Beyond Lowi's policy typologies

This week’s readings regarding McCool’s chapter 3 (policy process) and chapter 4 (policy typologies) present very fundamental topics in the study of public policy but there is little consensus on which policy processes and policy typologies have stronger explanatory power than others, why researchers need them in policy studies, how researchers should take advantage of them and interpret them in several diverse policy arenas, and how the interaction, interrelationship, or intra-relationship among several kinds of policy actors is revealed or manipulated in the policy-making processes. Although several scholars competitively introduce various policy processes and policy typologies to improve the understanding of public policy and it is worthwhile for them to be studied, we as scholars still have the following questions: Do we really need policy processes and policy typologies to understand the policy-making process? How do we set the relationship between the substances and the processes (or contexts) of public policy? Focusing on chapter 4 dealing with policy typologies and related debates, I will make some comments regarding them.

In chapter 4, one of the main discussion points is the relationship between related variables to make “good” policy typologies. To do so, McCool introduces Lowi’s typology based on four types of policies (distributive, regulatory, redistributive, and constituent) as a starting point for discussion. As we already read in McCool’s chapter 4, the main argument of Lowi’s article is that the types of policies determine or guide the types of politics. This is a reverse argument, compared with the existing studies dealing with power relationships in policy analysis. However, Steinberger and Greenberg et al., and Spitzer attempt to revise or criticize Lowi’s basic assumption regarding the relationship between the two variables with a more realistic viewpoint. I really enjoyed the arguments made by Steinberger and Greenberg et al., and I think they can be better alternatives to understand Lowi’s policy typologies, although they each make use of a different research methodology: a qualitative and phenomenological viewpoint of Steinberger’s article versus the empirical and systematic comparative viewpoint of Greenberg et al.

Greenberg et al. and Steinberger alike argue that public policy has an inherently complex and ambiguous nature, as Lowi mentioned, but most particular policies can be, and are, coherently understood and defined in a wide variety of ways, in terms of multiplicity of aspects and participants, interaction of policy actors and their environments (Greenberg et al’s argument), and socially constructed meanings (Steinberger’s argument).That is, Lowi’s policy typologies based on the categorization of a particular policy within four types of policies should be carefully understood under a contingent and multi-dimensional consideration of several explicit as well as implicit factors, according to them. I really agree with their arguments and think they have higher validity, probability, and reliability for understanding real policy analysis. In spite of their excellent arguments, some of their argument is a little confusing. For example, the concept of “requisite actions” and “autonomous actors” in Banfield’s theory under Greenberg et al. is very interesting but it is a little confusing regarding the exact meaning and the relationship between these concepts and the validity of Lowi’s argument.

Finally, I would like to comment on the condition of a “good” theory in policy studies, as McCool mentioned at the conclusion section of chapter 4. According to McCool, the researchers under today’s policy analysis face a dilemma, such as which factors should or could be emphasized over others: “economy” or “simplicity” vs. “testability.” Which one is selected or preferred depends on the researcher’s intention and understanding regarding how policy typologies should be used and why policy typologies are needed. I think there is no answer concerning this debate but the efforts to make consensus of it should be required for a better policy analysis.