One of the most distinctive developments in India’s post-independence political evolution has been the growing nexus between crime and politics.1 Indeed, a statistic often quoted in reference to India’s present political economy is that as many as one-third of elected Members of Parliament (MPs) face ongoing criminal cases. While many of these cases involve minor charges, one in five MPs face at least one case involving potentially serious infractions, ranging from murder to physical assault.2

The group of lawmakers (and alleged lawbreakers) include figures such as Pappu Yadav, a notorious bahubali who once bragged that there was no jail in his home state of Bihar whose insides he was not familiar with; Rajan Vichare, the Shiv Sena strongman hailing from Thane in Maharashtra, who faces a dozen cases winding their way through the courts; and the Congress Party’s Adhir Ranjan Chowdhury, the self-styled ‘Robin Hood of Baharampur’ in West Bengal. Despite the anti-corruption rhetoric of Bharatiya Janata Party (BJP) Prime Minister Narendra Modi, eight legislators in his first cabinet were connected to serious criminal cases. Indeed, the prevalence of criminal taint is widespread, touching all parties and reaching all corners of the country.

At first glance, the affinity between crime and politics paints a mental picture of democracy being subverted by rogue actors who regularly employ coercion or violence and whose actions defy the popular will. And coercion, to be sure, is very much a part of many criminal politicians’ standard repertoires. Indeed, a candidate’s coercive reputation can be viewed as an asset insofar as it helps to weaken or counterbalance political opposition from rival groups through (actual or threatened) violence and intimidation.

But criminality cannot be reduced to coercion or the perpetration of violence alone. Especially since the mid-1990s, when the ECI (Election Commission of India) came into its own as a powerful ‘referee’ institution, booth capturing and brazen intimidation at the polling booth have precipitously declined.3 Operating in environments where there are serious governance gaps, criminal politicians also rely on non-coercive mechanisms – such as redistribution, social insurance, and dispute resolution – to win over voters.

Therefore, while scholars must acknowledge that coercion is typically part of the modus operandi of serious criminal politicians, they must also devote sufficient attention to understanding how ‘tainted’ politicians do not simply survive, but thrive in some of the most competitive elections in the world. Like all politicians in India, those with criminal reputations also live and die at the ballot box. And, arguably, elections today are freer and fairer than they have been at other points in India’s post-independence history.

To that end, I argue that it is useful to view the relative success of criminal politicians as a byproduct of democratic practice, rather than its authoritarian antithesis. Indeed, under certain conditions, malfeasant politicians and democratic accountability can be compatible. Granted, this form of accountability is often partial, imperfect, or counterintuitive. Yet, it offers a corrective to the traditional narrative that the very existence of a large number of corrupt or criminal legislators is symptomatic of a breakdown in the standard chain of democratic accountability.

It is helpful to think about politicians, of all types, operating within an electoral marketplace.4 In elections, there are buyers (voters) and sellers (parties and politicians). As with any market, there are both supply and demand factors at work that allow the market to survive. The idea of supply can be disaggregated into two components: the decision of individuals to step forward as candidates, and the selection of candidates by political parties.

In India, the initial supply of criminality resulted from strategic decisions taken by criminal entrepreneurs who had long been active in the political sphere. Indeed, the historical record dating back to the early days of the republic is replete with evidence of politicians conniving with ‘antisocial’ or ‘lumpen’ elements, who functioned akin to free agents, in the early post-independence era. Indeed, even in the first general election of 1951-2, the ECI found scattered evidence of Congress Party politicians contracting with local strongmen to influence elections by coercing opponents, mobilizing supporters, distributing clientelistic handouts, and, in some extreme cases, stuffing ballot boxes.5 In providing assistance to politicians, criminals occasionally engaged in illegal activity, but they also used their social capital and local standing to bridge the divide between state and society in ways that political parties increasingly struggled to do, especially as their organizational foundations atrophied.

Over time, a number of trends pulled criminals into a more direct role in electoral politics; these included the gradual erosion of the Congress Party’s hegemonic status, rising social demands often expressed through identity politics, the deterioration of public sector institutions, and the collapse of the prevailing system of financing elections. However, criminals were also pushed into politics by their ultimate desire for self-preservation. As political competition intensified and the party system grew increasingly fragmented, criminals allied with the Congress could no longer take their political patrons’ re-election for granted. This uncertainty created huge new risks for criminals: without secure political protection, they would be subject to retribution, either at the hands of the state or from their political rivals.

