Four Theories of Public Policy Making and Fast Breeder Reactor Development

Author(s): Herbert Kitschelt

Source: International Organization , Winter, 1986, Vol. 40, No. 1 (Winter, 1986), pp. 65- 104

Published by: The MIT Press

Stable URL: https://www.jstor.org/stable/2706743

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected] Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at https://about.jstor.org/terms

, The MIT Press , Cambridge University Press and University of Wisconsin Press are collaborating with JSTOR to digitize, preserve and extend access to International Organization

This content downloaded from ������������107.182.72.224 on Sun, 13 Mar 2022 22:11:16 UTC�������������

All use subject to https://about.jstor.org/terms

Four theories of public policy making and fast breeder reactor development Herbert Kitschelt

The recent revival of the discipline of political economy challenges purely economic explanations of economic growth, technological innovation, and sectoral change. This approach recognizes that political actors, institutions, and strategies to organize the economic process together shape the economic development of industrial societies. Whereas economists have emphasized determinants of growth such as savings and investment rates, degrees of domestic and international competition in an industry, or the supply of labor, the new political economists view the political definition of property rights, the nature of state intervention in the economy, the resources of politically mobilized groups, and political actors' belief systems as critical determinants of economic transformations.' Both economists and political economists, however, share the assumption that actors are rational; they pursue their interests in a calculated manner within a given system of institutional constraints.

The commitment to rational-actor models and to a structuralist analysis of interests and institutions represents the smallest common denominator among modern political economists. Outside this conceptual core exists a wide variety of competing hypotheses, four of which appear in this article:

so-called sociological theories of policy making, political coalition theory, domestic regime structure theory, and international systems theory. Although theoretical and empirical work on these approaches has as yet been incon- clusive, recent research points to the compatibility and complementarity of different explanations, rather than a simple zero-sum competition between them.2 Single-factor theories are usually not rich enough to capture the dy-

For helpful comments on a first draft I thank Joseph Grieco and Peter Lange. 1. For a sophisticated historical reconstruction of economic and political development, see

Douglass C. North, Structure and Change in Economic History (New York: Norton, 1981). 2. In this vein Peter A. Gourevitch in "The Second Image Reversed: The International Sources

of Domestic Politics," International Organization 32 (Autumn 1978), pp. 881-91 1, links inter- national systems and domestic coalition arguments. Several authors have attempted to combine

International Organization 40, 1, Winter 1986 0020-8183 $1.50 ? 1986 by the Massachusetts Institute of Technology and the World Peace Foundation

This content downloaded from ������������107.182.72.224 on Sun, 13 Mar 2022 22:11:16 UTC�������������

All use subject to https://about.jstor.org/terms

66 International Organization

namics of complex processes of industrial transformations. Even careful research design-that is, the selection of difficult cases, and the analysis of crucial experiments in the perspective of a specific theoretical proposition- can rarely control all relevant intervening variables or provide sufficient data. Rather than resignation or an indifferent endorsement of theoretical pluralism and eclecticism, these problems should stimulate empirical investigations to

engage in more complex theoretical arguments and in a configurative analysis of public policy making. Testing the compatibility and interdependence of different theories prevents theoretical parsimony from leading to oversimplification.

In this article I will provide an example of how a complex configurative policy analysis can be constructed. The likelihood that multiple explanations of public policy will be found relevant increases if analysts employ one or any of the following four strategies in comparative analysis: survey a large number of cases; compare determinants of several different policies; measure the dependent policy variable at a high level of quantitative precision (interval scales), or at least distinguish analytical components of public policy; compare determinants of a specific ongoing policy using time series data.

Although I analyze a single policy-the development of fast breeder reactors (FBRs) in France, the United States, and West Germany -I break down the dependent policy variable into a number of analytical components. Moreover, I examine FBR policies over time to determine whether or not the causal structure of policy making remains the same.

