Bad news about candidate selection

Okonomos

Few in India know it but the affinity between India and Italy goes beyond Sonia Gandhi, not to mention Octavio Quatrocchi. The Italians, too, have been trying to cope with the problem of rampant political corruption and the need to get good, honest folk into the business of running Italy.

Now, as befits economists worth their salt, two Italians Matthias Messner and Mattias Polborn have a come up with a model that offers some more bad news. In a recent paper* they argue that not only "bad candidates are more likely to run for office than good candidates", but also that more often than not "only the least qualified candidate runs."

The starting point of their analysis is the not so unreasonable assumption that the decision to run or not depends on the financial rewards. As in India, public service as a reward in itself matters not at all.

Their model assumes two things, again not unreasonably. One, that potential candidates will incur private costs and two, that only they know these costs. So "entry decisions depend on their private cost parameter and their ranking among the other potential candidates in the eyes of the voters." Candidates differ in their competence only and it is assumed that all voters know the value embedded in this competence.

The fact that the opportunity costs for the bad candidates are less than those for the good ones has an interesting side effect: it allows, say the authors, candidates to free-ride on the bad ones. The reason is that once the good candidate declares his willingness to run, he knows that he will be elected. The reputation of the bad candidate reduces the costs of the good candidate somewhat. Not in India, though, where competence has never been critical in elections.

Another interesting insight is that "the expected quality of running candidates might actually decrease as the official remuneration increases." This has important implications for the argument that you can get better politicians if the pay for elected office is increased. But, though appealing, this is not always true because if the pay is higher all other candidates will be more willing to run.

In such an event, "the higher probability that other candidates run makes it more attractive for a competent candidate (who would choose to run if there were no one else to fill the job) to try to free ride on them, and so an increase in the remuneration might induce some competent candidates not to run." To overcome this indirect effect, it would be necessary to increase the official remuneration to a sufficiently high level. Readers will recognise the Ricardian theory of rent here.

Although there is a lot in this paper that is applicable to the Indian situation, there is a paper by a couple of Finnish economists as well that may be worth reading. P Poutvaara and T Takalo argue that if after a candidate decides, there is another selection stage, as happens in India with the distribution of tickets, the chances are high that the good candidates will be "crowded out" because a very large number of bad candidates chose to enter as their opportunity costs are lower. "The chance of a high-ability candidate to be admitted to the final election decreases and so does his incentive to enter the race and spend the campaign costs."

Another important reason for reading this paper is the rich bibliography. It shows how closely foreign economists, as opposed to their Indian counterparts, are involved in studying the political process. For example, several important political leaders in India are now on record as saying that, eventually, what matters is "winnability".

Some Indian economist should extend the arguments in this and the papers that it cites to account for this "winnability" factor. The Europeans and the Americans, for instance, would never allow candidates with criminal records to run or election.

*CEPR, NBER and Università Bocconi Working Paper Series, Paying Politicians, Working Paper 246, November 2003, IGIER (Innocenzo Gasparini Institute for Economic Research), Bocconi University Milano, Italy https://bsmedia.business-standard.comwww.igier.uni-bocconi.it