Reverse Revolving Doors: The Influence of Interest Groups on Legislative Voting

Lawmakers' main responsibility is to draft and pass legislation. Despite that, they might or might not have expertise on the subject under discussion relying on different sources of information and influences when deciding whether to support a piece of legislation. In this paper, we examine one particular type of influence: their peers' professional experience in an Interest Group. Using 15 years of recorded votes in the European Parliament, we show how Interest Groups, through their former employees turned politicians, play a role in shaping legislators' voting behaviour. This paper contributes to the literature by reconciling two long-standing areas of study within social science, the one on the voting behaviour determinants, and the one on lobbying. We rely on the alphabetic seating in the European Parliament as a quasi-exogenous process of network formation, identifying those legislators who used to work for an organization registered as having interest in the European Union policymaking. We utilize this type of network to quantify the impact of Interest Groups in the legislative voting process, an aspect that has generally been difficult to isolate in this literature. A key feature of our analysis is that we distinguish each voted procedure by whether its subjects are relevant for the Interest Group for which the politician used to work, as a way to determine situations in which the Interest Group's influence is more likely to be exerted. We use a dyadic surname adjacency measure to estimate the existence of significant diferences in voting behaviour across pairs of Members of the European Parliament (MEPs). We identify pairs of MEPs with and without professional experience in an Interest Group and compare their voting congruence. In addition, we examine rich data comprising biographical details of the MEPs allowing us to analyse the persistence of the effect conditioning on the time since the politician left the Interest Group. Furthermore, we identify the sector of activity and type of organization carrying out this practice. We also extend our analysis considering only politicians who used to work for an Interest Group in a managerial position and with significant tenure in the organization. Lastly, we perform a series of placebo tests to first support our identification strategy by using a sub-sample of politicians that do not sit in alphabetical order. Our analysis results will shed light on how legislators with professional background in an Interest Group affect their peers' voting behaviour. It will provide evidence to the current debates on the legislators' independence and on the implementation of cooling-off period policies to those legislators entering politics directly from interest groups.