A Review of Milan Vaishnav’s When Crime Pays: Money and Muscle in Indian Politics
India apparently enjoys the dubious distinction as the world’s most criminal democracy: a third of the members of its national lower house of representatives are currently under indictment for wrongdoing. And the crimes in question are typically not trivial, as Milan Vaishnav’s 2017 book, When Crime Pays, documents in vivid detail. Murder, arson, rape, and other violent transgressions figure prominently among the accusations against hundreds and hundreds of national and state level elected politicians. The obvious questions are how this could come to be, how to understand the implications for democratic politics, and what to do about it.
Vaishnav tackles these questions in a well-written, highly accessible study based on the assembly and analysis of a large original dataset, close readings of secondary and primary documents, and extensive first-hand observation. The dataset, which provides a window into the individual correlates of political criminality, draws on the sworn affidavits of all candidates to India’s state and national legislative elections after a 2003 Supreme Court ruling. In these affidavits, candidates must disclose past and present criminal proceedings, as well as their financial assets. Such mandated disclosures, especially those aimed at financial transparency, have become increasingly common around the world (Djankov 2010), and they provide substantially better information about candidate characteristics than previously available. To a quantitative analysis of affidavit information reported by the more than 60,000 candidates running in 35 state assembly and three national elections, Vaishnav adds a wealth of individual detail from stories collected on the ground. The insights that these details offer — insights about politicians, political campaigns, and voters — give When Crime Pays a rich texture and human interest that makes the book suitable for use in undergraduate teaching and also an effective tool of policy engagement.
The core of the book considers why criminals have an incentive to become politicians, why political parties select them as candidates, and why voters support them. On the first, Vaishnav traces the use by political parties, especially the Congress Party, of thugs and criminals in the 1950s and 1960s and the subsequent claims to political office by these very same thugs and criminals. This is a fascinating story that I had not previously understood. The shift of criminals from hired hands of political parties to autonomous political agents, Vaishnav contends, was driven by multiple factors, among which the erosion of Congress political hegemony in the 1970s and the resulting inability of Congress to provide ongoing political protection to the criminal elements it employed. As Congress dominance weakened, criminals struck out on their own as politicians in order to evade potential prosecution. They thereby became even more enmeshed in the political system, and increasingly reliant on the protections it provided as well as eager recipients of the status and prestige public office accorded.
But money is not the sole resource of criminal politicians. Criminals, Vaishnav writes, enjoy “an edge in elections extending above and beyond their financial prowess” (p. 160). The theoretical heart of When Crime Pays comes down to untangling this electoral edge. It is, he argues, genuine rather than merely a result of voter ignorance or misinformation. Voters affirmatively select criminal politicians because they offer group protection in settings characterized by ethnic conflict and weak rule of law. In such environments, where state resources, individual security, and public order are all provided at levels that are inadequate and in ways that are discretionary and unreliable, vulnerable voters turn to thugs who offer protection. Criminality serves as a candidate’s “signal [of] his ability to protect the interests of his community” (p. 172).
The empirical evidence presented in support of this claim is indirect and, to be entirely honest, somewhat slim. It appears to consist essentially of three elements. First, Vaishnav reports anecdotes in which criminal candidates boast about their wrongdoing rather than apologizing for it, thus illustrating the idea that candidate criminality might serve as a positive attribute for some voters. Second, he shows that electoral constituencies that are reserved for candidates who are members of scheduled castes or tribes have significantly lower rates of criminal candidacies, and he infers from this that criminality is an electoral asset only in contexts where ethnic conflict may be salient; i.e. non-reserved constituencies. Third, he reports results from a survey conducted across India prior to the 2014 elections in which some respondents (26 percent) stated that they would vote for a candidate who delivered benefits even if he were facing serious criminal charges and in which even more respondents (46 percent) stated that it was important to them that a co-ethnic win the seat in their constituency (pp. 233–35).
