Measuring Politicians’ Performance

Measuring the performance of politicians is no easy task. Given the general belief that politicians should be selfless individuals in service of the public, some might argue that such a pursuit is unnecessary. However, even if we embraced this notion, good intentions can often be hindered by incompetence. Furthermore, rather than as selfless public servants, politicians should be viewed as self-interested agents, just like other individuals. They are driven by a wide range of incentives such as the size of their salaries, non-monetary allowances (car, secretary, housing, etc.), respect of their peers, hunger for power and fulfillment of their ideology. On top of that, salaries might be only a small fraction of their income if they are prone to accepting bribes and extracting rents. With such motivation, the pursuit of their goals might be divergent from public welfare. But since their behavior might be significantly altered by their need to stay in office – to be re-elected – finding proper indicators to measure their performance could provide the necessary the transparency that is needed in order to ensure that their personal goals are more in line with the interests of their voters.

This is a file from the Wikimedia Commons. Information from its description page there is shown below.

Accomplishing such a task is difficult, given the ambiguous nature of the public welfare itself. If the very goals are unclear, how can we measure if they were met in a satisfactory fashion? Furthermore, this problem gets more prevalent the higher up we climb the political ladder, as the goals become more and more broad and ambiguous. Meaningful measurement of Parliament members’ performance thus seems almost impossible. Nevertheless, the few following paragraphs will try to overview some of the possible ways in which such a task might be accomplished, while discussing the underlining problems and providing some empirical illustrations using the data from the Chamber of Deputies of the Parliament of the Czech Republic for the election period of 2006 -2010. The Chamber of Deputies of the Czech Republic is one of the two chambers of the Parliament, the other one being the Senate. It consists of 200 members (Deputies) that serve four-year terms and are elected using the party-list proportional representation system.

While there have been quite a few different approaches to measuring the activity and competence of politicians, their relevance is often rather dubious and they are usually met with the problem of data unavailability. One of the more commonly used indicators is the number of submitted and approved bills, which, for example, was used by Ferraz and Finnan (2009) in their research of Brazilian municipalities.[1] However, such an indicator might prove to be problematic. By simply adding the bills, we lose the ability to differentiate whether the bills are actually good and proposed in order to increase public welfare (and thus in the interest of the electorate), or whether they are poor and just create distortions (whether unintentionally or in order to further the goals of some interest groups). Nevertheless, even if this indicator might not be a good way to measure politicians’ performance in pursuit of public welfare, it could be a relatively good measure of a particular politician’s competence, since the assumption that only the more capable ones are able to see their suggested bills passed seems to be quite reasonable. Unfortunately, using such a measure is further complicated in Czech conditions, because quite often there are several names attached to the proposed bills, making it impossible to select the real author and initiator of the bill. Very often, the deputies attach their names to particular bills without actually having actively participated in the process of their creation and approval. Furthermore, it is not uncommon that the initiator uses others to propose the bill. Given these issues, using such an indicator might not provide very reliable information for the evaluation of the members of Czech parliament.

Another quite frequently used measurement is the information about participation in voting on the proposed bills. By voting, the politicians can directly represent the interest of their electorate. For example, in the United States, the independent webserver provides detailed information about the voting behavior of Congress members such as the percentage of their absences during the voting process. Similar projects can be also found in the United Kingdom and France. In the Czech Republic this information is publicly accessible directly from the website of the Chamber of Deputies of the Parliament. We can differentiate between five different situations when it comes to voting – active participation (meaning that a member –Deputy – was logged into the voting system and gave either his ‘aye’,’ nay’ or abstained), he was not logged in, or he has excused himself from attending. Examining the statistics from the Chamber of Deputies of the Czech Parliament shows less than stellar attitude of its members towards representing the interest of the electorate via voting. On average, the members actively participated in voting in only 65 % of cases, and 42% of these members had below average active participation, with the most extreme individual cases being just 1.73 %, 19.03 %, 28.13 % (of cases when those individuals actively participated). These numbers seem rather alarming, but their interpretation should be made with caution, since they might to some limited degree just capture particular strategic voting behavior.