Traditionally, comparative public policy studies, especially with respect to economic and industrial policy making, have poorly defined and concep- tualized their dependent variables.3 In the case of FBR development policy, quantitative policy measures are difficult to construct. Instead, I distinguish among four analytical aspects of policy making:

1.The social groups that mobilize around a public policy. Here I am looking for an explanation of the structural position of actors in a pol- icy arena and the relevance actors attribute to a policy issue vis-a-vis their self-definitions of "interests."

domestic structure and political coalition theories; see, for example, Peter J. Katzenstein, "Con- clusion: Domestic Structures and Strategies of Foreign Economic Policy," in Katzenstein, ed., Between Power and Plenty (Madison: University of Wisconsin Press, 1978); Francis G. Castles, ed., The Impact of Parties (Beverly Hills: Sage, 1982); John Zysman, Government, Markets, and Growth (Ithaca: Cornell University Press, 1983); and Peter A. Hall, "Patterns of Economic Policy: An Organizational Approach," in Stephen Bormstein, David Held, and Joel Krieger, eds., The State in Capitalist Europe (London: Allen & Unwin, 1984).

3. Cf., as a good critique, George D. Greenberg, Jeffrey A. Miller, Lawrence R. Mohr, and Bruce C. Vladeck, "Developing Public Policy Theory: Perspectives from Empirical Research," American Political Science Review 71 (December 1977), pp. 1532-43. Studies of economic policy making often do not clearly distinguish output variables such as tax policies and welfare expenditures from policy outcome variables such as employment, inflation, and economic growth. The problem can be seen in Manfred Schmidt, Wohlfahrtsstaatliche Politik unter bufrgerlichen und sozialdemokratischen Regierungen (Frankfurt: Campus, 1982), pp. 121-23.

This content downloaded from ������������107.182.72.224 on Sun, 13 Mar 2022 22:11:16 UTC�������������

All use subject to https://about.jstor.org/terms

Fast breeder development 67

2.The specific institutional arenas of political decision making. Here I focus on the organizational rules of selectivity which facilitate or impede the access of actors to a specific arena. It is distinguished from the broad political regime and opportunity structure in a coun- try which features "policy styles" and institutions that remain similar across policy arenas. (I will elaborate this point below.)

3.The decision-making process. In the case of FBR policy a number of subgroups are closely enough meshed to merit treatment as a single complex of variables: the use of resources and the coalitions of actors preferring specific policy options; the choice of policy instruments to pursue an objective-public incentives, regulation, state investment, and so on; the extent to which these instruments are applied. The second and third subgroups are aspects of policy "outputs."

4.The economic, social, and political impacts of policy, that is, its "out- comes." Outcomes are determined by the effectiveness and the effi- ciency with which certain results are brought about, the unintended side effects of a policy, and the legitimacy that policies enjoy.

A public policy, then, is a cluster of actors, institutions, decision-making processes, and outcomes. Obviously, a causal relationship exists among the four components of policy making. The interplay among actors, decision- making processes, and outputs logically precedes the outcome. But the precise nature of the relationship may well be contingent upon broader constraints and inducements to policy formation. For FBR development in France, West Germany, and the United States, political actors and policy arenas do not directly covary with decision-making processes and policy outcomes. Sim- ilarly, in FBR policies in the 1970s, although actors are similar across coun- tries, policy arenas, processes, and outcomes differ.

Unfortunately, much of the empirical policy literature focuses on just one component of policy making-budget allocations, or measures such as in- flation, economic growth, social unrest, for example-without reconstructing the complexities of policy formation. This narrow focus promotes single- factor explanations.

In addition, time and timing in the ongoing (re)production and transfor- mation of social systems also have received little attention.4 Most theories treat time as a continuous, linear variable. It is then possible to look at the relative timing of a country's development within the context of international systems. Within countries, domestic structure theories point to the relative inertia of institutions; hence, the past predicts future policies most successfully. But within both approaches, change in public policy, and the role of time and timing for public policy, are explained in terms of invariant causal con-

4. In social theory this issue has been critically analyzed by Anthony Giddens, Central Problems in Social Theory (Berkeley: University of California Press, 1979), especially pp. 202-4.

This content downloaded from ������������107.182.72.224 on Sun, 13 Mar 2022 22:11:16 UTC�������������

All use subject to https://about.jstor.org/terms

68 International Organization

nections between variables. Relations between variables are expected to re- main stable over time and be applicable to a broad range of contexts. This is logically and theoretically assumed when policy analysts treat cross-sectional comparisons of public policies as if they revealed longitudinal patterns of policy formation. Contrary to this assumption and inference, causal rela- tionships in policy making may themselves change over time. Time, timing, and contextual boundary conditions of public policy making may limit the generality of theories about policy formation much more than the nomological version of policy theory leads us to believe. If theoretical propositions about policy making are only valid in very limited contexts, cross-sectional and longitudinal analysis can no longer be treated as equivalent. Conversely, we can imagine that time and timing can change the nature of causal relationships that are involved in public policy making. We may also have to assign a time index to theoretical propositions about policy formation to account for rupture. Cross-sectional and longitudinal analyses of public policy may then single out different determinants of public policy.5

