy,
Case-Studies, and Political Theory," World Politics, 16
(July, 1964), 677-715.
26
Lewis A. Froman, "The Categorization of Policy
Contents," in Ranney, op. cit., pp. 41-50.
27
Aaron Wildavsky, "The Political Economy Effici
ency: Cost-Benefit Analysis, Systems Analysis, and Program
Budgeting," Public Administration Review, 26 (December,
1966), 292-310.
28
Wayne L. Francis, Legislative Issues in the Fifty
States: A Comparative Analysis (Chicago: Rand McNally,
1967). _______________________________
9
29
content on policy-making processes." They attempt to I
study the process but through a different tact, i.e., from
the policy-end of it concentrating on the "impact" of the
policies.
The diagram on the following page summarizes some
of the representative literature in policy studies under
the four approaches just reviewed.
Table 1 is not an exhaustive listing of the
literature in the field. Nevertheless, it shows the range
of different approaches available for the study of public
policy at various levels of analysis.
Immediate Focus
Of the four levels of analysis, the public policy
study at the state level has begun to attract a great deal ;
of attention from social and political scientists in recentj
years. Because fifty American states provide the political|
analysts with an excellent laboratory for comparative |
i
political research. In the words of Richard Dawson,
. . . simply because there are sufficient common
characteristics, and because there is a much greater
reservoir of comparable data than that usually avail- j
able in cross-cultural analysis.30
29
Ranney, op. cit., p. 14.
30
Richard E. Dawson and James Robinson, "Inter
party competition, Economic Variables and Welfare Policies
in the American States," Journal of Politics, XXV (1963),
265.
10
TABLE 1
CLASSIFICATION OF THE GENERAL LITERATURE ON PUBLIC POLICY
Level of Analysis
Approache s General National State-Local Foreign
Actor Simon
Lasswell
Neustadt
Lane
Keeley
Flynn
Pye
Problem Dror Hitch
MeKean
Davis
Dobeare
Ostrom Epstein‘ S
Process Easton
Bentley
Dahl &
Lindblom
Truman
Key
Dawson
Lockard
Dye
Cutwright
kosenau
Content Lowi
Ranney
Froman
Wildavsky
Francis
32
Fifty states share in common national political
culture and history, written constitution, similar formal
political structure, and electoral systems, social
"^Leon D. Epstein, British Politics in the Suez i
Crisis (Urbana: University of: Illinois Press, 1^64).
32
Following are some of the works: Warner R. Schil
ling, "The Politics of National Defense: Fiscal 1950," in i
Warner R. Schilling, Paul Hammond, and Glenn Snyder (ed.)«
Strategy, Politics, and Defense Budgets (New York:
Columbia University Press, 1962), pp. 1-266; Vernon Van
Dyke, Pride and Power: The Rationale of the Space Program
(Urbana: University of Illinois Press, 1964); W. Phillips
Davison, The Berlin Blockade: A Study of Cold War Politics
(Princeton, N. J.: Princeton University Press, 1958).
11
institutions. Yet, they show wide variations in socio
economic environment within which state politics takes
place and in the way values and resources are authorita
tively allocated.
For instance, New Jersey spends over $172.00 per
33
capita on education while Mississippi spends $101.00.
California adopts a policy of increasing public assistance
to medically poor and indigent whereas Alabama doesn't
appear to consider this kind of policy. Welfare expendi
tures range from a low of $10 or $11 per capita in South
and North Dakota to a high of over $40.00 in Wyoming and
Nevada.^
Why do such differences occur? How could the
differences in these policy outcomes be explained? Could
they be the consequence of differences in the political
system characteristics or socio-economic environment, or
something else? If anything else, what could they be?
Do differences in the type of governor or state
legislature each state has have anything to do with varying
policy outcomes? What difference does it make in policy
outcomes if a state is one-party or two-party dominated?
Or how much influence does the level of education of
33
United States Statistical Abstract (Washington,
D. C.: U.S. Government Printing Office, 1968), p. 120.
^Dye, op. clt., p. 132.
12
a state have on policy outcomes? Or does the voter turnout
in state election make any difference?
These are some of the questions this study attempts
to investigate within the context of a conceptual model.
The model is constructed with a general systemic
notion that "any outcomes, including policy outcomes, are
the result of Forces brought to bear upon a System and
35
causing it to make particular Responses." Then there are
three basic cells making up the model:
FORCES SYSTEM RESPONSES
Figure 1
In this study, "forces" are construed as referring
to environmental inputs, specifically, socio-economic
development of a state. System refers to the characteris
tics of decision-making machinery primarily in regards to ;
!
political process. Responses refer to a particular type of
policy output, that is, decisions made with respect to such
things as education, highway, health, or air pollution, at
a given period of time. j
Given these descriptive labels in each cell or the j
i
i
!
35 I
Ibid., p. 3; graphic notations in subsequent pages!
are largely adaptation of the original models that appear j
in Thomas Dye, loc. cit.; a part of the model pertaining j
to "content" is drawn from Austin Ranney, op. cit., j
pp. 3-22. j
13
subsystem, the linkages depicting the relationships among
the three subsystems are shown in the schematic diagram
below:
INPUT OUTPUT
Feedback Loop (D4)
D2 D1
Policy
Outcomes
Decisional
System Charac
teristics
Socio-Economic
Development
D3
Policy Process Policy j
Content !
Figure 2
Environmental characteristics, the socio-economic
I
development, are considered to generate inputs which are ;
i
directed into the political system as represented by the j
direction, Dl, in the diagram. It is assumed that socio- j
economic variables are inputs which shape the political j
system and that the character of the political system in
turn determines policy outcomes, as Dl is led or being
mediated through the political system into the region of
output via direction of D2 in the diagram. j
Thus, linkages Dl and D2 suggest that political
system variables have an important independent effect on
policy outcomes by mediating between socio-economic
[conditions and these outcomes^ Linkage D3 assumes, however)
14
that socio-economic variables affect public policy directly,
without being mediated by political system variables. That
is, policy is still processed and formulated through the
political system, but the latter does not independently
influence policy outcomes. Therefore, linkage between
socio-economic inputs and policy outcomes is unbroken.
Both linkages D1 and D2 and/or unbroken linkage D3
are conceptual notations to analyze how particular policy
outcomes may have come about.
But they do not explain or describe whatever impact
the implementation of that particular policy may have had
i
on various components of the system and the system's
i
environment itself. Simply who got what and how is pre- !
eluded from the analysis suggested by the linkages of Dl,
i
D2, and D3. It is no less important to know what happened |
as the consequence of implementation of a policy than to |
analyze what effected a particular policy decision. j
i
Analytic linkages between a particular policy decision and |
policy impact are viewed as unbroken and continuous by
conceptualizing the impact as the forces engendered by the
implementation of that policy which are fed back into each
region of the system's environment as the feedback loop,
D4, indicates. Thus, these four linkages, i.e., the
hypothesized relationships between economic development and
political system characteristics, between political system
characteristics (Dl) and policy outcomes (D2), between
15
economic development and policy outcomes (D3), between
policy outputs and subsystems (D4), constitute the essence
of the proposed conceptual scheme of this study.
These linkages are, however, the assumptions at
this stage. Rationale of these assumptions is presented in
the sections to follox^ while the variables to be included
in each one of the three subsystems, i.e., economic
development, political system, and policy outcomes, are
enumerated and described.
Economic Development Variables
Economic development as defined here includes four
closely related variables: industrialization, urbanization,|
36
education, and income. Economists provide justification i
for the assumption that economic development influences i
i
political system characteristics and policy outcomes. j
Whether it may be planned or unplanned economy, j
industrialization is said to draw population from rural to j
urban centers because of expanding need for manpower and
traditional "pull" of the big cities.
Thus, industrialization almost invariably entails
36
See the following works: Charles P. Kindleberger,
Economic Development (New York: McGraw-Hill, 1958); Walter
Rostow, Stages of Economic Growth (New York: Norton, 1962);
Walter Krause, Economic Development (San Francisco: Wads
worth, 1961); Harold Groves, Education and Economic Growth
(Washington: National Education Association, 1961);
Benjamin Higgins, Economic Development (New York: Norton,
1959). ---------- ---------------------------------
16
urbanization. Such urbanization of a nation or region
based on expanding industrialization is presumed to produce
surplus wealth for subsequent capital formation which in
turn acts to increase the income of urban dwellers.
As the tempo of industrialization and urbanization
is accelerated, industries recruit their manpower resources
increasingly from the educated sector of the population
because of the requirements of new technology.
Industrial society also entails problems, i.e., ills
of industrial dislocation; crowded dwellings, parks, roads,
emergence of ghetto, and seasonal lay-offs. These social
difficulties all create the need for increasing new demands
upon the political system.
The higher the level of education and income of the
population affected by consequence of increasing industri
alization and urbanization, the more likely they will take
an initiative in placing demands on the political system
to respond.
As these demands are made on the political system,
the citizens in industrial society create their own channel
of expressing and aggregating their demands, i.e., advent
of interest groups, political factions, and alliance. In
short, they result in conditions conducive to political
pluralism.
This political pluralism determines in turn the
;character of the political system because the power holders
in the system have to be receptive to the demands of the
populace to stay in power. In democratic setting, this
means the emergence of two-party or multi-party systems
competing for the votes of various factions of politically
relevant citizen groups.
These linkages in economic development and political
changes are well supported by empirical research. Seymour
37
Lipset found in his studies of Western Europe and Latin
America that economic development variables and rates of
changes in these variables were highly related to stable
democratic government as opposed to unstable government or
38
dictatorships. Daniel Lerner used economic development
variables to explain the stability of political systems in
the Middle East and found that disproportionate development
in economics was related to political instability. Lyle
39
Shannon correlated indices of economic development with
self-government and non self-government and chose that the
latter type of political system is related to economic
underdevelopment.
"^Seymour Lipset, Political Man (Garden City, N. Y.:
Doubleday, Anchor Books, 1963), p. 64.
38
Daniel Lerner, The Passing of Traditional Society
(Glencoe, Illinois: Free Press, 1958), p. 60.
39
Lyle W. Shannon, "Is Level of Development Related
to Capacity for Self-Government?" American Journal of
Economics and Sociology, XVLL (1958), 367-82.
18
While Lipset, Shannon, and Lerner were concerned
with measuring the relationship of economic development
variables with the character of political systems, Phillips
Cutwright4^ explored in his cross-cultural studies the
relationship between system variables and policy outcomes.
In uis survey of seventy-six nations, Cutwright found that
policy outcomes in the area of social security were
strongly related to economic development. Cutwright
reports:
Nations with high levels of economic development but
with less than perfect' (i.e., democratic) political
systems had government activities highly similar to
those undertaken by democratic governments.4-1
Extrapolating these insights offered by Cutwright to
42
the study of American states, Thomas Dye hypothesized in
his study of policy outcomes in fifty American states that
variations in policy outcomes of American states may be
more a product of economic development than the political j
t
system characteristics. In this study, the linkage C, thatj
is, assumption of unbroken linkage between environmental
inputs and policy outcomes, is replication of Dye's formu
lation.
Thus, it appears that industrialization, j
^Cutwright, loc. cit.
41Ibid., p. 548.
42^
Dye, op. cxt.t p. 4.
19
urbanization, income, and education merit inclusion in the
definition of economic development which is conceptualized
as crucial inputs into the system.
Political System Variables
Despite similarities in constitutional framework,
the political characteristics of fifty states are highly
varied. Strength of Republicans and Democrats tends to
vary with region and sections within the same states;
conflict between rural and urban representation, while in
some other states dominance of a strong governor over the
legislature is more of a characteristic than distribution
of party strength. |
As it was with economic environment, so is it neces
sary to reduce highly complex political conditions to a
few variables considered to be most basic and crucial to j
i
the model in this study. Three sets of political variables
are chosen for inclusion in the political system part of
the model. They are inter-party competition, voter turnouti
and malapportionment rate.
/ "5
V. 0. Key first suggested that the degree of party
competitiveness is the variable on which we ought to focus, j
44
Key argued that the higher the competition between the
two parties to compete tor the support of the electorate,
4^Key, op. cit., pp. 298-314.
44Ibid. ___________________________________
20
the greater the probability that the existence of a two-
party system may make a difference in policy outcomes.
Key described how some of the Southern states, e.g.,
Mississippi, a virtually one-party stronghold, tend to
respond to upper socio-economic strata.
In the strong two-party states, Key proposed that
they are likely to respond to the demands of "haves" as
well as "have-nots" not because of different ideologies
but because of the necessity of avoiding retribution at the
poll.
Duane Lockard found that among the six New England
states the two-party states (Massachusetts, Rhode Island, |
!
and Connecticut), contrast to the one-party states (Maine,
New Hampshire, and Vermont), received larger proportions
i
of their revenue from business and death taxes and spent
more on welfare services. !
j
Neither Key nor Lockard, however, conducted any j
systematic studies of the relationship between socio- j
economic variables, party characteristics, and policy out
comes. It was Dawson and Robinson when they reported that
both economic variables and party competitiveness were
1
i
found to be positively related to policy outcomes. But
when they controlled the income or wealth variable constant,
45Ibid., p. 310.
4^Lockard, op. cit., p. 326.
21
party competition no longer appeared to be closely related
to policy outcomes.^
Thus, Dawson and Robinson concluded that economic
development, particularly wealth, is more responsible for
48
policy outcomes than party competitiveness. They
relegated the importance of political system characteris
tics to secondary position. Thomas Dye came to the same
conclusion in his study of fifty states in their respective
policy outcomes in education, welfare, highways, taxation
49
and regulation of public morality.
How definite is this conclusion? Inclusion of party
competitiveness as one variable in the model of this study
is to further investigate the validity of this claim.
The level of voter participation is another
important characteristic of political systems. Do states
with higher voter turnout effect substantially different
policy measures compared with the states with lower voter
participation? Can the differences in policy outcomes,
then, be attributed to the differences in the level of
voter participation?
47
Dawson and Robinson, op. cit., p. 266.
48T, . ,
Ibid.
49
Dye, op. cit., pp. 297-301.
22
Milbrath,*^ Campbell,"^ and Dye“ ^ found that higher
participation in elections is closely related to higher
level of education, income, and urbanism, while non-voting
is rather consistently related to lower level of education,
non-whites, lower social strata, and rural characteristics
of the region.
If these findings hold true, the states with lower
economic development may be likely to have one-party
dominated characteristics which may in turn result in lower
participation of eligible voters in the election. This
lower participation may then have a great deal to do with j
the differences in policy outcomes.
The third variable this study considers is malappor-
53
tionment. Charles Adrian reports that malapportionment
is found to have serious effect on the policy outcomes.
54 !
V. 0. Key also assigned a high value to reappor- j
tionment in upgrading the states' policy-making process.
i
I
i
"^Lester W. Milbrath, Political Participation
(Chicago: Rand McNally, 19657"!
51
Angus Campbell, Phillip E. Converse, W. Miller and
Donald Stokes, The American Voter (New York: Wiley, 1960).
52
Dye, loc. cit.
53
Charles Adrian, State and Local Governments (New
York: McGraw-Hill, 1960), p. 306.
■^Key, op. cit., pp. 265-71.
23 ,
The Commission on Intergovernmental Relations'*'* put
forth its recommendation for reapportionment because it may
make a difference in public policy outcomes in the states.
56 57
Herbert Jacob and Thomas Dy * studied the effect
of apportionment on the policy impact rather recently.
Jacob reports no appreciable differences in party competi
tion, welfare expenditures, highways, between well-
58
apportioned states and malapportioned states. Dye also
reports that relationship between we11-apportioned and
malapportioned is found to be rather insignificant in terms
59
of differences in policy outputs. !
Inclusion of malapportionment in the model is to !
i
explore whether the findings of Jacob and Dye or Adrian
would still hold true in a particular policy outcome this
I
study investigates. |
]
Policy Content and Outcome !
V. 0. Key and Duane Lockard set the early stage for j
i
■^Commission on Intergovernmental Relations, A
Report to the President for Transmittal to Congress
(Washington, D. C.: Government Printing Office, 1955),
p. 546.
****Herbert Jacob, "The Consequences of Malapportion
ment: A Note of Caution," Social Forces, XLIII (1964),
256-261.
"*^Dye, loc. cit.
58
Jacob, op. cit., p. 259.
59
Dye, op. cit., pp. 67-68.--- --------------
24:
the comparative studies in state politics primarily because i
of their suggestion that party competitiveness may have
close relationship with the variations in the states' public
. . . 60
policies.
Key suggested this conclusion without systematically
61
relating his contention with policy outcomes as such.
His was no more than carefully studied description of
Southern states. Lockard was more specific and systematic;
he attempted to compare two-party with one-party states in
New England on such policies as welfare, tax revenues, aid
62
to the blind, the aged, and dependent children. !
Dawson and Robinson dealt exclusively with welfare
to dependent children, the blind, etc. Cutwright studied
63
social security payments in seventy-six nations. Thomas
Dye selected five areas: education, welfare, highways, |
64 '
taxation, and regulation of public morality. j
What is common among these analysts is simultaneous |
i
treatment of several policies, with emphasis on welfare j
policies. There seem to be three reasons for it.
I
^^Key, loc. cit. ; Lockard, loc. cit.
C 1
Key, op. cit., pp. 267-89.
62
Lockard, loc. cit.
63
Dawson and Robinson, op. cit., p. 266.
64
Dye, op. cit., p. 1.
25
One of them is simply availability of the data in
this area under Lhe budget items in a ready-to-compare
form. Secondly, the data are available in a quantified
form because they deal with budgeted figures at a particu
lar given moment. Thirdly, they deal with several policies
en masse because of the rationale that the larger the size
of the sample, the higher the degree of "scientific"
generalization.
As Ranney,^ Froman,^ and Salisbury*^ correctly
argue, they were all concerned with explaining the "process'*
through which the variations in the policy outcomes may j
have been affected. Consequently, they took the policy
outcomes at a certain given moment as the "dependent"
variables to be "explained."
i
They did not consider it their scope of study to
follow through the impact of implementation of those :
policies. They deliberately took a uni-dimensional view of
the policy process because their concern was to focus on
the state of affair that is ex-post facto. They did not
care to examine the impact of the content of a particular
65
Ranney, loc. cit.
66
Froman, loc. cit.
^Robert H. Salisbury, "The Analysis of Public
Policy: a Search for Theories and Roles," in Austin Ranney
(ed.), Political Science and Public Policy (Chicago:
Markham^ 1967), pp. 151-78.
26
class of policies on the system.
68
In this study, it is proposed (as Ranney and
Salisbury^ do) that the linkage from Dl is not to be
broken off at D2 or D3 but to be conceptualized as unbroken
and continuous extending through D4. The impact of policy
outputs is proposed to be viewed in a form of feedback
completing its circular loop.
What kinds of impact or effects could the feedback
have brought about? Of course, it is important to account
for how the variations in amount of expenditures of a group ;
of policies may have come about taking a given set of j
budget figures as the data at a given time. But it is
i
equally important to account for what happened after those i
I
resources were expended, i.e., impact of the implementation;
of a policy.
i
This is because some policies may have caused an j
I
impact on the environment as well as political system in
such a way that the latter may have been altered. Or some i
other policies may not have brought any significant impact
on the system at all. That is, the effect may have been
nil as far as alteration of the system is concerned.
Accurate knowledge of impact or "content" may also
give us a better handle over the future direction of the
68
Ranney, op. cit., p. 14.
69
Salisbury, op. cit., pp. 164-70.
27
policies. It may permit us to assess the effectiveness of j
the policy in terms of meeting its implicit or explicit
objectives. Or how good has the policy been to the accom
plishment of its intent?
This study incorporates this "content" noti n into
the model by dealing with a single policy, instead of an
assortment of several policies.
Reasons for opting for a single policy are as
follows: (1) a single policy may lend itself more manageably
to the search for impact than a simultaneous attempt at
several policies, (2) the results obtained under more
manageable conditions may yield higher accuracy and reli
ability, (3) it may avoid inflicting injustice to the
importance of an individual policy as it might be the case
i
with treatment of several policies, and yet, (4) this singl^
policy can still be subject to Dawson or Dye's process j
analysis. j
That single policy to be dealt with in this study is j
the Medicaid Program as it has been adopted in various
forms by thirty-eight states in the Union as of the end of
1968.
Medicaid
Medicaid is state-administered health program
effected since 1966 as a result of amendments to original
70 !
Social Security Act of 1935. It is set up within the
framework of Federal-State assistance programs. This pro
gram is unlike earlier programs in several respects: (1) it
stipulates minimum benefit standards for state plans which
receive Federal support, (2) it enlarges the share of
Federal funds in approved state programs and offers Federal
aid to groups of persons--poor and medically poor--those
not previously aided, (3) it complements health insurance
provisions of Medicare by paying the deductible amounts for
needy, aged persons who are insured, and (4) it supplements
the insurance program by providing services not offered
71
through Medicare, and aid to persons under age sixty-five.
The states were given fairly wide latitude in
responding to this new Federal legislation. Each state was
I
given (1) the freedom to adopt it or not adopt, (2) once |
i
adopted, the freedom to decide on eligibility after meetingj
the minimum requirement as is legislated in Title 19, (3)
discretionary power as to the scope of coverage after |
meeting mandatory minimum requirement of the law, and (4)
i
U. S. Congress, House, Committee on Ways and
Means, Hearings, Presidents Proposals for Revision in the
Social Security System, 90th Cong., 1st Sess., 1967, pp.
^Advisory Commission on Intergovernmental Relations,
Intergovernmental Problems in Medicaid: A Commission
Report (Washington, D. C.: Government Printing Office,
1968), pp. 1-12.
29
72
mode of administration of the program.
Given such a wide discretionary power at the state
level, there was a great deal of speculation and uncer
tainty as to what the states might choose to do in response.
Would they implement the new law and set up Medicaid
programs? How rapidly would they act? How "liberal" would
they be in extending aid to large segments of the popula
tion, and how generous in the range of medical services
they would make generally available?
Two and one-half years have now elapsed since Medic
aid became effective, and the decisions and experience of
thirty-seven states to date provides data to some of these
questions. There are a great deal of variations in each
state’s decisions on eligibility, coverage, and degree of
generosity in extending the medical services to the state’s|
j
poor. These empirical data are considered as the concrete
i
policy outcomes in each state.
One major element of this investigation is to explain
why these state-to-state variations in Medicaid provisions
have come about. A second aspect of the problem is the
j
impact of Medicaid policy-implementation for the last two j
and one-half years.
Policy impacts are conceptualized as the forces
engendered by the implementation of a policy which are fed
Ibid.
30
back into the system's environment as the feedback loop D4
completes its circulation. These forces are viewed to have
generated new demand and support for the system.
The system is viewed as having responded to these
new inputs. The responses to the original policy are in
turn viewed as the value commitments of the political
system in terms of adjustments to new knowledge and learn
ing made possible by the feedback.
In summary, then, the conceptual model for this
study of public policy includes the following variables:
i
Environmental and economic inputs consisting in '
industrialization, income, education, and urban-1
ization;
Political system characteristics in terms of I
inter-party competitiveness, voter participa
tion, and malapportionment;
Policy outcome and content in the current status
l
I
of Medicaid program adopted by thirty-eight
states in the Union.
|
Thus, the full version of model containing all of the above j
variables in each subsystem like the following:
(1)
( 2)
(3)
31
Feedback D4 (Impact flow)
S -----------
Environmental
Input s
Dl
Political System
D2
V
Policy Output
Urbanization
Income
Industrial
ization
Education
Party Competition
Voter turnout
Malapportionment
Medicaid
Program in
thirty-eight
states
f /
Policy Process Policy Content
D3 ---)-----------------------
Figure 3
Statement of the Problem
The primary objective of this dissertation is to
investigate and test the following hypotheses derived from
I
the explanatory linkages in the model against the empirical j
data.
First of all, the model posits that economic develop-!
ment helps to shape the character of state political system
(linkage Dl in the model). This is converted into a hypo
thesis :
Hypothesis I. The higher the level of economic
development of a state, the higher the degree of
variations in its political system characteristics.
With the given variables in the environmental and
political subsystems, this first hypothesis may permit us
to explore such questions as: Is a certain state more
32 ,
competitive politically if it shows a higher per capita j
income or industrialization or urbanization? Or how much
relationship would urbanization have with malapportionment
rate? With voter turnout in elections? In general, the
hypothesis may permit us to say something about the deter
minants of political system characteristics in American
states. It suggests, also, that political system charac
teristics help to determine the policy outcomes in the
states (D2 in the model). In hypothesis:
Hypothesis II. The higher the degree of variation
in the political system characteristics in a state, j
the higher the variations in a states' policy out-
i
comes. ;
This second hypothesis will explore such questions
I
as: Does a two-party state tend to be more generous in
i
aiding medically indigent persons than a one-party state? j
Is a state with high political participation likely to j
allocate a higher portion of its resources to medical assis-j
tance program than state with low participation? Would
malapportionment rate of a state have any relationship with
lower medical assistance payment to its poor citizens?
The model also raises the question of whether or not
the economic conditions in the states independently influ
ence the policy outcomes (D3 in the model). Stated in
hypothesis:
33
Hypothesis III. The higher the level of economic
development of a state, the higher the degree of
variations in policy outcomes in a state, regardless
of the variations in the political system character
istics .
If this third hypothesis turns out to be tenable in
the light of empirical data, the result may permit us to
say something about the role of political system in the
policy outcomes. Does it really then support the notion
that the political system is merely an agent for mediating
the demands made upon it from the environmental pressure?
Is the political system itself impotent to take any initi
ative unless economic conditions are ready to exert
pressures on it? How does it challenge the concept of
democratic form of government?
Lastly, the model suggests that policy studies ought
to extend beyond merely explaining the process through
which the variations in certain policy outcomes may have
come about. It posits that the implementation of a policy
has a direct impact upon the political system and environ
ment itself (D4 in the model). Instead of converting this
assumption into a hypothesis, this study attempts to inves
tigate the "impact" in terms of significant decisions and
events that took place after the policy went into effect.
34
Limitations of this Study
There are two evident limitations in this study:
variables and methodology.
The study takes the single policy approach, that is,
taking various aspects of the Medicaid program as the main
policy outcomes to be investigated. This approach is weak
from the standpoint of process approach to the study of
public policy. Because the process approach is an attempt
to generalize the role of environmental conditions and
political system in making public policies.
Thus, within the theoretical framework of a process-
oriented approach, the sample of policy outcomes rather
than a single policy is more desirable as a basis for infer
ring generalizations about the American state political
system.
This study takes modest issue with the process
approach, though the bulk of this dissertation is based on
it. The issue is that the process approach neglects to
follow through the impact of policy once implementation is !
I
underway. j
The single policy approach came to be a choice !
because from the outset this study put a rather high j
i
priority on incorporating the need for knowing the impact J
of a policy outcome into the model. Perhaps the single j
policy, as this study opted, is a poor strategic choice. j
!
However, in the absence of any empirical research !
35
example of policy studies within the context of "content"
73
approach, the single policy approach was judged to be the
best possible compromise. This is simply because a single
policy lends itself more easily to the analysis of the
impact than several policies under the circumstance.
One of the important weaknesses or defects in the
variables included in the model is the absence of bureau
cracy or some administrative variables in the political
system. It has been rather well established in the litera
ture of administration that bureaucracy is an important
dimension in initiating and processing public policies.
Notwithstanding the importance of bureaucracy, it
was impossible to gather data on thirty-seven state bureau
cracies with limited time and resources.
Anoeher limitation of this study is in the nature of
its research. Since it is ex-post facto analysis, manipu
lation of the variables was not possible. Consequently,
this study included no way to detect extraneous effects
which variables other than those included in the model may
73
Refer to Chapter II on review of the literature in
which the lack of empirical research applying "content"
approach to the study of public policy is spelled out in
detail. Since Lowi's suggestion to take the policy as an
independent variable instead of "dependent" variable,
nothing very much has been done in terms of implementing
Lowi's ideas, provocative as it was. Reasons are also
analyzed in the review of literature.
36
have had on dependent variables, i.e., policy outcomes. |
Bureaucracy may have been such a variable.
Yet another limitation of this study is in the
nature of the data. Data gathered for the purpose of this
study were largely drawn from secondary rather than original
sources. This should be recognized in evaluating and
interpreting the data in any great depth. Analysis of the
impact of Medicaid presented the dilemma whether to be
descriptive or prescriptive. The notion of impact implied
in the "content" approach to the study of public policies
is to be descriptive rather than prescriptive, however. i
j
Because the purpose of the "content" approach is to j
i
bring out the effect of implementation of policies on j
various system parameters, evaluation of effects is con
sidered outside the scope of the study. Depending upon
the value preference of readers of this paper, absence of !
I
prescription may be considered a limitation.
Nevertheless, the importance of this study is not
to be overlooked in terms of its implication to the field
of comparative studies of politics.
Importance of the Study
Much has been written about the value of comparative
74
studies in political science. Aristotle was certainly a
^See Gabriel Almond, et al., "A Suggested Research
Strategy in Western European Government and Politics,"
37
student of comparative analysis in politics. Aristotle's
description and classification of the constitutions of
Greek city-states was founded, in spirit, in empirical com
parative analysis. ^
This Aristotelian tradition was not materialized into
modern empirical analysis of comparative politics until the
76
1950s, however. In 1953, David Easton pleaded for more
empirically grounded political theories by looking at
foreign governments primarily as the sources of data to
test broad general hypotheses.
Two years after Easton's plea, in 1955, Roy Macridis j
!
attacked the traditional approach to comparative studies forf
its failure to take the opportunities afforded it to con
tribute to empirical theory.^ In criticism, Macridis
offered his view on the maxim of comparative studies: !
American Political Science Review, XL1X (1955), 1042-49;
Dankwart Rustow, "New Horizons for Comparative Politics,"
World Politics, IX (1957), 530-49; Harry Eckstein and David
Apter (eds.), Comparative Politics: A Reader (New York:
Wiley, 1963); Lucian Pye, "The non-Western Political Pro
cess," Journal of Politics, XX (1958), 468-86; Arthur
Kallebert, "The Logic of Comparison: A Methodological Note
on the Comparative Study of Political Systems," World
Politics. XIX (1966), 69-82.
^Ernest Baker, The Politics of Aristotle (New York:
Oxford University Press^ 1946), p. 115.
^David Easton, The Political System (New York:
Wiley, 1953).
^Roy C. Macridis, The Study of Comparative Govern
ment (Garden City, N. Y.: Doubleday, 1955).
38
. . . comparative study entails the comparison of
variables against a background of uniformity either
actual or analytical for the purpose of discovering
causal factors that account for variations.^8
Fifty American states ideally fit into Macridis'
criterion of comparative study. All of the states display
wide variations among important socio-economic characteris
tics wherein state politics takes place, but show a great
deal of similarities in written constitutions, political
structure, and national culture.
In spite of such a great potential for comparative
treatment of American state politics, it was only after 195(j>
i
when state governments and politics began to be studied
within the framework of comparative studies.
79
A beginning was made when V. 0. Key published his
much celebrated Southern Politics in 1951. Eight years j
80
after, in 1959, Duane Lockard published his work, New i
England State Politics under the aegis of V. 0. Key. How
ever, the works of Key and Lockard fell short of fully !
exploiting the potential of resources offered by the j
American states. !
81 !
It was Richard Dawson and James Robinson who j
1
78Ibid., p. 2. |
79
Key, loc. cit.
8^Lockard, loc. cit.
81
Dawson and Robinson, loc. cit.
39:
studied the policy outcomes of fifty states in the spirit
and form of comparative study suggested by Roy Macridis.
Dawson and Robinson held the policy outcomes at a given
point of time constant against the variations in economic
variables and inter-party competitiveness.
Subsequently, Jacob,^ HofferbertDye,®^ and
85
Sharkansky followed suit. There is a common theme
running through all of these comparative analysts of
i
American state politics. That is, to account for the
variations in policy consequences among the states relative
to economic and political variables. The purpose of it is
to build a body of generalizations regarding some critical ;
I
features of American democracy, e.g., role of political
party, political participation, apportionment.
This study is, in part, replication of these tradi
tions already established by V. 0. Key, Dawson, Dye and
others. As it was indicated previously, the study attempts
to re-examine some of the claims made by the existing
82
Jacob, loc. cit.
8 3
Richard I. Hofferbert, "The Relation Between
Public Policy and Some Structural and Environmental Vari
ables in the American States," American Political Science
Review, LX (1966), 73-82.
84
Dye, loc. cit.
85
Ira Sharkansky and R. Hofferbert, "Dimensions of
State Politics, Economics, and Public Policy," American
Political Science Review. LXIII (1969), 867-879.
40
studies in accounting for the variations of policy outcomes I
in the states.
Implication of the comparative studies of American
state politics does not end, however, with the American
political scene. Findings in this area challenge the very
assumptions of democratic precepts, i.e., the representa
tive character of democracy.
Consensus of existing literature in the field is that
economic development is the independent determinant of both
political system and policy outcome. Evidence supports the
86
notion of functional prerequisites, especially economics, 1
i
to the emergence of democratic systems of government.
This is, in a way, revival of an old theme of eco
nomic determinism. What does it mean to the emerging
nations in looking up to American democracy as a source of j
