David Austen-Smith

Information and Influence: Lobbying for Agendas and Votes

This paper explores the extent and character of interest group influence on legislative policy in a model of decision making under incomplete information. A committee may propose an alternative to a given status quo under closed rule. Policies are related to consequences with ex ante uncertainty. An interest group is able to acquire policy—relevant information at a price, and has access to legislators at both the agenda setting stage and the vote stage. Lobbying is modeled as a game of strategic information transmission. The price of information is itself a private datum to the group, and legislators cannot observe whether the group elects to become informed. If the group is informed, then its information is likewise private. Among the results are: that not all informed lobbyists choose to try and influence the agenda directly; that there can coexist influential lobbying at both stages of the process; and that while informative agenda stage lobbying is genetically influential, the same is not true of voting stage lobbying.

Competitive lobbying for a legislator's vote

This paper develops a model of interest group lobbying based on the central premise that such lobbying is fundamentally an exercise in strategic information transmission. Lobbyists typically possess information that legislators do not and, inter alia, such information is relevant to legislators when it concerns the consequences — either policy or political — of supporting one bill rather than another. However, given that the interests of lobbyists do not necessarily coincide with those of legislators, the extent to which a lobbyist is able to persuade a legislator to act in his or her interest is moot. The paper explores the extent to which lobbyists can influence a legislative decision in such a setting; in particular, we are concerned with the incentives for interest groups to acquire costly information and lobby a legislator when there exist other groups that do not share the same interests. Among the results are that a legislator will on average make “better” decisions with lobbying than without, and that the more important is an issue to a special interest group, the more likely is the legislator to make the correct full-information decision.

Redistributing Income under Proportional Representation

Although majoritarian decision rules are the norm in legislatures, relatively few democracies use simple majority rule at the electoral stage, adopting instead some form of multiparty proportional representation. Moreover, aggregate data suggest that average income tax rates are higher, and distributions of posttax income flatter, in countries with proportional representation than in those with majority rule. While there are other differences between these countries, this paper explores how variations in the political system per se influence equilibrium redistributive tax rates and income distributions. A three‐party proportional representation model is developed in which taxes are determined through legislative bargaining among successful electoral parties, and the economic decision for individuals is occupational choice. Political‐economic equilibria for this model and for a two‐party, winner‐take‐all, majoritarian system are derived and compared.

Deliberation and Voting Rules

We analyze a formal model of decision-making by a deliberative committee. There is a given binary agenda. Individuals evaluate the two alternatives on both private and common interest grounds. Each individual has two sorts of private information going into committee: (a) perfect information about their personal bias and (b) noisy information about which alternative is best with respect to a (commonly held) normative criterion. Prior to a committee vote to choose an alternative, committee members engage in deliberation, modeled as a simultaneous cheap-talk game. We explore and compare equilibrium properties under majority and unanimity voting rules, paying particular attention to the character of debate (who influences who and how) and quality of the decision in each instance. On balance, majority rule induces more information sharing and fewer decision-making errors than unanimity. Furthermore, the influence and character of deliberation per se can vary more under majority rule than under unanimity.

Strategic Models of Talk in Political Decision Making

Talk in politics is effective only insofar as the speaker is able to persuade his or her audience of the relevance of some point, or the validity of some claim, contained in the speech. From a rational choice perspective, therefore, such speech making is a strategic activity in which speakers seek to influence the beliefs of decision makers. This paper is an informal introduction to rational choice models of political decision making involving the strategic use of speech.

Costly signaling and cheap talk in models of political influence

The motivation for the paper is that, insofar as agents can inflict self-imposed utility losses, purely cheap talk communication is never the only available instrument for information transmission. Given this and the importance of recent work applying cheap talk models to understand a variety of political phenomena, we explore two related questions for the theory. First, what are the equilibrium implications for cheap talk communication when an informed agent is free to use both costless and costly signals and, second, what are the consequences of allowing both sorts of signal for the widespread use of ex ante welfare comparisons as a basis for predictions on the degree of information transmission or institutional choice?

The Spatial Theory of Electoral Competition: Instability, Institutions, and Information

The author reviews the literature on the spatial theory of electoral competition, initiated by Downs. Two main lines of inquiry are distinguished. The first is concerned with the purely analytical properties of majority preference as an aggregation rule for mapping individual preferences into social preferences. And the second is devoted to providing explanations of the choices of political decisionmakers, and the consequences of these choices, within a simple plurality electoral system. These two lines are intimately related and in the review the author seeks to explore this relationship.

Individual contribution to public goods

Abstract This letter presents an attempt to provide a theoretically satisfactory rationale for expecting free-riding, in the presence of public goods, to be less extensive than the pure theory might suggest.

Sincere Voting in Models of Legislative Elections

An assumption of sincere voting for ones most preferred candidate is frequently invoked in models of electoral competition in which the elected legislature consists of more than a single candidate or party. Similarly, such an assumption is more-or-less implicit in many normative discussions of the relative merits of alternative methods for electing a representative assembly. Voters, however, have preferences over policy outcomes—which are determined by the ex post elected legislature—and not over candidates per se. This paper examines the extent to which the sincere voting assumption is legitimate in a wide class of strategic models of legislative elections. The finding is negative, and this has direct implications for the interpretation of conlusions drawn from models—formal or otherwise—which impose sincere voting as an assumption.

Party policy and campaign costs in a multi-constituency model of electoral competition

Summary and conclusionsIn this essay, a multi-constituency model of electoral competition was developed under the assumption that party candidates possessed no effective autonomy. Such a model describes the polar extreme opposite that of the more common single constituency framework, in which candidates have complete autonomy. The existence of a Nash equilibrium to the multi-constituency election game was established under the assumption that parties maximize the expected number of winning seats or behaved as pure independents. Under any other reasonable election goal (e.g., maximize probability of winning a majority of constituencies), this existence breaks down. The nature of any equilibrium was investigated in some detail and the influence of campaign costs — both statutory and non-statutory — on party behaviour analysed. This was found to be qualitatively important. In particular, differingeconomic resources between parties is sufficient to prevent convergence inpolicy space.Multi-constituency models raise a whole set of new questions for the spatial theorist. It is clear that the now highly sophisticated single constituency model is insufficient for analysing party behaviour in more general political systems with no proportional representation. This is particularly evident in view of the negative result of section 4. This result also questions the appropriateness of the Nash solution as an equilibrium concept for election games. The generally fragile nature of this solution is well-known: in the present context, though, there is another reservation. Elections take place over a finite time period and observed policies, at least in emphasis, are altered during the campaign. Since the polling day(s) are constitutionally predetermined, there seems no reason why the electionper se should occur when the parties offer equilibrium policies. And, since voter distributions change over time, we cannota priori expect parties to offer essentially identical (spatial) policies at each election. Perhaps, then, we should be looking for optimal policypaths over the electoral period and across elections.