It is possible to link theory and empirical investigation of public policy without falling into the traps of reductionism, eclecticism, or linear time analysis. Analysis of the development of a new and extremely sophisticated, research-intensive energy technology reveals that the theoretical arguments which provide the most powerful explanations of the four components of policy making will differ for a given period. Whereas sociological policy theories and coalition theories describe FBR policy from the mid- 1960s until the economic and political watershed of 1973-74 in the three countries I compare here, domestic regime structure and international systems theory provide the stronger explanation for the period after 1974.

The fast breeder reactor provoked intense political controversy in the 1970s. As a result, the case is methodologically relevant because it dem- onstrates the significance of timing. The energy crisis of the early 1970s and the gradual politicization of energy issues by environmental movements in- troduced new "intervening" variables into FBR policy arenas. The impact of these variables on public policy in the three countries differed considerably. Among France, West Germany, and the United States, the trajectory of FBR policy, shaped by the intervening variables, moved from greater similarity in the 1960s to greater dissimilarity in the 1970s and early 1980s. An em- pirically exhaustive treatment of FBR policy within the confines of this article

5. Only a few authors have conceptualized a historically changing structure of public policy making. See, for example, Martin 0. Heisler and B. Guy Peters, "Comparing Social Policy across Levels of Government, Countries, and Time: Belgium and Sweden since 1870," in Douglas Ashford, ed., Comparing Public Policies (Beverly Hills: Sage, 1978); Peter Flora and Jens Alber, "Modernization, Democratization, and the Development of Welfare States in Western Europe," in Flora and Arnold Heidenheimer, eds., The Development of Welfare States in Europe and America (New Brunswick: Transaction, 1981); and Manfred G. Schmidt, "The Role of Parties in Shaping Macro-Economic Policy," in Castles, Impact of Parties.

This content downloaded from ������������107.182.72.224 on Sun, 13 Mar 2022 22:11:16 UTC�������������

All use subject to https://about.jstor.org/terms

Fast breeder development 69

is not possible.6 Instead, I use the empirical case to make a point about the methodology and theory of comparative policy analysis rather than to prove each empirical assertion about FBR policy.

1. Determinants of public policy making

The four strands of public policy theory I review briefly do not exhaust the range of explanatory options but define variables that are especially relevant in "most similar systems" comparisons, such as between advanced capitalist democracies that share the same level of economic development, competitive party systems, and similar structures of consciousness and culture.

Sociological policy theory

The first explanatory approach argues that the nature of policy issues in a societal context determines the nature of political actors, decision-making structures and processes, and policy outcomes. In similar societies, we expect to find similar policies toward the same issues across political systems, varied policies across issues within the same system.

Neo-Marxist public policy analysis assumes that the structures of power and the interests in the economic system determine the capacity of political groups to organize the shape of political regimes and arenas, and, finally, of policy outcomes. Neo-Marxists emphasize differences in policy formation between "state functions" such as the provision of industrial infrastructure or social policies.7 Different styles of rationality emerge in political admin- istrations, depending on the policy arena,8 and different areas of state activity correlate with different organizational structures of policy making.9 The structure and dynamics of state policies thus vary over time and across policy arenas, for example, between repressive, economic, and ideological policy concerns'0, or between production and circulation issues."I

These theories all rely on an overly simplistic image of social structure

6. For a closer investigation of FBR policies in the context of the overall energy policies of the three countries, see Herbert Kitschelt, Politik und Energie (Frankfurt: Campus, 1983), chaps. 3-5.

7. See James O'Connor, The Fiscal Crisis of the State (New York: St. Martin's, 1973); Claus Offe, " 'Krisen des Krisenmanagements': Elemente einer politischen Krisentheorie," in Martin Janicke, ed., Herrschaft und Krise (Opladen: Westdeutscher, 1973).

8. See Claus Offe, "Rationalitaitskriterien und Funktionsprobleme politisch-administrativen Handelns," Leviathan 2, 3 (1974), pp. 333-45.