87 i
inspiration? As William Mitchell suggests in his recent j
I
study of American politics, does it drive us to take a hard
look at the old tradition of political economy which was
O r
There has been a number of works arguing for func
tional prerequisites to political development; see
Alexander Gerschenkron, Economic Backwardness in Historical
Perspective (Cambridge: Belknap Press, 1962); Albert
Hirschman, Journey Toward Progress (Englewood: Prentice-
Hall, 1965); S. N. Eisenstadt, Modernization: Protest and
Change (New York: Wiley, 1966).
87
William C. Mitchell and Joyce Mitchell, Political
Analysis and Public Policy: An Introduction to Political
Science (Chicago: Rand McNally, 1969), pp. 1-15.
41
prevalent in the nineteenth century?
Thus, the importance of this study may extend beyond
the confines of the American political system, in addition
to a specific contribution to the growing body of compara
tive data on American state politics.
Organization of the Remainder of this Study
This first chapter was introductory presenting the
setting and nature of the model to be tested in this study.
The specific problem of this study was to investigate a
set of hypotheses derived from the proposed model. The
i
J
second chapter reviews literature related to the develop
ment of comparative studies of public policies in American j
state politics.
The third chapter describes the research methods and;
procedures followed to obtain the data and test the hypo
theses in the model. The fourth chapter presents the j
|
analysis of findings on the relationship between economic
j
development and political system characteristics.
The fifth chapter presents the findings regarding
the interface of policy outputs in Medicaid with socio
economic measures and political system characteristics.
The sixth chapter deals with the impact of Medicaid pre
senting the data on the effect of two and one-half years of
implementation of the program. The last chapter discusses
the findings under concluding summaries and suggestions.
CHAPTER II
REVIEW OF THE LITERATURE
This second chapter reviews the current literature
related to comparative studies of American states. The
first part of the chapter presents the pertinent works in
chronological sequence; the second part is devoted to sum
mary and critique.
State Policy Studies
State policy studies begin with V. 0. Key's classic,
Southern Politics^ which set its avowed aim to be "compara
tive," i.e., to compare political system characteristics
and policy outcomes of one state with others to arrive at
some generalizations on the American political system. j
This task has been carried on primarily around three :
basic questions: '
1. What are the roles of such political institutions ;
as political parties and legislatures in formulating
and processing public policies? Is there any common
pattern in this regard among different states?
I
2. How may differences in policy outcomes in various
states be explained? Are they because of
■ * " V . 0. Key, Southern Politics in State and Nation
(New York: Knopf, 1951).
_______________________42_________ ____________________
43
differences in political system characteristics or
something else? If anything else, what?
3. Couldn't the policy(ies) itself be an independent
variable in making differences in political system
characteristics in the states, instead of the latter
having an impact on the former?
2
In his seminal work, Key set the stage for state
policy studies by reporting results of his examination of
relationships between political system characteristics and
3
policy outcomes in eleven Southern states.
The purpose of the study was to test the hypothesis 1
that "the best government results when there is free and
vigorous competition at the ballot box in contests in which
genuine issues are defined and candidates take a stand.
Over-all results of Key's study as reported confirmed this i
assumption. j
Key found that one-party states characterized by
multifactional politics tended to be especially conservative
in their public policies. Key reports that the reason for
this is that one-party systems do not provide adequate
institutional mechanisms for the expression of interests j
and demands of lower socio-economic elements.
2Ibid.
3Ibid.
^Ibid., p. 2.
44
The factional system simply provides no institution
alized mechanism for the expression of lower-bracket
viewpoints. By chance and by exertions of temporary
leaders and connivers, candidates are brought into the
field, but no continuing, competitive groups carry on
the battle.^
Thus, Key's contention is that more "liberal" poli
cies result from competitive political environments engen
dered by a virile two-party system. This isn't because of
the different ideological or philosophical stance of each
party but because of competition generated at the ballot
box. Key states that:
The great virtue of the two-party system is not i
that there are two groups with conflicting policy
tendencies from which the voters can choose, but that
there are two groups of politicians.6
He continues,
The fluidity of the factional system handicaps the
formation of two such groups within the Southern Demo
cratic party, and the inevitable result is that there
is no continuing group of "outs" which of necessity
must pick up whatever issue is at hand to belabor the
"ins."7
Key's hypothesis was more explicitly formulated and j
i
8 ■
investigated in Duane Lockard's study of New England state j
I
!
~*Ibid., p. 337. I
^Ibid., p. 336.
^Ibid., p. 338.
8
Duane Lockard, New England State Politics (Prince
ton: Princeton University Press, 1959).
politics. Building explicitly upon Key's work, Lockard j
sought to test more systematically the impact of party
competition upon policy outcomes, using the New England
states as the analytical units.
Q
Lockard reports that non-competitive New England
states (Maine, New Hampshire, and Vermont) place more tax
burden on lower income groups and provide less in services
for needy people than do the more competitive states
(Connecticut, Massachusetts, and Rhode Island).
Thus, Lockard came to the same conclusion as Key did
that "more competitive states are, more 'liberal' in their
i
public policies.”
Lockard concludes that
The political leaders in a two-party system are more
responsible to the preferences of the citizens, whether;
of high or low status in life, not so much out of ^
principles as out of fear of retribution at the polls, j
In 1963, Richard E. Dawson and James Robinson^- i
set out to test the hypothesis proposed by Key and Lockard
i
in a more systematic and rigorous way. Dawson and Robinson;
drew an important insight into the party-competitiveness
from the implications of Key and Lockard's findings.
Ibid., pp. 326-40.
10Ibid., p. 333.
■^Richard Dawson and James Robinson, "Inter-Party
Competition, Economic Variables and Welfare Policies in the
American States," Journal of Politics, XXV (1963), 265-289.
46
Both Key and Lockard found that one-party states are
rather consistently associated with less industrialized,
less urbanized, and lower per capita income states, while
two-party states are associated with highly industrialized
urban states.
On the basis of this finding, Dawson and Robinson
reasoned that "the level of competitiveness is most likely
related to various socio-economic conditions within the
12
polity in which it functions." They reasoned further
that existence of highly varying socio-economic interests
may foster a social diversity which may, in turn, create
the conditions conducive to the two-party system to aggre
gate widely varying interests.
Thus, Dawson and Robinson constructed three sets of
variables, i.e., indices of inter-party competitiveness,
socio-economic variables, and welfare policy outputs for
forty-six states. Then, correlations between these sets of
variables were computed.
In keeping with the propositions advanced by Key and
Lockard, Dawson and Robinson found that "the more competi
tive states tended to be more liberal in the adoption of
welfare programs, and also that the more highly industrial
ized, more urbanized, and especially the wealthier states
12Ibid., p. 404.
47
13
tended to be more liberal in welfare policies." All three
of the factors were found to be significantly correlated
with each other.
However, in order to examine the relative impact of
wealth and inter-party competition upon welfare policies,
they computed the correlation between per capita income and
policy outputs with inter-party competition controlled and
also between inter-party competition and policy outputs
with per capita income controlled.
Results were surprising: the relationships between
wealth and welfare policies were found to be substantially
higher than those between inter-party competition and
welfare outputs. On the basis of this finding, Dawson and
Robinson challenged Key and Lockard's proposition arguing
that policy outcomes are more a function of socio-economic
factors especially per capita income, wealth, than inter
party competition.
The authors concluded that "inter-party competition
does not play as influential a role in determining the j
scope of welfare policies as earlier studies suggested. j
j
The level of public social welfare programs in American j
j
states seems to be more a function of socio-economic |
factors, especially per capita income."*4
13Ibid., p. 409.
14Ibid.
48
In 1966, Thomas Dye^ proceeded to test the proposi
tion advanced by Dawson and Robinson more extensively.
Dye also constructed three sets of variables, i.e., poli
tical system characteristics (measured by rate of voter
participation, public office turnover, malapportionment),
socio-economic development variables (industrialization,
urbanization, education, income), and five of what Dye
considered as "the most important areas of state politics,
viz., education, health and welfare, highways, taxation
and revenue, and the regulation of public morality."^ ,
Dye employed the same correlational, techniques as |
Dawson and Robinson did to measure the relationship between ,
these three sets of variables throughout the fifty states
in the Union. Dye's conclusion generally confirms the
Dawson proposition that political party characteristics do |
l
not play influential a role in shaping or affecting the !
policy outcomes. Dye concludes that "the principal vari- j
ables with which we have been concerned traditionally, i.e.,
inter-party competitiveness, political participation,
public office turnover, do not explain the variations in
17
state policy outputs." Socio-economic variables are
i
■^Thomas Dye, Politics, Economics and the Public
(Chicago: Rand McNally, 1966).
16Ibid., p. 1.
17Ibid., p. 300.
49
responsible for the variations in policy outputs.
But Dye put forth an additional conclusion that the
socio-economic variables are also responsible for the dif
ferences in states' political system characteristics which
seems, however, to have no independent bearing upon the
policy outcomes in general. Thus, according to Dye,
In short, economic development affects political
system characteristics as well as policy outcomes.
Thus, the proposition reached by Dawson, Robinson,
19
Dye and others who followed suit is that socio-economic
development is responsible for both policy outputs and the
party development and its related political characteristics
in American states. It was only until recently that the
proposition reached by Dawson, Dye, and Robinson began to be
challenged.
20
Those who challenge, like Froman, for example,
argue not on the merit of the conclusion they reach but on
the consequence of approach they use. Froman says that
18Ibid., p. 285.
19
See, for example, Duncan MacRae, "The relation
between Roll Call Votes and Constituencies in the Massachu
setts House of Representatives," American Political Science
Review, XLVI (1952), 1046-55; Austin Ranney and Willmoore,
"The American Party System," American Political Science
Review, XLVIII (1954), 477-85; Lester Milbrath, Political
Participation (Chicago: Rand McNally, 1965).
20
Lewis A. Froman, "The Categorization of Policy
Contents," Austin Ranney (ed.), Public Policy and Political
Science (Chicago: Markham, 1968), p. 42.
50
these people typically take the given "governmental outputs"
(e.g., welfare, education, public assistance, highway) as
the "dependent" variable to be "explained" by the analysis
of process of institutional structures (e.g., party, appor
tionment, voter participation) and environmental character
istics (e.g., urbanization, industrialization, etc.).
Froman argues that this approach isn't really doing
very much in terms of increasing either the level of know
ledge or theoretical sophistication in policy studies.
Froman asks other policy scholars to heed seriously the
message of Theodore Lowi who suggests to reverse "the i
21
orthodox explanatory relationships."
22 23 ■
Lowi and Froman propose to "consider policy
content or output as the independent variable, instead of
dependent, and observe the differential impact of different '
24- 2 3 i
contents on policy-making processes." Lowi proposes
I
that there are three major categories of public policies, ;
namely, "distributive" as something setting a definite
21
Theodore Lowi, "American Business, Public Policy,
Case-Studies, and Political Theory," World Politics, XVI
(1964), 711. |
22
Ibid.
23
Froman, loc. cit.
24 . .
Austin Ranney, The Study of Policy Content: A
Framework for Choice," Austin Ranney (ed.), op. cit., p. 14.
25
Lowi, op. cit., p. 720.
51
beneficiary(ies), "redistributive,” i.e., taking from some
group and giving another, and "regulative” as something
restraining to provide a rule of game (e.g., licensing
procedures, etc.).
Lowi argues that each arena may tend to develop its
own characteristic political structure, political process,
26 2 7
elites, and group relations. Froman suggests two cate
gories, i.e., "constitutional" referring to policies setting
the framework for subsequent action and "substantive"
referring to those policies calling for a concrete action
28 t
or impact. Salisbury adds what he calls "self-regulative”
to Lowi's three categories of policies.
i
Froman and Salisbury argue that the categories they
respectively propose would disclose the political and its
related structures which otherwise may have been hidden
under the broad brush of Key, Dawson, Robinson, and Dye.
There is not a single work, however, which applied
this content approach advanced by Lowi, Froman or Salisbury ;
to empirical research.
There is a good reason for this which will be dis
cussed in the following section of the proposal. Those who
i
....■I.— — I , ■ I. — — ... - - — - M . |
26
Ranney, op. cit., p. 14.
27
Froman, op. cit., p. 47.
28
Robert Salisbury, "The Analysis of Public Policy:
A Search for Theories and Roles," Ranney (ed.), op. cit.,
pp. 151-78.
52
produced any empirical work employing this "content"
approach have done it in their own peculiar way. Vincent
29
Ostrom measures the impact of water resource policies in
terms of what he calls "political cost and benefit" based
30
on James Buchanan's model of "calculus of consent."
31
Wildavsky proposes to employ the cost and benefit
analysis (as Ostrom does) in terms of various political
system parameters, such as decreases in social hostility
and instability, political supports, interest group coali
tion pattern, and opportunity costs incurred.
32
Wayne Francis examines the impact of eight policy I
measures designated as the "most important" by 860 state
I
legislators responding to his questionnaire on legislative
decision-making structure in terms of (1) party caucus,
(2) floor, (3) substantive committee, (4) pre-legislative !
session, (5) policy committee, and (6) governor's office. j
Neither Ostrom-Wildavsky's nor Francis' way of J
29
Vincent Ostrom, "Water Resource Development: Some
Problems in Economic and Political Analysis of Public
Policy," Ranney (ed.), op. cit., pp. 123-50.
30
James Buchanan and Gordon Tullock, The Calculus of
Consent (Ann Arbor: University of Michigan Press, 1962).
31
Aaron Wildavsky, "The Political Economy Efficiency:
Cost Benefit Analysis, Systems Analysis, and Program Budget
ing," Public Administration Review, XXVI (1966), 292-310.
32
Wayne L. Francis, Legislative Issues in the Fifty
States: A Comparative Analysis (Chicago: Rand McNally,
; t967t:
53
application of content approach is even close to the policy
categories proposed by Lowi, Froman, and Salisbury. How
ever, the point they all push forward is important. That
is, "to use policy contents as independent variables to
learn more about policy outcomes, i.e., the impact of public
policies on the political system's environment and on the
33
system itself." As it previously was pointed out, this
plea for "content" away from "process" is implicitly moti
vated for scientific as well as professional or pragmatic
reason. Scientifically it has a prospect of raising the
level of generality of policy "theories" by employing j
abstract category(ies) of policies, such as Lowi's,
i
Froman's, or Salisbury's instead of pursuing nominal policy
categories (e.g., welfare, highway, etc.).
j
Pragmatically it may equip the political scientists !
i
as a professional with better tool to advise and recommend j
the policy changes in the '70s. This pragmatic reason
looms large in the decision announced by Social Science i
Research Council's Committee on Governmental and Legal
Processes in 1965 to launch public policy studies in a new
direction. The decision or consensus announced by the Com-
34
mittee was that:
33
Ranney, op. cit., p. 14.
34Ibid., p. 7.
54
. . . the most urgent and timely question now within
the committee's purview is: What professional expertise
and obligations, if any, have political scientists to
study, evaluate, and make recommendations about the
contents of public policies?
Summary and Critique
Summary
Over the two decades of scholarly effort beginning
from V. 0. Key's work of 1949, studies on state policy
progressed through three stages of development. First
stage was set by Key and Lockard.
Their purpose was to test the hypothesis that poli
tical institutional characteristics, especially political
party, in a state makes difference in the quality of the
state's public policy outcomes. They took the political
system characteristics as the variable affecting indepen
dently the outcomes of the state's public policy. Thus,
the policy outcomes are considered "dependent" upon the
functions of political party system.
This theoretical stance may properly be termed as
that of normativist. Because the very nature of their
hypothesis is embedded in the value of political party that
it serves as a linkage between the government and citizenry
pulling together the varied interests and demands of indi
viduals and groups and attempting to shape and integrate
them into some form of public policy.
The second stage of development emerged as the______
55
consequence of Dawson and Robinson's important finding that
socio-economic development is more responsible for the
variations in states' policy outcomes than inter-party
competitiveness or other related political characteristics.
This finding was reinforced by Dye's more definitive con
clusion that socio-economic development affects the states'
policy outputs as well as party development and its related
characteristics.
They effectively challenged the notion that the
political system characteristics affect independently the
states' policy outcomes. They rather turned around the
proposition proposing that both political system and policy
outcomes in the states are "dependent" upon the rate of
socio-economic development.
This second group may be termed as the environmen
talists because their concern is to account for the impact j
of socio-economic environmental characteristics upon the 1
I
political system and policy outputs in the states.
The third development as it is represented by Lowi,
Ranney, Wildavsky, Ostrom, and Francis is the recent idea
to focus on the policy outputs or contents as the indepen- j
dent variable. The purpose is to observe how different !
policies affect the system characteristics and its environ
ment as a means of enriching our knowledge of policy process
through the systematic examination of the impact of the
policies.
56
The reason for this switch of emphasis from process
to content is the dissatisfaction with the process-oriented
model which tend to gloss over the differential impact of
the policy contents upon the decision-making processes and
tend to make the political scientists as a professional
stay away from policy advising and recommending function.
Thus, this last group may be termed as the pragma
tists because of their explicit emphasis on the accurate
knowledge of policy impacts--who gets what and how as the
consequence of implementation of the public policies. The
three stages of development is summarized in a schematic
form below:
Sub-System Variables
Approach Socio-
Economic
Political
System
Policy
Outcome
Normativist Independent--•♦Dependent
Environment
alist
♦Dependent m a e p e n - ■
dent - ------- - ^Dependent
Pragmatist Dependent *---------------
Dependent f--•-Independent
Figure 4
Critique
Both normativists and environmentalists depended for
57
their development upon the test of a single major hypo
thesis that:
The higher the level of variations in the political
and environmental system characteristics, the more
"liberal1' or (varied) a state's policy outputs.
For the purpose of this study critique will be
directed at two major variables, i.e., the notion of "liber
ality" as a measurement of variation and that of "policy."
What Key, Lockard, Dawson, Robinson, and Dye mean by
"policy" is very simple one: they all refer to the objects
of allocation of public expenditures in a state. Such
things as education, highway,— items that come under the
purview of budgetary decisions--are referred to as "poli
cies."
The term "liberal" refers to the comparative dif
ferences in the amount of dollars allocated by each state
to one or a group of these expenditure items. The greater i
the amount of dollars a state X allocated to policy Y, the ;
more "liberal" that state is with respect to that policy
Y relative to all other states enforcing the same policy Y.
(
There are all the virtues of simplicity in this
i
definition, but three considerations make it less than j
satisfactory. j
35
This hypothesis is my own synthesis drawn from
Dawson, Dye, and Key's findings as each of these works was
cited above. See Dawson, op. cit.. particularly, p. 314
and Dye, op. cit.. p. 287; Key, op. cit., p. 327.
58
First, it has to assume the uniform existence of
policy Y or any other policy in question in all of the
states under the survey, because the objective is to explain
the state-to-state variations in the amount of dollars
allocated to that particular policy(ies).
Consequently, it has to leave some other important
policies out of consideration just because they are not
found to be uniformly in existence in that group of states
at a given point of time. For example, the policies on
francise, racial integration, or equal opportunity are more
significant measurement of "liberality" of a state than
welfare, health, or highway as these are the policies j
considered by Key, Dawson, and Dye. I
The former policies are not considered in Key, j
Dawson, and Dye's studies simply because they are not found}
I
to be adopted in all of the states they surveyed. The j
I
latter group of policies were selected for the study becaus^
they were uniformly adopted by all of those states and
shown in budget items, although they are less significant
measures of "liberality."
Secondly, when one adds time dimension to the obser
vation, one may find that a certain policy which may have
been found to be non-existent in a certain state at one
point may have been later adopted. Thus, instead of look
ing at things as a straight dichotomy between existence and
non-existence at a certain point of time, one may have to
59
look at them as continuous variables observing such things
36
as "speed and facility of adoption" over a period of
time.
Thirdly, concern with expenditure variations results
in selective treatment of the policies in some other way;
it tends to consider only those policies highly amenable
to quantification. This tendency also unduly restricts the
range of policies to be considered.
For instance, Dawson and Robinson deals exclusively
with welfare policies in the fifty states of the Union for
two reasons: (a) they are found to be in existence in one
form or another in all of the fifty states and (b) they are
!
shown in easy-to-compare budget figures in each state.
Dye's selection of five basic policy areas--educa-
tion, health and welfare, taxation and revenue, highway,
and public regulation--is an improvement over that of
i
i
Dawson and Robinson. Nevertheless, it appears that Dye
also selected these categories of policies primarily be-
i
cause of their malleability to quantitative variations. j
i
Regardless of the motive of these policy analysts in j
selecting the policies and the forms of data for analysis, |
however, one may raise another related question: Why should
the policies be associated with such things as education,
36
Salisbury, op. cit., p. 125.
60
welfare, or highways at all? Couldn't the policy be some
thing else?
37
David Easton says that policy is "the authorita
tive allocation of values for the whole society," distin
guishing the process from the output. Berelson, Lazars-
38
feld, and McPhee attempt to distinguish "specific and
concrete" actions from "abstract" and "general" policies.
39
Lasswell and Kaplan define policy as a projected pro
grams of goal values and practices."
Related to Lasswell and Kaplan's view is the con
ception that political behavior is understood in terms of
its interest or purposive orientation. Carl Friedrich says
that policy is:
. . . a proposed course of action of a person, group or
government within a given environment providing ob
stacles and opportunities which the policy was proposed
to utilize and overcome in an effort to reach a goal or
realize an objective or a purpose. 0
What emerges to be important from these views of
"policy" is that "there is a difference between specific
37
David Easton, A Systems Analysis of Political Life
(New York: John Wiley, 1965), p. 120.
38
Bernard R. Berelson, etal., Voting (Chicago:
University of Chicago Press, 1954), p. 109.
39
Harold D. Lasswell and Abraham Kaplan, Power and
Society (New Haven: Yale University Press, 1950), p. 71.
^Carl Friedrich, Man and His Government (New York:
McGraw-Hill, 1963), p. 79.
61
decisions or actions and a program or course of action,
and that it is the latter to which the term 'policy' re-
t . ..41
fers.
That is, it is "patterns of behavior rather than
i»42
separate, discrete acts which constitute policy.
Defining "policy" as a discrete object of expenditure does
not do very much to explaining this pattern of behavior.
Rather it may aid in analyzing a specific act or conse
quence of it within the limited range of given policy
"label."
Explanation based on nominal label of policies as
they are given in the real world may, thus, have a limited
consequence for generalizing policy-making behavior in a
polity. One may have to search for more abstract typolo- :
gies of policies "without regard to the names the real '
/ 3 i
world assigns to its activities" to raise the level of
theoretical generality.
The analysis based on nominal categories is the s
j
consequence of our hitherto held notion that the policy is
something to be explained, that is, to take it as the
"dependent" variable to be explained. This notion led the
^Salisbury, op. cit., p. 153.
42Ibid.
/ *5
Salisbury, loc. cit.
62 ,
analysts to expend considerable effort in elaborating the |
independent variables assumed to have brought about the
policy outcomes.
For example: absence of inter-party competitiveness
tended to have a "conservative" bias in policy outputs
(Key); industrial states (independent variable) are more
likely to have higher per capita expenditure on welfare
(dependent variable) than poorer states (Dawson and Robin-
x 44
son).
This approach assumes that as long as one does the
utmost in the description and analysis of these independent !
variables, one has explained the dependent variable, i.e., '
i
policy outcomes. Such preoccupation with the description
of "independent variables" justified a negligence in
i
examining the consequence that follows from how one views !
the "policy" in the first place. Thus, the result was to
I
take the most convenient notion of policy, i.e., the way
they bear the "labels."
Lowi and others began to offer an alternative in the
content approach to it arguing that nominal category-based
description falls short of explaining the behavioral conse
quence the actions separately or collectively taken under
each "label" brings about. In short, they claim that the
results based on nominal categories contribute little in
44
Froman, op. cit., p. 44.
63
building a general theory of policy-related behavior. j
The way out of these impediments to theory building
is, the content advocates propose, to view policy as an
independent variable that helps to explain something else.
Rather than treating policy as an outcome of the
political process, we are asking whether policy itself
is a variable that may affect the political process.
To answer this question we would certainly need some
categories of policy which could then be related to
variations in political process.
Focus on the impact of policy was, then, believed to
be most productive in creating more abstract categories of
policy bearing on the behavior, thus, away from nominal
approach. This led Lowi^ to propose his three categories '
of policy, i.e., "distributive," "regulative" and "redis
tributive."
i
Lewis Froman proposed "areal" referring to those j
i
policies affecting whole community (e.g., city-manager I
i
plan, fluoridation) and "segmental" referring to those
affecting less than total population (e.g., urban renewal,
welfare). Salisbury offered "constitutional" and "sub-
• i . 47
stantive categories.
!
No doubt these categories are more abstract and
broad-gauged than nominal categories; they are highly
45
Froman, op. cit., p. 44.
46
Lowi, loc. cit.
^Salisbury, loc. cit.
64
suggestive. However, on close examination, they contain j
serious defects in (a) concepts and (b) methodology.
Conceptually, the categories are built specifically
to propose that each of their respective categories serves
to explain and discern the different types of consequence
the policy brings to the political process and its environ
ment .
However, they all fail to show how their respective
categories go about in discerning the impacts. For
instance, to say that fluoridation has "areal" or "regula
tive" impact, the theory ought to yield a proposition as to
why it is likely to be "areal" or "regulative" rather than
I
"segmental" or something else.
How is each category proposed to discern the impacts
upon which aspects of the political process? Are the
differences in the categorical attributes, e.g., "regula-
i
tive" vs. "distributive," to be due to differences in the
degree of impact it discerns in the different aspects of
the system or the differences in the way the policy af
fected the same aspects?
In order for their categories to be a viable concep-
j
tual framework, they should yield some hypotheses on these j
questions in some fashion to facilitate a systematic
research.
Even if there were some hypotheses available, one
still runs into a knotty methodological problem. How to
65
measure the impact? What ought to be the criteria of
impact? Shall we rely on perceptual data, or attitudinal
data of the actors involved? beneficiaries? Or some
impersonal quantitative measurement? When would the impact
of a policy be felt? after a year or two?
Ostrom and Wildavsky opted to use exclusively the
quantitative measures as the criteria of impact. Conse
quently, they deal with the issues^ that are quantifiable.
Their conceptual model, i.e., cost-benefit analysis is,
itself a scheme most amenable to quantitative data. It
can't aid much in measuring the qualitative impacts of a
policy (e.g., poverty program, school aid, moral regula
tion) .
49
Wayne Francis relies entirely on perceptual data
in determining the impact of seven groups of policies on
legislative decisional structures. How reliable are the
perceptual data? Is the perceptual image of the impact of
a policy given by a group of actors, in this case, the
state legislators the "objective" world?
Theoretically suggestive as the content approach may
appear, it has yet to formulate a definite pattern in both
48
Ostrom deals with water resources and Wildavsky
with budgetary problems, both of which are highly quanti
fiable. See Ostrom, loc. cit. and Wildavsky, loc. cit.
49
Francis, loc. cit.
66
concepts and application. There are, thus far, no unifying)
concepts or methodological direction in the ideas proposed
by the "content" advocates.
Thus, each approach seems to have the virtue as well
as weakness of its own. No one approach commands a clear
superiority over the others for the moment. Environmental
ists suffer from the forms of data they utilize: they are
the data drawn from too simple a notion of policy based on
nominal categories.
Nevertheless, environmental approach has the virtue
in its simplicity, thus, lending itself readily to research
!
operation. The content approach still remains at the
conceptual stage and shows little prospect for meaningful
operation.
In summary, the following observations may be derive4
from this review of the current status of state policy j
studies:
1. There are still some good reasons for continuing to j
invest in the analysis of policy cast within nominal
categories. For one thing, the object of analysis
may not always be the building of more general
theory. Secondly, as Salisbury points out, "we are
not so rich in descriptive statements that we can
afford to stop seeking them, even if we were agreed,
as we are not, that we are never interested in
explaining, say, variations in educational policy
because we are interested in educational p o l i c y . j
2. Focus on a single policy may be a more productive
alternative to simultaneous treatment of several
policies.
3. To be productive, however, this single policy
approach is to be utilized within a context which
combines Dawson and Dye's type of process explana
tion of policy-making with that of content approach
focused on the search for the impact of policy.
4. This combination approach is suggested for several
reasons: |
51
(a) As Ranney and others correctly observed, the |
I
process explanation of public policy-making in
i
the past seriously neglected in analyzing the j
impact of the implementation of policy. Thus,
the content approach may have to be incorpo- i
rated into any future model of policy studies j
!
even just for pragmatic reason alone, that is, j
to gain more accurate knowledge of the impact
of important social policies.
51
See Lewis Froman, loc. cit.; Vernon Van Dyke,
"Process and Policy as Focal Concepts in Political
Research," Ranney (ed.), op. cit., pp. 23-39; probably,
Yehezikel Dror may be most vocal and emphatic in this
regard; see Yehezkel Dror, Public Policymaking Reexamined
(San Francisco: Chandler Publishing Co., 1968).
68
!
(b) A single policy approach may lend itself more i
easily to the search for the impact of policy
than a simultaneous attempt at several policies
(particularly in the absence of workable hypo
theses as the analysis of content approach
revealed). First, it is simply more manageable
to handle a single case than many. Second,
the results obtained under more manageable
conditions may yield higher accuracy and relia
bility. Third, it avoids inflicting injustice
to the importance of an individual policy as
it might be the case with simultaneous treat-
i
ment of several policies.
(c) The sum total of more accurate and reliable
data on the process as well as on the impact of
j
various policies conducted by different scholar^
j
on a single policy basis may eventually lay j
more important ground in inventing abstract
typologies of policy, which is prerequisite to
building a more general theory of public policy„
CHAPTER III
RESEARCH METHODOLOGY
Any empirical research involves a set of procedures,
methods and techniques. This third chapter presents the
methods and procedures followed in the study under the
following headings: the research design, the hypothesis and
variables, the sample, data-gathering procedure, and the
statistical techniques.
Research Design !
The research design this study takes is primarily an |
I !
expost-facto analysis. That is, the accomplished facts at
a given moment are the object of investigation focused on
!
the explanation of the way they became the facts. I
Thus, the decisions made by each of thirty-eight
states with respect to various provisions of Medicaid are
the object of this study. The hypotheses derived from the j
proposed conceptual model are brought into the test to J
explain or account for the variations in these decisions
^"Expost-facto research is perhaps the most feasible
design in social science because of practical difficulty in
setting up experimental and control groups in social phe
nomena. It is, of course, not the most ideal situation but
sensible one at the moment. See, for example, Fred N.
Kerlinger, Foundations of Behavioral Research (New York:
Holt, Rhinehart, 1965), pp. 359-74.
I ____________________________ ^ ________________________________________
70
among thirty-eight states.
Minor portion of this study devoted to the analysis
of impact of Medicaid is, however, designed to be descrip
tive and exploratory. Absence of proper hypotheses and
adequate data in this regard determined the nature of
"impact" analysis to be a descriptive exploration.
Hypotheses and Variables
In Chapter I, the hypotheses and variables were set
forth on some rational ground within t.ie context of the
conceptual model. The variables included in the model are
now operationally defined.
First, the environmental characteristics were
defined to be economic development consisting in four
closely related variables, i.e., urbanization, industrial-
i
!
ization, income, and education. j
Urbanization refers to the size of population living i
in U.S. Bureau of Census defined standard metropolitan
statistical areas in each state.
Industrialization refers to the number of employees
engaged in nonagricultural establishments defined by the
U.S. Bureau of Census. This is a traditional index of i
i
i
industrialization in the United States Bureau of Census.
Income refers to per capita income of each state
which is the total amount of personal income divided by the
number of population in a state.
71 i
Education refers to median school years completed by
all of the residents in each state. The median school
years completed is again the standard procedure in asses
sing educational level of a region or state in the social
science literature.
The political system consists in three variables:
inter-party competition, voter participation, and apportion
ment rate. Inter-party competitiveness index is derived
2 3
from Ranney's adaptation of Dawson-Robinson's measure of
party competition.
For each state, the percentages of the two-party
i
popular vote for governor received by each party in each
election and percentages of the seats in each house of the
legislature held by each party in each session were com- j
puted.
From these tables Ranney computed four basic figures:
I
(1) the average per cent of the seats in the state senate ;
!
held by the Democrats; (2) the average per cent of the
seats in the state house of representatives held by the
Democrats; (3) the average percent of the popular vote won
2
Austin Ranney, "Parties in State Politics," Herbert
Jacob and Kenneth Vines (eds.), Politics in the American
States (Boston: Little, Brown, 1965), pp. 61-65.
3
Richard Dawson and James Robinson, "Inter-Party
Competition, Economic Variables and Welfare Policies in the
American States," Journal of Politics, XXV (1963), 271.
721
by the Democratic gubernatorial candidates; and (4) the per|
cent of all terms for governor, senate, and house in which
the Democrats had control.
Then, all these four percentage figures are averaged
out to produce an index of competitiveness for each state
carried to four decimal places with a range of .000 (total
