Parliamentary speech is an important and highly visible feature of legislatures in all democracies. Time in parliament is scarce and not every member of parliament (MP) can speak on all topics they want to. Party leaders play a crucial role deciding who gets to speak on which topic in many countries; in others MPs individually request access to the parliamentary floor. As legislative speech is mostly used to send policy signals, party leaders have incentives to closely monitor who they pick for which topic – and MPs have incentives to present their potentially diverging opinions. While a number of scholars have researched the descriptive characteristics of speakers in parliament in recent years, their policy positions have usually at most been mentioned in passing in the literature on speech allocation. This paper looks into the relationship of policy differences between MPs and their party leaders, and the allocation of parliamentary speeches within parties. It also analyzes, whether institutional differences in regards to floor access matter.

We use textual data from Twitter, a social media platform popular with politicians where they can publish direct and unsolicited political statements with little party control, to estimate the positions of MPs on multiple policy dimensions using correspondence analysis. As data from social networks are very noisy, this step is preceded by exhaustive data cleaning and sorting out of non-political content using a machine learning algorithm. We then use these positions to model floor participation in parliament in multiple European democracies with different institutional settings.