The obvious solution to this dilemma was for criminals to take matters into their own hands and join electoral politics. Much like private firms seeking to ‘vertically integrate’ their operations, criminals decided to cut out the politician middleman in order to maximize control over their own survival and protection by becoming politicians.6 For example, a gangster such as Suryadeo Singh – a huge player in the Dhanbad coal mafia – parlayed his rising clout into a successful career in politics, contesting and winning assembly elections in 1977 and, eventually, a seat in the Lok Sabha in 1984.

This transition played out in the 1970s and 1980s; by the 1990s, the predominance of elected representatives facing criminal charges, including those of a serious nature, had been locked in. For instance, an estimated eight per cent of Uttar Pradesh Members of the Legislative Assembly (MLAs) elected in 1984 assumed office while under criminal scrutiny. In 2012, that share stood at 45 per cent.

The supply of criminally linked politicians, however, cannot be reduced to a story of individual incentives alone. In nearly all democracies, parties are the key gatekeepers that decide which politicians eventually get a spot on the ballot. Candidates always have the option of running as independents, but the success of those choosing this option is rather limited, in India and elsewhere. Indeed, according to ECI data, of the 4,300 MPs elected between 1977 and 2014, a mere 1.4 per cent (or 82 in total) lacked a formal party affiliation.

Why parties recruit candidates linked to wrongdoing is not obvious, but parties’ underlying quest for ‘rents’ suggests one possible explanation.7 In settings where elections are costly yet parties are weakly organized, seeking out self-financing candidates is an attractive option. Candidates with deep pockets can contribute financial rents in numerous ways; in addition to plugging election funding gaps, they can directly provide financial payments to party leaders or engage in other kinds of rent-seeking behaviour that could ultimately benefit the party. Indeed, qualitative evidence suggests, and quantitative analyses confirm, that money is a prime motivation for why parties are willing to embrace ‘muscle’.8 Candidates with greater financial resources (read: spending capacity) are much more likely to win elections. And candidates linked with criminality, in turn, often have better access to resources. Data from the last three general elections reveal that the richest quintile of candidates (the wealthiest 20 per cent in terms of personal financial assets) were more than twenty times as likely to win elections when compared to candidates in the poorest quintile.

What remains a mystery, however, is why voters might support candidates with serious criminal reputations. Scholars who have studied the connection between malfeasance and democratic politics have often assumed that voters do not willingly support so-called ‘bad politicians’ – shorthand for politicians involved in illegal or unethical behaviour – but instead do so inadvertently due to a lack of information.9

In a poor country such as India, where widespread illiteracy and uneven media access are a reality, it is certainly plausible that many voters make their decisions on election day with a heavy dose of ignorance about the qualifications of their local candidates. If this were the case, the election of suspected criminal candidates would be incidental, rather than a genuine expression of latent demand.

It is also conceivable that so-called ‘low information’ voters could be more susceptible to coercive threats by strongmen politicians. As one recent paper on India’s criminal politicians suggests, ‘literate voters are endowed with a whole panoply of resources, including access to information and access to police protection, that together offers them greater resilience in the face of electoral violence.’10

Such arguments often portray voters as unwitting pawns in a game rigged by political elites. As such, they do not leave much space for ‘affirmative’ sources of support for criminal candidates. However, there is an alternative explanation that suggests voters can have an underlying strategic logic for supporting criminal candidates when two conditions are present.

In places where the rule of law is weak and social divisions are rampant, politicians can use their criminality as a signal of their enhanced capacity and willingness to do whatever it takes to protect their supporters’ interests. This ‘protection’ typically involves substituting for a state administration that is unable (or unwilling) to effectively and impartially fulfil its basic functions, such as guaranteeing public security, resolving disputes and providing core public services.

The constituent ‘interests’ politician’s pledge to protect are often cast in terms of preserving the status of the ethnic community (or communities) in question, which further allows a candidate to spin his criminal reputation as doing whatever is necessary to ‘defend’ his allies. According to this logic, information about a candidate’s criminality is not only well known to many voters, but it is also intrinsic to their voting behaviour.

Many politicians therefore wear their outlaw bona fides as a badge of honour. Examples abound, but consider a strongman politician such as Brij Bhushan Sharan Singh of Uttar Pradesh. Sharan Singh, a one-time student activist turned Thakur caste leader, at one point had racked up as many as forty criminal cases, ranging from murder to kidnapping and even abetting terrorism. The defiant neta once claimed his controversial actions are done with the intent of helping the helpless, remarking: ‘I have never troubled the poor and the decent and I have never shied away from taking on those who oppress others.’ Once, when launching a protest over the inaction of the local police, Sharan Singh was asked about his tendency to create a ruckus for political gain: ‘I am a mafia man. How can I be Gandhi? I have forty cases lodged against me,’ the MP stated. ‘How can I be any good? I try to do my best. This is the public. I want to be of some help to them.’ The voters, for their part, have been quite helpful in return; in 2014, Sharan Singh rang in his fourth term as an MP.