Republican success) to 1.000 (total Democratic success)
and .5000 representing even two-party competition.
Voter participation refers to the data^ on voter
turnout for gubernatorial and senatorial election in non-
presidential years, 1950-65. Nonpresidential years are
i
selected because it may remove the halo effect of usually
I
intensive and dramatic presidential campaign on the poli- j
tical participation at the state-local level. Thus, voter
turnout for gubernatorial and senatorial elections in these
nonpresidential years could be construed as reflection of j
genuine intent of voters to participate politically at j
I
I
state level. I
The apportionment score this study employs is drawn |
5
from Schubert-Press index of malapportionment. The
4
Voter participation data in this regard are derived
from: Lester Milbrath, "Political Participation in the
States," Jacob and Vines, op. cit., p. 40.
^Glendon Schubert and Charles Press, "Measuring
Malapportionment," American Political Science Review,
LXIII (1964), 302-27.
73
indices are based on the degree to which the population
assigned to each legislative seat deviates from the popula
tion figure derived from the total population of a state
divided by the number of seats in Chamber.
On the basis of these deviations, Schubert and Press
rank the states from we11-apportioned to "worst" appor
tioned. ^
The policy outcomes this study selects as empirical
referents are three aspects of decisions made by thirty-
eight states operating Medicaid programs as of the end of
1968. Those three aspects of the decisions are (a) the
scope and extent of services each state decided to offer,
(b) the number of beneficiaries to be eligible for these
services, and (c) the size of total expenditures each
state allocated to Medicaid.
As the subsequent chapters on findings show,^ there
are a great deal of variations in the above three aspects
from state to state. Accounting for these variations is
the primary objective of the investigation in terms of the
g
hypotheses developed in the model.
6Ibid.
^See Tables 28 and 24; also Chart 1 in Chapter V.
g
Refers to "Statement of the Problem" in Chapter I,
which is a set of problems derived from the basic model of
this study.
74;
Sampling Procedures j
Selection of Medicaid as the basis of empirical
sources of data on policy outputs determined a number of
things pertaining to the sampling procedures in the study.
First of all, it set a limit to the size of sample;
all of the thirty-eight states which participated in the
operation of Medicaid as of the end of 1968 constituted
the universe. Decision was to study the whole universe,
that is, one-hundred per cent.
Secondly, it also delimited the scope of investiga
tion of relationship between socioeconomic development and j
!
political system characteristics. Since this investigation;
i
is conducted to explain the variations in Medicaid decision^
of the thirty-eight states, relationship between economic
and political variables cover only these thirty-eight j
states leaving the other twelve states out of consideration!
I
Thirdly, time dimension selected with respect to
economic and political variables was also dictated by the
time period reflected in the operation of Medicaid. Since
Medicaid became effective in January, 1966, it had roughly
about two and a half years of experience.
I
However, the year the states began to operate the
program varies from group to group. Some states (e.g.,
California, New York, Michigan) operated the program from
January, 1966, when Medicaid was formally put into effect
by Congress. Some other states began to operate much
75
later; Illinois, South Dakota, and North Dakota in 1968.
In view of such a difference in the effective date
of the program operation in the states, the time this study
selects as "benchmark" in examining the effect of political
and economic variables on Medicaid decisions is 1967.
Because it is felt that the time falling in-between the
earliest 1966 and the latest 1968 may fairly be represen
tative of the conditions of economic and political system
characteristics in terms of their probable effect on the
policy decisions made in the few years counted either back
ward or forward.
Data-Gathering
Since the Medicaid program was adopted by the states
on different dates, it led the study to some practical
9
difficulty in obtaining the data. A certain set of data
is drawn, for example, from less than thirty-eight states,
because only available data are those obtained at a time,
when less than thirty-eight states participated in the
Medicaid program.
Whereas, some other data, for instance, on the size
of expenditures are available on all of the thirty-eight
9
This refers to the data presented in Chapter VII in
which Tables 23, 29, and 33, for instance, contain the data
on less than thirty-eight states because of the reasons
stated.
76 ;
i
states. Because this particular data came from a source j
obtained during 1968. These secondary sources of the data
on Medicaid are the reports published by Health Insurance
Institute,^ Advisory Commission on Intergovernmental
11 12
Relations. Tax Foundation, Ninetieth Congressional
Hearings on Proposals for Revision in the Social Security
•1 o * » /
Program, and U.S. Bureau of Census. Data on 1969 and
1970 are simply not available in any useful form for the
purpose of this study at this moment.
The reason is that compilation of such data is
usually made available a year after the event took place.
I
Thus, the data on 1969 will be made public some time this
summer of 1970; 1970 data will be available in the summer
of 1971. Considering then, that this study began from
Health Insurance Institute, 1969 Source Book of
Health Insurance: Data (New York: Health Insurance Insti- I
tute, 1969). !
^Advisory Commission on Intergovernmental Rela- j
tions, Intergovernmental Problems in Medicaid (Washington,
D. C.: U.S. Government Printing Office, 1968).
12
Tax Foundation, Medicaid: State Programs After Two
Years (New York: Tax Foundations, Inc., 1968).
^U.S. Congress, House, Committee on Ways and Means,
Hearings, President’s Proposals for Revision in the Social
Security Program. Part 1, 90th Congress, 1st Sess., 1967.
14
U.S. Bureau of Census, Statistical Abstract of the
United States, 1968 (98th edition); U.S. Department of
Health, Education, and Welfare, Health, Education, and
Welfare Trends: 1966-67 (Washington, D. C.: U.S. Govern
ment Printing Office, 1968).
77
February, 1969, it may legitimately claim on the recency of
the data.
Data on socio-economic development of thirty-eight
states are entirely drawn from the United States Statistical
Abstract.^J They are all treated as interval data relying
on 1960 census data as they are recorded in the Abstract.
Data on inter-party political competitiveness was
derived from Austin Ranney's adaptation of Dawson-Robinson
measure as the former is recorded in Ranney's article,
"Parties in State Politics.Data on political partici
pation in this study are the data Lester Milbrath used in
17
his Political Participation in the States. Measures
on malapportionment are drawn from Schubert-Press' article,
"Measuring Malapportionment," published in American Politi-
18
cal Science Review in 1964.
Statistical Procedures
Data found in the documents, reports, and journals
mentioned above were selected out and transferred to a
master data sheet. The nature of data was found to meet
the assumptions of normal curve, linearity of bivariate j
I
15Ibid. I
18 1
Ranney, loc. cit.
■^Milbrath, loc. cit.
18
Schubert and Press, loc. cit.
78
distribution, and homescedasticity, i.e., equal variances,
of scatter along each regression line, which justified the
19
use of Pearson product-moment analysis.
Thus, they were read into conventional Fortran
coding sheet and fed into a program pre-designed for cor
relational matrix analysis. This matrix analysis yielded
cross-correlation of fourteen factors which composed the
total number of variables considered in this study.
While correlation coefficient is the primary infer
ential statistic employed in this study, other descriptive
statistics are also used to illustrate and analyze certain
bivariate relationships. That is, simple percentages,
ranges, means, and frequency distribution are liberally
used to analyze, especially, the relationships between
socio-economic measures and political system characteris
tics.
Simple two-by-two or two-by-three matrix analysis |
was also found to be very useful for highlighting critical
I
!
relationship between economic development and political j
!
system characteristics. j
I
Analysis of interface of policy outcomes in Medicaidj
i
19
See Herbert Blalock, Social Statistics (New York:
McGraw-Hill, 1960), pp. 115-210; John G. Peatman, Introduc
tion to Applied Statistics (New York: Harper & R o \ T , 1963),
pp. 78-89; Mr. Ron Langley, Resident Computer Programmer at
Fresno State College Computer Center rendered his expertise
in examining the nature of the data this study collected.
79 ,
with economic and political system variables was, however,
carried on entirely by means of simple correlation coeffi
cient techniques. Neither matrix nor simple percentage
analysis was employed, because the purpose was primarily to
compare the degree of strength of relationship between
economic development and policy outputs, on one hand, and
between political system variables and policy outputs on
the other.
On the basis of comparative strength of relationship,
tenability of hypothesis of whether it is economics or
politics that has the direct impact on policy outputs was
to be determined. In determining the degree of strength to
be statistically significant, this study followed the con-
20
ventional .05 level of significance.
Analysis of the impact of Medicaid entirely relied 1
on such descriptive statistics as percentage, means, range,
and rank-order. In the absence of significant hypotheses iri
this regard, the investigation of impact of Medicaid meant
to be exploratory and descriptive. j
i
Summary
Research design this study takes is ex-post facto
It is almost an established convention among
social scientists to adopt .05 level of significance; see
Blalock, op. cit., p. 117; Peatman, Ibid.; Thomas Dye,
Politics, Economics, and the Public (Chicago: Rand McNally,
1966), p. 39.
80 ,
analysis, with exception of minor aspect of "impact"
analysis which is basically descriptive and exploratory.
Operationalization of the major variables constituting
economic development and political system characteristics
followed the pattern of existing literature in the field
specifically to build the findings upon some cumulative
basis.
The data are all derived from the secondary sources
which were made available for public use in form of docu
ments, reports, and informational journals. Selection of
samples was pretty much determined by the nature of policy
outcomes the study undertook to investigate, that is, the
status of the operation of Medicaid program as of the end
of 1968.
'Thirty-eight states which adopted the Medicaid pro- |
gram in one form or another by the end of the fiscal year j
of 1968 was, therefore, the basis of the entire universe. i
Statistical techniques employed are combination of simple
correlation coefficient analysis and such descriptive j
statistics as percentage, range, means, and rank-order
i
analysis. J
t
Findings derived through these methods and procedures
are now presented in immediately following three chapters.
CHAPTER IV
ECONOMIC DEVELOPMENT AND POLITICAL
SYSTEM CHARACTERISTICS
This chapter presents the data gathered for the
first hypothesis of this study. The hypothesis was that
economic development helps to shape the character of poli
tical system in a state. Converted into a hypothesis: the
higher the level of economic development in a state, the
higher the degree of variations in the political system
characteristics in the state.
Considering the political system characteristics as
the dependent variables, the data are presented in follow
ing order:
(1) Relationship between inter-party competitiveness
and four variables in economic development, i.e.,
industrialization, urbanization, education, and
income;
(2) Relationship between political participation and all
of the above four variables in economic development;
(3) Relationship between rate of malapportionment and
the four variables in economic development.
The purpose is to find not only how closely the
general character of political system is related to the
rate of economic development but also to discern which of
81 _____________
82
the four variables is more closely related to which of the
political variables.
Inter-Party Competition and Economic Variables
It is suggested that strong two-party government is
desirable for a healthy democracy because policy alterna
tives may be available for the citizenry by virtue of compe
tition between the two parties.'*'
Without weighing the merit of this position, the
first thing is to know what makes one state more competi
tive than the other. When we know about the forces which
shape this particular characteristic of political system,
we may be in a better position to control the course of
events.
In the literature on state politics, it is proposed
thai- the socioeconomic development influences the way a
2
state political system takes a certain characteristics.
To test, then, how economic development variables
■*"See V. 0. Key, Southern Politics (New York: Knopf, |
1951); Duane Lockard, New England State Politics (Prince- !
ton: Princeton University Press, 1959). j
2 1
See Richard Dawson and James Robinson, Inter-party
Competition, Economic Variables and Welfare Politics in the
American States," Journal of Politics, XXV (1963), 265-289;
Phillips Cutwright, "Political Structure, Economic Develop
ment, and National Security Programs," American Journal of
Sociology, LXX (1965), 537-50; Robert Golembiewski, "A
Taxonomic Approach to State Political Party Strength,"
Western Political Quarterly, XI (1958), 494-513.
83
affect one of the political system characteristics, i.e.,
inter-party competitiveness, this study utilizes the inter
party competitiveness index devised by Ranney's adaptation
3
of Dawson-Robinson's measure as it was described in the
preceding chapter.
The indices of thirty-eight states' inter-party
competitiveness are shown in Table 2.
The index score shows the range from the extreme
one-party Republican state (South Carolina) with an index
of .0001 and the extreme one-party Democratic state
(Georgia) with a .9915.^ Twenty-one states of the total of
thirty-eight, that is, 80 per cent, fall into the two-party
category.
Then, the relationship between these three cate
gories of the states, i.e., one-party Democratic, two-party
Austin Ranney, "Parties in State Politics," H.
Jacob and Kenneth Vines (eds.), Politics in American
States (Boston: Little, Brown, 1965), p. 65.
^Rationale behind Ranney's adaptation of Dawson-
Robinson' s indices of inter-party competition was described
in the chapter on methodology. A brief recapitulation: for
each state, the percentages of the two-party popular vote
for governor and percentages of the seats in each house of
the legislature held by each party in each legislative
session were computed. The percentate figures were, then,
averaged out to produce an index of competitiveness for
each state carried to four decimal places with a range of
.000 (total Republican success) to 1.000 (total Democratic
success) and .5000 representing even two-party competition.
TABLE 2
THIRTY-EIGHT STATES CLASSIFIED BY THE DEGREE
OF INTER-PARTY COMPETITIVENESS (1967)
INTER-PARTY COMPETITIVENESS
States
One-Party
Republican States
Two-
Party States
One-Party
Democrat ic
S. C. .0001 N. Y. .3173 N. M. .7023
Vt. .1760 Wyo. .3470 Md. .7137
N. D. .1860 Ohio .3523 W. V. .7223
S. D. .2320 Ore. .3545 Kent. .7650
Me. .2405 Mich. .3770 Okla. .8193
Kan. .2415 Idaho .3780 Tex. .9590
Iowa .2495 111. .3847 La. .9867
N. H. .2680 Neb. .3875 Ga • .9915
Wis. .2997 Cal. .3930
Penn. .4056
Conn. .4420
Utah .4605
Minn. .4610
Mont. .4695
Hawaii .4897
Mass. .5227
Nev. .5263
Del. .5420
Wash. .5647
R. I. .6327
Mo. .6603
9 21 8
Source: Austin Ranney, "Parties in State Politics," Herbert
Jacob and Kenneth Vines (eds.), Politics in Ameri-
can States (New York: Little, Brown, 1965), p. 65.
one-party Republican dominant, and each of four economic
development variables, i.e., industrialization, education,
income, and urbanization is examined.
Each of these economic development variables is
85
broken into the following three categories, high, medium,
and low, as the data on industrialization show below.
TABLE 3
INDUSTRIALIZATION IN TERMS OF THE NUMBER OF POPULATION
ENGAGED IN NON-AGRICULTURAL SECTOR IN
THIRTY-EIGHT STATES (IN THOUSANDS)
RATE OF INDUSTRIALIZATION3
Low Medium High
Wyo. 0097 R. I. 0317 Mass. 2019
Vt. 0122 Neb. 0416 Mich. 2674
N. D. 0146 W. V. 0477 Tex. 2925
S. D. 0152 Kan. 0600 Ohio 3364
Nev. 0157 Ore. 0607 111. 3864
Idaho 0178 Okla. 0648 Penn. 3915
Mont. 0181 S. C. 0686 Calif 5800
Del. 0184 Iowa 0755 N. Y. 6520
N. H. 0217 Kent. 0759
Hawaii 0219 Wash. 0897
N. M. 0263 La. 0906
Me. 0295 Conn. 1033
Utah 0301 Md. 1060
Minn. 1082
Ga. 1257
Wis. 1332
Mo. 1476
^ean of thirty-eight states; is 1,260; S. D.=15.51
Source: U.S. Bureau of the Census, Statistical
Abstract of the United States: 1968 (89th edition),
p. 220.
Before examining the correlational measure between
the industrialization and inter-party competitiveness, the
table below analyzes the relationship between three types
[Of inter-party competitiveness and three ranges of the____
86
rate of industrialization. According to hypothesis, those
states showing high industrialization are also to reflect
high variations in party competition characteristic.
TABLE 4
RELATIONSHIP BETWEEN INTER-PARTY COMPETITIVENESS
AND INDUSTRIALIZATION OF THIRTY-EIGHT STATES
PARTY COMPETITIVENESS
Industrial
ization
One-Party
Republican
Two'-Party One-Party
Democratic
Low (4) (7)
(1)
Vt.
N. D.
S. D.
Me.
Wyo.
Idaho
Utah
Mont.
Hawaii
Nev.
Del.
N. M. 12
Medium (4)
S. C.
Kan.
Iowa
Wis.
(7)
Ore.
Neb.
Conn.
Minn.
(6)
Md.
W. V.
Kent.
Okla.
La.
17
High
(1)
N. H.
(7)
N. Y.
Ohio
Mich.
111.
Cal.
Penn.
Mass.
(1)
Tex.
9
9 21 8 38
Simple Correlation Coefficient between industrialization
and party competition, r=.05.
Of twenty-one states showing two-party competitive
characteristic, only seven or about 33 per cent show high
;rate of industrialization. The rest of fourteen states are
87 ,
equally divided between the states with low and medium rate
of industrialization.
With respect to nine states of one-party Republican
dominance, four of nine or almost 45 per cent of the total
show medium rate of industrialization. Only New Hampshire
with high rate of industrialization belongs to the category
of Republican dominated one-partyism. The same number of
the states, i.e., four come from low to medium industri
alized states.
With respect to Democratic dominated one-partyism,
a bulk of the states, that is, six of them or 80 per cent
come from the states with medium industrial states, instead
i
of low industrialized states. Texas and New Mexico stand
as exceptions on the contrary to the hypothesis. Particu
larly the instance of Texas with high industrialization
score practicing Democratic dominated one-partyism has its j
roots in something other than economic characteristics.
1
The reason lies in its historical and political culture;
Texas has been a traditionally Democratic stronghold as
the leader of Southern states' resistance against the
Eastern liberal power.
Consequently, the overall relationship between
industrialization and inter-party competitiveness is
insignificant in this study as the Pearson product-moment
correlation is shown as .05. Hypothesis postulated in
this study in this regard is not tenable in the light of
88
these data. However, this result is contrary to the find- j
ings of some other studies.
Ranney and Kendall3 and Golembiewski^ report a high
correlation between industrialization and party competition;
The reason this study shows extremely low correlation may
be attributable to the fact that the sample left out a
number of Southern states that are known to be one-party
dominated and lowly industrialized. For instance, Missis
sippi, Arkansas, Alabama.
Even with the absence of extreme Democratic one-
party states in the South from the sample, this particular |
correlation between industrialization and party competition!
!
is not too divergent from Dye's finding^ which reports the
Pearson r between these variables to be .03. Dye found, I
however, that income and education to be highly correlated, j
In this study, about the same results are found to
be the case; to examine the relationship between education
and party competition, thirty-eight states are broken down
into three ranges of educational achievement.
Three ranges of educational accomplishment are now
Austin Ranney and Willmoore Kendall, "The American
Party System," American Political Science Review, XLVIII
(1954), 477-85.
0
Golembiewski, loc. cit.
^Thomas Dye, Politics, Economics, and the Public
(Chicago: Rand McNally, 1966), p. 58.
89
TABLE 5
SCHOOL YEARS COMPLETED BY THE
POPULATION 25 YEARS OR OLDER
IN THIRTY-EIGHT STATES
EDUCATION3
Low Medium High
S. C. 08.7 Penn. 10.2 Del. 11.0
Kent. 08.7 Minn. 10.3 Conn. 11.0
W. V. 08.8 S. D. 10.4 N. M. 11.2
La. 08.8 Okla. 10.4 Iowa 11.3
Ga. 09.0 Tex. 10.4 Hawaii 11.3
N. D. 09.3 Wis. 10.4 Mont. 11.6
Mo. 09.6 Md. 10.4 Neb. 11.6
111. 10.5 Mass. 11.6
N. Y. 10.7 Kan. 11.7
Mich. 10.8 Idaho 11.8
Vt. 10.9 Ore. 11.8
N. H. 10.9 Me. 11.0
Ohio 10.9 Wyo. 12.1
R. I. 10.0 Wash. 12.1
Cal. 12.1
Nev. 12.1
Utah 12.2
^ea^lO. 7; S. D.=10.256; the above data should be read
with a note on second decimal on each figure.
Source: U.S. Bureau of Census, Statistical Abstract of the
United States: 1968 (89th edition), p. 113.
analyzed in Table 6 in terms of its relationship with three
categories of inter-party competitiveness.
Of twenty-one two-party states, ten or forty-six per
cent of the total are the states in the category of high
educational achievement of its population ranging from
eleven years to 12.2 years of school completion. Eight
90
TABLE 6
RELATIONSHIP BETWEEN INTER-PARTY COMPETITION AND
EDUCATIONAL ACHIEVEMENT IN THIRTY-EIGHT STATES
PARTY COMPETITIVENESS3
Education One-Party
Republican
Two-Party One-Party
Democratic
Low (4)
S. C.
N. D.
S. D.
Wis.
(3)
Penn.
Minn.
Miss
(7)
Md. La.
W. V. Ga.
Kent.
Okla.
Tex. 14
Medium
(3)
Vt.
Iowa
N. H.
(8)
N. Y.
Ohio
Mich
111.
Conn.
Hawaii
Del.
R. I.
(1)
N. M.
12
High (2)
Me.
Kan.
(10)
Wyo.
Ore.
Idaho
Neb.
Cal.
Utah
Mont.
Mass.
Nev.
Wash.
(0)
12
9 21 8 38
Simple correlation coefficient between education and party
competition, r=.31. i
states (New York, Ohio, Michigan, Illinois, Connecticut,
Hawaii, Delaware, Rhode Island) or thirty-eight per cent
are the states with medium educational record. Only six
teen per cent or three states (Pennsylvania, Minnesota,
and Montana) are the states showing low educational
91
achievement as hypothesized.
A bulk of low education states, that is, seven of
fourteen or fifty per cent are Democratic dominated one-
party states (Maryland, West Virginia, Kentucky, Oklahoma,
Texas, Louisiana, and Georgia). Four of seventeen in this
low educational category or twenty-eight per cent are
Republican dominated one-party states.
Interestingly enough, no states in the category of
high educational accomplishment is found in Democratic
dominated one-party states but two of them (Maine, Kansas)
are found in Republican dominated one-party states. It
supports partly the belief that Republican party draws its
strength from middle and upper middle sector of general
population thus representing the interest of upper strata
of the society.^
That the Republican party is represented by better
educated sector of American society is further supported by
examining the states in the second row of the medium educa
tion group. Three states out of twelve or twenty-five per
cent belong to Republican dominated one partyism while only
one state (New Mexico) in this group is found in the
g
See the following works in this area: Key, op. cit.,
pp. 223-56; Lester Milbrath, Political Participation
(Chicago: Rand McNally, 1965Y; Robert K. Lane, Political
Life (New York: Free Press, 1959), pp. 299-317; Dye,
op. cit., pp. 78-80; Herbert Jacob, "State Political
System," Jacob and Vines (eds.), op. cit., pp. 3-17.
92
Democratic dominated one-party column.
Overall correlation between the combined measures of
party competition and education is .31, which is signifi
cant enough to warrant a generalization that party compe
tition is in part likely to be the product of the level of
educational achievement of a state.
Income is another measure that is closely related to
the characteristics of party competition. Data on income
are also divided into three ranges of high, medium, and low.
TABLE 7
PER CAPITA INCOME OF THIRTY-EIGHT
STATES (IN DOLLARS)
PER CAPITA INCOME*
a
Low Medium High
S. C. 2052 Tex. 2542 N. H. 3019
W. V. 2176 Vt. 2595 R. I. 3047
La. 2277 Utah 2617 Ohio 3056
Ga • 2379 Mont. 2623 Minn. 3111
Kent. 2387 Wyo. 2739 Hawaii 3124
S. D. 2420 Mo. 2817 Md. 3204
Idaho 2445 Kan. 2864 Wash. 3222
N. M. 2462 Neb. 2905 Mich. 3269
Okla. 2462 Ore. 2908 Nev. 3497
Me. 2477 Penn. 2968 Del. 3529
N. D. 2485 Wis. 2973 111. 3532
Iowa 2992 Cal. 3660
N. Y. 3726
Conn. 3865
^ 630=2851
Source: U.S. Bureau of the Census, Statistical
Abstract of the United States: 1968 (89th ed.),
p. 311.
Three categories of per capita income level are
examined in relation to the level of party competitiveness
in Table 8.
TABLE 8
RELATIONSHIP BETWEEN PER CAPITAL INCOME
AND INTER-PARTY COMPETITIVENESS
PARTY COMPETITIVENESS3
Income One-Party
Republican
Two-Party One-Party
Democratic
Low (4)
S. C.
N. D.
S. D.
Me.
(1)
Idaho
(6)
N. M.
W. V.
Kent.
Okla.
La.
Ga.
11
Medium (4)
Vt.
Kan.
Iowa
Wis.
(7)
Wyo.
Ore.
Neb.
Penn
(1)
Utah
Mont
Miss.
Tex.
12
High (1) (13) (1)
N. Y. Conn.
Ohio Minn.
Mich. Hawaii
111. Mass.
Cal. Nev.
R. I. Del.
Wash.
9 21 8 38
£
Simple correlation coefficient between income and party
competition. r=.30
Of fifteen states in the high income category,
thirteen states or eighty-six per cent of the total (New
York, Ohio, Michigan, Illinois, California, Connecticut,
94
Minnesota, Hawaii, Massachussetts, Nevada, Delaware, and !
Washington) are found in the column of the two-party states,
whereas only one state is distributed in each of the other
one-party dominant columns, that is, New Hampshire in the
Republican column and Maryland in the Democratic column.
In the medium income category, seven of twelve or
fifty-eight per cent (Wyoming, Oregon, Nebraska, Pennsyl
vania, Utah, Montana, Missouri) are contributing to thirty-
three per cent of the states found in the two-party column.
Only one state (Texas) in this medium income category is
found in the Democratic column while four (Vermont, Kansas,!
|
Iowa, Wisconsin) of twelve states contribute to forty-four j
per cent of the total composing Republican dominated one-
party states.
i
Of the eleven poor or low income states, six (New 1
j
Mexico, West Virginia, Kentucky, Oklahoma, Louisiana,
i
Georgia) or fifty-four per cent contribute to seventy-five
per cent of the total number of the states in Democratic
dominated one-party column, whereas lesser number, four
states in this low income category or thirty-six per cent
contribute to forty-four per cent of the states in the
Republican dominated one-party state column.
In the medium income category, again the strength of
the Republican column is much stronger than that of the
Democratic. Four (Vermont, Kansas, Iowa, Wisconsin) of
twelve states in this medium category are found in the
95
Republican column while only one state (Texas) in this
group is found in the Democratic dominated one-party
column. The majority of the states in this medium income
group, that is, seven of twelve is found in the two-party
states while only one state as low income state (Idaho)
contributes rather exceptionally to the column of two-party
states.
These data definitely show that the states with high
to medium per capita income tend to have the two-party
system. The data also support the view, as it was the case
with respect to education, that the Republican party draws
its strength from the richer Mid-western and some North- j
eastern states which in turn is indicative of representing
the interest of the wealthier group of the society. One-
party Democratic states are heavily populated with the
Southern states with low per capita income with the excep- s
i
tion of Maryland and Texas. Again explanation for these !
i
exceptions lies in historical and political culture rooted I
9
in racially ingrained political practice.
Last variable in economic development, i.e., urban
ization, is also broken down into three ranges of high,
medium, and low.
9
See Key, loc. cit.; Lockard, loc. cit.; Bruce
Morris, Problems of American Economic Growth (New York:
Oxford, 1961), pp. 161-172; Dawson and Robinson, loc. cit.;
Dye, loc. cit.; Robert Salisbury, "State Politics and
Education," Jacob and Vines, op. cit., p. 341.
96
TABLE 9
RATE OF URBANIZATION OF THIRTY-EIGHT STATES
IN TERMS OF PERCENTAGE OF POPULATION
LIVING IN URBAN AREAS
RATE OF URBANIZATION3
Low Medium High
N. D. 352 Minn. 621 Tex. 750
W. V. 382 Ore. 622 Hawaii 765
vt. 385 Okla. 629 Conn. 783
S. D. 393 La. 633 111. 807
S. C. 412 Wis. 638 Mass. 837
Kent. 445 Del. 652 N. Y. 854
Idaho 475 N. M. 657 Cal. 864
Mont. 502 Mo. 666 R. I. 864
Me. 513 Wash. 681
Iowa 531 Nev. 704
Neb. 543 Kan. 710
Ga. 553 Penn. 716
Wyo. 568 Md. 727
N. H. 583 Mich. 734
Ohio 734
Utah 749
aMean=62.08; the above data are percentages.
Source: U.S. Bureau of Census, Statistical Abstract
of the United States: 1968 (89th edition), p. 17.
These three ranges of rate of urbanization are
examined in terms of its hypothesized relationship with
inter-party competition, that is, higher the level of
urbanization, higher the degree of inter-party competitive
ness .
TABLE 10
RELATIONSHIP BETWEEN INTER-PARTY COMPETITIVENESS AND
RATE OF URBANIZATION IN THIRTY-EIGHT STATES
POLITICAL COMPETITIVENESS3
Urbani
zation
One-Party
Republican
Two-Party One-Party
Democratic
High (0) (6)
N. Y.
111.
Cal.
Hawaii
Mass
R. I.
(1)
Texas
7
Medium (2)
Kan.
Wis.
(10)
Ohio
Ore.
Mich.
Penn.
Utah
Minn
Nev.
Del.
Wash.
Mo.
(4)
N. M.
Md.
Okla.
La.
16
Low (7)
S. C.
vt.
N. D.
S. D.
Me.
Iowa
N. H.
(5)
Wyo.
Idaho
Neb.
Conn.
Mont.
(3)
W. V.
Kent.
Ga.
15
21 8 38
Simple Correlation Coefficient between urbanization and
party competition, r=.ll
Six states (New York, Rhode Island, Illinois, Cali
fornia, Hawaii, Massachusetts) out of seven with high rate
of industrialization are found in the two-party column; this
is overwhelming majority rendering some tenability to the
hypothesis that high industrialization may lead to healthy
98 ;
state of political party characteristic. Only exception to
it is one state, viz., Texas that is highly urbanized state
but shows Democratic dominated one-party characteristic.
Distribution of political party characteristics
among sixteen medium urbanized states is also supportive of
the hypothesis. Ten states out of sixteen or sixty-two per
cent of the total show strong inter-party competitiveness,
that is, two-party characteristic. Four states (New Mexico,
Maryland, Oklahoma, Louisiana) are Democrat dominated one-
party states, although they are the states with medium
urbanization. !
Of fifteen states of low urbanization, three states j
I
(West Virginia, Kentucky, Georgia) are the states dominated
by Democratic one-partyism. When these three states and
the other four states in medium and one (Texas) in high |
urbanized category are taken together, they are all I
Southern states or near the border of South, i.e., Maryland
and West Virginia.
This confirms the historical pattern that the states
down in the South have been traditionally Democrat dominated
one-party states.
It is suspected that Democrat dominated one-partyism
found in the eight Southern states in this study could have
been the function of urbanization as well as political
history or culture. Perhaps, it may be more of the latter
than the former.
99
Six "low" urbanized states (Vermont, North Dakota,
Maine, Iowa, South Dakota, New Hampshire) are Republican
dominated one-party states. They are all Northern or
North-western or Eastern states. Again, these six states
are traditional stronghold of the Republican party.
Five states out of the total of fourteen low urban
ized states are the two-party states (Wyoming, Idaho,
Nebraska, Connecticut, and Montana), while six states in
this low urbanization category are Republican dominated
one-party. Therefore, distribution between two-party and
one-party characteristic among low-urbanized states is
close to even division rather than skewed distribution.
Thus, it could be interpreted that urbanization as
one of economic development variables does seem to influ
ence inter-party competitiveness in the states. However,
this influence seems to be stronger as the level of
urbanization becomes higher, because at the lower level of
urbanization, the impact appears to be rather negligible.
Perhaps because of this negligible impact at lower
level of urbanization, the over-all correlation coefficient
between urbanization and party competition is low .11. In
Table 11, the over-all strength of relationship between
I four economic development variables and inter-party com
petitiveness are presented.
It can then be summed up at this point that economic
development, in general, tends to affect one important
100
TABLE 11
CORRELATION COEFFICIENT BETWEEN COMBINED
MEASURES OF INTER-PARTY COMPETITION
AND ECONOMIC DEVELOPMENT
Industrial
ization
Education Income Urban
Party Competition -.05 31* .30* .11
Asterisks refer to statistically significant relationship
at .05 level of significance.
characteristic of state political system, that is, politi
cal party competition. Specifically, however, party com
petition may be more of the function of education and
income, rather than industrialization or urbanization. i
Second characteristic of political system, political
participation, is now investigated in relation to four j
variables of economic development following the same proce- !
dures as was with party competition. j
!
Political Participation and Economic Development
Political participation takes a variety of forms;
running for public office, contribution of time and money
to campaign fund; participation in marches and demonstra
tion, attending rallies and listening to political speeches;
and following certain issues in the meeting or over the
news media. Of these activities, voting for the candidates
and issues at election may perhaps the most expressive of
political participation. All the other activities may be
101
considered as preliminaries to culmination in physical
participation in election.
Election drawing highest percentage of voter turnout
is still the presidential campaign; about 65 per cent of
the American people participate in presidential election.^
Since this study is primarily interested in state politics,
the voter turnout data do not include the presidential
elections because this may not reflect directly the citi
zens' interest in their home state.
Thus, this study uses the data on voter turnout on
gubernatorial and senatorial election in nonpresidential
years, 1950-65. Selection of nonpresidential years may
remove the halo effect of usually intensive and dramatic
presidential campaigns on the participation at the state-
level. Participation in gubernatorial and senatorial ;
elections in nonpresidential years could be construed as
i
genuine intent of the voters to participate in state-
related elections. j
t
Data on thirty-seven states' participation in these '
gubernatorial and senatorial elections are presented in the
table below. They are broken down into three ranges of
I
high, medium and low participation.
These three ranges of data on voter turnout are now
compared and related to the three ranges of industrializa
tion as the latter was presented in Table 3.
"^Dye, op. cit., p. 59; Milbrath, op. cit., p. 89.
102
TABLE 12 ;
THE AVERAGE PERCENTAGE3 OF VOTER TURNOUT IN
GUBERNATORIAL AND SENATORIAL ELECTIONS
IN NONPRESIDENTIAL YEARS (1952-65)
POLITICAL PARTICIPATION
States Low States Medium States High
Hawaii 000 Okla. 40.9 Wyo. 60.5
Ga • 10.7 Md. 42.6 Del. 61.4
S. C. 11.9 Mo. 43.7 S. D. 62.4
La. 12.3 Ky. 44.0 R. I. 62.7
Texas 13.6 N. M. 44.4 Conn. 63.4
Me. 45.6 Utah 64.3
Neb. 48.7
N. Y. 50.9
Iowa 50.9
Vt. 51.0
Mich. 51.2
Wis. 51.3
Ohio 52.3
Wash. 52.6
Kan. 52.7
Cal. 53.4
Penn. 54.6
N. H. 55.0
W. V. 55.3
111. 55.3
Nev. 55.7
Ore. 56.9
N. D. 56.5
Mass. 58.8
Min. 58.8
Mont. 59.6
Idaho 59.9
5 27 6
Source: Lester Milbrath, "Political Participation in the
States," in H. Jacob and K. Vines (eds.) , Politics in
American States (New York: Little, Brown, 1965) , p. 40.
aMean=46 .2, Hawaii did not become a state until after 1956
election; thus, Hawaii is left out of the sample this study
used in other political system variables » e.g., party com-
petition This reduces the size of the sample from the
original thirty-■eight to thirty-seven in this instance.
103
TABLE 13
RELATIONSHIP BETWEEN VOTER TURNOUT AND INDUSTRIALIZATION
IN THIRTY-SEVEN STATES
POLITICAL PARTICIPATIONa
Industrial High Medium Low
High (0) (7)
N. Y.
Mich.
Ohio
Cal.
Penn
111.
Mass.
(1)
Tex.
8
Medium (2)
R. I.
Conn.
(12)
Okla.
Md.
Mont.
Ky.
Neb.
Iowa
Wis.
Wash.
Kan.
W. V.
Ore.
Minn.
(3)
Ga.
S. C.
La.
17
Low (4)
Wy.
Del.
S. D.
Utah
(8)
N. M.
Idaho
Vt.
N. H.
Nev.
N. D.
Mont.
Me.
(0)
12
6 27 4 37
Simple Correlation Coefficient between industrial and
political participation, r=.03.
On the contrary to the hypothesis, four states or
thirty-three per cent of the twelve low industrialized
states are found to be contributing sixty-six per cent of
the strength to the high voting category.
And also eight states (New Mexico, Idaho, Vermont,
New Hampshire, Nevada, North Dakota, Montana, Maine) or
104
sixty-nine per cent of twelve low industrialized states are
contributing thirty-three per cent to the category of
medium voting strength, while none of the low industrialized
states is found in the low voting column.
Rather three states (Georgia, South Carolina,
Louisiana) of seventeen medium industrialized states con
tribute seventy-five per cent to the low voting column.
Furthermore, Texas as high industrialized state is low
voting state, instead of high voting. More confounding is
that not even a single high industrialized state is found
in high voting column; seven states (New York, Michigan,
Ohio, California, Pennsylvania, Illinois, Massachusetts) or ;
eighty-seven per cent of the eight high industrialized
states are medium voting states. These data show that
industrialization has very low correlation with the rate of
j
political participation in the states; correlation coeffi- j
cient between industrialization and political participation
is very insignificant .03. |
Just as it was the case with inter-party competition,
industrialization is found to have no significant influence
on political participation characteristic of the states'
i
political system in this study.
This result is rather consistent with the findings of
other studies. Dye also reports'^ very low correlation
^Dye, op. cit., p. 60.
105
between economic development and voter participation, that
12
is, .05; Richard E. Dawson reports it to be .09.
Voting participation is now analyzed in the light of
variations in per capita income of the thirty-seven states.
Three ranges of per capita income of the states as they
were presented in Table 7 are related to the voter partici
pation rate in Table 14.
Fourteen high income states are all distributed over
high to medium voting range and none of the states in this
high income category is found in low voting column. The
majority of twelve medium income states, that is, nine of
them, are medium voting states with two in high voting and
one (Texas) in low voting category.
Of eleven low income states, only one (South Dakota)
is in the high voting column and three of these are low
voting states with seven in the medium voting category.
Four of the low voting states are all from the states with
low to medium income with three of them or seventy-five per
cent in the low income category.
Twenty of twenty-seven medium voting states or
seventy-four per cent come from the states with medium to
high income; eleven are from high and nine are from medium
income category. The majority of the high voting states
