Why do voters support corrupt politicians? This question is of vital importance if it is believed that corrupt politicians are often widely popular and that corruption has a detrimental effect on development. The answer to the question shows that democracy is not necessarily a palliative to corruption and that reducing corruption may require political changes that go far beyond the administrative reforms of the "good governance" variety.
This is not to suggest that corruption itself is popular. Indeed, we encounter a widely observed paradox: unpopular corruption and popular corrupt politicians. Despite the apparent aversion to corruption, many voters do support Silvio Berlusconi in Italy, Chart Thai in Thailand, the PRI in Mexico and the Congress Party in India. The paradox deepens when we consider that the "game" of corruption is likely to be negative-sum, thus decreasing the overall gains potentially available to the "players." Why would a majority of voters support politicians that are likely to impoverish their community, which may often include the majority of these voters themselves?
The question is not a widely discussed one. Indeed, one not infrequent ploy is to expunge it by defining democracy in such a way that excludes corruption by definition. Nevertheless, the literature yields a number of suggestions why voters might support corrupt politicians. Much of the clientelism literature attempts to explain the support for corrupt politicians by pointing to voters' desire for particularistic benefits - politicians "helping the people." At the same time, it is acknowledged that the majority of those supposedly helped by such practices are impoverished by the corrupt practices of the same politicians. A further twist is added by the claim that patronage systems serve the interest of voters whose income is low and variable. Although plausible, this claim is empirically unsubstantiated. There is little evidence that such "subsistence insurance" influences voting behavior, since it does not seem to have much practical relevance in modern patronage politics. Cultural explanations have similar weaknesses. They are unable to explain the survival of customs that impoverish precisely those who are supposed to