Paying Politicians for Performance

The issue of how much Ukrainian MPs should be paid has received quite some attention lately. In contrast to the Ukrainian MPs in 2009 who voted to halve their own salary, some of the new generation of MPs would like to see their salaries increased.

The argumentation in favor of higher salaries typically consists of two parts: first, it is argued that jobs that offer high salaries should attract more and better candidates, that is, there is a selection effect. The second argument goes that if you have a well-paid job, you should be willing to work hard in order to keep that well-paid job, that is, there is an incentive effect. Both these arguments lead to the conclusion that the wages of Ukrainian MPs should be increased. After all, who would be against having better MPs who work harder?

Of course, as any good economist knows, good decision making means we should not look at whether or not we can expect benefits from our decision but rather that we should compare the size of the benefits to the size of the costs[1]. Computing the amount of extra cost is simple – we multiply the wage increase by the number of MPs. Estimating the size of the benefits, that is, how much a wage increase will improve the quality and the work effort of the MPs, is much harder, however. Luckily, the experience of some other countries who changed the wages of their MPs can be helpful in this respect.

Three recent studies for example evaluate how Members of the European Parliament (MEP) reacted to a change in their wage in 2009. Before 2009, the salary of a MEP was equal to the salary a member of parliament in his/her home country parliament. For example, a Hungarian MEP earned about 10000 Euro a year, a Spanish MEP would earn about 40000 Euro, while an Italian MEP would earn more than 140000 Euro a year (in addition, they can also earn a couple of 100 euros per diem for each day they are present in the parliament). From 2009 onwards however, all MEPs earn the same, about 90000 Euro a year. So the MEPs of some countries (mainly Eastern European countries) experienced a dramatic increase in wage, while for others there was a significant decrease. How did this affect these MPs behavior?

Despite the huge change in salary (on average, the salary tripled), the impact hasn’t been impressive.

First, the impact on the kind of people that made it into the parliament was rather modest. A study by Thomas Braendle of the University of Basel showed that the wage increase did not lead to an increase in the share of MPs with higher education while a study by Raymond Fisman of Columbia University and three of his colleagues indicated that doubling the wage of the MEP makes it only 15% more likely that MEPs will have attended a top university[2]. MEPs were more interested in retaining their job, however, as a doubling the wage of an MEP made it 40% more likely that the MEP would run for re-election and made it somewhat less likely that the MEP would quit before the end of her term. The increased salary also increased the number of parties that propose candidates for the parliament.

While the selection effects were modest, the effect on the MEPs’ performance on the job were even smaller. The wage increase for example did not increase MEPs’ presence in the parliament – in fact, 2 out of 3 studies found that the wage increase somewhat increased absenteeism – as their guaranteed wage had increased, the MPs became less interested in earning per diems for being present in the parliament.

The effect of the wage increase on the activity-level of MPs while in parliament also remained limited – Raymond Fisman and his colleagues found that the wage increase did not reduce the incidence of MPs failing to vote while being in the parliament. Thomas Braendle’ study did find a positive but only small effect on the number of speeches and declarations – doubling the wage increased those by about 15% each. But a study by Naci Mocan (Louisiana State University) and Duha Altindag (Auburn University) found that the wage increase reduced the total number of written and oral questions of MEPs, and had no effect on speeches delivered, motions filed or reports written.

Of course, the main performance effect Ukrainians are hoping for is that a wage increase for MPs will decrease corruption. The only study of the effect of MP’s wages on the level of corruption is a study by Mitchell Hoffman (University of Toronto) and Elizabeth Lyons (University of California, Davis). Studying data from US state governments, they found no effect of higher salaries on the number of government officials being convicted for corruption, nor on how corrupt a state government is perceived. A recent [non-scientific] article in the Guardian further showed, using a small sample of developed countries, that there is almost no relation between the level of corruption in a country and the wage of MPs in those countries. Italy, for example, is relatively corrupt country despite the fact that wages of MPs there are extremely generous.

At the same time, there is some indirect evidence for the impact of higher salaries on corruption from studies that focus on civil servants rather than MPs. Three economists of the University of Groningen found that countries with relatively higher wages for public servants (relative to wages in the private sector) have lower levels of corruption – but also, that this was only true for relatively poor countries, with GDP per capita of up to about 9000$, like Ukraine.

And in a 2005 review on the academic literature on corruption, Jakob Svensson cites the example of Sweden which successfully got rid of reputation as one of the most corrupt countries in the 17^th and 18^th century but, among other things, increasing wages of civil servants and deregulating the economy. Svensson also concludes his review of the literature that anticorruption wage policies can only be effective if there is at the same time an effective monitoring of the civil servants.

This last conclusion points naturally to the following solution for Ukraine – Ukraine needs to have a pay-for-performance system. Pay-for-performance is not really common for MPs, and I have to admit there is no research on the effectiveness of pay-for-performance for politicians[3], but there are examples. The New York State legislature at some point introduced a “No Budget, No Pay” clause that stipulated that if the budget was not approved in time, state legislators could not receive salaries before the budget actually passed. And the California state legislature introduced a clause where the state governor and MPs could not receive a pay raise in case the state was expected to run a deficit.

Given there is a broad consensus on the burden corruption puts on Ukraine and that getting rid of corruption is of the highest priority for Ukraine, a major increase in salary of the MPs thus should be made part of, and conditional on, the passing of a comprehensive set of reforms that tackle corruption.  In this way, the pay increase for the current MPs gets explicitly linked to their performance. In addition, one might want to make the future salary increase conditional on being present during the parliamentary sessions, and even on voting-in-person. The growth dividend such reforms will bring for Ukraine will easily cover the increased cost of paying MPs.

Notes