A new study finds that legislators who worked for interest groups before taking office influence the voting behavior of their colleagues when the motion under consideration is relevant to their former employer’s economic activity.


Interest groups engage in multiple types of activities to directly persuade legislators of their position. Their actions range from drafting reports containing arguments in favor or against specific motions to conducting meetings with lawmakers with the aim of instructing them how to vote. As a matter of fact, in a series of surveys conducted by S. Hix, R. Whitaker and G. Zapryanova, members of the European Parliament (MEPs) report receiving a weekly average of at least 21 meeting requests from interest groups, while 59 percent of them admitted attending at least one of those per week. As a result, 89 percent of the interviewed MEPs report receiving voting instructions directly from interest groups on specific motions.

A subtler practice that is also available to interest groups is the so-called reverse revolving doors. This practice refers to the flow of individuals who were formerly employed by interest groups into active politics, constituting a passive form of lobbying since it can potentially place industry insiders with a hidden private agenda inside democratically elected institutions. Indeed, 22 percent of all surveyed MEPs admit having been encouraged to stand as a member of the European Parliament by an interest group representative.

In our study, we investigate to what extent the voting behavior of the members of the European Parliament is influenced by the presence of fellow legislators who worked for an interest group before being elected into office.

The main empirical challenge in measuring the causal effects of reverse revolving doors on the legislative process is to obtain an appropriate measure of connection between legislators: A measure that needs to be decisive in the development of intrapersonal relationships; but that, which at the same time, cannot directly explain how legislators vote. We address this hurdle by using the seating arrangement of the members of the European Parliament, by which non-leader members of the main political groups sit in alphabetic order in the chamber. The existence of such a norm is primal for our study as lawmakers that sit next to each other during plenary sessions are more likely to interact and are found to influence each other’s views.

To identify all those legislators with previous professional experience in an interest group, we match all the résumés submitted by the MEPs upon taking office with the list of organizations registered as exhibiting interest in the European Union policymaking, known as the EU Transparency Register. We find that 28 percent of the MEPs elected between 2004 and 2019 worked at some point before taking office for an organization listed as an interest group. The positions we found these individuals were holding range from short working spells on regional NGOs to high-level consulting jobs in lobbying firms.

We further enrich our sample by linking the economic activity of the interest groups that employed MEPs with the subjects of each motion that was voted at the European Parliament at any point during our studied period. For instance, we link motions on energy policy with those interest groups we detected in the legislators’ résumés that operate in the energy sector; legislation related to capital markets with interest groups operating in the financial sector, such as banks and insurance companies, and so on and so forth. Connecting the interest groups with the subjects of voted legislation is a crucial step in the process of evaluating the impact that reverse revolving doors can have on the lawmaking process. Specifically, it allows us to disentangle the effects emanating from a legislator being influenced by colleagues who used to work for interest groups in two very different situations: when the interest groups’ economic activity is related to the motion being voted and when it is not.

The legislation enacted at the European Parliament covers a wide range of subjects: from policies affecting agriculture, transport, and other crucial industries; to regulations on consumers’ protection and the Union’s single market. For instance, a piece of legislation that we find to be relevant to the energy interest groups represented in our sample is the 2010 regulation on state aid to facilitate the closure of uncompetitive coal mines. Similarly, a regulation that was deemed to be relevant for thee interest groups operating in the financial sector is a 2010 directive on supplementary supervision of financial entities.

We find that legislators whose sitting neighbor used to work for an interest group are 2.4 percent more likely to coincide in their ballots when the voting motions are related to the interest group’s economic activity, compared to those sitting beside a legislator with no experience in an interest group. The magnitude of such effect corresponds to 57 percent of the influence exerted by those legislators in charge of drafting the motions voted upon—the so-called rapporteurs, in the European Parliament’s jargon. Contrarily, we detect no statistically significant effect of sitting next to a former interest group employee when the vote is not related to the interest group’s economic activity.

“Legislators whose sitting neighbor used to work for an interest group are 2.4 percent more likely to coincide in their ballots when the voting motions are related to the interest group’s economic activity, compared to those sitting beside a legislator with no experience in an interest group.”

However, not all legislators are affected in the same way by their colleagues’ professional experience in an interest group. We find that female MEPs, first-time elected members, and members lacking expertise on the subject being considered are the main groups driving the increase in the likelihood of casting the same ballot as former interest group members during motions related to the groups’ economic activity. Contrarily, legislators that are themselves former interest group members are not affected by the influence of their neighboring colleagues who also worked for a similar type of organization. Similarly, the influence is not found to be working across European Parliament group lines, as members sitting next to former interest group members from other political groups are not more likely to cast the same ballot.

The influence that former interest group members exert on their colleagues is found to be extremely important in high-stake votes. Legislators seating beside former interest group employees are 5 percent more likely to cast the same ballot in motions containing big public expenditure decisions, such as those on the European Union budget. The main effect of reverse revolving doors is found to be driven by national interest groups as opposed to those based in the de facto capital of the European Union, Brussels. Importantly, no differential influence is found along with the interest group’s business nature—namely, whether it is a for-profit or a not-for-profit organization.

Interestingly, not only does the subject of the motions being voted on plays a role in determining the influence exerted by former interest group members to their seating colleagues, but also whether they are deciding on amending the text under consideration or on the whole text as amended. When voting on amendments the influence is detected for motion subjects that are relevant to the interest groups’ business activity, which is no longer the case in the motion’s final vote. When casting the final vote on a piece of legislation, former interest group members are found to also influence their seating colleagues in casting their same ballot independently of the relevance of the subject to its past employer. We view these results as providing supporting evidence for our hypothesis that interest groups use reverse revolving doors to shape specific parts of the legislation voted in Parliament.

Finally, we shed light on how the legislators’ ballots are actually influenced by exploring their voting mobilization. We find that former interest group members influence their colleagues into reducing their abstention ballots when the motions are relevant to their former employers, as well as increasing their peers’ presence in parliament for all types of motions. Further results suggest, however, that legislators slowly learn from their seating neighbors’ leanings and preferences, and accordingly start accounting for them by progressively reducing the number of ballots in which they agree.

To the best of our knowledge, this is the first study providing clear and solid evidence of the distorting effects generated by reverse revolving doors on the legislative process. We consider our findings to have important implications for policymaking, as they shed light on a relatively overlooked practice used by interest groups to achieve their goals by placing their insiders in democratically elected institutions. Our results support the hypothesis that revolving doors have a meaningful impact on the legislative process also when working in reverse.