The existence of a marketplace for criminal politicians does not mean that one should gloss over the corrosive effects the fusion of crime and politics can have in democratic systems. To the contrary, the symbiosis between crime and politics should be a cause for concern on several counts.11 First, the fact that many of India’s leading lawmakers are also among its foremost (alleged) lawbreakers has serious implications for the sanctity of the rule of law. If citizens believe that laws can be openly flouted without repercussion, these perceptions can create a self-fulfilling reality, thereby encouraging widespread disrespect for the rule of law.

Second, one recent study finds that constituencies with an elected representative who has an ongoing criminal case experience significantly lower rates of economic growth. This negative effect is even larger when the incumbent faces serious cases, cases involving financial crimes, or hails from states with endemic corruption and weak institutions. The authors of the study surmise that this is a byproduct of lower investment in public goods provision.12

Third, when it comes to social welfare, it is true that there is some segment of the electorate that stands to gain from the presence of criminal politicians – namely, their core supporters. Yet these improvements are not likely to be evenly distributed; there may be gains for certain segments of society, but politicians who tout their criminal reputations thrive by exploiting – not resolving – social divisions. Their modus operandi relies principally on catering to narrow segments of the electorate, rather than the population at large.13

But even for their supporters, criminal politicians arguably have a mixed impact. On the one hand, candidates known for flouting the law regularly campaign on pledges of delivering benefits of various sorts to their adherents. While these can be material in nature, they also involve things like physical protection or dispute resolution and justice. But while politicians suspected of wrongdoing appear to plug a governance vacuum, they also benefit from the persistence of a broken public sector machinery. If the state were able to consistently and effectively carry out its core duties, these candidates would certainly lose much of their relevance. This way of functioning finds an analogy in the operations of mafia groups in other parts of the world.

In his landmark work on the Sicilian Mafia, sociologist Diego Gambetta argues that a lack of social trust in Italy first gave rise to the mafia. What mafia members could offer – to all parties involved – was the credibility that a transaction would be honoured, thus overcoming the social trust deficit. But, as Gambetta explains, the ‘Mafioso himself has an interest in making regulated injections of distrust into the market to increase the demand’ for the very protection he offers.14 After all, if citizens were able to organically develop bonds of trust among themselves, the mafia would be out of business.

From the standpoint of democracy, the intimacy of crime and democratic politics confounds many commonly accepted notions. After all, many of the leading lights from political science and political philosophy who have written eloquently about democracy over the past several centuries suggest that what distinguishes democracy from autocracy is that the former engenders accountability because elections offer voters the opportunity – at regular intervals – to punish lawmakers who betray the public trust. This mechanism is at the heart of the democratic project and is thought to provide the carrot and the stick by which ‘bad’ politicians can be weeded out and ‘good’ politicians ushered in.

Later, scholars came to the realization that democracy did not always work as advertised. On the contrary, they postulated that democracy could facilitate accountability only when voters participating in democratic processes had access to information about the behaviour of their representatives. Armed with this information, voters could then make rational decisions about whether to re-elect or reject their incumbent politician, thus facilitating true accountability (Figure 1).

Based on this reasoning, it is no surprise that most studies of wayward lawmakers contain an assumption that the political success of bad politicians in democratic countries is inherently symptomatic of a breakdown in the chain of democratic accountability. This is taken as a given; an alternative scenario in which bad politicians and accountability might coexist is typically not envisioned.

In a few instances, some scholars actually begin with this premise and then work in reverse. Once the accountability failure is identified, scholars work backwards to identify the primary cause. In many instances, information is often believed to be at the heart of this breakdown – the missing ingredient which, when absent, creates the space for bad politicians to creep in under the cover of darkness. This process of reverse induction (Figure 2) is more likely subliminal than intentional.

Whichever causal logic is pursued – that depicted in Figure 1 or Figure 2 – the ignorant voter thesis is a compelling narrative and has been shown to be true in a great many cases. But it is not a universal truth. Assuming away the possibility that connections to illegal behaviour can sometimes be an asset, rather than a liability, for politicians in consolidated democracies can be deeply problematic.

This turns the standard logic on its head: well informed voters may, in fact, have good reasons to lend their support to bad politicians, thus creating a victory for democratic accountability (Figure 3). Granted, the ensuing form of democratic accountability is partial, imperfect, and flies against many common normative conceptions. Furthermore, it is not without its adverse side effects. Yet, it offers a corrective to the narrative that voters, and parties, are saddled with malfeasant politicians. To the contrary, they might have rational, affirmative reasons to give them support.

This article was originally published in Seminar.


