If you want better politicians, pay them less

By Kate Ancell     
December 09, 2013

From: Magazine

Take a poll about whether politicians should get a raise, and the result is likely to be a resounding, “No.” Now there’s empirical research to support that response.

According to a study by Professor Emir Kamenica, along with colleagues from Columbia University, Princeton University, and DONG Energy, raising politicians’ salaries is unlikely to result in better governance. In fact, it may accomplish the opposite.

Perhaps in theory, and in a rational market, it would make sense to pay politicians more. Doing so would generate increased competition between candidates, attract better applicants, and secure better politicians. But politicians—not market forces—are responsible for setting their own salary levels, one reason why the political labor market veers toward irrationality.

The researchers consider the effects of a salary increase on some members of the European Parliament (MEPs). In 2009, the European Parliament implemented a flat-salary payment system, and all MEPs began to earn identical annual salaries of €90,000, equal the researchers note, “to 38.5% of a salary of a judge at the European Court of Justice.” This new payment methodology righted a striking imbalance: under the former system, which started in 1979, when the European Parliament held its first general election, each members’ home country paid him or her the same salary paid to members of the home country’s lower chamber of parliament. For instance, Italian representatives received €144,084 per year, while Hungarian representatives earned a mere €10,080.

The researchers analyze the change in composition and behavior of MEPs who received a higher salary as a result of the parliament’s new payment system. They find that a salary increase led to politicians with less education. Doubling an MEP’s salary decreased by 15% the chance that an MEP had attended a college ranked in the top 500 in the world. The researchers equate graduates of lower-ranked colleges with lower-quality legislators.

Why did a salary hike have the opposite effect intended? On one hand, that doubling increased the “logarithm of the number of parties that field a candidate by 41% of a standard deviation,” thereby providing the voters with a broader choice of political platforms, the researchers write. It also reduced the likelihood that a candidate would quit before the end of the term.

But on the other hand, the raise highlighted a divide—some politicians, namely those from lower-ranked schools, were more influenced by money. The higher salary increased the likelihood that an MEP in this group would run for reelection. But the increase didn’t similarly influence the reelection decisions of higher-quality MEPs, who had better job prospects outside of politics. As a result, a raise gave lower-quality MEPs a greater incentive to get into office and, once there, to stay put.

Maybe a salary increase would at least inspire politicians to work harder? Unfortunately, the increase apparently did little to improve how much effort politicians put into their jobs. The researchers constructed a variable to define shirking—noting a member’s voting record as well as how often he or she signed the daily attendance register but then left without attending the legislative session. They conclude that salary had an insignificant impact on both shirking and attendance.

Work cited

Raymond Fisman, Nikolaj A. Harmon, Emir Kamenica, and Inger Munk, “Labor Supply of Politicians,” Working paper, July 2013.