Do “vote-buying” activities of locally elected legislators in developing countries crowd out the representation of mass policy preferences? If a legislator can buy a citizen's vote with a material benefit, does he really have an incentive to represent the interests of his constituents in legislative decision-making? Imagine the following setting: A slum neighborhood near Bangalore city center in Southern India. A group of women in worn-out sari-s, domestic servants who clean houses for a living, sit chatting outside their one-room concrete shacks with tin roofs. They have no bathing or cooking facilities in their shacks. They share the same water tap with hundreds of neighbors. Their toilet is the nearby railroad track. A well-dressed man exits a car and approaches:

Good day. My name is Rama and I am running on the BJP party ticket for the upcoming local election. Are you registered to vote? [The women nod that they are registered]. Can I offer to help you improve your life? In exchange for your vote, the BJP party not only has sewing machines, but we will teach you how to use them and help you set up your own tailoring shops. Are you interested?

The women nod in agreement. The politician has just purchased the votes of these women in exchange for the promise of a sewing machine. This activity, frequently referred to as “vote-buying” in comparative politics literature, is a common practice in countries throughout the developing world. From Mexico and Argentina to Asian countries of Thailand and India, politicians and political parties maneuver through crumbling streets of low-income neighborhoods promising and providing material incentives to possible voters in exchange for political support (Helmke and Levitsky 2006; Kitschelt and Wilkinson 2007; Shaffer 2006).

When politicians engage in vote-buying activities, they alter the structure of incentives in their relationship to their constituencies. These activities can constrain how voters respond to political leaders over time by changing the expectations voters have of political leaders, and more generally of their formally established democratic institutions. Vote-buying activities can undermine the accountability of formal institutions for both legislative decision-making and resource allocation.

My research uses original public opinion data I collected from surveys of 1700 households and 40 political leaders in Bangalore, India to assess how vote-buying activities alter our conventional understanding of the “rules of the game” for democratic policy representation. During one year of field research in Bangalore from May 2006 to May 2007, I frequently encountered stories such as the one highlighted above. In the household survey 43.5 percent of Bangalore respondents report that they have experienced some form of vote-buying in which political leaders in their community have offered them material benefits. I seek to understand how these activities, particularly when we address different types of vote-buying activities, influence dyadic policy representation—a form of democratic representation, borrowed from literature in American Politics, that is defined by how well political leaders represent the policy preferences of their constituents in legislative decision-making (Hurley and Hill 2003; Miller and Stokes 1963).

I question how the common practice of vote-buying influences democratic policy representation. A number of studies assess how such activities influence electoral outcomes, yet it is widely assumed that vote-buying activities undermine core institutional guarantees of democracy, rights and liberties promised in representative democracies and outlined in such work as that of Robert Dahl (1971). Little research to date, however, has provided empirical evidence about the veracity of this assumption.

This is the puzzle my research currently addresses. Vote-buying activities may constrain the policy representation of some citizens relative to others because some prospective voters may express policy opinions contrary to their actual preferences to receive material inducements from politicians (Eisenstadt 2006; Kitschelt 2000). Low-income voters in need of material benefits appear especially vulnerable to this risk (Stokes 2004). We might also expect that political leaders engaged in vote-buying activities have an incentive to shirk their legislative responsibilities once in office, knowing that they can purchase the votes of citizens as opposed to being concerned about representing their preferences.