All of these types of evidence are pertinent and informative. But none is entirely well-suited to demonstrate a causal link between candidate criminality and voter choice. Anecdotes are, obviously, only anecdotes. They illustrate, and they do so nicely — but it is impossible to know how general the phenomenon is; to his credit, Vaishnav also reports instances in which supporters of criminal candidates apologize for or excuse their wrongdoing (e.g. p. 193), thus documenting that criminality provides a mixed blessing. Moreover, none of the stories he recounts appears explicitly tied to issues of caste electoral mobilization, so they do not get to the heart of Vaishnav’s theory. The evidence that criminals contest more often in regular rather than reserved constituencies is very strong, but the interpretation offered is only one among a number of those possible; there are many ways in which the two types of constituencies may differ from each other, and above all many ways in which the processes of candidate selection differ. Without more detailed, systematic information about how candidate selection operates in the two types of constituencies, it’s difficult to know whether Vaishnav’s interpretation holds water. Finally, although Vaishnav states that survey respondents who prefer their own co-ethnics be elected are also more likely to support criminal candidates who provide benefits, the proportion of respondents who endorse both views are not provided; in addition, it’s difficult to make causal inferences on the basis of simple survey data.
Where the book excels is in delineating the policy relevance of the political involvement of so many accused criminals. Vaishnav devotes two chapters to discussing the implications of the phenomenon he studies for democratic accountability in India and for possible policy solutions. As he rightly notes, there is no magic bullet to clean up the Indian political arena, but there are nonetheless numerous reforms that might cumulatively erect higher barriers to political entry by persons accused of severe criminal activities. One he only touches on is the speed of the judicial process in India. One reason the accused get elected and then potentially even reelected maybe because they can rightly contend that they have not been convicted of any crimes. It takes years and years for cases to wind their way through the courts, thereby facilitating the ongoing political involvement of the accused. Investing in the judiciary, offering a fast track to the judgement of legislative candidates, and then barring those convicted of serving in public office could have a large impact.
When Crime Pays thus offers an intriguing idea. It may well be that some voters actually embrace criminal candidates precisely because of their tough-guy Robin Hood demeanour. What kind of evidence might conclusively demonstrate this argument to a social scientist? One strategy would be to study whether criminal politicians actually deliver more benefits to their co-ethnic constituents than comparable non-criminals. Thus far, evidence relevant to this suggests the reverse may be the case. Indeed, as Vaishnav notes, one recent publication documents that constituencies with criminal representatives are marked by substantial reductions in consumption by society’s most vulnerable (p. 254) and another still unpublished paper reports that constituencies with criminal representatives experience lower economic growth than other constituencies. This evidence accords with other investigations that show that criminal politicians show up less often in parliament than other legislators; rather than protecting their co-ethnics, many criminals seem to use their time in office to further their own interests. But these scattered pieces of evidence are incomplete and indirect; the main question that Vaishnav raises is important and still awaits well-designed field research.
Before concluding, I want to ask two questions that Vaishnav’s fascinating study raised in my mind. The first regards the empirical generalizability of the Indian case. There are many countries that are marked by high levels of ethnic conflict and weak rule of law — including others in South Asia and most of sub-Saharan Africa — but that do not appear to have generated as criminal a political class as India’s. India, in fact, seems extreme in the degree of direct political access gained by indicted criminals. Why is this? What has made Indian political institutions so vulnerable to penetration by accused criminals? Vaishnav sets his study in comparative context, discussing a range of other countries in which criminal elements have also penetrated politics. But from another angle, India looks almost unique, and we really do not understand how it became so saddled with criminal politicians.
The second question is theoretical and concerns how to understand the decisions by voters to endorse known criminals. This requires theoretical work that would necessarily be undertaken by scholars with other types of expertise: those who study the political psychology of voting, on the one hand, and those who write formal models of strategic interactions, on the other. Vaishnav’s book introduces an important theoretical problem that transcends its country confines, and that ought to inspire complementary investigations by others.
Miriam A. Golden is Professor of Political Science at the University of California at Los Angeles and the author of Corruption: What Everyone Needs to Know (Oxford University Press, 2017), written with economist Ray Fisman. During the 2018–19 academic year, she will be a Fellow at the Center for the Advanced Study of the Behavioral Sciences at Stanford University, where much of her time will be spent writing a study of political selection and economic development entitled Bad Government.