Corruption is endemic in most developing countries, and it is widely believed to inhibit their economic development. Most people’s firsthand experience with corruption usually involves individual government officials or courts, and those interactions are the most frequently studied aspect of corruption. Yadav believes that most studies of corruption have seriously neglected the role of elected legislatures. She argues that legislative procedures strongly influence the extent of corruption, particularly through lobbying by business groups. Her analysis reveals that when a legislature is organized around political parties that can set the policy agenda, determine amendments to legislation, and discipline party members who step out of line, there is significantly more corruption than when parties are weaker and legislators are more independent. This careful empirical work focuses in detail on Brazil and India -- large, complex countries with federal structures. But it also draws on evidence from 62 other democracies in the developing world. Yadav finds that contested elections do not eliminate legislative corruption in the form of financial payments made to parties and politicians by business groups to assure policy outcomes favorable to their interests. On the contrary, costly elections create more opportunities for corruption.