1. This article is based on edited excerpts from the author’s book, When Crime Pays: Money and Muscle in Indian Politics. HarperCollins India, New Delhi, 2017. The author is grateful to Rebecca Brown for research assistance and Kanchan Chandra for her comments.

2. Data compiled by the Association for Democratic Reforms; for more information, see

3. E. Sridharan and Milan Vaishnav, ‘Election Commission of India’, in Devesh Kapur, Pratap Bhanu Mehta and Milan Vaishnav (eds.), Rethinking Public Institutions in India. Oxford University Press, New Delhi, 2017, pp. 415-461.

4. Elections and markets are, of course, not completely analogous. For instance, the electoral market is shaped by social realities and imperfections that prevent it from consistently operating as a ‘free market’ in the purest sense. These caveats aside, there is a utility to employing the market analogy.

5. Election Commission of India, Report on the First General Elections in India, 1951-52. Election Commission of India, New Delhi, 1955, (accessed 5 July 2013).

6. ‘Vertical integration’, a concept popularized by the Nobel economist Oliver Williamson, refers to the ‘substitution of internal organization for market exchange’, or the merging together of two businesses that are at different stages of production. A classic example is an auto manufacturer that decides to make its own tyres in-house rather than contract with a third party to manufacture them. See Oliver E. Williamson, ‘The Vertical Integration of Production: Market Failure Considerations’, American Economic Review 61(2), May 1971, pp. 112-23.

7. ‘Rents’ are typically defined as an economic payoff where the ultimate beneficiary incurs no costs in producing that payoff. When economists think of rents, they typically think of illicit acts of corruption. But a more expansive definition could include funds candidates provide to cover the expenses of contesting elections or contribute to party coffers.

8. For a qualitative exploration of the link between money and muscle, see Christophe Jaffrelot, ‘Indian Democracy: The Rule of Law on Trial’, India Review 1(1), 2002, p. 98; James Manor, ‘Changing State, Changing Society in India’, South Asia: Journal of South Asian Studies 25(2), 2002, pp. 231-56. For a recent quantitative study, see Bhaskar Dutta and Poonam Gupta, ‘How Indian Voters Respond to Candidates with Criminal Charges: Evidence from the 2009 Lok Sabha Elections’, Economic and Political Weekly 49(4), 25 January 2014, pp. 43-51.

9. Alicia Adsera, Carles Boix and Mark Payne, ‘Are You Being Served? Political Accountability and Quality of Government’, Journal of Law, Economics, and Organization 19(2), 2003, pp. 445-490; Claudio Ferraz and Federico Finan, ‘Exposing Corrupt Politicians: The Effects of Brazil’s Publicly Released Audits on Electoral Outcomes’, Quarterly Journal of Economics 123(2), 2008, pp. 703-45; and Eric C.C. Chang, Miriam A. Golden and Seth J. Hill, ‘Legislative Malfeasance and Political Accountability’, World Politics 62(2), April 2010, pp. 177-220.

10. Toke Aidt, Miriam Golden, and Devesh Tewari, ‘Criminal Candidate Selection for the Indian National Legislature’, unpublished paper, Department of Political Science, University of California, Los Angeles, November 2015, single-post/2015/11/11/Criminal-Candidate-Selection-for-the-Indian-National-Legislature (accessed 18 January 2017).

11. Indeed, numerous official documents, and various committees and commissions reveal that the Government of India itself is very concerned by the interaction of crime and politics. Indeed, the government has repeatedly identified the need to rid politics of criminals as a key public policy objective. The most well known of these reports is that of the Vohra Committee. See Ministry of Home Affairs, Government of India, Final Report of the Vohra Committee. Government of India, New Delhi, 2005, 20REPORT_0.pdf (accessed 20 April 2016).

12. Nishith Prakash, Marc Rockmore and Yogesh Uppal, ‘Do Criminally Accused Politicians Affect Economic Outcomes? Evidence from India’, unpublished paper, Department of Economics, Youngstown State University, October 2015, (accessed 23 January 2016).

13. Given India’s first-past-the-post electoral system, in which elections at the constituency level are regularly won with substantially less than a majority of votes, winning coalitions can be quite narrow. For instance, in the 2014 general election, 37 per cent of candidates won a majority of votes cast in their respective constituencies. That number was actually a significant increase from 2009, when just 22 per cent were ‘majority winners’. See Milan Vaishnav and Danielle Smogard, ‘A New Era in Indian Politics?’ Carnegie Endowment for International Peace, 10 June 2014, (accessed 15 June 2014).

14. Diego Gambetta, The Sicilian Mafia: The Business of Private Protection. Harvard University Press, Cambridge, 1996, p. 25.