come from the states with high to medium income states with
12
Dawson and Robinson, op. cit., p. 248.
106
TABLE 14
RELATIONSHIP BETWEEN VOTER PARTICIPATION AND PER
CAPITA INCOME OF THIRTY-SEVEN STATES
POLITICAL PARTICIPATION3
Income High Medium Low
High (3)
Del.
R. I.
Conn.
(11)
Md.
N. Y.
Mich.
Ohio
Wash.
Cal.
N. H.
111.
Nev.
Mass.
Minn.
(0)
14
Medium (2)
Wyo.
Utah
(9)
Mo.
Neb.
Iowa
Wis.
Kan.
Penn.
Ore.
Mont.
Vt.
(1)
Texas
12
Low
(1)
S. D.
(7)
Ky.
N. M.
Idaho
Me.
W. V.
N. D.
Okla.
(3)
Ga •
S. C.
La.
11
6 27 4 37
aSimple correlation coefficient between income and political
participation, r=.32.
three from high and two from medium.
Thus, in general, increasing income is found to be
quite regularly associated with increasing or higher voting
participation. Thus, correlation coefficient between the
two variables is shown to be statistically significant, i.e.,
.32.
107 '
13 '
This result is in accord with other findings in
14
this regard. For instance, Dye found the relation
between voting participation and income to be .52 and Daw
son reported^ it to be .66.
Traditionally it has been rather well established
that income has a direct correlation with voter participa
tion. About the same finding in this study in this respect
is, therefore, no surprise. Rather relationship this study
finds, that is, .32 is somewhat weaker than the results of
the others as cited above. Nevertheless, the direction it
shows is clearly consistent with what the literature in
this area has been saying all along the way, i.e., higher j
I
the income, higher the rate of voter participation. !
Another finding that has been rather well estab- j
|
lished over the years is the high correlation between voter j
13
See Angus Campbell, et al., The American Voter
(New York: Wiley, 1960), pp. 65-89; Lloyd A. Fress and
Jack Dennis (eds.), The Political Beliefs of Americans
(New Brunswick, N. J.: Rutgers University Press, 1968);
David Wallace, First Tuesday, A Study of Rationality in
Voting (New Yorici Doubleday, 1964); Kenneth Langton,
Political Socialization (New York: Oxford University
Press, 1969); V. 0. Key, Public Opinion and American Dem
ocracy (New York: Knopf, 1961).
14
Dye, op. cit., p. 59.
Dawson and Robinson, op. cit., p. 215.
108
participation and education.^ xn this study, median
school years completed by the population in thirty-eight
states were divided into three ranges (see Table 5).
Those three ranges of educational achievement of
the states are analyzed in relation to the level of voting
participation as shown in Table 15.
Of twelve states in the high level of educational
achievement, none of them is low voting state; all of the
twelve are medium to high voting states with two (Wyoming
and Utah) in the high voting category. This shows how the
states with high educational resource tends to be more
politically active. :
Of eighteen medium educated states, an overwhelming
i
majority, that is, seventeen out of eighteen are medium to
high voting states with exception of one state (Texas)
|
belonging to low voting category. Texas again proves to j
be a maverick state as it was mentioned in the previous |
section; explanation for Texas' deviation from the general
pattern may lie in the history and culture of that state
i
embedded in limited franchise relative to racial bias and
1 f t
See Lane, loc. cit.; Dye, op. cit., p. 73; Mil-
brath, op. cit., p. 67; James Barber, Citizen Politics
(Chicago: Markham, 1969), p. 13; Bernard Berelson, et al.t
Voting: A Study of Opinion Formation in Presidential
Campaign (Chicago: University of Chicago Press, 1954);
Gabriel Almond and Sidney Verba, The Civic Culture (Prince
ton: Princeton University Press, 1963).
TABLE 15
109
RELATIONSHIP BETWEEN VOTER PARTICIPATION
AND EDUCATION OF THIRTY-SEVEN STATES
POLITICAL PARTICIPATION3
Education High Medium Low
High (2) (10)
Wyo.
Utah
Idaho
Me.
Neb.
Wash.
Kan.
Cal.
Nev.
Ore.
Mass
Mont
Del.
(13)
Okla. Wis.
S. D. Md. Ohio
R. I. N. M. Penn
Conn. N. Y. N. H
Iowa 111.
Vt. Minn
Mich
(0)
12
Medium (4) ( 1)
Texas
18
Low (0) (4)
27
W. V,
Mo.
Ky.
N. D.
(3)
Ga •
S. C,
La.
7
37
Simple correlation coefficient between education and
political participation, r=.33.
concentrated wealth in a few industries.^
Ten per cent of eligible voters who are minority
Key, Southern Politics (New York: Knopf, 1959),
pp. 187-213; Austin Ranney, op. cit., pp. 61-68; Lockard,
loc. cit.
110 j
citizens are still barred in all practical sense from full
i
18
participation in Texas politics. Wealth in a few hands is
not equitably distributed over the general population,
although per capita income may show otherwise because it is
a measure of gross income divided by the number of popula
tion. Such per capita figure does not, therefore, mean
that each citizen enjoy that particular amount of annual
income.
Thus, when one is deprived of political franchise
for the reason of low personal disposable income, full j
i
political participation can not be accomplished.
Of seven low educated states, none of them is high
voting state; three Southern states (Georgia, South Caro- j
j
lina, Louisiana) or forty-three per cent are low voting i
states and four are medium voting states. In general,
therefore, low educated states tends toward medium to low
voting participation category.
When one looks at high voting states, all of the six
states in this category are either high or medium education
states; none of them is a state with low education. With
respect to medium voting states, twenty-three of twenty-
seven or eighty-five per cent come from the states with
educational level of medium to high. Rather small percen
tage, i.e., seventeen per cent or four states of twenty-
18
Barber, op. cit., p. 87.
Ill ,
seven in this medium voting category are the states with j
low level of education. Of four low voting states, none of
them is the state with high level of education; the majority
of them, that is, three of four come from the states with
low level of education.
The overall result tends to support the hypothesis
proposed in this study as well as established by other
studies that education has a positive effect on political
awareness and political participation. In this study,
simple correlation coefficient between education and voting i
participation is .35 which is significant enough to support [
I
the assumption that one of inputs of socioeconomic develop- j
ment, i.e., education, has a direct influence on one of the
i
characteristics of political system, voting participation.
Urbanization is another variable included in the J
i
socio-economic development. Rates of urbanization of j
thirty-seven states were presented in terms of three ranges j
j
of high, medium, and low in Table 9. Those three ranges
are analyzed (see Table 16) to see how closely they are
related to three ranges of voting participation.
The majority of high urbanization states, that is,
six out of seven are medium to high voting states. Of
sixteen states of medium urbanization, thirteen or eighty-
one per cent of sixteen are medium voting states; only two
states (Delaware, Utah) are in high voting and one (Louisi
ana) is in low voting category.
TABLE 16
RELATIONSHIP BETWEEN POLITICAL PARTICIPATION
AND RATE OF URBANIZATION OF
THIRTY-SEVEN STATES
POLITICAL PARTICIPATION* 1
Urbanization High Medium Low
High (2)
Conn.
R. I.
(4)
111.
Mas s.
N. Y.
Cal.
(1)
Texas
7
Medium (2)
Del.
Utah
(13)
Minn.
Ore.
Okla.
Wis.
N. M.
Mo.
Wash.
Nev.
Kan.
Penn.
Md.
Mich.
Ohio
(1)
La.
16
Low (2)
S. D.
Wyo.
(10)
N. D.
W. V.
Vt.
Ky.
Idaho
Mont.
Me.
Iowa
Neb.
N. H.
(2)
S. C.
Ga.
14
6 27 4 37
Simple correlation coefficient between urbanization and
political participation, r=.ll.
Of fourteen states with low urbanization, ten of them
or seventy-one per cent are medium voting states with two
in each of the other remaining high and low voting cate
gories. Of twenty-seven medium voting states, ten of them
are the states with low urbanization which is larger than
expected. Six high voting states are drawn equally from
113
all of the three ranges of urbanization, that is, two
states from each of high, medium, and low categories of
urbanization.
Thus, there appears to be some relationship between
urbanization and voting characteristic but strength of
relationship is very negligible; overall strength is only
.11. This result is not very much different from Dye's
finding which shows the correlation between urbanization
19
and voting participation to be .11. Urbanization doesn t
appear to have any direct influence on voting participation.
Summary results of correlation between political partici- j
pation and four variables of economic development are pre- !
sented below. '
TABLE 17
CORRELATION COEFFICIENT BETWEEN POLITICAL
PARTICIPATION AND ECONOMIC DEVELOPMENT
ECONOMIC DEVELOPMENT
Indus. Income Education Urban.
Political
Participation .03 .32* .35* .11
Asterisks refer to statistically significant relationship
at significance level of .05.
It appears that socio-economic development in
general does seem to exert influence in making a difference
19
Dye, op. cit., p. 62.
114;
! I
in one of the important characteristics of the states'
political system, i.e., voter participation. However, this
generalization may have to be qualified in the light of
very low correlation with respect to industrialization
(.03) and urbanization (.11). Thus, economic development
may have contributed its inputs to making a difference in
the rate of voter participation in American states only to
the extent that education and per capita income were
relevant factors. One other basic political system attri
bute, that is, character of representativeness manifested
in the rate of malapportionment is the next characteristic !
to be investigated.
In the words of the Supreme Court: ". . .as nearly i
i
as practicable, one man's vote is to be worth as much as j
20 21
another's." Until Baker vs. Carr in 1962, American j
state legislatures were not equally apportioned. However, |
in practical sense, Baker vs. Carr did not remove much |
inequity prevailing in many of American state legislatures,
i.e., underrepresentation in urban regions vs. overrepre
sentation in rural area.
Since fifty states are not alike in population gains
and loss, some states are more unequally distributed than
I some others. These differences are certainly one of the
90
Westberry v. Sanders, 84 S. Ct. 526 (1964).
21Baker v. Carr. 369 U.S. 189 (1962).
115
important characteristics of states' political system with j
i
serious impact on policy outcomes.
In this study, these differences in the rate of
equity of apportionment among the states were hypothesized
earlier in the first chapter of this study to be the conse
quence of differences in the rate of socio-economic devel
opment .
The degree of malapportionment in the states'
political system is hypothesized in this study to be the
consequence of the differences in the rate of socio-
i
economic development in each state as the differences in
the other two political system characteristics were just
i
!
dealt with in like manner.
Economic Development and Malapportionment
As it was described in methodology section of Chap-
22
ter three, like David and Eisenberg, Glendon Schubert and
23
Charles Press developed a technique for assessing how
well the states have managed the apportionment problem.
Schubert-Press came up with a set of index and ranks
of states, from we11-apportioned to the worst or mal-
22
David Paul and Ralph Eisenberg, Devaluation of the
Urban and Suburban Vote (Charlottesville: Bureau of Public
Administration, University of Virginia, 1961).
23
Glendon Schubert and Charles Press, "Measuring
Malapportionment," American Political Science Review, LXIII
(1964), 302-27.
116
apportioned, as indicated by selected population measures.
The indices developed by Schubert-Press that are presented
below are based on the degree to which the population
assigned to each legislative seat deviates from the popula
tion figure derived from: Total Population of State/Number
of Seat in Chamber.
TABLE 18
SCHUBKRT-PRESS RANKS OF APPORTIONMENT
IN THIRTY-EIGHT STATES
APPORTIONMENT
States Well-apportioned States Mal-apportioned
Rank Order Rank Order
Ohio 01 Montana 26
Oregon 02 Missouri 27
New Hampshire 03 Idaho 28
Massachusetts 04 Michigan 29
Utah 05 Rhode Island 32
Maine 06 Nebraska 33
Pennsylvania 08 Louisiana 34
New York 10 Texas 35
West Virginia 12 North Dakota 37
South Dakota 13 Maryland 38
Hawaii 14 Connecticut 39
Washington 15 Nevada 40
Wyoming 16 New Mexico 41
Wisconsin 18 California 42
South Carolina 20 Iowa 44
Delaware 22 Kentucky 45
Illinois 24 Oklahoma 47
Vermont 25 Kansas 48
Minnesota 49
Georgia 50
18 20
1
J Source: Glendon Schubert and Charles Press, "Measuring
Malapportionment," American Political Science Review, LXIII
|(1964), 315. Mean-25.31.
117
Thirty-eight states are divided into two groups,
i.e., well-apportioned and malapportioned according to
Schubert-Press's category. These two categories are
related to three ranges of industrialization as the latter
are presented in Table 3.
TABLE 19
RELATIONSHIP BETWEEN INDUSTRIALIZATION AND
APPORTIONMENT IN THIRTY-EIGHT STATES
APPORTIONMENT'
a
Industrial We11-Apport ioned Mal-apportioned
High (5)
Ohio 111.
Mass.
Penn.
New York
(3)
Mich.
Texas
Cal.
8
Medium (5)
Ore
W. V.
Wash.
Wis.
S. C.
(12)
Mo.
R. I.
Neb.
La.
Md.
Conn.
Iowa
Ky.
Okla.
Kan.
Minn.
Ga. 17
Low (8)
N. H.
Utah
Maine
S. D.
Hawaii
Wyo.
Del.
(5)
Mont.
Idaho
N. D.
Nev.
N. M.
13
18 20 38
Simple correlation coefficient between industrialization
and malapportionment, r=.08.
As can be readily seen, the relationship between
industrialization and apportionment is very erratic; of
seventeen states with medium rate of industrialization,
twelve are malapportioned states. Of eight highly indus
trialized states, three or thirty-seven per cent are mal-
apportioned states. Five of eight states are well-
apportioned states.
These five states, however, constitute only about
one third of eighteen states that are we11-apportioned.
Rather a good percentage or forty-four per cent of the
i
eighteen we11-apportioned states or eight (New Hampshire, i
|
Utah, Maine, South Dakota, Hawaii, Wyoming, Delaware) are !
i
the states with low industrialization.
Such large proportion of the we11-apportioned statesj
with low industrialization shows that industrialization is
not associated in any way with apportionment rate. With j
respect to malapportioned states, twelve of twenty mal-
apportioned states come from the states with medium rate of
industrialization. About this many states in this malap
portioned category ought to come from the states of low
industrialization to sustain the hypothesis that industrial
ization is an input in shaping apportionment consequence.
Only five states of twenty or twenty-five per cent are the
states with low rate of industrialization. Overall correla
tion coefficient is thus negative, -.08. Apparently
apportionment is not the function of industrialization in
119
American state political systems.
Education doesn't show very much relationship either
as Table 20 shows.
TABLE 20
RELATIONSHIP BETWEEN EDUCATION AND APPORTIONMENT
IN THIRTY-EIGHT STATES
APPORTIONMENT3
Education Well-Apportioned Mal-apportioned
High (8)
Ore.
Mass.
Utah
Maine
Hawaii
Wash.
Wyo.
Del.
(9)
Mont.
Idaho
Neb.
Conn.
Nev.
N. M.
Cal.
Iowa
Kan.
17
Medium (8)
Ohio
N. H.
Penn.
N. Y.
S. D.
Wis.
111.
Vt.
(6)
Mich.
R. I.
Texas
Md.
Okla.
Minn.
14
Low (2)
W. V.
S. c.
(5)
Mo.
La.
N. D.
Ky.
Ga • 7
18 20 38
Simple correlation coefficient between education and
malapportionment, r=.12.
Of twenty malapportioned states, fifteen are the
states with medium to high level of education; five are the
states with low education score.
The same erratic relation is shown with respect to
120 i
I
we11-apportioned states. Sixteen of eighteen well- j
apportioned states are medium to high level of education.
Two are the states in the category of low level of educa
tion. This portion alone may appear to support the
hypothesis that equitable representation has to do with
level of education of the citizens in the states.
However, this conclusion is not tenable at all when
fifteen of twenty malapportioned states are also the states
of high level of educational score.
The relationship between income and apportionment is j
I
not significant either as shown in Table 21. j
Fifteen states of high per capita income are just
about evenly divided into well-apportioned and malappor-
j
tioned categories with eight and seven respectively. With |
respect to the states of medium per capita income, twelve
states are distributed evenly over the two categories of
apportionment with six in each column.
These results amply show that the states' income
does not have anything to do with apportionment character
istic because we find as many malapportioned states as
well-apportioned in medium to high income states. If
income is to have any effect on apportionment, we may have
more of malapportioned states among the medium and far less
than seven among the high income states. Relationship if
any is highly negative.
The way eleven low income states are distributed
121
TABLE 21
RELATIONSHIP BETWEEN PER CAPITA INCOME AND
APPORTIONMENT IN THIRTY-EIGHT STATES
APPORTIONMENT3
Income We11-apport ioned Mal-apportioned
High (8)
Ohio Wash.
N. H. Del.
Mass. N. Y.
Hawaii 111.
(7)
Mich.
R. I.
Md.
Conn.
Nev.
Cal.
Minn.
15
Medium (6)
Ore.
Utah
Wyo.
Wis.
Vt.
Penn.
(6)
Mont.
Mo.
Neb.
Texas
Iowa
Kan.
12
Low (4)
Maine
W. V.
S. D.
S. C.
(7)
Idaho
La.
N. D.
N. M.
Ky.
Okla.
Ga.
11
18 20 38
£
Simple correlation coefficient between per capita income
and malapportionment, r=-.0029.
over the two apportionment characteristics does seem to
show positive relationship; seven of eleven or sixty-four
per cent of the total are malapportioned states. That is,
it indicates with some show of strength that low income
states tend to cluster around malapportioned characteristic
in political system.
However, this relationship is immediately nullified
122
by negative results shown among twenty-seven medium to high i
income states. Pearson correlation coefficient, -.0029,
shows that the overall relationship between income and
apportionment characteristics is insignificant. It is,
therefore, rather clear that apportionment characteristics
must have been the function of the variable other than
income.
That other variable is by no means urbanization,
another economic variable included in the study. Urbaniza
tion does not show any systematic relationship with
apportionment characteristic, as is shown in Table 22. i
With exception of sixteen medium income states, j
i
eight of high income states and fourteen of low income |
i
states are evenly divided into the we 11-apportioned and
malapportioned categories with four and seven in each of |
the respective categories. |
i
Sixteen medium income states appear to be an excep- j
tion; seven of sixteen states in this medium urbanization
category are we11-apportioned states and the rest of nine
states are malapportioned. Uneven as it is, the relation
ship is contrary to what is hypothesized. If the hypo
thesis that higher urbanization leads to better apportion
ment is to be tenable, there ought to be more of well-
apportioned states than seven and far less than present
nine of sixteen medium urbanized states in malapportioned
column.
I
123
TABLE 22
RELATIONSHIP BETWEEN URBANIZATION AND
APPORTIONMENT IN THIRTY-EIGHT STATES
APPORTIONMENT4 1
Urbanization We11-Apport i oned Mai-Apportioned
High (4)
Mass.
N. Y.
Hawaii
111.
(4)
R. I.
Texas
Conn.
Cal.
8
Medium (7)
Ohio
Ore.
Utah
Penn.
Wash.
Wis.
Del.
(9)
Mo.
Mich.
La.
Md.
Nev.
N. M.
Okla.
Kan.
Minn.
16
Low (7)
N. H.
Me.
W. V.
S. D.
Wyo.
S. C.
Vt.
(7)
Mont.
Idaho
Neb.
N. D.
Iowa
Ky.
Ga.
14
18 20 38
£
Simple correlation coefficient between urbanization and
malapportionment, r=.0160.
That there are more of medium urbanized states in j
the malapportioned category than we11-apportioned does not
lend any support to the assumption that higher urbanization
may lead to better apportionment in the states' legislature.
As expected, therefore, the overall correlation coefficient
between urbanization and apportionment is negative, i.e.,
-.0160.
124 ,
Summary !
This chapter presented the analysis of the data
which attempted to test the first hypothesis of the study.
That is, the higher the variations in the states' socio
economic development, the higher the degree of variations
in their political system characteristics.
Four variables included in the definition of eco
nomic development were examined in terms of their respec
tive strength of correlation with three variables of
political system characteristics. Four economic development
variables considered were (1) industrialization, (2) educa
tion, (3) income, and (4) urbanization. Three political
system variables were (1) interparty competition, (2) vot- j
ing participation, and (3) rate of apportionment.
Strength of relationships between the four economic
variables and three political variables are reviewed in
Table 23 in capsule form.
The values of correlation coefficient do not support
the general hypothesis that economic development has direct
influence on political system characteristics. Only cer
tain aspects of the general hypothesis are tenable; educa
tion and income as part of economic development do seem to
have significant effect on the degree of states' inter
party competitiveness.
On the other hand, industrialization and urbanization
as the measures of economic development seem to have no
125
TABLE 23
RELATIONSHIP BETWEEN ECONOMIC DEVELOPMENT
AND POLITICAL SYSTEM CHARACTERISTICS
Economic Development
Variables
Political Syst.
characteristics
Industri
alization Education Income
Urbani
zation
Interparty
competition .05 .31* .25* .11
Political
Participation .03 .33* .32* .11
Apport ionment
Rate -.08 -.12 -.01 -.01
‘Asterisks refer to statistically significant relationships j
at the level of significance of .05.
significant impact on the states' party competition. Indus-i
trialization and urbanization are found to be consistently ;
* I
unrelated to all of the three political system character-
!
istics; political participation and apportionment are j
apparently the function of neither industrialization nor
urbanization.
Political participation as one of the characteris
tics of political system is shown to be strongly related to
income and educational level of the states. Thus, education
and income emerge to be critical variables of economic
development in influencing two characteristics of political
system, i.e., interparty competitiveness and political
participation.
126
Significant as education and income are with respect
to party competition and political participation, they did
not show any relationship with apportionment characteristic
of the political system. In general, all of the four
economic development variables are unrelated to the appor
tionment characteristic. In fact, the relationship is not
only insignificant but also negative.
This negative relationship of malapportionment with
the economic development measures in this study is not an
exception. Dye, for instance, found the relationship
between apportionment score and socio-economic measures to
be very low. In order of urbanization, industrialization,
income, and education, correlation coefficients are
reported to be .01, .14, .14, and .13 respectively in Dye's
- ^ 24
study.
Thus, it appears that apportionment characteristic
of American state political system is not the function of
socio-economic development but something else. What could
it be? This will be dealt with in the last chapter of this
study on interpretation and discussion of the findings.
Next concern of the study is to see whether it is politics
or economics that has the most influence on policy out
comes .
^Dye, op. cit., p. 68.
CHAPTER V
INTERFACE OF POLICY OUTPUTS WITH
ECONOMICS AND POLITICAL SYSTEM
Two explanatory linkages were suggested in the model
in regards to interface of policy outcomes with economics
and politics. One of the linkages suggested that political
system characteristics may have direct effect on policy
outcomes irrespective of differences in economic develop
ment because the political system is, after all, the place
where allocation of values is being determined.
The other linkage suggested that economic variables
may exert an independent influence on policy outcomes
irrespective of political system characteristics. Because
j
it may be that political system is there merely to legiti- |
i
mize the policy decisions as socio-economic forces indepen-j
j
dently bring pressures on the former. Therefore, the poli
tical system's influence on policy outcomes is not direct
but dictated by the independent forces of socio-economic
environment.
This chapter presents the data which were gathered
to investigate the tenability of the above two explanatory
hypotheses. In order to analyze the relative strength of
relationship of politics and economics with policy outcomes,
political and economic variables are separately subjected
____________ . . . 127________________________________ _
128 ,
to correlational analysis in relation to the following
three provisions of Medicaid programs: scope of coverage,
nature and extent of services offered, and size of expendi
tures in thirty-eight states.
Scope of Coverage
As it was previously described, Medicaid is a pro
gram which Federal and State governments share the costs of
medical care for the needy and medically needy. Its
ultimate aim is to make medical care of high quality avail-
able to all of those unable to pay for it.
I
i
As it has been put into effect, however, the scope
of coverage varies from state to state. Because individual
states are given considerable discretion as to whom and how j
many their respective program is to cover beyond the '
2
minimum mandatory coverage required by Federal legislation.
Thus, there are two broad categories in coverage
plan, viz., minimum mandatory coverage as a condition of
participation of a state in the program and optional
coverage left to the discretion of each state.
^"According to Section 1903(e) of original amendment
to Social Security Act of 1935 creating Title 19, all
states in the Union are to commit to comprehensive medical
care for substantially all the needy and medically needy by
1975; see Advisory Commission on Intergovernmental Rela
tions, Intergovernmental Problems in Medicaid (Washington,
D. C.: Government Printing Office, 1968), p. 58.
2Tbid.
129
Until enactment of Medicaid, the Federal government
helped finance medical care for the needy by sharing in the
cost of the categorical assistance programs which consisted
of those receiving aids under OAA (Old Age Assistance),
AFDC (Aid to the Families of Dependent Children), AB (Aid
to the Blind), and APTD (Aid to the Permanently and Totally
Disabled).
Under mandatory clause of Medicaid of 1966, each
state must include everyone receiving aid under these four
"categorically needy." In addition to this categorically
needy group, the states may include at their option the j
I
following groups who are lumped in this study as "others"
3 |
meaning other than mandatory coverage:
. Families with an unemployed father in the home in states
i
not making AFDC payments to such families and people
i
considered permanently and totally disabled under the j
j
Federal definition;
I
. People whose income and resources are large enough to j
cover the daily living expenses but are not adequate to
pay for their medical care and who are aged, disabled,
blind, or members of families with dependent children;
j
. All medically needy people under age 21 even though
they are not eligible for financial assistance under
federally aided public assistance program;
3Ibid., pp. 28-31.
130
. People who are receiving or are eligible for general
assistance under a statewide program (noncategorically
related needy).
. Those between the ages of 21 and 65 who have enough
income and resources for daily needs but not for
medical expenses and are not blind, disabled, or mem
bers of families with dependent children; this group is
referred to as noncategorically related medically
needy.
The scope of coverage under Medicaid plan looks,
then, as follows.
TABLE 24
MEDICAID COVERAGE GROUPINGS
Mandatory
minimum "Others"
Categorically 1. Certain categorically related
Needy (OAA, AFDC, needy
AB, APTD) 2. Categorically related medically
needy
3. Certain noncategorically related
medically needy
4. Noncategorically related needy
Source: Advisory Commission on Intergovernmental Relations,
Intergovernmental Problems in Medicaid (Washington, D. C.:
U.S. Government Printing Office, 1968), p. 12. |
i
i
As of the end of 1968, the states were by no means
uniform in the scope of coverage with respect to mandatory
and optional categories; all of the participating states
did adopt some features of categories other than minimum
131
mandatory. However, the total number of recipients covered
in effect show a wide range of variations. Table 25 gives
the number of persons covered by categorically needy manda
tory and optional coverage of "others."
State-to-state differences in the number of persons
covered by the program are the reflection in differences of
value commitment of each state to the medical program.
Particularly the number of persons covered under the
optional discretion is reflective of degree of "generosity"
or "liberality" of each state.
The purpose of this study is to analyze why these
differences may have occurred; is it due to differences in
the nature of political system in each state or differences
in economic development? For instance, Texas which is high
income state shows only 32 per cent of the total which is
committed to helping those other than minimum mandatory
category; Maryland which has low voting record shows
highest percentage, i.e., 56 per cent of the total com
mitted to categories beyond minimum requirements.
In order to see which of the political system char
acteristics and socio-economic development are responsible
for these variations in policy outcomes, the three cate
gories of data, i.e., mandatory, optional "others," and
the total were separately correlated with economic vari
ables and political system variables.
132
TABLE 25
NUMBER OF PERSONS COVERED BY MEDICAID
PLANS IN THIRTY-SEVEN STATES (1968)a
States
Mandatory
Minimum 'Other* Total
Other as
of Tota!
Cal. 518,400 167,400 607,000 18
Conn. 24,000 17,700 41,600 42
Del. 5,000 980 6,000 16
Ga • 39,600 76 39,700 -
Hawaii 5,200 3,100 8,300 37
Idaho 3,700 1,700 5,400 31
111. 172,000 38,700
210,000 18
Iowa 16,730 2,670 19,300 13
Kan. 25,800 6,700 32,500 21
Ky. 59,400 29,900 89,300 33
La. 58,400 12,100 70,500 17
Me. 3,600 1,800 5,400 33
Md. 40,472 53,628 94,100 56
Mass. 80,000 109,000 189,000 57
Mich. 77,500 22,500 100,000 22
Minn. 41,000 28,200 69,200 40
Mo. b b
Mont. 5,000 1,600 6,600 24
Neb. 10,400 5,100 15,500 33
Nev. c c
N.H. 4,700 2,300 6,900 33
N.M. 20,600 910 21,500 4
N.Y. 302,200 481,800 784,000 61
N.D. 5,570 2,630 8,200 32
Ohio 90,300 14,100 104,000 13
Okla. 4,600 4,600 d
Ore. 14,100 4,600 18,700 24
Penn. 69,500 53,800 123,800 43
R.I. 15,800 13,900 29,700 46
S.D. 2,400 2,200 4,600 47
Texas 13,400 6,300 19,700 32
Utah 9,060 2,340 11,400 20
Vt. 3,800 1,400 4,200 33
Wash. 34,500 32,900 67,400 48
W.V. 37,800
-
37,800
-
Wis. 30,000 35,800 65,700 54
Wyo. 800 160 960 16
133
Key--Table 25
a— Figures may not add because of rounding.
b--Program initiated in November; no payments made.
c--Data not reported.
d--Oklahoma data for mandatory minimum included in the
'Other' column.
Source: Advisory Commission on Intergovernmental Relations,
Intergovernmental Problems in Medicaid (Washington, D. C.:
Government Printing Office, 1968), p. 92.
TABLE 26
RELATIONSHIP BETWEEN POLITICAL SYSTEM CHARACTERISTICS
AND SCOPE OF COVERAGE IN TITLE 19 STATES
Political System
Characteristics
Scope of
Coverage
Interparty Voter
Competition Participation
Apportion
ment
Mandatory .05 .10 -.11
Optional
Others -.15 .10 -.13
Total (mandatory
+ Optional) -.04 .11 -.13
Figures are simple Pearson correlation coefficients.
As the value of the above correlation shows, politi
cal system characteristics as a whole do not show any
significant relationship with the outcomes in the scope of
coverage of recipient groups. Voter participation may be
regarded as something tending to show some positive rela
tionship but not at significant level.
134
Apportionment shows complete negative relationship j
in all of the three aspects of coverage; so does inter
party competition. However, economic development shows
amazingly high correlation with three aspects of recipient
categories.
TABLE 27
CORRELATION COEFFICIENT BETWEEN ECONOMIC DEVELOPMENT
AND SCOPE OF COVERAGE IN TITLE 19 STATES
Economic Development Variables
Scope of
Coverage Urban. Income Education Indust.
Mandatory
J.
.48 .42 .12 .84*
Optional
Others .41* .38* .04 .71*
Total (mandatory
+ Optional) .50* .44* .08 .87*
'Asterisks refer
at the level of
to statistically
significance of
significant
.05.
relationship
Except education, the rest of variables are highly
related to all of the three categories of the scope of
Medicaid recipients. Industrialization is found to be one
variable which shows the strongest relationship; next to
industrialization is urbanization with .48, .41, and .30
correlation coefficients.
Income is not as strongly related as urbanization or
industrialization. Nevertheless, its degree of strength of
,relationship is at a significant level, that is, .42, .38,
135 ,
.44 respectively in three categories of coverage. Defer
ring full interpretation of these findings to the concluding
chapter, it may suffice to note here that one of the policy
outcomes in Medicaid, viz., scope of coverage, appear to be
the function of economic development rather than political
system.
Range of Services
Services to be offered in states' Medicaid plan are
also organized in the same manner as the scope of coverage.
There are two basic categories, i.e., mandatory minimum and
optional. As Table 28 shows, such services as inpatient
hospital, outpatient hospital, laboratory X-ray, skilled
nursing, and physicians are the basic minimum services any |
i
state plan must include in compliance with Federal law.
Services beyond these minimum five are the conse- j
quence of each state's decision to commit their resources
to as many different kinds of services as possible. Maxi- j
mum number of services over and beyond five minimum extends |
up to fifteen as Table 28 shows. !
As of July 1, 1968, as Medicaid law required, all of
the thirty-eight states are uniform in offering the basic
five minimum services. However, there are a great deal of
variations in the optional services offered; California
and Connecticut offer all of the fifteen optional services;
4Ibid.
136
TABLE 28
TYPES OF SERVICES PROVIDED UNDER MEDICAID
Code No. Services Type
1 Inpatient Hospital M E
2 Outpatient Hospital
o 3
4 - 1 E
3 Lab. X-Ray
c d
x) C
4 Skilled Nursing Home
c - h
is s
5 Physicians
6 Podiatrists
7 Chiropractors c
8 Naturopaths
o
t O t H
9 Home Health Care
0 1 - U
O 0 )
1 0 Private Duty Nursing
■ r ^ U
> o
L t o
C D • - ‘
1 1 Clinic
1 2 Dental
m ^ 3
1 3 Physical Therapy
»— i *
c d t o
G C D
1 4 Eyeglasses
1 5 Prosthetic Devices
O 4 - J
• r H id
16 Prescribed Drugs
4 J 4 - >
(XtO
1 7 Audiology
o
4 - >
18 Hearing Aids
t d
1 9 Other Diagnostic
2 0 Ambulance
Source: Advisory Commission on Intergovernmental
Relations, Intergovernmental Problems in Medicaid
(Washington, D. C.: Government Printing Office,
1968), p. 96.
New York offers fourteen; Texas which is relatively
wealthy state offers only ten while Wyoming offers none
of the optional services (see Table 29).
These variations are reflective of policy outcomes
of each state consequent to respective differences in
committing to the value of publicly assisted medical
program.______ _________ ___________________
137
TABLE 29
OPTIONAL SERVICES PROVIDED UNDER MEDICAID
BY THE STATES, JULY, 1968
Optional Services
States
a
6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Total
Calif. X X X X X X X X X X X X X X X 15
Conn. X X X X X X X X X X X X X X X 15
Del. X I
Oa . X X X 3
Hawaii X X X X X X X X X X X 11
Idaho X X 2
ill. X X X X X X X X X X X X X 13
Iowa X X X X X X X X X X X 11.
Kansas X X X X X X X X X y
Kent. X X X 3
Lei • X k X k X k 6
Me. X X 2
Md . X X X X X X X X X X 1U
Mass. X X X X X X X X
V
X X X X 13
Mich. X X X X X X X /
Minn. X X X X X X X X X X X X X X 14
Mo. X i
Mont. X X X X X X X /
Neb.
p
X X X X X X X X X X X
■ t t
Nevada X X X X X X X X X X X X 12
N ".'H . X X X X X X X X X X X X X X 14
N'.M . X X X X X X X X X X X X X X 1 4 "
N.V. X X X X X X X X X X X X X X 14
NT.D. X X X X X X X X X X X X X X 14
Ohio X X X X X X X X X X X X X X 14
Okl. X X X X X X X 7
Oregon X X X X X X X X X X X X X IT""
Penn. X X X X X 5
ITT....... X X X X X X X X X y
S.C. X X X X X X 6
s.b. X X X X X X X X X X ITT
*Texas X X X X X X X X 8
Utah X X X X X X X X X X 'ITT"
V t. X X X X X X 6
Wash. X X X X X X X X X X X X X 13
W.V. X X X X X X X X X X 10
WIs . X X X X X X X X X X X X 12
Wyo. 0
Numbers refer to the services coded in numerical order in
Table 28. Source: Advisory Commission on Intergovernmen
tal Relations, Intergovernmental Problems in Medicaid
(Washington, D. C.: Government Printing Office, 1968),
p. 96.
138
Mean of the number of services offered is 9.6;
range is spread from the highest number of fifteen (Cali
fornia, Connecticut) to the lowest, zero (Wyoming).
In order to explain or account for these state-to-
state variations, political system variables and economic
variables are again brought into the test to compare and
analyze the relationship of each set of variables with the
number of services offered. Simple correlation values are
presented below:
TABLE 30
CORRELATION COEFFICIENT BETWEEN NUMBER OF SERVICES
OFFERED AND POLITICAL AND ECONOMIC VARIABLES
Political System Variables
Party
Competition
Voter
Participation
Apportion
ment
Number of
Services .01 .17 -.25
Economic Development Variables
Urban. Income Education Industrial
ization
Number of
Services .40* .42*
n r *
.2j .31*
Asterisks refer to statistically significant relationships
at the level of significance of .05.
139
None of the political variables shows any relation
ship with variations in the number of services offered by
the states. In fact, apportionment shows a negative rela
tionship.
On the other hand, economic variables show very high
relationship with the scope of services offered. Particu
larly income shows a value of .42, the highest positive
value of the four economic variables.
In fact, all of the four variables hold significant
relation with the scope of services, while all of three
political system variables show low insignificant correla
tional values. Thus, these data provide a reasonable
support to the hypothesis that political system may be
there to merely translate the socio-economic environmental
inputs into a set of policy outcomes.
Regardless of the variations in such political
variables as party competition, voter participation, or
apportionment, the system may not be activated unless it
feels the pressure or demands from the environment. Eco
nomic environment, particularly, state's wealth or income,
appears to be the single most important variable impacting
independently upon the political system in making the dif
ferences in policy outcomes.
140
Expenditures in Medicaid !
Related to the scope of coverage and services is the
money involved in financing the program in the states. Of
course, the states that cover greater number of people with
more variety in the services spent more money than the ones
simply covering less number of people.
Table 31 shows the range of expenditure commitment
5
of each state in terms of state-local share of the costs
with Federal government and per capita. Total expenditures
range from the highest of $666.8 million (California) to the
lowest of $1.1. million (Wyoming).
One can readily observe in Table 31 that West Vir- j
ginia is the state showing the lowest per capita cost of |
$3.39 and California showing the highest per capita cost of
$35.49. Michigan ($4.97), New Hampshire ($6.40), are some
of the low range states in per capita cost; Kansas ($34.57),
New York ($23.87), and Rhode Island ($18.49) are the high
range states. !
j
These variations in money expenditure are another
“*For the purpose of this study, local government is
assumed as an extension of state government. Therefore
local share as such is not treated separately from state
share; they are lumped together. Title 19 as amended
requires that States must pay at least 40 per cent of the
