WHO DECIDES? POLICY RIGHTS AND COALITION GOVERNMENT Lucy Goodhart Columbia University [email protected] February 2007 ________________________________________________________________________ I thank David Rueda for comments on an earlier version of this paper and Julien Demoulin-Smith for research assistance. All remaining errors are my own.1 ABSTRACT Who decides policy in a coalition government? The question is important to political scientists and presumably to voters. It plays a central role in a literature on government accountability and economic voting. Empirical analysis of policy-making, however, faces considerable challenges. This paper circumvents those challenges and infers macro-economic policy using the actual record of the macro-economy in a specification based on Alesina’s rational partisan theory (Alesina, 1987). The testing strategy relies upon the variation we see in party control of different cabinet ministries under coalition government. According to rational partisan theory, partisan policy preferences will only have real effects on the economy after a surprise change in the macro-economic policy-maker, as after an election. By tabulating changes in party control of different ministerial positions with changes in the real economy, therefore, we can infer which cabinet positions actually confer policy rights and are decisive in macro-economic policy. In particular, we can test whether the Minister of Finance has full power over macro-economic policy, as posited by Laver and Shepsle’s theory of ministerial discretion (1996). The analysis is implemented by extending Alesina’s model of rational partisan theory using a set of interaction terms that isolate the effect of party change in different positions and is estimated using data on government composition and ministerial appointments from Woldendorp, Keman and Budge (2000) and macro-economic data from the IMF and OECD.2 I. INTRODUCTION As Martin and Vanberg (2004) note, although great progress has been made in understanding coalition government, the majority of analysis has focused on cabinet formation and termination, leaving the question of how coalition cabinets function in office relatively unexplored. Much of the information we possess on coalition government is thus demographic in nature – a record of births and deaths – leaving the legislative activities and policy coordination of coalition governments a black box. We still do not fully understand how the policy preferences of different cabinet members are aggregated to produce government policy, the key quantity for voter welfare. The lack of attention given to policy coordination in coalition governments is connected to the opacity of the policy-making process. Coalition negotiations over policy are not directly observed so that hypotheses on the power and influence of different parties are not readily tested. Many analysts fall back on the task of attempting to infer individual party input into policy from the collective record of cabinet decisions. A cabinet’s collective decisions, however, are not simply a source of data. For cabinet members, they are a crucial resource and potential liability, as the output of coalition governments directly affects their constituents. Delegation of specific cabinet jurisdictions to different parties creates a principal-agent problem, since individual ministers have an incentive to exploit “slack” in coalition contracts to push policy in their favored direction. The most prominent theory of cabinet decision-making rejects the possibility that cabinet members from one party can ever monitor or control the actions of ministers from another. Coalition contacts, if they exist, are moot. Instead, the parties negotiating the coalition treat policy outcomes in each dimension as following exactly the preferences of the party holding the relevant ministry. Laver and Shepsle (1996) note the long tradition of ministerial discretion in cabinet government and the deference given to technical expertise within separate departments. In addition, they point out that ministers decide which issues to bring to cabinet and that individual departments have significant power over the shape of proposals that are presented in that forum. Given the obstacles to transmission of information, and the gate-keeping power of ministers, Laver and Shepsle theorize that coalition partners resign themselves to ministerial defection and grant full autonomy to individual ministers in their particular jurisdiction, a theory3 known as “ministerial discretion”.1 Since political parties under most parliamentary systems are cohesive, this implies that parties assume full rights over the policy areas attached to individual departments. The bargaining over portfolios that takes place between political parties in a coalition, therefore, is exactly equivalent to negotiating a coalition policy contract. In more recent years, scholars of coalition behavior have noted that coalition members possess mechanisms by which they can both monitor the policy choices of other coalition members and penalize them for defection from an implicit coalition contract. Thies (2001) analyzes the role of junior ministers as “watchdogs” while Martin and Vanberg (2004, 2005) show how the threat of parliamentary scrutiny and delay can discourage policy slack by coalition partners. Evidence from hiring patterns and the duration of parliamentary debates supports the claim that these mechanisms are put into practice in ways consistent with theory. Finally, narrative accounts of cabinet policy-making contain diverse examples of inter-party coordination via cabinet sub-committees, including the “concertation” procedure in Austria (Gerlich and Müller in Blondell and Müller, 1997) and the practice in Norway under Willcook’s coalition cabinet of not bringing “tricky” issues to cabinet until they had been discussed by the Prime Minister and the chairmen of the other two coalition parties (Eriksen in Blondell and Müller, 1997). The existence of monitoring and control mechanisms,