Critiques of Kingdon’s multiple-streams framework

Nevertheless, the multiple-streams framework is not without criticism. First, it is criticized for ignoring institutional constraints on agenda-setting (Mucciaroni 1992; Zohlnhofer et al. 2016). The multiple-streams framework is grounded in the behavioral tradition of political science (Spohr 2016: 257). It conceives of politics as the ‘political climate’, referring to such things as swings in national mood and party turnover in Congress and emphasizes the fluid, changeable aspects of politics (Mucciaroni 1992: 465). It views the role of institutions almost exclusively in situational terms, which constitutes part of the changing ‘political climate’, and thereby neglecting those aspects of political life that either change

Population ageing, financing, and ETC in Asia 11 very slowly, change infrequently, or do not change (Mucciaroni 1992: 465-6). Second, the framework is criticized for being ahistorical (Mucciaroni 1992; Weir 1992). It focuses on recent trends and immediate events in the problem or political stream, which serve as catalysts for pushing items onto the agenda (Mucciaroni 1992: 471). But problems evolve over time, and ‘[i]tems that reach the agenda in any given period have their roots in the conditions, events and choices of the past’ (Mucciaroni 1992: 470). It is insufficient to simply say that the issue becomes salient when indicators signal trouble or when a crisis occurs and hence historically grounded explanations should be considered (Mucciaroni 1992: 471). Third, the framework is criticized for viewing the problem, policy, and political streams as independent streams that do not converge until windows of opportunity are open (Weir 1992; Zahariadis 2007). A historical perspective would show that the streams are linked in important ways over time (Weir 1992: 191). ‘The conception of what problems arc and how they should be defined very often depends on previous policies, which establish some groups as authoritative voices in a particular field’ (Weir 1992: 191). It means that ‘changes in one stream can trigger or reinforce changes in another, making coupling much less fortuitous and the process more purposive’ (Zahariadis 2007: 81). Hence, it is more appropriate to view streams as interdependent (Zahariadis 2007: 81).

A refined version of multiple-streams framework

The stages model of policy-making, which is also known as the ‘linear model’ (Thomas and Grindle 1990), the ‘sequential model’, or the ‘heuristic stages model’ (Kula? and Özgür 2017: 146) ‘has been used extensively to understand the policy process and its key phases’ (Dovlo et al. 2016: 216). The model ‘theorises that policy processes are rational, balanced, objective and analytical’ (Dovlo et al. 2016: 216). There are numerous variants of the models, with the number of stages varying from four to nine (Anderson 2015; Brewer and deLeon 1983; Foyle 1999; Hogwood and Gunn 1984; Ripley 1985). Tn recent work, a five-stage model of the policy process has been most commonly used’ (Howlett and Giest 2015: 17), which consists of the stages of agenda-setting, policy formulation, decision-making, policy implementation, and policy evaluation (Howlett and Giest 2015: 17).

The stages model has two main advantages. The first advantage is that it ‘facilitates the understanding of public policy-making by breaking the complexity of the process into a limited number of stages’ (Howlett and Ramcsh 1995: 12). Another advantage is that the stage model ‘is not culture bound’ (Anderson 2015: 6) and ‘can readily be used to examine policymaking in foreign political systems’ (Anderson 2015: 6) or ‘make a comparison between different countries in case of each stage’ (Kula? and Özgür 2017: 149). But the main disadvantage of the stage model is that it fails to reflect the real policy-making processes, and it is treated as a textbook approach (Nakamura 1987). In reality, ‘policy-making processes are seldom linear and stage-like’ (Auer 2007: 542) because they aremessy and highly political (Dovlo et al. 2016: 216). The stages ‘are often compressed or skipped, or followed in an order unlike that specified by the logic of applied problem-solving’ (Howlett and Ramesh 1995: 12).

The multiple-streams framework was originally developed to examine public policy agenda-setting in presidential democracy in the US. In 1991, Sabatier (1991: 151) suggested that the multiple-streams framework should be expanded to include the entire policy process. Since then, scholars have examined the subsequent stages of the policy process with some modifications of the framework. This study aims to examine the processes of health financing and LTCI reforms in the agenda-setting, formulation, decision-making, and implementation stages. While previous studies suggest that every stage comprises three streams, this study suggests another interpretation of three streams in different stages of policy-making. In the agenda-setting stage, problem recognition is the most important. The focus should be on the problem stream. In the formulation stage, analysis and identification of preferred alternatives is the most important. The focus should be on the policy stream. In the decision-making and implementation stages, there are negotiations involving multiple stakeholders and the predominance of certain ideas. The focus should be on the political stream. This study emphasizes that the problems, policy, and political streams are linked in important ways over time due to historical forces. To increase the explanatory power of the multiple-streams framework and its applicability to other political systems or other countries, the author decontextualizes the framework by enriching some important concepts in the original framework as well as bringing the theoretical lens of historical institutionalism into the framework. This study aims to examine how the combined forces of policy entrepreneurs, ideas, windows of opportunities, history, and institutions in different streams shape the processes of health financing and LTCI reforms.