non-Federal share of costs and thereafter must pay either
the entire non-Federal cost or must distribute State and
Federal funds in such a way that inadequacy of local funds
will not cause lowering of the scope or quality of the
program; see Ibid., p. 17.
141
TABLE 31
COSTS OF
THIRTY-
MEDICAID AS ADOPTED BY
■SEVEN STATES (1968)
States
Total
(millions)
Fed. Share
of Costs %
State-Local
Share %
Per
Capita
Ga • $ 41.6 81 19 $ 9.35
Idaho 7.1 71 29 10.18
La. 36.5 76 24 10.10
Me. 8.0 70 30 8.13
Mo. 28.0 74 26 6.13
Mont. 7.0 64 36 9.96
Neb. 17.5 60 40 12.16
Nevada 5.0 50 50 11.60
N.M. 5.7 71 29 5.70
Ohio 35.3 52 48 3.40
Oregon 18.9 54 46 9.60
S.D. 8.9 73 27 12.91
Texas 123.2 80 20 11.47
Vt. 6.3 68 32 15.22
W.V. 20.8 74 26 11.51
Wyoming 1.1 59 41 3.39
111. 87.9 50 50 8.15
Ken. 36.3 77 23 11.40
N.H. 4.3 60 40 6.40
N.D. 9.6 67 33 14.85
R. I. 16.6 56 44 18.49
Conn. 30.6 50 50 10.63
Del. 4.4 50 50 8.57
Iowa 35.0 60 40 12.68
Kansas 78.5 58 42 34.51
Mass. 139.9 50 50 25.89
Minn. 69.5 60 40 19.45
Olcla. 72.5 70 30 29.26
Cal. 666.8 50 50 35.46
Hawaii 5.8 53 47 8.03
Md. 33.3 50 50 9.33
Mich. 42.1 50 50 4.97
N.Y. 434.7 50 50 23.87
Penn. 113.7 54 46 9.80
Utah 8.8 66 34 8.78
Wash. 31.9 50 50 10.49
Wis. 62.1 58 42 14.90
Source: Tax; Foundation, Medicaid: State Programs After Two
Years (New York: Tax Foundation, Inc., 1968), p. 67.
142
important dimension of differences in policy outputs of
individual states. These expenditure data are also put to
correlational analysis in relation to political and eco
nomic variables just as the scope of services and benefits
was dealt with.
TABLE 32
CORRELATION COEFFICIENT BETWEEN THREE MEASURES
OF TITLE 19 STATES' EXPENDITURE AND
POLITICAL AND ECONOMIC VARIABLES
Political System
Variables
Expenditure Party Voter Apportion
Categories Competition Participation ment
Federal Share .04 •12 -.12
State-Local .06 •13 -.17
Per Capita
1
00
o
•
•15 .02
Economic Development
Variables
Expenditure Urbani Per Capita Educat ion Industri
zation Income alization
Federal Share
_ , &
.01 .43* .14 .85*
State-Local
00
•
.50* .12 .84*
Per Capita .40* .45* .05 .30*
Asterisks refer to statistically significant relationship
at the level of significance of .05.
Political variables show very low relationships with
expenditure categories, although voter participation tends
toward some positive direction. But strength of this
positive direction is negligible. Economic variables
except education are highly related to all of the three |
expenditure categories.
The higher the income, urbanization, and industri
alization, the higher the degree of commitment of a state
to medical expenditures. Particularly industrialization is
most strongly related to Federal and State-Local shares.
That economic development is found to be strongly
related to expenditure decisions may indicate that growing
industrialization plus expanding urbanization in a state is
more of an important direct input in shaping the nature of
policy than a handful of legislators or electorate.
Summary
The purpose of this chapter was to test the last two
hypotheses proposed in the model of this study. One of the!
I
two hypotheses was that the political system may be an j
independent variable mediating between socio-economic !
environmental forces and policy outcomes. ;
0
This high correlation of Federal share of cost with
economic development variables except education is rather
expected because such a close relationship is built into
Federal share formula in Medicaid legislation. Medicaid
costs, including health insurance premium, are reimbursed
from Federal funds at the rate of the State*s medical
assistance percentage. Federal matching varies from 50 to
83 per cent--inversely in relation to a State's per capita
income, with the highest Federal matching going to States
with the lowest per capita income. A state receives about
55 per cent Federal matching funds if its per capita income
is equal to the national average; see Ibid., p. 15.
That is, no matter how strong the environmental |
pressures may be, they may not be able to activate the
political system unless the latter takes an initiative inde
pendent of the environmental inputs. An opposite view
was, however, presented in the second hypothesis that
environmental forces are the inputs causing the political
system to act. However active the political system may
wish to be, it can't act independent of the environmental
forces in shaping the policy outcome. The system may
merely process the demands coming from socio-economic pres
sures; these pressures in form of inputs directly determine)
the size and nature of decision outputs, not the political
system which merely functions as a transmission belt with
no independent effect on the outcome.
I
There were three aspects of the decisions which !
generated the policy data for this study: (a) the scope of j
beneficiaries covered, (b) scope and nature of services !
offered, and (c) monies expended in supporting the program.;
The data on these three aspects of the program were
regarded as "dependent variables." To explain the depen
dent variables, the political system and economic develop
ment variables were put to correlational analysis relative
to the above three aspects of the program.
Explanation was, of course, implied in the hypo
theses, that is, the dependent variables may have been the
consequence of function of political system characteristics
145
or the function of environmental inputs independent of the j
!
political system.
To resolve the competing claim of political system
and economic development variables in relation to policy
outcomes, comparative analysis of correlational strength of
both against each other as well as against the individual
variables was carried on. Summary of the results is shown
in Table 33.
The correlational coefficient values of political
system variables relative to all of the three aspects of
policy data, i.e., scope of coverage, benefits, and costs, j
i
were consistently insignificant; they are in the range of I
-12 to .17. I
On the basis of such low correlational values
i
between policy outcomes and political system variables, j
!
then, the hypothesis that political system may have had an
independent effect on the policy outcome (figure 3), is ;
- j j
considered untenable. Indeed, as Dawson-Robinson and
O
Dye claim, the political system may be there merely to
transmit the demands emanating independently from the envi
ronment .
^Richard Dawson and James Robinson, "Inter-party
Competition, Economic Variables and Welfare Policies in the
American States," Journal of Politics, XXV (1963), 288-89.
g
Thomas Dye, Politics, Economics, and the Public:
Policy Outcomes in the American States (Chicago:Rand
McNally, 1966), p. 285.
!
TABLE 33
CORRELATION COEFFICIENTS BETWEEN SEVEN MEASURES OF MEDICAID POLICY
OUTCOMES AND POLITICAL AND ECONOMIC VARIABLES
Economic Development
Variables
Political Svstem
j
Variables
Policy Outputs Urban. Income Ed. Indus.
Political
Competition
Voter
Turnout
Apportion
ment
Mandatory
minimum .48* .42* .12 .84* .05 .10 -.11
Optional
Others .41" .38* .04 .71* -.15 .10 -.13
Total (mandatory
+ Optional) .50" .44" .08
CO
•
-.04 .11 -.13
No. of Services
Offered .40*
JU
.42" .25 .31* .01 .17 -.25
Federal Share
JL
.51" .43* .14 .85* .04 .12 -.12
State-Local Share .48* .50* .12 .84* .06 .13 -.17
Per Capita Cost .40* .45* .05 .30* .08 .15 .02
Asterisks refer to statistically significant relationships at the level of significance
of .05.
This latter conclusion is well supported by gener
ally high correlation between economic development variables
and seven related parameters of policy output in Medicaid.
As summary of the data in Table 33 shows, industrialization,
urbanization, and income show significant relationship
ranging in correlational values from .30 (per capita cost
and industrialization) to highest value of .87 (total
number of beneficiaries and industrialization).
Education as a variable of economic development
tends to show rather low correlation ranging from low value
of .04 (optional categories) to .25 (number of services
offered). It appears that a state with higher rate of
urbanization, income, and industrialization may be likely to
allocate higher amount of resources to health program,
regardless of the level of education of the citizenry.
Thus, wealth manifested in high rate of industrial
ization, high income and urbanization seems to be the
strongest force causing the differences in policy outputs.
Therefore, the hypothesis that socioeconomic forces exert
an independent influence on the policy outcome (D3 in
Figure 3) may be considered tenable in this study.
CHAPTER VI
IMPACT OF POLICY OUTPUTS: MEDICAID
The preceding chapter presented the analysis of the
data obtained to test the hypothesis proposed to account
for the variations in policy outcomes in the states. The
data supported the view that the variations in policy out
puts are the function of variations in the rate of socio
economic development in the states rather than political
system characteristics.
This was the 'process' study of the public policy, !
because the focus was on analyzing the process through
which the policy decisions were made. It was argued, how
ever, in the earlier part of this study^ that explaining j
the 'process' alone is not enough in public policy studies. |
Knowing the impact of implementation of public |
i
policies was argued to be as important as explaining the
'process' for two reasons: theoretical and pragmatic. It
was argued that the process approach took the policy out
comes as 'dependent' variables to be explained. Conse
quently, it held the socio-economic development and
■^The reader is referred to the section of first
chapter of this study on explanation of the model; see also
Austin Ranney (ed.), Public Policy and Political Science
(Chicago: Markham, 1968), pp. 3-22.
148
149
j
political systems as 'constants' or independent variables.
Emphasis on the impact of public policies was
advanced on the proposition that the theoretical perspec
tive of the 'process' approach should be reversed. That is,!
to look at the environment and political systems as some
thing varying under the impact of the policies rather than
holding them as constants. This new perspective was termed
t 1
the content approach.
Pragmatically it was suggested that knowing what a
certain social policy has done to the intended target group |
i
or the system is just as much critical, if not more, as
being told how the policy decisions were made. j
Policy impacts are conceptualized in this study as j
the forces engendered by the implementation of a policy. I
These impacts are, in turn, visualized as having been fed
back into the system completing its feedback loop.
This chapter presents the data on the impact of two
and a half years of implementation of Medicaid Program.
Presentation is organized under the following four head
ings: (1) brief history of Medicaid, (2) fiscal effects of
Medicaid, (3) non-fiscal effects of the program and (4) the
feedback of the effects into the system in the last section
on "Summary and Analysis."
Medicaid: A Brief History
In the early colonial period, common law established
150
that care of the poor including medically poor was funda
mentally a local problem. As the nation came to assume
federal form of government, it became the responsibility
2
of the state government to care for the medically poor.
It was not until the 1930s when the Federal govern
ment began to assume this responsibility. From 1933 to
1935, the Federal government through the Federal Emergency
Relief Administration for the first time made available to
the States funds to pay the medical expenses of the needy
unemployed.^
This FERA arrangement was very restricted in its
i
scope of services; it covered physicians fees, some drugs,!
j
nursing services. It excluded hospitalization from the
4 ^
coverage. However, FERA set a precedent in sharing the (
medical finances between Federal and State governments.
Federal-State partnership became more firmly estab- j
!
I
lished in the Social Security Act of 1935, in which the I
i
Federal Government shared with the States the cost of
providing maintenance to the needy aged, families with
dependent children, and the permanently and totally
2
Advisory Commission on Intergovernmental Relations,
Intergovernmental Problems in Medicaid (Washington, D. C.:
Government Printing Office, 1968), p . 3.
^Ibid.
4Ibid.
151 i
I
J '
disabled. In 1950, Congress broadened the definition of i
assistance to include money for 'vendor payments'--direct
payments to the suppliers of services, rather than welfare
g
recipient himself. These vendor payments, however, were
to be made within the existing maximums on Federal sharing,
then on an individual basis.
President Truman in 1945 attempted to introduce
bills which might have considerably expanded the scope of
coverage. But the bill met a great opposition. Eisenhower
also backed in 1953 a bill to be based on Social Security
Old Age and Survivors Insurance. This again was not
approved by either House.'7
Subsequently Kerr-Mills Bill was passed as a compro- .
I
mise piece which became a precursor to Medicare and Medic- [
aid. It differed in several important respects from
previous public assistance programs:^
(1) For the first time medical care was to be offered to
i
persons not on the regular welfare rolls, but who
were considered medically indigent, needing financial
aid in paying their medical bills but able to support
3Ibid.. p. 7.
^Ibid.
^Ibid., p. 8.
g
Eugene Feingold, Medicare: Policy and Politics
(San Francisco: Chandler Publishing Co., 1966), pp. 85-96.
........... 152 i
themselves otherwise;
(2) Federal standards were prescribed in the form of
some institutional and non-institutional care as
conditions of Federal sharing in the costs;
(3) The Federal government was to share with the states
in the total cost of the program on an open-ended
basis, without limitation on individual payments or
total state expenditures.
Implementation of the Kerr-Mills program was left up
to the states. Some forty-three states adopted the program
in one form or another. Under this influence, public !
!
medical vendor payments tripled after 1960, rising to some |
$1.5 billion by 1965 and accounting for over fifteen per
9
cent of public expenditures for health and medical care. .
Enormous as it may seem, it was only three per cent j
of the total medical costs of the nation, that is, $40 i
billion.^ It was not a comprehensive national health care
I
program by any means. Thus, in 1964, President Lyndon B.
Johnson proposed in his annual message that the nation
should make an effort in inaugurating a comprehensive
medical and health care program for both the aged and those
unable to provide adequate care themselves.
g
Advisory Commission Report, op. cit., p. 9.
10Ibid.
Direct outgrowth of this proposal by Johnson came to
be known as Title 18 and 19 amendments to the original
Social Security Act of 1935. Title 18 is what is referred
to as Medicare which is essentially a health insurance
program for those over 65.
The latter, Title 19, is the Medicaid, which covers
those under 65; they may be either those who are medical
indigents or those who are unable to provide any care
medically but able to support themselves otherwise.
General method of financing of both programs is Federal
matching with the contribution of the state government.
I
I
The details of these two programs are spelled out in;
i
I
the following sections dealing with the impact of Medicaid.
Fiscal Impact j
Impact of Medicaid in terms of specific figures in
dollars and cents is manifested in three sectors of the
j
nation's health and medical care system. Those three !
sectors are (a) consumer groups, (b) providers of services,
and (c) public revenue sources.
Consumer Groups
One can raise a simple question here: how much
benefit did Medicaid bring to those groups whom the pro
gram intended to reach? As Dr. I. J. Brightman of the New
York State Department of Health said:
154 ,
I
The effect of Medicaid in terms of the amount of j
care it has provided is difficult to measure.^
Neither inputs nor outputs--costs and benefits—
lend themselves to neat quantitative measurement. Because
it is difficult to isolate the effects of a single program
such as Medicaid under any circumstances.
There is a multiplicity of other programs operating
12
simultaneously and other concurrent influences on health.
A variety of Blue Shield and Cross, employer-sponsored I
health insurance, advances in living standards, and a (
pattern of personal health habits are some variables enter-;
ing into overall effect on longevity and general state of
health.
i
However, there are some data on national health j
expenditures financed by private and public sources. There
are interesting variations in the size of expenditures for
personal health care before and after Medicaid became
effective in 1966.
As Table 34 shows, the major effect of the program
■^Tax Foundation, Medicaid: State Programs After Two
Years (New York: Tax Foundation, Inc., 1968j, p. 55.
12
In 1969, some 1,785 private insuring organizations
in the United States made health insurance available to the
public. This total included 1,405 insurance companies, 75
Blue Cross plans, 75 Blue Shield plans, and nearly 600
independent-type health insurance plans; see, Data: 1969
Source Book of Health Insurance (New York: Health Insur-
ance Institute, 1969), pp. 1-8.
TABLE 34
PERSONAL HEALTH CARE EXPENDITURES BY SOURCE
OF FUNDS (FISCAL YEARS 1960-67)
Amount (billions)
Percent of
Total
Fiscal Y Total Private Public Private Public
1960 $23.4 $18.4 $ 4.9 78.9 21.1
1961 25.7 19.2 5.4 77.9 22.1
1962 26.3 20.5 5.7 78.1 21.9
1963 28.2 22.1 6.2 78.2 21.8
1964 30.9 24.2 6.7 78.4 21.6
1965 33.9 26.8 7.1 79.0 21.0
1966 36.8 28.6 8.1 77.9 22.1
1967 41.5 28.8 12.6 69.5 30.5
Source: Tax Foundation, Medicaid: State Programs After Two
Years (New York: Tax Foundat i on, Inc., 1968), p. 54.
has been in the shift in the financing of care from the
private to the public.
During the fiscal 1967 , national personal health
care outlays show the largest absolute gain on record,
rising from $4.7 billion to $41.5 billion in 1967. Such
a shift may be interpreted as lessening the financial
burden of private sources.
However, this shift in financing from private to
public does not have any direct bearing on the additional
benefits that may have been generated. Because increasing
medical care prices and population absorbed much of the
gain in absolute size of total expenditure during fiscal
1967 (see Tables 36 and 37). Thus, the resulting real per
13
capita improvement is reported by the Tax Foundation to
be less than five per cent which is, in fact, a smaller
percentage than the annual gains in two previous years in
the 1960s (see Table 35).
TABLE 35
PER CAPITA PERSONAL HEALTH CARE
EXPENDITURES IN CONSTANT 1967
DOLLARS FY 1960-1967
FY
Amount
Per
Capita
Percent Change
Over the Previous
Year
1960 $159.88
1961 161.00 +0.6
1962 163.82 +1.7
1963 169.22 +3.3
1964 178.86 +5.7
1965 189.20 +5.8
1966 196.69 +4.0
1967 206.31 +4.9
Source: Tax Foundation, Medicaid: State
Programs After Two Years (New York:Tax
Foundation, Inc., 1968) , p. 55. j
Thus, in spite of absolute increase of personal
health care expenditure from both private and public source?
in 1967 from previous 36.8 billion dollars to 41.5, per
capita health care improvement rather decreased from pre
vious two years prior to Medicaid, that is, 1964 and 1965
13
Tax Foundation Report, op. cit., p. 55.
157
with percentage of 5.7 and 5.8 respectively to 4.0 and 4.9
per cent in 1966 and 1967.
Providers of Services
It was just noted that an absolute increase in
expenditures during Medicaid effected years of 1966 and
1967 did not really contribute to the improvement of health
care. Rather per capita health expenditure decreased from
previous years.
As Tables 36 and 37 show, major effect Medicaid
brought about is the rapid rise in medical care prices and
costs, by pushing up the demand more rapidly than supply
resources could be augmented.
TABLE 36
PERCENT OF PHYSICIANS REPORTING
FEE INCREASE (FY 1966-67)
Percent of Physicians
Procedures Reporting Fee Increase
Average Percent
age of Increase
in Fees
Family doctor,
Office visits 32.4 23.3
Family doctor,
House calls 27.2 20.7
In-hospital care 26.2 18.0
In-hospital Surgical
care 35.8 16.6
Source: Tax Foundation, Medicaid: State Programs After
Two Years (New York: Tax Foundation, Inc., 1968), p. 44.
158
j
Twenty-six to thirty-five per cent of the doctors i
surveyed report fee increases in office visits to in-
hospital surgical treatment ranging in average from sixteen
per cent to twenty-three per cent during 1965 and 1967
(see Table 37).
During the period of twelve months ending in June
1967, medical care prices rose 7.3 per cent, in comparison
with a 2.7 per cent rise in the overall consumer price
index. The medical care services component of the index,
however, rose 9.2 per cent, and daily hospital service
i
charges by an alarming 21.9 per cent (see Table 37). j
, I
bven though physicians fees over-all went up only j
7.3 per cent, nearly one-third of the physicians increased
their fees for regular procedures during the twelve month ;
period by amounts ranging from seventeen to twenty-three ,
per cent (see Table 36). j
It is a truism that excess of demand over supply |
tends to push the cost up. Shortages of supply relative
to demand in medical and health field are of several kinds;
personnel, facilities, to name a few. However, it was
widely reported right after Medicaid became effective that
less than honest charge for the costs was rampant among the
physicians and health service providers. They saw a
chance to profiteer in the government-supported medical
program which defies a clear accountability.
TABLE 37
MEDICAL CARE PRICE CHANGES ON SELECTED ITEMS
AND PERIODS (JUNE 65-DEC. 67)
Index, 1957-59=100 Annual rate of change
Items
June
1965
June
1966
June
1967
Dec.
1967
June 65-
June 66
June 66'
June 67
June 67-
Dec. 67
Consumer price 110.1 112.9 116.0 118.2 +2.5 +2.7 +3.8
Medical care,
Total 122.2 127.0 136.3 140.4 +3.9 +7.3 +6.1
Medical care
Service 127.0 133.0 145.2 150.4 +4.7 +9.2 +7.3
Physicians
fees 121.1 128.0 137.3 141.0 +5.7 +7.3 +5.5
Dentists' fees 117.4 120.9 126.9 130.7 +3.0 +5.0 +6.1
Daily hospital
charge 152.5 162.2 200.1 211.4 +7.7 +21.9 +11.5
Drugs & Pre
scriptions 98.1 98.6 97.7 98.1 +0.5 -0.9 +0.8
Source: Tax Foundation, Medicaid: State Programs After Two Years (New York:
Tax Foundation, Inc., 1968), p. 43.
159
The Tax Foundation devised a questionnaire"^ in 1967
out of such widely reported news information. The ques
tionnaire was essentially a list of those problem areas
pertinent to price-rise and then asked twenty-six states to
check the items in accordance with applicability to each
individual state's experience.
Parts of the results relevant to factors behind the
rise of medical care prices are presented in Table 38.
The above survey shows that a majority of the states
that is, sixteen out of twenty-six, checked on general
cost-push inflation; nine and eight checked on shortage of
personnel and facilities.
Widely rumored 'racketeering' on the part of medical
practitioners has met the response from only one state.
Obverse of the medical practitioners' racketeering is the
abuse of "free-service" by the general consumers, i.e.,
care recipients; making unwarranted use of the services in
the form of visiting a doctor just for a common cold, etc.,
was experienced by only one state.
However, considerable number of the states, that is,
eight of them, experienced the cases of 'profiteering' in
the form of taking advantage of maximum allowance. This
concept of maximum allowance practically eliminates the
"^Tax Foundation Report, op. cit., p. 42.
161
TABLE 38
SOME FACTORS UNDERLYING THE RISE OF
PRICES DERIVED FROM CRITICISM OF
PROGRAM OPERATIONS (JUNE, !
MEDICAL CARE
TITLE 19
1967)
N=26
Factors Number of States
Checked checking
Shortage of Personnel--
doctors, nurses, auxiliary 9
Shortage of facilities 8
Availability of medical care had led
to unwarranted use or over-use 1
"Profiteering," i.e, raising of fees
to take advantage of maximum allowable 8
Increase in charges for hospital to take
into account allowances for depreciation
of building and equipment 5
General cost-push inflation, e.g.,
rising salaries 16
"Racketeering" on part of medical
practitioners (e.g., reported payments
for doctors' visits never made) 1
Source: Medicaid (New York: Tax Foundation, Inc., 1968),
p. 42.
economy incentive which results in excessive charge and
care costs.
Thus, the rise in medical prices appears to be due
to two kinds of causes, justifiable and unjustifiable.
Perhaps, such factors as shortage of personnel, facilities,
and general inflation may be legitimate variables causing
the cost-push. But other such factors as profiteering and
i
i 162
excessive charges in allowance for depreciation of build- !
ings and equipment may not be economically or socially
justifiable.
i
Impact On Public Revenue
Sharply rising prices and costs naturally fell on
the shoulders of the Federal and State agencies involved in
financing the program once committed to it. Was this
increase expected among the State governments? How were
the costs and prices met? '.That fiscal problems were encoun-j
tered and how have they been dealt with by State and local
governments?
As late as six months before the end of the fiscal
year 1967, the budget estimated Federal Title 19 cost in i
that fiscal year at $884 million, the numbers to be served
15
at 6.4 million, with an expected 30 states participating.
The actual costs turned out to be forty-four per
cent higher, $1,173 million, despite participation of
fewer states and about nineteen per cent fewer recipients
16
than estimated in January.
The 1969 Federal budget shows a revised estimate of
$1,761 million for Title 19 costs to the Federal government
|in fiscal 1968, about 50 per cent higher than estimated a
^Tax Foundation Report, op. cit., p. 30.
Ibid.
TABLE 39
FEDERAL EXPENDITURES FOR TITLE 19 PROGRAMS
ACTUAL AND ESTIMATED, FISCAL YEARS 66-69
Year and Date of
Estimate
Federal Costs
in Millions
No. Aided
(millions)
No. of States
Participating
Fiscal 1966, Actual3 $ 202. 1.6 9
Fiscal 1967:
Jan., 1967 Estimate 884 6.4 30
Actual 1173 3.2 29
Fiscal 1968
Jan.,1967 Estimate 1162 8.0 48
Jan.,1968 Estimate 1761 7.3 43
Fiscal 1969:
Jan., 1968 Estimate 2121 8.5 48
The first six state programs began operations at the
middle of the fiscal year (January 1, 1966).
Source: The Budget of U.S. Government, fiscal years end
ing June 30, 1968 and June 30, 1969.
year earlier, even though the estimates of the number of
states and persons participating were revised downward.^7
VThy such a discrepancy between the estimates and
actuals in Federal budget? Since Federal involvement in
state Medicaid program is primarily cost-sharing partner
ship on percentage reimbursement basis, the ultimate
sources of miscalculation lie in the characteristics of the
individual participating states plus general cost-push
17 Ibid.
164
inflation in the economy. According to the Tax Foundation j
survey:
The errors in the advance national estimates were
apparently due largely to the major discrepancies
between expected and realized costs in California and
New York.lo
The Tax Foundation reported further that:
Advance national estimates have vastly understated
the per recipient cost of Title 19 aid, by nearly
three-fourths in the case of fiscal 1967 (with six-
month lead time) and by as much as two-thirds or more
in the January 1967 estimate for fiscal 1968, although
the latter is subject to further change. . . . 19
For individual states, comparable data are still ;
not available. However, the Tax Foundation found on its |
own that eight states out of twenty-six states which had I
been operating a year or longer at the time of the survey
did not report any significant discrepancy in the number j
of eligibles projected a year before. |
i
However, the rest of the states except for two j
(Iowa and Nevada) experienced under-estimation of the num- j
bers eligible with Hawaii ranging from an estimate of 4.7
per cent to an actual 7 per cent and Rhode Island from
20
8 per cent to 21 per cent.
Spotty as the data on estimated eligibles and other
18Ibid.
19Ibid.
20Ibid., p. 31.
165
comparable data are, there are a set of cost figures which |
shows the expenditures of the states for public medical
assistance before Medicaid and the size of expenditures
incurred after Medicaid.
TABLE 40
COMPARISON OF RECENT RATES OF PER CAPITA MEDICAL
VENDOR PAYMENTS WITH PRE-TITLE 19 EXPERIENCE
BY SCOPE OF COVERAGE
Scope of Coverage
Amount of Vendor
per capita (a)
1965 (b) 1967 (c)
Percent change
1965 to 1967
All States $7.63 $14.76 +93.4
States with
Medicaid (d) 9.10 18.65 +104.9
States without
Medicaid 3.80 4.69 +23.4
a. Coefficient of dispersion was 51.4 per cent in 1965
and 55.1 per cent in 1967.
b. Calendar year.
c. Annual rate of payments in third quarter of 1967,
divided by population of the respective states with
or without Medicaid.
d. Thirty-five states were operating Medicaid Programs
as of September 1967.
Source: Medicaid After Two Years (New York: Tax Founda
tion, Inc., 1968), p. 33.
As it is expected, the data show that the states
with Medicaid program incurred upon themselves about three
times more expenditures than those without. That burden
is, in effect, one-hundred per cent increase from the
amount of money thirty-five participating states spent for
166
public medical assistance in 196j, the year when lledicaid
was not effected.
However, the impact of those one-hundred per cent
increase in expenditure on individual thirty-five partici
pating states is not uniform. There are state-to-state
variations in the change of expenditures for public assis
tance medical payments from pre-Medicaid year to post-
Medicaid years.
As the data indicate, the Medicaid programs in Cali
fornia and New York, i.e., the populous states, have a
substantial influence on the national totals, reflecting
both the size of their total population as well as the
liberality of their programs in terms of coverage and bene
fit levels (see Table 41).
Together they account for nearly one-half or 48.5
per cent of payments to medical vendors in the fall of
1967, and for some thirty-seven per cent of all recipients
of medical public assistance medical care. Iowa, Nevada,
West Virginia, Wyoming, and New Hampshire actually spent
less in 1967, when Medicaid was in force, than 1965, when
Medicaid was not effected (see Chart 1).
Since the above five states are Medicaid-states
offering only minimum service and coverage in terms of
categorically needy, they temporarily were able to cut some
expenditures in this area. This is perhaps one of the
side effects of Medicaid, that is, permitting a state to
TABLE 41
MEDICAL VENDOR PAYMENTS BY STAGE OF TITLE 19 IMPLEMENTATION AND STATE 1965 AND 1967a I
Percent change
Amount of vendor payments 1965 to 1967
STAGE OF IMPLEMENTATION Total (millions) Per capita (b)
AND STATE I960 1967 1965 1967 Total Per capita
ALL STATES $1,472.7 $2,908.9 $ 7.63 $14.76 97.5 93.4
STATES WITH TITLE
19 PROGRAMS 1,268.5 2,651.4 9.10 18.65 109.0 104.9
Idaho 5.3 5.4 7.57 7.73 2.8 2.1
Louisiana 31.3 37.7 8.81 10.30 20.4 16.9
Maine 8.2 9.6 8.30 9.89 17.7 19.2
Montana 6.2 6.9 8.79 9.82 11.6 11.7
Nebraska 12.S 18.8 8.76 13.07 46.9 49.2
Nevada 3.5 2.6 8.07 5 .95 -24.3 -26.3
New Mexico 5 .4 10.3 5.29 10.28 92.2 94.3
Ohio 50.0 67.4 4.87 6.45 34.9 32.4
Oregon 13.8 15.3 7.10 7.67 11.5 8.0
South Dakota 4.3 6.8 6.28 10.07 57.1 60.5
Texas 38.6 73.5 3.66 6.77 90.5
85.0
Vermont 3.4 5.9 8.47 14.16 73.0 67.2
West Virginia 12.4 8.4 6.80 4.68 -31.9 -31.2
Wyoming 1.5 1.2 4.67 3.70 -24.7 -20.8
Illinois 91.4 128.9 8.59 11.84 41.5 37.8
Kentucky 20.6 27.2 6.49 8 .54 32.1 31.6
New Hampshire 3.4 2.0 5.06 2.92 -41.2 -42.3
North Dakota 7.7 9.9 11.89 15.56 28.4 30.9
Rhode Island 10.3 19.4 11.57 21.56 88.3 86.3
TABLE 41--Continued
I
I Amount of vendor payments
! STAGE OF IMPLEMENTATION Total (millions) Per cap
! AND STATE 1965 1967 1965
Connecticut 30.4 41.6 10.71
Delaware .5 2.5 1.02
Iowa 25.5 20.6 9.24
Kansas 17.8 26.6 7.91
Massachusetts 98.9 157.5 18.45
Minnesota 59.1 79.8 16.62
Oklahoma 27.7 56.0 11.26
California 196.8 563.8 10.69
Hawaii 4.5 6.7 6.28
Maryland 12.1 44.6 3.43
Michigan 66.6 147.4 8.00
New York 223.5 776.7 12.34
Pennsylvania 77.0 125.4 6.65
Utah 6.6 9.3 6.71
Washington 41.0 47.4 13.76
Wisconsin 50.7 88.0 12.22
STATES WITHOUT
19 PROGRAMS,
TITLE
TOTAL 204.1 257.4 3.80
Alabama 17.3 14.9 4.95
Alaska 1.4 1.2 5.08
Arizona .9 1.7 .56
Arkansas 15.9 20.3 8.19
Colorado 23.7 24.1 12.17
Percent change
1965 to 1967
(b)
1967 Total Per capita
14.21 36.8 32.7
4.84 390.7 374.5
7.47 -19.4 -19.2
11.68 49.5 47.7
29.06 59.2 57.5
22.28 35.0 34.1
22.46 102.6 99.5
29.44 186.6 175.4
9.04 49.8 43.9
12.11 268.3 253.1
17.17 121.4 114.6
42.36 247.6 243.3
10.78 62.8 62.1
9.12 40.5 35.9
15.36 15.8 11.6
21.02 73.7 72.0
4.69 26.1 23.4
4.22 -13.5 -14.7
4.24 -15.0 -16.5
1.01 86.5 80.4
10.32 27.6 26.0
12.18 1.6 .0
168
TABLE 41--Continued
Percent change
Amount of vendor payments 1965 to 196/
STAGE OF IMPLEMENTATION Total (millions) Per capita (b)
AND STATE 1965 1967 1965 1967 Total Per capita
Florida 19.9 23.2 3.44 3.87 16.3 12.5
Georgia 14.4 23.5 3.29 5.21 62.5 58.4
Indiana 18.2 27.1 3.72 5.42 49.1 45.7
Mississippi 1.9 3.1 .80 1.33 67.2 66.3
Missouri 15.2 15.0 3.37 3.25 - 1.2 - 3.6
New Jersey 30.0 52.7 4.42 7.52 75.4 70.1
North Carolina 15.9 18.7 3.22 3.71 17.4 15.2
South Carolina 9.0 8.1 3.50 3.12 - 9.3 -10.9
Tennessee 12.1 15.4 3.16 3.96 26.9 25.3
Virginia 8.4 8.6 1.89 1.90 2.9 .5
a. Annual rate of payments in third quarter of 1967.
b. Total payments divided by population of each state.
Source: Medicaid: State Programs After Two Years (New York: Tax Foundation, Inc.,
1968), p. 70.
CHART 1
170
Per Capita Public Assistance Medical Payments
1965 and 1967
State and rank
in 1967
1.
2 .
3.
4.
5.
6 .
7.
8 .
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
2 2.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
$10 $15 $20 $25 $30 $35 $40 $45
N. Y.
Cal.
Mass.
Okla.
Minn.
R. I.
Wis.
Mich.
N. D.
Wash.
Conn.
Vt.
Neb
Md.
111.
Kan.
Penn.
La.
N. M.
S. D.
Maine
Mont.
Utah
Hawaii
Ky.
Idaho
Ore.
Iowa
Texas
Ohio
Nevada
Ga.
Del.
W. V.
Wyo.
Mo.
S. C.
N. H.
< I c Z 7 2 Z Z .
' 7 7 7 7 7 7 7* Z3ZZ«ZrZn
' S t '/ / / / >
7 7 7 7 7 7 7 7-
7 7 7 7 7 7 7 7 7
7 V s 7 7 7 7 s 0 9 .
/7SS77777-
7ZZvZZsX
r Z J C T 7 7 ?
mr~
WBKi
HS^R/22
m.
S
H8P
W«P
irrmr
mrr
WTO*
w r
5771
m am mm
196 7
196
Source: Tax Foundation, Medicaid: State Programs After Two
Years (New York: Tax Foundation, Inc., 1968), p. 34.
171
participate in the program offering the benefits only to
those in welfare assistance categories.
By so doing, they were able to achieve two things:
jointly they cut the fund spent for previous medical assis
tance programs under the various standing legislations
(e.g.), Kerr-Mills, Aid to Families of Dependent Children,
other state welfare measures by limiting the new Medicaid
to those in categorically needy which resulted in smaller
number of recipients.
Secondly, they obtained Federal matching grant by
having adopted the Medicaid program which in turn helped
them reduce the size of expenditures.
This phenomenon is, however, temporary because any
state adopting Medicaid is bound by law to extend the
coverage as well as the benefits to all those medically
needy by 1975. Therefore, these states are bound to show
increasing financial burden in the future in order to
justify continuing participation in the Medicaid program.
New York. Costs of the New York Medicaid program,
which began April 1, 1966, have repeatedly exceeded
advance budget estimates. Original estimates of $350 mil
lion in fiscal 1967 reached $461 million at the end of
: 1967 exceeding the estimates by the amount of $110 million.
In 1968, estimation went up to $658 million and later
172
21
revised it upward to $687 million.
These results in New York occurred despite the fact
that the number certified for Medicaid services was less in
1968 than half the official estimate of numbers who would
be eligible if they applied. Such an alarming increase of
cost moved the New York state legislature to make material
22
changes in the New York Medicaid Act. The new law is to
result in a reduction of about $300 million a year in pro
gram costs and the cutting off of about one million persons
from the 3.5 million already certified for Medicaid ser-
23
vices. i
I
i
Michigan. The Michigan Medicaid program, initiated |
!
in October, 1966, very quickly outran its original approp-
2 4 -
riation. It was outrun by the amount of $70 million. J
For 1968 costs, the state still underestimated by the j
2 5 J
amount of $40 million. j
Except those five states (Iowa, Nevada, West Vir
ginia, Wyoming, and New Hampshire) which managed to spend
less for their Medicaid programs, what is the impact of
excessive costs on the tax system of the rest of the other
^ Ibid., p. 35.
^ Ibid., pp. 37-38.
23-rv • i
Ibid.
^ Ibid., p. 37.
25TV. ,
Ibid.
173
states? Considerable impact on tax sources is reported.
California is anticipated to devote a portion of
1-cent increase in the state general sales tax to the sup
port of the MediCal program. Hawaii is reported to have
increased state taxes to support much larger budget for
2 6
increased medical care. The state take-over of welfare
costs in Massachusetts is reported to be a major factor in
27
the $94 million tax increase program.
In Pennsylvania, $264 million tax increase is attri
buted to medical assistance costs which resulted in a
28
general fund deficit in 1967-68 fiscal years.
New York had to increase state and local sales and
property taxes from thirty-six to fifty per cent to meet, j
. 29;
in part, the need of the state s medical assistance costs.
Idaho, Maine, Louisiana, North Dakota, Ohio, and Oregon
iiiitially intended to expand the program but subsequent
funds were not made available largely because of reluctance
to raise taxes.^ Thus, these states operate Medicaid
programs which provide only minimum coverage and scope of
services.
26tk. 1
Ibid.
27TV. ,
Ibid.
28 r,. ,
Ibid.
^ Ibid., p. 38.
30-r-, . ,
Ibid.
174
i
Six of thirteen states not operating Medicaid as of i
January, 1968 (Alabama, Arkansas, Colorado, Florida, New
Jersey, Tennessee) indicated that new revenue sources
31 i
would be required if the program were adopted.
Non-Fiscal Impacts
Apart from fiscal impact, lion-fiscal impacts of the
Medicaid program are considerable. These non-fiscal
effects are some operating problems and difficulties
reportedly encountered by the participating states.
The problems were widely reported in the newspaper
columns and other news media shortly after the program
j
became effective. For instance, a Wall Street Journal !
editorial of August 8, 1967 is read like this:
How do you begin the broadest and perhaps one of
the most expensive government medical programs in the i
nation's history?-^
Having raised such a question, the editorial offers !
a cynical observation which was quite accurate as a reflec
tion of reality at the time: j
i
Start with a total lack of preparedness, move
quickly into chaos, add a little apathy and then stir
up some fights, with the doctors, dentists, and pharma- j
cists. Much of the trouble with Medicaid, say both its !
administrators and critics, stems from the fact that
almost from its inception it has been a spur-of-the
Ibid.
~^Wall Street Journal, August 8, 1967, p. 28.
175
moment program with few of those involved realizing its j
vastness.33
For the purpose of the format, the operating prob
lems experienced by Medicaid are described and analyzed in
terms of two groups of organizational units involved in
the administration of the program: (a) providers of ser
vices in delivering the service to the recipients and (b)
Federal-State intergovernmental relations.
Providers of Services
It was widely reported by news media that the
doctors and laboratories were intensely competing for par- j
i
ticipation in the program seeing in the latter an oppor-
i
tunity to "profiteer." Data obtained by the Tax Foundation j
and Commission on Intergovernmental Relations show the
fact to be otherwise. j
i
As of June, 1968, more than half of the states j
1
surveyed are reported to have experienced that "for one
reason or another, some physicians and other suppliers of
medical services had been unwilling to serve Medicaid
Ibid.
3 ^
These two sources are the best known in the field