9. Linking Marxist political theory to organization theory is Goran Therborn, What Does the Ruling Class Do When It Rules? (London: NLB, 1978).

10. Compare Nicos Poulantzas, Political Power and Social Classes (London: NLB, 1973), and Classes in Contemporary Capitalism (London: NLB, 1975).

11. See Gosta Esping-Anderson, Roger Friedland, and Erik Olin Wright, "Modes of Class Struggle and the Capitalist State," Kapitalistate 4/5 (1976), pp. 186-220.

This content downloaded from ������������107.182.72.224 on Sun, 13 Mar 2022 22:11:16 UTC�������������

All use subject to https://about.jstor.org/terms

70 International Organization

which takes economic class into account but treats sectoral, territorial, and

cultural differences as politically insignificant. Consequently, policy analyses often resort to ad hoc categories such as class factions, societal categories or nonclass actors, and multiclass actors in empirical investigations. Moreover, because actors do not always readily define their interests in class terms, policy structures, processes, and outcomes cannot be directly deduced from a political-economic class analysis.

A second version of policy theory shares with neo-Marxism the assumption that determinate societal "interests" shape the processes and outcomes of policy formation, but rejects class analysis as the sole foundation of such interests in favor of a more flexible, inductive approach that links the actors' perception of the costs and benefits of policy options to the nature of a given policy process. In this vein, James Q. Wilson proposes that political groups will organize and mobilize more or less vigorously depending on the perceived

costs or benefits of a policy.'2 Accordingly, governments can easily adopt policies with concentrated benefits and dispersed costs, but it is almost im- possible for them to act on policy issues with the reverse configuration. Policies with both distributed costs and benefits can easily be institutionalized, whereas policies with highly concentrated costs and benefits lead to protracted and intense conflicts with the affected actors.

Although Wilson avoids economic determinism, his approach raises some questions. Are the actors' definitions of costs and benefits grounded in and explainable in terms of the social structures that generate the decision-making problems? Or, are the perceptions themselves the final anchor of the theory?

The first alternative leads to a historically and structurally refined macro- sociological theory. By implicitly choosing the second alternative-a sub- jective, actor-oriented definition of interests and stakes-Wilson risks depriving the theory of content. Circular reasoning and ad hoc assumptions can render the theory tautological by attributing the perceptions of costs and benefits to actors after the fact, based on observed patterns of policy making.

Similar questions exist with respect to Theodore Lowi's well-known policy theory of public policy making, which distinguishes four types of political arenas.'3 In more recent work, Lowi rejects both a sociological-structural and a subjectivist definition of policy issues. Instead, he adopts a statist perspective, treating the formal, legal provisions of enacted policies as a "formal classification of the functions of the state" and of the intentions of the rulers.'4 A semantic analysis of laws then generates predictions about policy processes that are associated with specific legal patterns.

12. James Q. Wilson, Political Organization (New York: Basic, 1979), chap. 16. 13. Theodore Lowi, "American Business, Public Policy, Case Studies and Political Theory,"

World Politics 16 (July 1964), pp. 677-715, and Lowi, "Four Systems of Policy Politics and Choice," Public Administration Review 32 (July-August 1972), pp. 298-310.

14. See Theodore Lowi, "The State in Politics: An Inquiry into the Relation between Policy and Administration" (ms., Cornell University, 1982), p. 1 1. In this more recent formulation, Lowi's approach is no longer far removed from another statist policy theory that uses properties

This content downloaded from ������������107.182.72.224 on Sun, 13 Mar 2022 22:11:16 UTC�������������

All use subject to https://about.jstor.org/terms

Fast breeder development 71

Correlations between semantic structures of law and policy formation, though interesting, fall short of the expectations Lowi's original formulation raises. The statist formulation cannot explain why new policy issues that are not yet legally codified produce certain patterns of policy formation.'5 More- over, other than by reference to the rulers' volitions, the statist approach cannot explain systematic changes of legal codifications of a policy issue over time. Lowi's approach disconnects the link between social structure and policy formation. The operational content and the impact of policy on politics and society are, as Lowi himself confirms,'6 irrelevant for his approach.