Besides the voting behavior, another useful indicator might be active participation of the politicians in plenary sessions. Such an indicator is for example used in evaluation of the parliament in Uganda by the AFLI (African Leadership Institute)[2]. Particularly, the active participation in the discussion of issues could be represented by the number of given speeches. While the overall number of speeches might indeed indicate the grasp of particular issues by the speakers and their willingness to represent the interests of their voters, it does not allow us to differentiate between speeches of actual substance that contribute to the discussion, and those void of any relevant information and opinions. The average member of the Czech Chamber of Deputies during the observed period gave 78 speeches (with median being 51). However, 5.36 % of them did not participate in the discussion even once, 12.5 % less than ten times and 66.52 % had participation below average.

Perhaps much more relevant information, especially in the Czech Republic, could be provided by examination of the politicians’ activity during committee sessions (again used for example by ALFI), since they are the key body where the particular bills are mostly discussed and amended. In the Czech Republic, the committees are created by the Chamber of Deputies of the Parliament and their sessions are supplemented by the attendance roll. Even though the committee members are not obligated to sign these attendance rolls, it is a customary and common practice, so it should provide relatively good information about their attendance. Of course, mere attendance does not necessarily mean active participation during discussion, but free-riding should be more difficult in comparison with the plenary sessions, given the small number of members (usually 18). All deputies (besides the chairman and deputy chairmen) can be part of at most two committees (on average they are members of 1.74 committees). On average, the committee members have been absent during 28.11 % of meetings. 21 deputies have at least been absent at 50 % of the meetings and 75 for at least 30 %. The least exemplary were members of the Petition committee with the average absences being 45.47 % and the most stellar were the members of the Public health committee with 18.46 %. Looking at individual attendees gives even more disconcerting numbers, with cases of absences of over 70 % and in some cases even 100 %.

While the above discussed indicators are the more prevalent ones, there are also a number of others that might provide quite useful and relevant information. For example, the UK website uses several quantitative indicators such as participation in public debates, the number and frequency of responses to the question of voters and many others. Another interesting indicator, at least in the Czech Republic, might be the number of times the particular deputy was chosen to be media representative for discussed issues. This should provide information about both activity (with these numbers being positively correlated with the committee attendance) and competence, since we can assume that only the more competent and knowledgeable deputies are usually chosen. For example, in the examined period the average deputy was appointed five times for this post, while 21 % of them were not chosen even once.

As was already discussed above, finding good and relevant indicators might increase transparency in the political process and ensure that the interests of voters are properly represented. Unfortunately, there might be some downsides as well. Most of the known indicators are very crude and can paint only a very sketchy picture of the activity and competence of politicians. If they are sufficiently well known and accepted, using these indicators can lead to some distorting behavior. Politicians might just divert their attention to improve their standing in the particular indicators at the expense of using their energy and time on other activities that lead to increased public welfare but that are not captured by these indicators. In fact, several such occasions when politicians changed their behavior (giving a number of speeches void of content, etc.) in order to boost their rankings were noted in the past in France and the UK.[3]

In conclusion, finding meaningful measures of politicians’ activity and competence is a rather difficult task. Even though some of the indicators that are used can provide quite useful information, they are very often plagued by a number of limitations. When used separately, their usage and correct interpretation can be problematic; however, using a great number of indicators together might provide illuminating information about politicians’ behavior. For example, the statistics from the Chamber of Deputies of the Parliament paint a rather grim picture of the state of Czech politics.[4]

[1] Ferraz, Claudio, and Frederico Finan. 2009. Motivating Politicians: The Impacts of Monetary Incentives on Quality and Performance. Working Paper. National Bureau of Economic Research.

[4] The numbers presented here are just some interesting highlights, much more detailed analysis (in the Czech language) can be found here and here.

Jonas Rais