of Medicaid. Tax Foundation is non-profit organization
interested in disseminating objective information to the
public. Advisory Commission is a public organization to
be concerned with an analysis of whatever problems whose
nature may be intergovernmental.
patients."^0 The reasons for this unwillingness are pri
marily three: fees held to be inadequate, too slow payment,
and excessive "red" tape (see Table 42).
TABLE 42
SOKE CRITICISM OF MEDICAID PROGRAM OPERATIONS
AS REPORTED FROM 26 STATES
Nature of criticism
Number of states
reporting criticism
Unwillingness of physicians or
other suppliers of services to
participate. Reasons reported:
a. Fees are held inadequate 14
b. Payment is too slow 11
c. Excessive "red" tape 7
d. Other 10
Source: Medicaid After Two Years (New York: Tax Founda
tion, Inc., 1968), p. 42.
Reimbursement plan for Medicaid is by law a system
which pays on a fixed-fee schedule. This fixed-fee
schedule is determined by the concept of comparability in
costs. That is, comparability in terms of "reasonable and
customary fees" prevailing in a given area.
Thus, each state commissioned a fiscal carrier,
usually a private insurance agency, to conduct a survey of
the fees charged under a variety of conditions and come up
with an average profile of costs which subsequently served
35
Tax Foundation Report, op. cit., p. 45.
j_
177
as the standard of "reasonable and customary" fees. This
brought about a great deal of rigidity and inflexibility in
adjusting the fees to each service provided.
Fixed-schedule also led to complaint that it in
effect heaps a bonus on rural doctors who customarily
charge lower rates than urban doctors who are in turn
36
short-changed in the process. Consequently, the ones
motivated to participate are the doctors who are marginal
in reputation and personal income, thus, providing only
"second class" medical care.
Seven of twenty-six states attribute the unwilling-
I
ness of the physicians and suppliers of services to parti- j
i
cipate to excessive red tape. There are two-hundred pages I
37
of rules and regulations governing the billing process;
sorting out the beneficiaries in accordance with five |
different categories requires twenty different forms to be
filled out.~*®
Eleven states attribute the unwillingness to parti
cipate to "slow payment." Because of excessive "red tape"
involved in processing the billing, the time lag between
the date of billing and payment is considerable; six months
36t> • i
Ibid.
37
Interview with District Director of Social Security
Administration, Los Angeles, September, 1968.
3 8 t i • j
Ibid.
178
delay is rather common. According to the hew York Times \
report, six months after Medicaid became effective, about
half of participating 5,500 of the total 15,000 doctors in
39
the city had quit because of their payments being delayed.
The other operating problems are in shortage of per
sonnel, facilities, and lack of knowledge of Medicaid among
the potential recipients. Again the data provided by the
Tax Foundation gives some empirical basis (see Table 43).
As the data show, one-third of the twenty-six states
indicate shortage of medical personnel. Acute shortage is
particularly felt in nurses more than any other personnel, j
!
Facilities shortages are reported to be acute in such long- ,
term care facilities as nursing homes and home care pro- j
i
[
grams, facilities for ambulatory and emergency patients. '
Seventeen of twenty-six states or half of the total
I
feel that some medically eligible members of the population I
have not made application for aid. The data show that the
"unawareness of potential eligibles and the wish to avoid j
|
the 'welfare' stigma associated with receiving aid" are
just about equal in their influence.
Other complaints are reported to be like the follow-
j
ing: "eligibility standards are too liberal"; "administra
tive mix-ups and repeated delays"; "general lack of
39
The New York Times, April 12, 1967, p. 51.
179
TABLE 43
EXPERIENCES OF MEDICAID STATES3
AS OF SUMMER, 1968
Nature of Experience
Number of States
Reporting
Shortage of personnel--doctors,
nurses, auxiliary medical helpers 9
Shortages of facilities and
equipment, such as hospital beds,
clinics, nursing homes, etc. 8
Failure of "medically indigent"
to enroll and establish eligi
bility. Reasons reported:
a. They are unaware of the fact
that they are eligible. 9
b. They wish to avoid the
"welfare" stigma which they
feel is associated with
receiving aid. 8
c. Other complaints. 8
aN=26 states. Source: Medicaid After Two Years (New York: ;
Tax Foundation, Inc., 1968), p. 42. Title of the table is
worded by the writer to make it fit to the purpose of this 1
study. Content of the data is, however, as they are pre
sented in the original study of Tax Foundation. |
understanding of the program by the public and by providers
of services.These impacts in form of problems and i
i
i
complaints are, in effect, the feedback generated by the j
implementation of the program.
These complaints and problems are those the states
experienced in their relations to the service providers and
40
Tax Foundation Report, op. cit., p. 42.
180 ,
clientele. They are the operating problems in properly |
linking the benefits with the ultimate target group. Doc
tors and other providers are an important variable in pro
cessing the linkage with the recipients.
Related to these linkage problems is an additional
dimension involved, that is, inter-governmental relations
to completely understand the process from the control
center up to delivery points.
Impact of Medicaid Upon Intergovernmental Relations
Since Medicaid is the creature of Federal legisla-
i
ture involving the state and local governments as partners,
the program raises a number of problems which are basically j
intergovernmental. ,
In September, 1968, the Advisory Commission on |
41 !
Intergovernmental Relations circulated a questionnaire
in which the Commission asked the state governors and !
i
legislative leaders to list the intergovernmental problems i
j
they experienced in administering the Medicaid. The
results are presented in Table 44.
The data indicate that a major crit icism by gover
nors and legislative .leaders of twenty-six states is on the
unclear relationship between Medicare and Medicaid. When
Congress voted on Amendments to Social Security Act in
41
Advisory Commission Report, op. cit., p. 51.
181
TABLE 44
MEDICAID PROBLEMS OTHER THAN FISCAL POLICY AS
REPORTED BY STATE GOVERNORS AND LEGISLATIVE
LEADERS, SUMMER, 1968
Nature of Problems
Number of
by
Governors
Times Mentioned
by
Legislative
Leaders
Coordination of policy and
administration between Medicaid
and Medicare 14 6
Guidelines Unclear 3 1
Tardiness of Guidelines 3 -
Insufficient decentralization
of authority and communication
to regional offices 4 4
Inadequate advance consultation
with States in developing policy
changes 2 2
Guidelines not specific enough 1 -
N=26. Source: Advisory Commission on Intergovernmental \
Relations, Intergovernmental Problems in Medicaid (Washing- j
ton, D. C.: Government Printing Office, 1968), p. 51.
Format of the original date was altered to make the table
fit to the purpose of this study.
1965, it created two separate measures of medical assis
tance program, viz., Title 18 and Title 19.
The former is popularly called "Medicare" which is
the federally administered program of health insurance for
people aged sixty-five and over. It offers two kinds of
benefits: hospital insurance--hospitalization and related
care often referred to as Part A and supplementary medical
182 ,
^ 2 j
insurance--physicians' services referred to as Pare B. i
The hospital insurance or Part A is financed by
deductions from employees* wages and matching taxes paid
43 . i
by employers and the self-employed. Medical insurance
or Part B is voluntary program financed by monthly premium
paid by the individual and by a matching fund from the
Federal GovernmentThe entire program is administered
by the Social Security Administration.
The latter, Title 19 or "Medicaid" comes into a
relationship with Medicare when it deals with those over
j
sixty-five in medically needy category. That is, those
aged people whose wage history is not sufficient to finance
hospital insurance or Part A of Medicare.
i
Medicaid is by law obligated to pay all or part of
the deductible and coinsurance amounts for these medically
needy aged. Furthermore, states are encouraged to "buy-in"
to the voluntary medical insurance or Part B of Medicare
42xu. ,
Ibid.
/ 3
U.S. Congress, Senate, Subcommittee on Health of
the Elderly, Special Committee on Aging, Medical Assistance
for the Aged: The Kerr-Mills Program 1960-63, 88th Cong.,
1st sess., October, 1963, p. 3.
^For more detailed explanation, see Herman M. and
;Anne R. Somers, Medicare and the Hospitals: Issues and
Prospects (Washington, D. C.: The Brookings Institution,
1967;; Margaret Greenfield, Health Insurance for the Aged:
The 1965 Program for Medicare (Berkeley:Institute of
Governmental Studies, University of California, 1966).
183
j
for those sixty-five and over under the Medicaid plan. I
I
State health departments administer Medicaid program.
Thus, coordination between Medicaid and Medicare
necessitates a flow of information and intimate administra
tive link between the two different agencies administering
the programs. Apparently the amount of paperwork involved
in processing this coordination is enormous and burdensome.
Consequently, coordination is very poor.
Jn the questionnaire circulated by the Advisory
Commission on Intergovernmental Relations, Illinois
responded as follows: I
|
Our relationship with the Medicare carriers has i
created a vast administrative problem--handling of i
paper--because of paying deductions and coinsurance
under Medicaid. It would be wise to have a study '
made of this whole area as it may be less costly to
have Medicare pay for these deductions and coinsurance |
than the State having the tremendous administrative ;
job of paying for them under Title 19. . . i
The solution suggested is to transfer the coinsurancej
j
part to the Federal government entirely or consolidate into [
i
one source.
A spokesman from Ohio said:
There are still many problems in relating Medicaid
to the provisions of Medicare. The amounts of money
involved, especially in connection with hospitaliza
tion, probably do not justify the administrative time
required. 4-6
^Advisory Commission Report, op. cit., p. 52.
46Ibid.
Wyoming expressed the same opinion:
Medicare covers part of the medical care for eligi
bles over 65 and Medicaid covers the balance. Payment
from one source would greatly reduce administrative
cost,4'
Such a cumbersome administrative arrangement appar
ently results in very poor communication between the
agencies administering Medicare and Medicaid. Hawaii's
spokesman described the situation:
One of the problems which we have experienced and
which apparently exists in all states which have
implemented Medicaid is a need for more coordination
and a freer interchange of information between those
administering Medicaid and Medicare. In Hawaii, the
agency administering these two programs are willing to
provide such an interchange of information but are
hampered by restrictions imposed by the Social Security
Administration on the release of information under
Title XVIII.48
This spokesman for Hawaii, thus, lamented:
It is difficult to understand why two federally
funded programs as closely related as these two titles
of the Social Security Act and under the same federal
agency should not be able to provide each other with
full and prompt information about experience, rates,
utilization and future planning.4 9
Louisiana had about the same feeling as Hawaii:
We believe also that since Medicare is so closely
allied to Medicaid, it is essential that the State
Public Welfare Department be kept advised of any policy
^ Ibid., p. 53.
^ Ibid., p. 52.
Ibid.
[ ____
185
changes anticipated. It would improve the administra
tion of Medicaid in relation to services to the persons
65 and over if the Public Welfare Department could
obtain accurate information on what recipients are
maintaining payments in Medicare.
Nevada echoed Louisiana:
It is very difficult to understand why the Medicare
program is so reluctant to completely cooperate with
the Medicaid program. It would appear that the same
tax dollar should allow for the mutual sharing of
information without additional costs.
Medicare is essentially an extension of Medicaid
which covers the citizens under the age 65 and who are
I
categorically and medically needy. However, blending
I
Medicaid withi Medicare in relation to those over 65 in j
needy in paying for health coinsurance makes the former j
potentially more radical and comprehensive program. i
Besides this blending from fiscal standpoint, both j
programs provide just about the same services; hospitaliza- j
I
tion, extended care facilities, nursing homes, and physi
cians' services. Thus, administrative integration makes
much sense as the following statement of a spokesman for
the state of Washington attested:
It is difficult, if not impossible, to coordinate
the health care programs as provided under present
Medicaid and Medicare, resulting in problem. . . .
Programs need to be coordinated philosophically and j
~^Ibid., p. 53.
186
fiscally in order to provide greater benefit to the
patients.52
As it stands now, however, the rules and regulations
governing their relationship are so complicated and mixed
up that an observer of California's version of Medicaid,
viz., "Medi-Cal" commented as follows:
One of the objectives of Medi-Cal was to consolidate
the various medical care programs offered at the state
and local level. . . . However, in accomplishing this,
the program became mesmerized by federal and state law
and county rules and regulations. This has resulted in
a situation in which less than a handful of individuals
(out of a state of approximately 20 million) really
understand the program.
Summary and Analysis
The data presented in this chapter permits us to
make the following observations with respect to the impact
the two and one-half years of implementation of Medicaid
program brought to the health care system in the United
States:
1. Medicaid definitely shifted the burden of medical
care cost from private to public sources; the shift
has occurred through a fifteen to twenty per cent of
increase in public medical assistance compared with
a comparable decrease in private medical expendi
tures .
52T, . ,
Ibid.
“^Dave L. Cox, "Medi-Cal or Medi-Quandry," Tax
Digest, LXV (1967), 33.
i
187
2. The effect of the increase of public expenditures
for medical assistance was largely, however, infla
tionary; it helped medical care prices to rise 7.3
per cent while over-all consumer price index rose
only 2.7 per cent.
3. This inflationary effect put a great strain on
revenue sources of state and Federal governments;
the actual expenditures for Medicaid outran the
estimated cost by forty-four per cent or $1,173
million, despite the participation of fewer states
and nineteen per cent fewer recipients than esti
mated in January, 1966.
4. Notwithstanding the nation's investment in health
program at such a rate, return on the investment was
disappointing because of (a) total lack of prior
planning, (b) poor arrangement for payments of
physicians' fees, (c) delay in payment, (d) exces
sive "red tape" and (e) lack of public knowledge of
the availability of medical assistance.
5. kxcessive "red tape" in the form of sundry rules and
regulations at state and Federal level left the
relationship between Medicaid and Medicare in utter
confusion; solutions tended to be suggested in terms
of consolidating medical and health programs fis
cally and philosophically into a single package.
188
These observations are, however, a mere historical
description of the effects unleashed by two and one-half
years of implementation of Medicaid programs.
These effects are meaningless, however, unless they
are properly fed into the system, so that the latter may
make corrections and adjustment. There are some data
indicating that the system did, indeed, respond to the
feedback in one way or the other, although the specific
content of the responses is not available to date.
Congress took action in December, 1967, to remove
some of the ill effects of Medicaid for the future opera
tion. This Congressional action resulted in policy revi
sions .
Among the changes relevant to the focus of this
study were provisions (a) limiting Federal matching support
in state Medicaid programs to persons below certain income
levels, and (b) tightening the procedures and controls of
Medicaid services and costs in the states, and (c) imposing
a "freeze" on future increases in the families with depen
dent children group.
Limiting Federal Matching Support
The 1967 amendments provided that Federal sharing
will not be available for families whose income
exceeds by more than one-third the highest amount a
state ordinarily pays to a family of the same size in
the form of money payments under the aid for families
with dependent children. For state plans already
approved, the limit of Federal sharing would be 150
189
Percent effective July 1, 1968; 140 percent on Janu
ary 1, 1969, and 133 1/3 percent on January 1, 1970.
This provision helps to cut some number of eligible
recipients in the states like New York and California which
set the eligibility at very liberal level which was lower
than one-third the amount paid out to the dependent fami
lies .
Tightening State Procedures and Controls
The 1967 amendments required the states to establish
methods and procedures to safeguard against unnecessary
utilization of health care and services, and to assure
that payments not exceed reasonable charges and that
they be made on a basis consistent with efficiency,
economy, and quality of c a r e . 55 |
i
The law also sets conditions for participation in I
Medicaid programs on licensed nursing homes, requiring
a professional medical audit for periodic medical ;
evaluations of the appropriateness of care given to j
Medicaid patients in nursing homes, mental hospitals,
and other institutions.56
I
This second provision is a direct response to the j
feedback that (1) patients tend to over-use free medical
services, (2) some "profiteering" was prevalent among the
rural physicians, sub-standard nursing homes, and (3)
general cost-push.
"^Tax Foundation Report, op. cit., p. 49.
55T, . ,
Ibid.
Ibid.
190 ■
"Freeze" Imposed on Families with Dependent Children j
The amendments set a limitation on Federal financial
participation in the AFDC program related to the pro
portion of the child population under age 18 because
of the absence from the home of a parent. Federal aid
would not be available for any excess above the percen
tage of children of absent parents who received aid, to
the child population under age 18 in a state as of
January 1, 1968.5?
Since Medicaid includes AFDC families as one of the
categorically needy groups, setting a limitation to Federal
aid to these groups, in terms of flat percentage to the
child population under age 18 is a way of curbing the num
ber of eligibles for Medicaid categorically needy.
This limitation is an indirect way of cutting the
number of potential recipients. This measure was necessary
because AFDC category was the largest and most rapidly
growing of all public assistance groups, constituting forty-
three per cent of all medical assistance recipients in
1967.58
Thus, the above three revisions in policy are economy
measures intended to control rapidly rising costs in medical
assistance. Response is rather unimaginative and stereo
typical, viz., a simple retrenchment in face of rising
expenditures.
~^Ibid., p. 50.
58t- , . ,
Ibid.
There are, however, a few fresh approaches the state
governments took to remove some operating problems, especi
ally in expediting payments to the providers of services
(see Table 45).
TABLE 45
STATE REPORTS ON MEASURES TO DEAL WITH CERTAIN
OPERATING PROBLEMS IN MEDICAID
Measures Taken No. States
Use of fiscal intermediary to
handle billing and collections (6)
Calif., La., Mich.,
N.M., Penn., Vt.
Other arrangements to expedite
payments (7)
Idaho, 111., Md.,
Mass., Ohio, Ore.,
R. I.
Measures to increase supply
of personnel and facilities (7)
Hawaii, 111., La.,
Mass., N.M.., Ore.,
Penn.
Steps to ease red tape associ
ated with enrollment in the
program (8)
Calif., Hawaii, 111.
Mass., Mich., Ohio,
Ore., Penn.
Measures to encourage enroll
ment and remove welfare stigma (5)
Conn., Hawaii, Mass.
Mich., N.Y.
N=26. Source: Medicaid After Two Years (New York: Tax
Foundation, Inc~ 1968), p. 47. Format of the table was
slightly altered to fit it to the purpose of this study.
Content is, however, intact.
Fiscal intermediary is a new concept apparently
borrowed from the pattern of defense industry.It is an
59
Contract administration is becoming an increas
ingly accepted mode of operation of many social programs
192 ,
administrative set-up based on a contractual arrangement \
between the public agencies and private insurance corpora
tions. The latter assumes the responsibility for making
payments to the providers of service drawing on the funds
deposited in their trust by the Federal and State agencies.
The rationale is to take advantage of high quality
skills of the private industry in carrying out insurance-
related administrative tasks, instead of sagging already
overloaded public bureaucracies with additional tasks.
Effectiveness of this third party arrangement is yet to be
assessed, however. i
i
Other arrangements in expediting payments is the
provision that the doctors and providers of other services
I
may bill directly the patients rather than welfare or health
agencies, so that the patients may later be reimbursed by j
the latter.
Seven states took steps to increase supply of per-
j
sonnel and facilities (see Table 45). Eight states report
that they have lessened the "red tape" involved in enroll
ment in the program, "through such measures as simplified
application forms.Five states made some attempts to
these days; see, for example, Stefan Dupre and Eric Gustaf
son, "Contracting for Defense: Private Firms and the Public
Interest," Alan A. Altshuler (ed.), The Politics of the
Federal Bureaucracy (New York: Dodd, Mead, 1968), pp. 251-
262.
^Tax Foundation Report, op. cit., p. 47.
193
disseminate the information pertaining to the eligibility j
and availability of medical assistance among the medically
indigent.
For instance, the New York State Department of
Social Services has reportedly distributed more than eight
million pieces of Medicaid literature in both English and
Spanish.^ The Department also issued numerous press
releases and TV-radio spot announcements; they also waged
a subway poster campaign and a "mailing which reached
6 2
every rural resident in the state."
In response to the intergovernmental squabble that !
took place around the problem of coordination between !
Medicaid and Medicare, Department of Health, Education, and I
!
Welfare inaugurated a use of common procedure for billing j
under State public assistance, Medicaid and Medicare pro- j
63
grams.
According to HEW officials, "this procedure was
designed to eliminate dual billing by physicians, reduce
delays in the disposition of claims by the fiscal inter
mediary and in the payment of physicians, and clarify
deductibles, coinsurance, and determination of claims for
Ibid., p. 48.
62tk. ,
Ibid.
' ^ A d v i s o r y Commission Report, op. cit., p. 53.
194
those age sixty-five or o l d e r . i
HEW officials further contend in the Commission
report that the states which adopted this common billing
procedures have either reduced or eliminated the delays in
payments and confusion surrounding the coordination problem
between Medicare and Medicaid.However, on the contrary
to HEW claim, the Advisory Commission on Intergovernmental
Relations report that only six states adopted this HEW
prescribed common billing procedures as of July, 1968.^
Less than enthusiastic support for this new common
procedure among thirty-eight states operating Medicaid I
j
programs is attributed to (1) lack of "push” by HEW regional
offices, (2) States' objection to the processing charge i
involved, and (3) the belief that existing arrangements |
with fiscal intermediaries were better than the proposed
67
system.
j
Thus, response to the intergovernmental problem j
involved in coordination between Medicare and Medicaid was
less than satisfactory. It is unfortunate that HEW seems
to take the problem as something procedural in billing;
the problem is deeper. The problem demands re-examination
k^Ibid.
65Ibid.
66Ibid.
^Ibid.
195
of distribution of authority and responsibility between the
state and Federal agencies.
The Federal government has to come up with a philo
sophical as well as practical arrangement in regulating
Federal-State relationship to properly deliver the service
to the rank and file. The problem is typical of any other
national program involving intergovernmental relations,
viz., how to balance concentration of authority at the
Federal level with preservation of sufficient autonomy of
the State and local governments within the Federal system
of governments.
CHAPTER VII
DISCUSSION AND CONCLUSIONS
Explaining the policy outcomes in American state
politics was the central focus of this study. It was sug
gested that policy outcomes are the function of the level
of economic development and political system characteris
tics.
Explaining the policy outcomes consisted, therefore,
in describing the independent effect of these economic and
political phenomena on particular policy outcomes. How
ever, it was contended that explaining how the particular j
policy outcomes were processed is not enough; knowing the j
i
impact of the policy outcomes was argued to be as much
important and critical as "explaining" the process.
Thus, this study attempted to examine the policy
outcomes from both perspectives; from the standpoint of
"explaining" or describing the "process" as well as explor
ing the "impact" brought about by the specific content of
the policy outcomes on the system.
Before sorting out the independent effect of eco
nomic and political system characteristics upon policy
outcomes, this study considered the investigation of the
relationship between the level of economic development and
political system characteristics as a prerequisite to the
i 196
197
analysis of policy outcomes.
Because it was felt that a knowledge of the linkage
between economic development and political characteristics
may later help the study tell more meaningfully about the
effect of each on the policy outcomes than otherwise.
Thus, this study investigated a set of three related
problems:
(1) Relationship between economic development and
political system characteristics in the states;
(2) Relationship between the policy outcomes and the
level of economic development on one hand, and I
I
political system characteristics, on the other;
i
(3) Investigation of the impacts of the policy outcomes.
j
Lconomic development consisted of four variables, j
viz., industrialization, education, per capita income, and |
urbanization. Political system characteristics referred to
three variables: inter-party competitiveness, voter parti
cipation, and apportionment. j
The policy outcomes serving empirical referents in
this study consisted in the decisions made by thirty-eight
states operating Medicaid programs as of 1968, with respect
to (a) scope of recipients covered, (b) extent of services
provided and (c) size of expenditures appropriated for the
program.
Impacts of Medicaid policy were analyzed in terms of
|fiscal and non-fiscal effects of the implementation of the
198
policy for the last two and one-half years. The revisions !
and modifications in the subsequent policies of Medicaid
was analyzed to describe how the system responded to the
feedback generated by the effects of implementation.
Simple correlation coefficient and percentage analy
sis of three-by-two or three-by-three tables were the main
statistical techniques the study utilized. However, it
had some limitations; data were obtained from secondary i
sources. Ex-post facto analysis precluded the possibility
i
of setting up rigorous experimental and control groups.
It was pointed out that these limitations should be
taken into consideration when one attempts to generalize on
the basis of findings presented in this study. Neverthe
less, the findings are authentic and reliable enough to
generate following discussions of the whole study.
Discussion
Relationship Between Economic Development and
Political System Characteristics
Of four variables included in the concept of "eco
nomic development" education and per capita income were
found to be significantly related to inter-party competi
tiveness (.31 and .25 respectively).
Industrialization and urbanization were not signifi
cantly related (.05 and .11 respectively). This finding
is rather consistent with findings of Dawson-Robinson and
199 i
Dye. Richard Dawson reported"^ in his study that inter
party competition was closely related to per capita income
and education but insignificantly related to industrializa- 1
!
tion. :
2
Thomas Dye found that party competition was highly I
related to income and education (.52 and .57 respectively),
while industrialization and urbanization amounted to only
.26 and .03 respectively relative to party competitiveness.
On the basis of this finding, Dye concluded that !
. . . socioeconomic development does not consistently |
favor one party or the other. Rather, high levels of !
income and education tend to foster party competition.^
Dye observes further:
Both one-party Republican and one-party Democratic
states tend to be states with populations with lower
income and educational attributes.^
It was also found in this study that thirteen states
out of fifteen high per capita income states were found to
be two-party states (see Table 3) . Of eleven low income
states, four were one-party Republican states and six were
^■Richard E. Dawson and James Robinson, "Inter-party
Competition, Economic Variables and ’ welfare Policies in the
American States," Journal of Politics, XXV (1963), 285-89.
?
Thomas Dye, Politics, Economics, and the Public:
Policy Outcomes in the American States (Chicago:Rand
McNally, 1966), pp. 54-58.
^Ibid., p. 58.
^Ibid.
200
I
one-party Democratic states. Idaho was the only state !
which is two-party state with low income.
Thus, it may be stated that party competition is
more of a function of education and income than industriali
zation or urbanization. As Daniel Lerner observed in his
Passing of Traditional SocietyJ that literacy supported by
higher standard of living appears to be conditioning factor
for political pluralization, education and income appear
to be important variables in American states.
Education and income were also significantly related
to voter turnout for senate and gubernatorial elections j
during non-presidential years in this study (.33 and .32
respectively). A close relationship between socio-economic i
level and voter participation has already been well estab- j
lished in the literature of political science.
For instance, Lester Milbrath found in his study
I
that education and income had a definite relationship with j
frequency and intensity of participation in voting as well
as other forms of political participation.^ Robert Lane
found just about the same as Milbrath that income and edu
cation are the most critical resources in facilitating
! ^Daniel L. Lerner, The Passing of Traditional
Society (Glencoe: Free Press, 1958), p. 60.
Lester Milbrath, Political Participation (Chicago:
Rand McNally, 1965), pp. 155-89.
201
7
electoral participation in politics.
V. 0. Key also reported in his Public Opinion and
g
American Democracy that electorate is "stratified" in
accordance with socio-economic composition largely reflect
ing educational and economic correlates. Thus, it is no
surprise to find in the study that education and income are
highly related to voter participation.
however, industrialization and urbanization were not 1
found to be related to voter turnout in any significant way
(.21 and .11 respectively). It is contrary to popular
notion that people in cities are more likely to participate |
!
in politics than persons living in the country. ,
It may have so appeared because more educated popu- j
I
lation tend to cluster around the urban center with indus-
I
trial base than rural areas. A mere fact of living in the !
I
city or country per se may not be a guarantee for more or
less propensity to participate in politics. Education and |
j
income may be the basic forces underlying regional pattern j
I
in politics.
Thomas Dye also reports:
. . . political participation is significantly related
| ^Robert Lane, Political Life (New York: Free Press,
1959), pp. 57-66.
. 0. Key, Public Opinion and American Democracy
(New York: Knopf, 1961), p. 96.
202
g
to income and educational levels in fifty states.
But Dye finds very low correlation of industrializa
tion and urbanization with voter turnout. In Dye's words,
. . . among the fifty states there is no discernible
relationship between urbanization and voter turnout.
Nor does the degree of industrialization, as measured
by non-agricultural employment, appear to influence
voter participation.-^
It should also be recalled that inter-party competi
tion was also highly related to education and income but
insignificantly related to urbanization and industrializa
tion. Thus, it seems reasonable co conclude that party
competition and voter participation are the function of
education and income in the American states.
Existing literature'*''*' also supports this conclusion.
Reasonable as this conclusion may appear, it seems to con
tain some flaws either in logic or data. A more rigorous
interpretation of the data seems to be called for especially
in terms of the relationships among the economic develop
ment variables.
9
Dye, op. cit., p. 62.
Ibid.
^See the following works: Joseph Schlesinger, "The
Structure of Competition for Office in the American States,"
Behavioral Science, V (1960), 197-209; Robert T. Golembiew-
ski, 'A Taxonomic Approach to State Political Party
Strength," Western Political Quarterly, XI (1958), 494-513;
iAustin Ranney and Willmoore Kendall, The American Party
System," American Political Science Review, XLVIII (1954),
477-85; V. 0. Key, American State Politics (New York:
, Knopf, 1956), p. 99. ______
203
A question may be raised: how do the education and
income come about in the first place? Couldn't it be that
education and income are the function of industrialization
and urbanization?
12 13
In fact, Lerner and Shannon argued for their
notion of economic prerequisites to politicization by
relating the industrialization to the need for a sophisti
cated level of education which leads to higher income.
This reasoning leads, then, to further search for relation
ship among the four variables of economic development to
see how closely they are related to each other.
TABLE 46
RELATIONSHIP AMONG URBANIZATION, INDUSTRIALIZATION,
INCOME, AND EDUCATION
Urban. Indust. Income Educ.
Urbanizat ion
Industrialize!ion
Income
1.0
.6?;:
.66 f
l.C ,
.59? 1.0 ,
Education .41“ .21“ .74“ 1.0
Source: Thomas Dye, Politics, Economics and the Public
(Chicago: Rand McNally, 1966), p. 33. The above figures
are simple correlation coefficients for the fifty states;
asterisks indicate statistically significant relationships.
■^Lerner, loc. cit.
13
Lyle W. Shannon, "Is Level of Development Related
to Capacity for Self-Government?" American Journal of
Economics and Sociology, XVII (1958), 367-82.
The data show that urbanization is significantly j
related to all of the three economic variables, viz.,
industrialization, income, and education ( .67 , .G6 , .41,
respectively).
Industrialization is significantly related to income
but not to education. Income and education show, however,
the highest correlation (.74). It can be inferred, then,
that the states with high level of education and income
are likely to be also the states with high level of indus- !
trialization and urbanization at the same time.
i
Low correlation between industrialization and educa-j
tion could be spurious because urbanization and industri
alization are highly related (.67). At the same time,
urbanization and education are highly related (.66).
Thus, the fact that industrialization and urbaniza
tion are not significantly related to voter participation
and inter-party competitiveness may not necessarily
invalidate the economic-resource explanation for voter
turnout or party competition. Because industrialization
and urbanization appear to be the conditioning factors for
higher level of education and income.
Logical as this conclusion may seem, it still does
not appear sufficient enough to generalize American state
political system characteristics based on economic-resource
alone.
Texas, South Carolina, and Tennessee, for example,
are the states with moderate to high levels of economic j
development but either with lower voting record than
national average or one-party orientation (see Table 6).
Explanation for these variations may have to be
sought in the context of unique historical and political
culture of a particular region or states. For instance, in
the South, voter participation is affected by exclusion of
Negroes through social, political, and economic barriers.
Lower educated groups are barred by complicated registra
tion procedures, literacy tests, and residence requirements^
Political culture embedded in racism still creates
i
artificial hindrance against active participation of large
potential electorates, although Voting Rights Act of 1965
and Civil Rights Act of 1968 somewhat improved the situa-
14 |
Lion. |
i
Efficacy of the Voting Rights Act and Civil Rights j
Act is yet to be tested. Enforcement of the acts is being
hampered by open defiance of the law on the part of several
southern and border states. Thus, Kelley, Ayres, and
Bowen concluded, after studying the effect of the Voting
■^Largely as a result of Voting Rights Act of 1965
and Civil Rights Act of 1968, in eleven Southern states,
for instance, the proportion of registered voting-age
Negroes increased 43 per cent in 1964 to sixty-two per cent
for years later. However, this figure is still below the
seventy-eight per cent of whites of voting age registera;
see, William Ebenstein, et al., American Democracy (New
York: Harper <5 Row, 1970), p. 245.
i 206
Rights Act of 1965, that:
Study confirmed what many had believed: Differences
in voter turnout are related to rates of registration,
and these in turn reflect _o a considerable degree
local differences in the rules governing, and arrange
ments for handling, the registration of voters.15
Yet, there is another factor to be considered before
hastening to conclude that the political participation is
the effect of the economic resources and unbridled institu
tional guarantees, e.g., the Voting Rights Acts, speedy
registration procedures.
That factor is the problem of motivation or the basic
attitude of the electorates toward the political process. |
One has to view the participation in political process as
a positive exercise of some consequence to take a part in
the system. How does, then, a certain level of political
motivation come about?
The economic-resource explanation leads us to
I
believe that a given level of socio-economic resources
engenders a corresponding level of political motivation.
This doesn't always seem to be the case, however. Dahl's
1 f i
study of the Negroe voters in New Haven shows that a high
15
Stanly Kelley, Richard Ayres, and William Bowen,
["Registration and Voting: Putting First Things First,"
American Political Science Review, LXXI (1967), 373-74.
■^Robert A. Dahl, Who Governs? Democracy and Power
in an American City (New Haven: Yale University Press,
1961), p. 115.
I
207 .
level of political motivation can come about independent of I
prior economic resources largely because of a certain
political culture in a particular region.
For instance, Robert Dahl reported in his study of
New Haven, Connecticut that a large number of Negro popu
lation in New Haven was participating in politics at a
level higher than national average of the Negroe voters.
Forty-four per cent of the Negroes were among Dahl's "high"
and "highest" campaign participants, compared with only
twenty per cent of the whites.^7 ,
