Abstract We exploit seating rules in the European Parliament to estimate causal peer effects in legislative voting. We find that sitting next to each other reduces by 13 percent the likelihood that two Members of the European Parliament (MEPs) from the same party differ in their vote. Using variation in seating across the two venues of the Parliament (Brussels and Strasbourg), we show that this effect reflects persistent peer influence: a pair of MEPs who have sat together in the past are less likely to disagree on a vote even if they do not sit together during that particular vote.