In lieu of an abstract, here is a brief excerpt of the content:

CHAPTER 3 Strategic Choice and Progressive Ambition in American Politics: An Examination of Rohde's Model Introduction A common problem in political science is the usage of decision theory , implicitly or explicitly, to examine strategic interactions. Depending on the question being asked, using a decision theoretic focus may yield vastly different results than a game theoretic one will. By using decision theory instead of game theory to study political situations involving strategic interactions, scholars may ignore the importance of the strategic interactions between sets of actors. For example, much of the literature on campaign donations by political action committees has focused on the committees' choices whom to provide with contributions. Scholars who focus on who gets how much money from political action committees usually consider only the characteristics of those receiving the money. These attributes include incumbency, seniority, committee membership, chairmanships of relevant committees, and commitment to the views supported by the political action committee. On the other hand, little work has been done concerning the ways in which choices by one political action committee affect and are affected by other political action committees . The strategic interaction part of this problem has not been addressed. Are conservative committees more or less likely to provide funds to a candidate running against a competitor who may receive substantial funds from liberal groups? Do liberal committees coordinate their spending in order to insure that most if not all liberal candidates are adequately funded? By focusing on the characteristics of the candidate and not the potential actions of other political action committees, these scholars may have missed part of the analysis. Let us consider this problem more concretely. Each political action committee has a finite amount of money to divide between a finite number of candidates. For simplicity's sake assume that there are only two political action committees. One is liberal, and the other is con61 62 Games, Information, and Politics servative. There are six types of political candidates who possess combinations of two political characteristics, ideology and incumbency . These candidates are either liberal, moderate, or conservative, and they are either incumbents or challengers. How should the political action committees ration their resources? Clearly, the choice of any committee depends on what the other committee is doing. For example, if the liberal committee chooses to divide its funds evenly among liberal incumbents, the conservative committee may choose to provide token funding to conservative incumbents and use the rest of its money on moderate incumbents, moderate challengers, and conservative candidates who are challenging liberal incumbents. Given this strategy choice by the conservative committee, the liberal committee may choose to provide fewer funds to liberal incumbents and more funds to liberal challengers of conservative incumbents. Strategic considerations need to be considered in these situations.' In this chapter, our focus is on another situation in which the strategic interaction between actors has not been taken into account. Rohde (1979) considers under what conditions incumbent officeholders will strive for higher office. Briefly, his theory argues that an incumbent will only try for higher office if the expected value of running for higher office is greater than the expectation of not running for higher office. His analysis is problematic because the potential actions of other actors, including other potential challengers and the incumbent in the higher office, are not explicitly taken into account. Actions of these other actors would surely have some effect on the incumbent's decision to pursue the higher office. We focus on the problems that arise from modeling strategic situations using decision theory instead of game theory. In doing so, the differences between decision theory and game theory are noted. In addition, we explore the case studied by Rohde in order to illustrate why one should use game theory instead of decision theory to study a situation involving strategic interaction. This chapter proceeds as follows. First, a brief introduction to decision theory is provided in which we contrast game theory with decision theory. Second, a more complete discussion of Rohde's model of progressive ambition is presented. In the discussion of his 1. Of course, one may object that this situation becomes too simplified by examining only two political action committees. Why is doing this to capture strategic interaction necessarily better than assuming n actors, treating n - i as the environment for e and focusing on i's decisions? The answer to this question is "it depends." It depends on the question you are asking. The question, of course, frames the manner in which...

Share