Freedom of speech, deterrence, and compellence in the parliament

In most countries Parliamentary immunity protects lawmakers from civil or criminal charges while in office, and it shields them from prosecution for their political speech or political actions. This paper presents the first empirical analysis in the literature of the impact of Parliamentary immunity on the behavior and performance of politicians. Leveraging a Constitutional Amendment, the adoption of which lifted the immunity of 132 of the 550 members of the Turkish Parliament, we find that immunity from prosecution impacts how the Members of the Parliament (MPs) act and perform their duties in the Parliament. Losing immunity (and the resultant presumed fear of prosecution) pacifies the MPs of the opposition parties. They become less diligent in the Parliament (drafting fewer pieces of legislation, initiating fewer investigation inquiries, delivering fewer and shorter speeches) and become less aggressive (interrupting other MPs less frequently). They also reduce their tendency to cast dissenting votes against the government. MPs of the opposition parties who lose their immunity are less likely to get re-nominated by their parties in the next election, and they are less likely to get re-elected. We find no evidence that more outspoken and active opposition MPs or those who are more valuable for their parties have been targeted for immunity revocation. There is no evidence that the MPs, who retained immunity, have increased their Parliamentary efforts in reaction to their same-party colleagues losing immunity. We find that laws are passed faster after the Constitutional Amendment was adopted, possibly as a consequence of reduced opposition and deliberation. Using Eurobarometer surveys, we find that citizens' reactions to the revocation of MP immunity are polarized. An individual's trust in the Parliament is decreased or increased based on whether an MP from the individual's province lost immunity and if that MP subscribes to the same or opposing ideology as the individual.