This higher rate of participation of Negroes in New j
Haven did take place in spite of the fact that their income i
I
and job levels were markedly lower than those for whites. :
I
Thus, Negroes participated in New Haven politics far more j
than a strictly economic-resource explanation would lead on^
to expect. !
Dahl attributes this deviant phenomenon to "oppor
tunity and motivation" nourished by a particular political
culture in New Haven. Dahl points out that the Negroes
met discrimination in housing and in hiring for private
enterprise but relatively no discrimination against poli
tical participation.
In Dahl's words,
17 Ibid.
208
J
In contrast to the situation the Negro faces in the j
private socio-economic sphere, in local politics and
government the barriers are comparatively slight.
There is no discrimination against Negroes who wish to
vote; they have participated in elections for genera
tions. Tartly because of their votes, Negroes are not
discriminated against in city employment; they have
only to meet the qualifications required of white
applicants. . . . Negroes also share in city patronage,
city contracts, and other favors.18
Because of tradition of the absence of political
discrimination in New Haven, therefore, low educated and
relatively poor as the Negroes were, they were motivated
to participate in electoral process. j
I
Almond and Verba also report about the same findings J
19 I
in their cross-cultural studies, The Civic Culture. j
I
Educated and high income professional groups in
Germany, Italy, and Mexico were found to have basically
20
passive attitudes toward politics and government. Con
sequently, the rate of political participation among these
professional people in three countries was lower than
comparable groups in the United States and United King
dom.21
Almond and Verba attribute this phenomenon to the
lack of democratic culture in the respective history of
1^Ibid., p. 294.
19
Gabriel Almond and Sydney Verba, The Civic Culture
(Princeton: Princeton University Press, 1963).
20Ibid., p. 274.
21Ibid.
209
Germany, Italy, and Mexico. That is, they all lack in
civic culture.
Thus, it appears that economic-resourcc explaiiation
22 23 24 25
as advanced by Dye, Dawson, Shannon, and Lerner,
explains only a part of the total picture of party compe
tition or political participation. It has to be supple
mented with proper attention to unique political culture,
historical and institutional variables underlying a par
ticular political phenomena.
It seems, then, that political participation is the
effect of composite of factors: motivation, institutional
structure, political culture, and economic resources. i
Closely related to institutional structure is another
factor that may ultimately affect the political partici
pation in the American states. That is the phenomenon of I
malapportionment.
A continuing sense of inequity in representation
will undermine the confidence of the electorate in the
j
political system. Notwithstanding the historical decisions
26 27
of Baker vs. Carr of 1962 and Reynolds vs. Sims of
i
22
Oye, loo, cit.
23
Dawson and Robinson, loc. cit.
24
Shannon, loc. ext.
Lerner, loc. cit.
^ Baker v . Carr, 369 U.S. 189 (1962).
^Reynolds v. Sims, 377 U.S. 606 (1964).
210
1964, malapportionment between rural and urban areas is
still a continuing phenomenon. How could one account for
this persisting problem of malapportionment? Does it have
anything to do with the level of economic development of
a state?
Variations in terms of apportionment and malappor
tionment in the states were found to be not related at all
to the four variables of economic development in this
study. Correlation between apportionment score and indus
trialization was -.08; with education, -.12; with income,
-01, and with urbanization, -.01.
In Dye's study of 1966, correlation between malap
portionment score and industrialization was -.19; with
education, -.19; with income, -.21; and with urbanization,
-.24. Dye's results are just about the same as those of
this study.
Thus, Dye concludes that
There is no relationship between economic develop
ment and malapportionment. The legislatures of rural
farm states are just as likely to be unrepresentative
as the legislatures of urban industrial states. The
southern states are no more mal-apportioned than the
non-southern states.23
For instance, Delaware with highest per capita
income in the nation is one of the worst apportioned states
New York and California with high industrialization and
28., . . r Q
Dye, op. cit., p. 68.
income are as much malapportioned as southern less Indus- i
Lrialized Mississippi, Alabama, and Georgia.
Therefore, apparently malapportionment is not the
function of economic development, explanation must be
found somewhere else. liisenberg offers an explanation
saying that urban underrepresentation in states took place
despite adherence by most of them to constitutional stan
dards for legislative apportionment.
Constitutional biases often are responsible for the
situation of inequitable representation in both legis- ,
lative houses or in at least one house.29
Eisenberg, then, cites that
For example, states whose representation system is j
the "federal" form are obligated to reapportion only
one house of the legislature and that house only withinj
guarantees of minimum representation per county.30 ;
Such provisions leave rural states solidly entrenchecj
in control of one house and overrepresented in the other,
even where urban areas control majorities in the second
1
house.
Even those states which use population as the stan
dard eventually end up with unequal representation.
Because they do so "within limits that reward the least
29
Ralph Eisenberg, "Power of Rural Vote," Howard D.
Hamilton (ed.), Legislative Apportionment; Key to Power
(New York: Harper & Row, 1964;, p. 14.
30-r-U • A
Ibid.
212
populated counties proportionately more than the most
populated counties."
Georgia is cited as a good example of this:
In Georgia, in the lower house each of the 159
counties has one representative but additional repre
sentation is given to the 38 most populous counties;
yet three is the maximum number that any county may
have.31
Another group of states simply ignores relatively
clear constitutional instructions to distribute represen
tation by population or takes advantage of general consti
tutional language by enacting only token reapportionments.
\Jhy are the states, then, so reluctant and slow in
responding to the requirement of equal representation based
on population? Continuing malapportionment directly vio
lates the democratic canon of "one man, one vote." There
is no valid reason that rural votes should carry more
weight than city-voters.
Apparently the basic motive underlying the states'
reluctance to reapportion equitably is political. Dye
expresses best this view:
Entrenched rural and small-town interests are
reluctant to surrender control to city populations
with their large blocks of ethnic, laboring,
Catholic and Negro voters. Rural interests are often
allied with those urban interests who share this view
of the urban electorate.32
31T- , • i
Ibid.
32
Dye, loc. cit., p. 66.
213
i
There is also another political ethos which goes i
back to the Jeffersonian notion of democracy based on small
agrarian community. This rural heritage as a part of
American political culture persists and serves as a basis
of defenses of rural overrepresentation.
Somehow it is still believed by some that rural
farm people are "pure" and "genuine"; they are the vanguard
of American democracy. Again Dye expresses this sentiment
well when he says that
Often revealingly frank defenses of rural over
representation are made on the ground that rural and
small-town people are "better" or "safer" than big-
city voLcrs . - J~ J
i
i
In addilion, there may be natural reluctance to i
censure themselves among the legislators. I
There is natural reluctance for legislative bodies
to perform major surgery on themselves; to expect a
legislature to redistrict itself is to expect legis
lators to threaten their own incumbency.
Malapportionment as one of the characteristics of
state political system is, thus, an effect of composite of
constitutional bias, political compromise, and political
history in the United States. Socioeconomic measures or
level of economic development appear to have a little to
do with the way American state legislatures are apportioned.
X U1U •
34t,. ,
Ibid.
214
i
Investigation of the socio-economic variables rela
tive to the variations in the states' political system
characteristics was carried out in this study not just for
its own sake. Political party competition or political
participation is hardly worth in themselves unless it has
an impact on the product of the system.
That is, knowledge of the process of political
character formation of each state political system is
valuable only when it can help us explain policy differ
ences among the states. Since the political system is the
place where the scarce resources are authoritatively alio- i
cated, would the fact that one state is one-party Demo
cratic dominated, experiences low voter turnout and malap-
portioned have anything to do with the way she allocates ,
her resources to various social programs? |
|
Interface of Policy Outcomes with Political and I
I
I
Economic Variables |
In this study, the relationship between political
system characteristics and policy outcomes in Medicaid was
found to be very low (see Table 33). i
Three measures of Medicaid decisions made by thirty-
j
eight states, viz., scope of recipients, extent of ser
vices, and size of expenditure, did not bear any signifi
cant relationship with three variables of political system
characteristics, namely, party competitiveness, voter
215
I
participation, and apportionment.
3 3
Dye also reports a very low correlation of wel
fare expenditure per capita and health care expenditure
with political system characteristics, tarty competition
scored .01; political participation, -.30; and malappor
tionment, .04.
These low correlational values in Dye's study are
rather close to the findings of this study. Correlation
coefficient between Medicaid cost per capita income and
party competition was .08; with voter participation, .15;
and malapportionment, .02 (see Table 33). !
36 37 !
However, Dawson and llofferbert found that their j
|
respective measures of political system characteristics j
showed moderate to high correlation with various policy
outcomes. For instance, when Dawson and Robinson corre-
i
lated inter-party competition and political participation
with per capita welfare expenditure of fifty states,
3 8
correlation coefficients were .57 and .39 respectively.
Notwithstanding such a high correlation between
political system variables and welfare expenditures,
3~ )Ibid. , p. 131.
36
Dawson and Robinson, loc. cit.
37
llofferbert, loc. cit.
38
Dawson and Robinson, op. cit., p. 218.
216
Dawson and Robinson broke a new ground in their influential
39
article, "Inter-party Competition, economic Variables,
and Del fare bolides in the American btates," by having
concluded that economic development, instead of political
system, independently influences the extent of welfare
expenditures in the United dates. Dawson and Robinson
were able to come to a conclusion discrediting the inde
pendent role of the political system because of another
high correlation they found in relation to economic vari
ables.
When Dawson and Robinson correlated four economic
development variables, i.e., industrialization, income,
education, urbanization, with welfare expenditures, the !
correlation coefficient was found to be significantly j
, . . 40 S
high. ;
Faced with high positive correlation relative to j
both political and economic variables, they examined cor- |
relations between economic variables and policy outputs
with political system variables controlled, and also
between political system variables and policy outputs with
economic development variables controlled.
With these controls, the relationships between
economic variables, especially per capita income and
39Ibid.
40Ibid.
217
welfare policies were found to be substantially higher j
than those between political variables and welfare out
comes. On the basis of this analysis, Dawson and Robinson
offered the conclusion: '
The level of public social welfare programs in the
American states seems to be more a function of socio
economic factors, especially per capita income. High
levels of inter-party competition do not seem to pos
sess the important intervening influence between
socio-economic factors and liberal welfare programs
that our original hypothesis and theoretical scheme
suggested .4-1
Dye used the same controlling procedures, i.e., J
partial correlational analysis in his study of five policy j
I
areas, namely, highway, education, public regulation, wel- j
fare, and taxation. Although Dye found the relationship
i
between political variables and health-related welfare !
outputs to be low, he found the political variables to be
42
highly related to all other remaining four policy areas. j
With controls, Dye subsequently found the relation- j
ship between economic variables and the above four policy j
t
I
outputs to be substantially higher than that between
43
political system variables and four policy outcomes. In
addition, Dye found that the four economic variables,
especially, income and education were highly related to
41 t v • j
Ibid.
^Dye, op. cit., pp. 282-285.
43T. . ,
Ibid.
44
inter-party competition and political participation.
Therefore, Dye was able to conclude that socio
economic development shapes policy outcomes as well as
45 1
political system characteristics in the American states.
This study also came to draw just about the same conclusion
as Dawson and Dye did. The procedures were different,
however, because of very low correlation between political
system variables and the measures of Medicaid program
ranging from -.01 to .15 (see Table 33).
In contrast to this low correlation between politi
cal variables and policy outcomes in Medicaid, the corre- i
j
lation coefficients between economic variables and policy
outcomes of Medicaid were significantly high. Sxcept ;
education, industrialization, income, and urbanization
showed significant relationships with Medicaid measures
ranging in values from .40 (scope of services offered and |
urbanization) to .87 (total number of recipients covered
and industrialization)(see Table 33).
i
Since the relationships between economic variables, i
except education, and Medicaid policy outputs were sig
nificantly high and those between political system vari
ables and Medicaid measures were low, there was no need
for controlling the effect of either political variables
44T. • j
Ibid.
^~*Ibid., p. 285.
on economic variables or vice-versa relative to policy
measures, i.e., Medicaid.
Thus, tliis study was able to conclude directly with
out going through statistical "controlling" procedures
that liberal Medicaid benefits in the American states are
the function of economic development rather than politi
cal system characteristics. Furthermore, this study
concluded that the level of economic development directly
influences the character of the states' political system
as well.
Because inter-party competition and political parti-j
cipation were found to be significantly related to per
capita income and education which were, in turn, at least
inferentially (see Table 37) determined to be the function
i
of industrialization and urbanization. That education is J
I
found to be significantly related to party competition and j
I
voter participation but conspicuously unrelated to Medic- |
aid policy outputs further supports the latter conclusion. S
Because this means that the factor found to be the
most crucial determinant of political characteristics is
the least important one in affecting the policy outputs.
What else could one conclude, in this case, but that the
political system is nothing but a belt through which
socio-economic inputs are transmitted to produce policy
outputs?
It partly, then, validates the notion of economic
220
determinism relegating the role of political process to ;
something subordinated to and dependent upon economic
resources. Its validity is, however, tenable only within
the context of assumptions and variables included in the
mode1.
The model does have some limitations. These limita
tions are of three kinds primarily related to the way the
model dealt with political system characteristics.
First of all, such an important parameter in politi
cal process as bureaucracy was left out. Interest group is
another dimension which was somewhat implied in the treat- !
ment of inter-party competition but not adequately treated. ,
Existing literature also notably lacks in the treatment of ;
j
bureaucracy and interest groups as aggregates, correspond
ing to the treatment of other variables in this study.
An additional dimension to the character of politi
cal system in the states is the role of central Federal j
government particularly in all social improvement pro
jects. ^ Massive flow of Federal grant-in-aid funds in the
past decade may have considerably undermined the autonomy
of the state governments.
46
Amount of money Federal grants alone to state and
local governments for projects ranging from natural
resources through education to welfare and health increased
from $8,634 million to $20,297 million in 1969. See:
U.S. Bureau of the Census, Statistical Abstract of the
United States: 1968 (89th edition), p. 410.
221 i
Intergovernmental fiscal relations has brought about I
a great deal of osmosis changing the traditional character
of state-local political systems. Intergovernmental
theories^ may have to be considered as a viable alterna
tive to Dye or Dawson's type of economic-resource explana
tion in the future.
Secondly, the impact of dynamic personalities or
strong traditions as a part of political culture upon [
policy outcomes was played down in this study. The elite,
organizational variables, and values embedded in social- <
cultural environment of a particular state or region are
legitimate variables to be included in the political system
characteristics. j
Thirdly, there perhaps couldn't be a single answer
to the question: Is it politics or economics that has the
greatest impact upon public policy? Because the answer
may vary with the dimensions of policy phenomena at issue.
j ~ j
For example, Jack L. Walker proposes to study
differences in policy outputs among the states within the
context of theories of "reference group." Walker argues
that there appears to be a leader and follower relation
among the states in terms of geographical contiguity,
instead of influence of economic resources alone. Fol
lower states adopts the policies in the light of what the
leader state(s) do first, independent of own economic
resources; see Jack Walker, "The Diffusion of Innovations
Among the American States," American Political Science
Review, LXIII (1969), 880-88; Grodzin's marble-cake notion
is another instance of intergovernmental theory, Horton
Grodzin, The American System (Chicago: Rand McNally, 1966),
222
A 3
Charles Cnudde and D. J. McCrone propose their
so-called hybrid model of policy studies in their recent
article, arguing that whether politics plays a role inde
pendent of economic inputs depends upon a particular class
of policies one happens to investigate. Ira Sharkansky
echoes the view of Cnudde and McCrone in his very recently
published article, '‘Dimensions of State Politics, econo
mics, and Public Policy" by reporting that:
Welfare-education policies relate most closely
with the Competition-Turnout dimension of state poli
tics and with the Affluence (Income) dimension of the
economy. Highway-Natural Resource policies show their
closest (inverse) relationships with the Industriali
zation dimension of the state economy. ^
This directly leads to the need for conceptualizing
the public policies based on some notion of taxonomy in
terms of probable impact of a class of policies on either
economic or political system or both.
The content approach this study incorporated into
the model is exactly a response to the plea for conceptu
alization of public policies in terms of proper categories
of impacts, rather than for measuring their relationship
Charles F. Cnudde and Donald McCrone, "Part}'’
Competition and Welfare Policies in the American States,"
American Political Science Review, LXIII (1969), 858-66.
4 9
Ira Sharkansky and Richard llofferbert, "Dimen
sions of State Politics, Hconomics, and Public Policy,"
American Political Science Review, LXIII (1969), 878
223
to economic or political system characteristics.
However, categorizing policies in terms of probable
impact is admittedly a difficult task. As it was noted in
the second chapter of this study, the "content" approach
is sLill in its formative stage. Thus, what this study
did in terms of content approach falls short of an ideal
operation of ideas and concepts implied in it.
Impact of Medicaid Policy
Medicaid evolved out of government's historic role
in caring for the nation's poor. This responsibility
largely remained as a state-local problem until the advent |
of the Social Security Act of 1935. The Act allowed the
Federal government to share with the States the cost of
providing the "categorical" assistance to the needy aged, |
i
blind, and families with dependent children, which included]
the medical costs in the total payment.
This provision of the original Social Security Act
of 1935 for helping medically only those in "categorically"
needy remained virtually unchanged until 1965. In 1965, the
concept of medical care on a nation-wide basis came to
extend substantially to all those who are not only poor,
i.e., "categorically" needy but also medically needy cul
minating in the enactment of Medicaid and Medicare.
It was found that policy-makers at all levels of
government were largely unprepared for the magnitude of
224
Medicaid. This unpreparedness became apparent soon after
the program's initiation in 1966. Lack of preparation and
projection of the future was apparent particularly in the
fantastic increase of the costs involved.
Federally assisted medical vendor payments rose
from $1,358 million in calendar year 1965 to an estimated
$4,184 million in fiscal year 1969.^ For the Federal
government this was an increase from $756 million to an
estimated $2,040 million in three and a half years;J^ for
state and local governments, a rise from $602 million to
52
an estimated $2,145 during the same period.
Dramatic as the rise was from the aggregate stand
point of all of the thirty-eight participating states as
of July, 1968, there were wide state-to-state variations
in scope of the program and its fiscal impact:
Thirteen states authorized medical care for the
categorically needy and additional categorically
related needy.
Twelve states authorized care for a broader range
of needy as well as medically needy, and only a few
50
Advisory Commission on Intergovernmental Relations,
Intergovernmental Problems in Medicaid (Washington, D. C.:
Government Printing Office, 1968), p. 56.
51TK.j
Ibid.
52tk.^
Ibid.
225
53
(e.g., California, New York) covered non-
categorically related medically needy for whom
Federal matching funds were not available.
Seventeen states provided to some extent at least
eleven of the fourteen types of statutorially spe
cified medical services for both needy and medically
needy.
For twenty-seven Medicaid states in effect for all
of 1967, the change in total medical vendor payments
between 1965 and 1967 varied from an increase of
$487 million in New York to a decrease of $1.7
million in West Virginia.
Of the twenty-three states providing medical care,
eleven were forced to cut back or otherwise adjust
their eligibility standards to stay within the
bound of fiscal capability.
In a few states (e.g., New York, California, Massa
chusetts), new or higher state taxes were in part
attributed to the rising cost of Medicaid.
Assumption of such an increase of medical cost by
state and federal government resulted in shift of medical
finance from private sources to public sources since the
program became effective in 1966. Medical prices increased
6.6 per cent in 1966 and 6.4 per cent in 1967, compared to
5 3
Ibid.
226
i
rises of 3.3 and 3.1 per cent in the overall consumer price
index.
While Medicaid has suffered from industry-wide
general inflation, it has contributed to that rise out of
proportion to the number of persons it served because of
its sudden injection of billions of additional dollars into
the demand for medical services. Its lack of effective
control over expenditures has also added to the pressure
on medical prices.
Among the non-fiscal problems in the Medicaid pro- ,
gram most concerning governors and legislative leaders j
I
I
I
were: !
(a) Difficulties in coordinating the administration of
I
Medicaid and Medicare;
(b) Difficulties in imposing adequate controls over
charges for service.
It was found that the feedback was fed into the
i
system. The system appeared to have responded to the feed
back. Fiscal intermediary was adopted as a means of doing
something about the delay in payment to the providers of
i
services. A common billing system was instituted by HEW
specifically to simplify the coordination between Medicaid
and Medicare.
However, the revisions of 1967 legislation pertain
ing to Medicaid were largely a policy of retrenchment in
face of rising costs of Medicaid at both Federal and state
227
level. I
Investigation of the impact of Medicaid was the
primary purpose of Lhe chapter six as a part of implemen
tation of the "content" approach in policy studies.
Conelusions
Thus far, this chapter discussed the summary of
findings on three related problems of this study, i.e.,
relationship between economic development and political |
system characteristics, interface of policy outputs in
Medicaid with economic and political system variables, and
i
analysis of the impact of Medicaid.
In retrospect, this study revealed the defects and
j
deficiencies prevalent in the study of public policies j
rather than a dramatic finding. The study concludes with j
i
the following suggestions for a new direction of policy
research in the future.
1. There has to be a serious search for more relevant
political variables to be included in the concept of
political system characteristics; political culture,
values, personalities, and administrative variables
ought to be reflected in any future model of public
policies.
2. There is an urgent need for developing a taxonomy of
public policies for two reasons:
228 ,
(a) because politics may be found to play an inde
pendent role, after all, in policy process
depending upon the policies one investigates;
thus, it is important to discern the type of
policies susceptible to political process as
opposed to those susceptible to direct influ
ence of socio-economic development;
(b) because a solid taxonomy of policies is pre
requisite to implementing the perspective of ■
"content" approach to the study of public !
i
policy. That is, to hold policies as indepen
dent variables or constants taking the politics
and economics as varying factors is to assume a j
certain fixation of a set or class of policies.
3. Impact of Medicaid ended in this study with descrip
tion of rising medical costs, intergovernmental
fiscal relations, and other operating problems.
Even in the absence of proper taxonomy of policies,
future study of the impact of Medicaid is suggested
to focus more specifically on political system
characteristics and socio-economic environment. For
example, how did rising medical costs alter "health"
politics in Congress and state legislatures? How
did the push for a sudden demand for more health and
medical care affect those socio-economic variables
relevant to forming subsequent political systems and
other policy outputs? How could the data on the
impact of Medicaid be used in inventing policy
categories, say welfare policies as opposed to
defense or transportation policies?
BIBLIOGRAPHY
Books
Adrian, Charles. State and Local Governments. New York:
McGraw-Hill Book Company, 1960.
Baker, Lrnest. The Politics of Aristotle. New York:
Oxford University Press, 1946.
Bentley, Arthur. The Process of Government. Chicago:
University of Chicago Press, 1908.
Berelson, Bernard R., et al. Voting. Chicago: Univer
sity of Chicago Press, 1954.
Blalock, Hubert M., Jr. Causal Inferences in Non-
experimental Research. Chapel Hill: University
of North Carolina Press, 1964.
________ . Social Statistics. New York: McGraw-Hill
Book Company, 1960.
Blaustein, A. P. and C. C. Ferguson, Jr. Desegregation
and the Law. New Brunswick: Rutgers University
Press, 1957.
Buchanan, James and Gordon Tullock. The Calculus of
Consent. Ann Arbor: University of Michigan Press,
1962.
Campbell, Angus, Phillip E. Converse, Warren E. Miller,
and Donald E. Stokes. The American Voter. New
York: John Wiley & Sons, Inc., 1960.
Campbell, Angus, Gerald Gurin and Warren E. Miller. The
Voter Decides. Evanston: Harper & Row Publishers,
1954.
Cyert, Richard, et al. A Behavioral Theory of Firm.
Englewood Cliffs, N. J.: Prentice-Hall, Inc., 1963.
230_____________________________
231
Dahl, Robert, et al. Politics, Economics and Welfare.
New York: Harper & Row Publishers, 1953.
David, Paul T. and Ralph Eisenberg. Devaluation of the
Urban and Suburban Vote. Charlottesville: Bureau
of Public Administration, University of Virginia,
1961.
Davison, W. Phillips. The Berlin Blockade: A Study of
Cold War Politics. Princeton, N. J.: Princeton
University Press, 1958.
Dror, Yehezkel. Public Policy Making Reexamined. San
Francisco: Chandler Publishing Company, 1968.
Durkheim, Emile. The Division of Labor in Society.
Translated by George Simpson. New York: The Free
Press, 1964.
Duverger, Maurice. Political Parties. New York: John
Wiley & Sons, Inc., 1958.
Dye, Thomas R. Politics, Economics and the Public.
Chicago: Rand-McNally & Co., 1966.
Easton, David. The Political System. New York: John
Wiley & Sons, Inc., 1953.
________ . A Framework for Political Analysis. New York:
Prentice-Hall, Inc., 1965.
________ . A Systems Analysis of Political Life. New York:
John Wiley & Sons, Inc., 1965.
Eckstein, Harry and David Apter, eds. Comparative
Politics: A Reader. New York: John Wiley & Sons,
Inc., 1963.
Eisenstadt, S. N. Modernization: Protest and Change.
New York: John Wiley & Sons, Inc., 1966.
Elazar, Daniel J. The American Partnership. Chicago:
University of Chicago Press, 1962.
232
Kps Lein, Leon 0. British Politics in the Suez Crisis.
Urbana: University of Illinois Press, 1964.
Farell, Barry, ed. Approaches Lo Comparative and Inter
national Politics. Pvanston, Illinois: North
western University Press, L966.
Fine, Sidney. Laissez Fen re and the General welfare aLate.
Ann Arbor: University of Michigan Press, 19 j6.
Flynn, Thomas. Governor Freeman and the Minnesota Budget.
New York: The Bobbs-Merril Co., Inc., 1937.
Francis, Wayne L. Legislative Issues in the Fifty States:
A Comparative Analysis. Chicago: Rand-McNally
N Co., 19b7.
Friedrich, Carl. Man and His Government. New York:
McGraw-Hill Book Company, 196 3.
Galbraith, John K. The Affluent Society. Boston:
Hought on-Mi f f1i n Company, 19 3 6.
Gerschenkron, Alexander. Fconomic Backwardness in liistorl-
cal Perspective. Cambridge: Belknap Press, Inc.,
1962.
Gerth, H. H. cind C. Wright Mills. From Max Weber. New
York: Oxford University Press, 1946.
Grant, Daniel and H. C. Nixon. State and Local Government
in America. Boston: Allyn 6 / Bacon, Inc., 1963.
Greenstein, Fred I. The American Party System and the
American People. Knglowood Cliffs, N. J.:
Prentice-Hall, Inc., 1963.
Herbert, Jacob and Kenneth N. Vines, eds. Politics in the
American States. Boston: Little, Brown and
Company, 1963.
Hirschman, Albert. Journey Toward Progress. Fnglewood
Cliffs, N. J.: Prentice-Hall, Inc., 1963.
_____j
233
Hitch, Charles and Roland McKean. The Economics of Defense
in the Nuclear Age. Cambridge: Harvard University
Press, 1961.
Keeley, John. Moses on the Green. New York: The Bobbs-
Merril Co., Inc., 1957.
Kcrlinger, Fred N. Foundations of Behavioral Research.
New York: Holt, Rinehart & Winston, Inc., 196 3.
Key, V. 0., Jr. American State Politics. New York:
Allred A. Knopf, Inc., 1956.
________. Southern Politics in State and Natjon. New
York: Alfred A. Knopf, Inc., 1951.
________. Politics, Parties and Pressure Groups. 4th ed.
Crowell Collier and Macmillan, Inc., 1958.
Lane, Robert E. Political Life. New York: The Free Pressj
1959.
________. Political Ideology. New York: The Free Press,
1962.
Lasswell, Harold D. and Abraham Kaplan. Power and Society.
New Haven: Yale University Press, 1950.
Lasswell, Harold D. Psychopathology and Politics. New
York: The Viking Press, Inc., 1960.
Lasswell, Harold D. and Daniel Lerner, eds. The Policy
Sciences. Stanford: Stanford University Press, j
1951. j
Lerner, Daniel. The Passing of Traditional Society.
Glencoe, Illinois: The Free Press, 1958.
Lockard, Duane. New England State Politics. Princeton: j
Princeton University Press, 1959. I
Macridis, Roy C. The Study of Comparative Government.
New York: Doubleday & Company, Inc., 1955.
Milbrath, Lester W. Political Participation. Chicago:
Rand-McNally & Co., 1965.
234
Mitchell, William C. and Joyce Mitchell. Political Analy
sis and Public Policy: An Introduction to Political
Science. Chicago: Rand-McNally & Co., 1960.
Morris, Bruce K. Problems of American Economic Growth.
New York: Oxford University Press, 1961.
Neustadt, Richard P. Presidential Power. New York: John
Wiley < 9 Sons, Inc., 1963.
Peatman, John G. Introduction to Applied Statistics.
New York: Harper A Row Publishers, 1963.
Pendleton, Herring. The Politics of Democracy. New York:
Holt, Rinehart A Winston, Inc., 1940.
Pye, Lucian. Politics, Personality and Nation Building.
New Haven: Yale University Press, 1962.
Ranney, Austin. The Doctrine of Responsible Party
Government. Urbana: University of Illinois Press,
1934.
Rostow, W. W. The Process of economic Growth. 2nd ed.
New York: W. W. Norton A Company, Inc., 1962.
Schattschneider, L. L. Party Government. New York:
Parrar, Stroud A Giroux, Inc., 1942.
Schilling, Warner R., Paul Hammond 4 Glenn Snyder, eds.
Strategy, Politics and Defense Budgets. New York:
Columbia University Press, 1962.
Seymour, Martin Lipset. PoliLical Man. New York:
Doubleday A Company, Inc., 1960.
Shannon, Lyle W. Underdeveloped Areas. New York:
Harper A Row Publishers, 1937.
Simon, Herbert, ed. Models of Man. New York: John
Wiley & Sons, Inc., 1957.
Truman, David. The Governmental Process. New York:
Alfred A. Knopf, Inc., 1951.
233
Van Dyke, Vernon. Pride and Power: The Rationale of the
Space Program. Urbana: University of Illinois
Press, 1964.
________ . Political Science: A Philosophical Analysis.
Stanford: Stanford University Press, 1960.
Vasey, Wayne. Government and Social Welfare: Roles of
Federal State and Local Governments in Administer
ing Welfare Services. New York: Holt, Rinehart &
Winston, Inc., 1938.
Wahlke, John C. and Heinz, Fulau. Legislative Behavior.
Glencoe, Illinois: The Free Press, 1939.
Articles and Periodicals
Almond, Gabriel, eL a1. "A Suggested Research Strategy
in Western European Government and Politics,"
American Political Science Review, XLIX (1933).
Bishop, George A. "Tax Burden by Income Class," National
lax Journal, XIV (1961).
Cnudde, Charles F. and Donald J. McCrone. "Party
Competition and Welfare Policies in the American
States," American Political Science Review, LXIII
(1969).
Cutwright, Phillips. "National Political Development:
Measurement and Analysis," American Sociological
Review, XXVIII (1960).
"Political Structure, Economic Development,
and National Security Programs," American Journal
of Sociology, LXX (1963).
Dauer, Manning J. and Robert G. Kelsay. "Unrepresentative
States," National Municipal Review, XLIV (1955).
236
Dawson, Richard, E. and James A. Robinson. "Inter-Party
Competition, Economic Variables and Welfare
Policies in the American States," Journal of Poli
tics, XXV (1963).
Dye, Thomas R. "Malapportionment and Public Policy in the
States," Journal of Politics, XXVII (1965).
Easton, David. "An Approach to the Analysis of Political
Systems," World Politics, IX (1957).
Fisher, Glen W. "Determinants of State and Local Govern
ment Expenditure: A Preliminary Analysis," National
Tax Journal, XIV (1961).
Flinn, Thomas R. "Party Responsibility in the States:
Some Causal Factors," American Political Science
Review, LVIII (1964).
Francis, Wayne L. "Influence and Interaction in a State
Legislative Body," American Political Science
Review, LVI (1962).
________ . "The Role Concept in Legislature: A Probability
Model and a Note on Cognitive Structure," Journal
Of Politics, XXVII (1965).
Golembiewski, Robert T. "A Taxonomic Approach to State
Political Party Strength," Western Political
Quarterly, XI (1958).
Goodman, L. A. "Ecological Regression and the Behavior of
Individuals," American Sociological Review, XVIII
(1953).
Hofferbert, Richard I. "The Relation Between Public Policy
and Some Structural and Environmental Variables in
the American States," American Political Science
Review, LX (1966).
________ . "Classification of American State Party
Systems," Journal of Politics, XXVI (1964).
237
________ . "Ecological Development and Policy Change in the
American States," Midwest Journal of Political
Science, XXV (19667T
Jacob, Herbert. "The Consequences of Malapportionment: A
Note of Caution," Social Forces, XL1II (1964).
Kolleberg, Arthur. "The Logic Of Comparison: A Methodo
logical Note on the Comparative Study of Political
Systems," World Politics, XIX (1966).
Keefe, William J. "Comparative Study of the Role of
Political Parties in State Legislatures," Western
Political Quarterly, IX (1956).
Lowi, Theodore J. "American Business, Public Policy,
Case-Studies, and Political Theory," World
Politics, XVI (1964).
MacRae, Duncan. "The Relatjon Between Roll Call Votes and
Constituencies in the Massachusetts House of
Representatives," American Political Science Review,
XLV1 (1952).
McClosky, Herbert, et al. "Issue Conflict and Consensus
Among Leaders and Followers," American Political
Science Review, L1V (1960).
Pye, Lucian. "The Non-Western Political Process,"
Journal of Politics, XX (1958).
Ranney, Austin and Willmoore Kendall. "The American Party
System," American Political Science Review, XLVIII
(1954).
Robinson, W. S. "Ecological Correlations and the Behavior
of Individuals," American Sociological Review,
XV (1950).
Schlesinger, Joseph A. "A Two-Dimensional Scheme for
Classifying the States According to Degree of
Inter-Party Competition," American Political
Science Review, XLIX (1955")~!
238
________. "The Structure of Competition for Office in the
American States," Behavioral Science, V (1960).
Schubert, Glendon and Charles Press. "Meastiring Malappor
tionment," American Political Science Review,
LXIII (1964"J"1
Seymour, Martin Lipset. "Some Social Requisites of
Democracy: Economic Development and Political
Legitimacy," American Political Science Review,
LII (1959).
Shannon, Lyle W. "is Level of Development Related to
Capacity for Self-Government?" American Journal of
Economics and Sociology, XVII (1958).
Shapiro, Sherman. "Some Socioeconomic Determinants of
Expenditures for Education," Comparative Education
Review, VI (1962).
Sharkansky, Ira and R. Llofferbert. "Dimensions of State
Politics, Economics and Public Policy," American
Political Science Review, LXIII (1969).
Wildavsky, Aaron. "The Political Economy Efficiency:
Cost-Benefit Analysis, and Program Budgeting,"
Public Administration Review, 1966.
Documents
Advisory Commission on Intergovernmental Relations.
Intergovernmental Problems in Medicaid. Washington,
D. C.: U.S. Government Printing Office, 1968.
U. S. Bureau of Census. Statistical. Abstract of the
United States, 1968, 98th edition.
U. S. Congress. House Committee on Ways and Means. Hear
ings , President's Proposals for Revision in the
Social Security Program. Part I, 90th Congress,
1st session, 1967.
U. S. Department of Health, Education and Welfare. Health,
Education and Welfare Trends: 1966-67. Washington,
D. C.: U.S. Government Printing Office, 1968.
239
ax Foundation, Medicaid: State Profircims AfLer Two Years.
New York: Tax Foundation, Inc., 1968.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Project Management And The Use Of Authority: A Study Of Structure, Role, And Influence Relationships In Public And Private Organizations