Short of a statist or a subjectivist approach to policy theory,'7 I see only two avenues to a policy theory that links social structure to politics and avoids a reductionist conceptualization. An inductive approach can always test the hypothesis that, within structurally similar societies, specific policy issues are associated with similar patterns of policy making and cost-benefit perceptions by actors. Assuming this hypothesis to be true, one can work backwards to reconstruct the underlying dynamics of interest formation in a society. Second, we can deductively explain the actors' perception of costs and benefits in a given social structure and make predictions about policy patterns. Although a substantive analysis of societal cleavages-of the emer- gence of preferences or values and of changes in cognitive and normative orientations-requires a more far-ranging macrosociological foundation than space allows, I will propose three formal hypotheses about the logic of interest mobilization in modern capitalist societies which political-economic theories of collective action and resource mobilization elaborate, and which empirical studies confirm: the magnitude and distribution of material gains or losses through a policy decision determines the level and aggregation of political mobilization in conflicting and coalescing groups; actors discount the future, hence they will mobilize more vigorously in response to policies with short- term impacts than those with long-term impacts; within this logic of social mobilization of interests, Wilson's hypotheses about the ease of policy in- novation, institutionalization, and conflict aggregation are valid.

Domestic regime theory

A second theory of policy formation directly opposes issue-based and sociologically based explanations, and argues that policy patterns within

of decision processes to predict the nature of political actors, conflicts, and outcomes: Jiirg Steiner, "Decision Process and Policy Outcome: An Attempt to Conceptualize the Problem at the Cross-National Level," European Journal of Political Research 11 (September 1983), pp. 309-18.

15. In a sense Lowi has thus confirmed the criticism made by discussants of his earlier work that-contrary to the statement "policy determines politics-the "policy type is rather an ex- planandum than an explanans of public policy." See Greenberg et al., "Developing Public Policy Theory," p. 1542.

16. Lowi, "The State in Politics," p. 1 1. 17. This subjectivist turn has been advocated by Peter Steinberger, "Typologies of Public

Policy: Meaning Construction and the Policy Process," Social Science Quarterly 61 (September 1980), pp. 185-97.

This content downloaded from ������������107.182.72.224 on Sun, 13 Mar 2022 22:11:16 UTC�������������

All use subject to https://about.jstor.org/terms

72 International Organization

countries across policy arenas are more similar than those across countries within policy arenas."8 Building on the comparative study of political insti- tutions, political economists and policy analysts have updated this approach. In the most general sense, domestic regime and opportunity structures of politics are expected to shape the participation, organization, and processes in all policy arenas of a country. The specific national "policy styles" that emerge are based on complex institutional patterns that govern entire political systems.'9 Such patterns are the principles, norms, rules, and decision-making procedures that have been firmly institutionalized over time and that survive fluctuations of power and coalitions among political actors in a country.20 These domestic regimes are thus relatively impervious to sudden changes in the domestic balance of power. Regime theories explain new policy pro- cesses and outcomes, if these can be predicted on the basis of knowledge about recurring patterns of policy making which are typical across policy arenas within a country.

Political economists distinguish two not always complementary theories of domestic political structure. One focuses on the interaction between state and society, specifically, political articulation and interest aggregation. It distinguishes pluralist patterns of interest intermediation in which multiple, overlapping, decentralized interest groups that arise spontaneously vie for the attention of policy makers, from a neocorporatist pattern of more orderly, sectorally monopolistic, and comprehensive interest groups that work in policy making through firmly established channels of communication; are represented equally in decision procedures; and are attributed a semiofficial

participation status by the government in policy formation.2' In contrast to corporatist systems, pluralist systems tend to permit a broader representation of newly mobilized political actors with innovative political claims.

The other domestic regime theory is concerned with the state's capacity to impose policies and implement them consistently. It highlights variables such as the territorial and functional centralization of the executive branch, the domination of the executive over the legislature, the control of material and informational resources by the state, and the ability of policy instruments

18. This argument is developed in Douglas E. Ashford, "The Structural Analysis of Policy or Institutions Really Do Matter," in Ashford, Comparing Public Policies.

19. The analysis of national policy styles is attempted in Jeremy Richardson, ed., Policy Styles in Western Europe (London: Allen & Unwin, 1982).

20. The definition of domestic structures and international regime structures rests on similar methodological and conceptual choices. For a definition of international regimes along lines similar to the definition of domestic structures see Stephen D. Krasner, "Structural Causes and Regime Consequences: Regimes as Intervening Variables," International Organization 36 (Spring 1982), pp. 185-205.

<p