PDF

Bureaucracy And The One-Party State: Politics And Administrative Transformation In Ghana

PDF

Changing Patterns In British Administration In Pakistan And India: A Study In Bureaucratic Attitudes

PDF

Program Direction Through Systems Integration: A New Frontier For Management

PDF

Some Correlates Of Friction: A Study Of Staff-Line Relations

PDF

Coordination And Control Of Computer Technology In State Government: Constraints And Strategies

PDF

Systems Theory And Organizational Change

PDF

Water Utility Performance: Some Factors Related To Water Resource Management In A Small City

PDF

Prediction In The Social Sciences: An Application Of The Delphi Techniqueto Public Personnel Administration

PDF

A Study Of Problems In The Control Of Depreciation Basis By The Government

PDF

Computerization In The Rcd-Countries (Regional Cooperation For Development Countries): Its Prospects And Problems

PDF

A Typology Of Search For The Contextual Variables Surrounding Public Administrative Policy-Making

PDF

Pay Utility In The Public Sector

PDF

Organizational Evaluation And The Government Laboratory

PDF

A Case Study Of The Impact Of Research Money Upon The Administration Of The University Of Washington

PDF

The Problem Of Highway Location And Contract Policies In The State Of California

PDF

Attacking Poverty In The Watts Area: Small Business Development Under Theeconomic Opportunity Act Of 1964

PDF

A Comparison Of The Conceptual Basis, Organization, And Planning, Programming, And Budgeting Activities Of The National And Local Defense Establishments In The United States

PDF

Attitudes On Mobility In The Public Service - The Junior College Case

PDF

Law Enforcement And The Juvenile: A Study Of Police Interaction With Delinquents

Asset Metadata

Creator Cha, Marn Jai (author)

Core Title A Study Of Process And Content Of Public Policy In American States: With Reference To The Medicaid Program

Degree Doctor of Philosophy

Degree Program Public Administration

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag OAI-PMH Harvest,Political Science, public administration

Language English

Contributor Digitized by ProQuest (provenance)

Advisor Storm, William Bruce (committee chair), [L, Joel F.] (committee member), Ofman, William V. (committee member), Siegel, Gilbert (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c18-444205

Unique identifier UC11362317

Identifier 7107705.pdf (filename),usctheses-c18-444205 (legacy record id)

Legacy Identifier 7107705

Dmrecord 444205

Document Type Dissertation

Rights Cha, Marn Jai

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA