Corruption is notoriously difficult for political scientists to capture empirically, and understanding its causes thus requires significant theorizing and methodological innovation. This is perhaps especially true in developing countries with high levels of corruption. Individual elected officials and state bureaucrats may be corrupt, yet are also frequently embedded in institutions where corruption is endemic, being part of the fabric or culture of the institutional environment. It becomes a challenge to distinguish institutions that embody active corruption from those that more passively enable corruption due to weakness in their ability to constrain individual behavior. Vineeta Yadav explores important questions about the nature and causes of corruption in Political Parties, Business Groups, and Corruption in Developing Countries, a well-researched and thoughtful book that will contribute substantially to several literatures.

Yadav asks why corruption is higher in some countries than others, with a particular focus on business groups and their cash payments to legislative actors. Her argument is rooted in legislative institutions, with the key independent variable being the strength of political parties. She comes to the intriguing conclusion that corruption is higher in countries such as India where parties exercise greater leverage over legislators, and lower in countries such as Brazil where political parties are weaker. The logic is predicated on business lobbies choosing to address their demands to the actors they perceive to have control over policymaking and legislative outcomes. In strong party systems, business groups lobby parties that have a relatively high ability to deliver benefits to business. In weaker party systems, meanwhile, businesses choose to lobby individual legislators, yet these have less control over the legislative process, as each individual legislator has a higher degree of agency. The scope and scale of corruption is more extensive and systemic in systems where parties can deliver on outcomes. This argument is provocative, and contributes to the growing number of findings that suggest that strong party systems—despite being seen as crucial to the consolidation of democratic institutions—do not necessarily contribute to improved governance.

Yadav extensively documents three specific sets of legislative rules that are indicators for party control over legislation: agenda-setting power, control over the legislative amendment process, and the power of party leaders to expel from the legislature those copartisans who do not vote on party lines. There is a clear distinction between cases where parties control these powers (India) and other cases where parties have little such leverage (Brazil). It is interesting to note that Yadav's selection of the Brazilian and Indian cases for qualitative comparison does not depend upon variations in electoral rules: While open-list proportional representation factors into party weakness in Brazil, the Indian plurality system is not seen as contributing to noticeably greater party strength (p. 61). Contrasts in party strength thus emerge inside the policymaking process, although the third indicator (the power to expel legislators) links clearly back to the electoral and political ambitions of individuals within a partisan framework.

Beyond these three indicators of party power, Yadav offers several mechanisms to explicate the causal links between party-based systems and higher corruption. Two are termed “demand-side” mechanisms, as they explain why strong party systems would generate higher demands than weak party systems for business contributions to political agents. These are the scale of a party's demand for resources, which Yadav argues is much greater than demand by individual legislators, and the desire of parties for fungible cash resources from lobbyists over other sorts of resources (such as information). Two other mechanisms emerge on the “supply side”: the ability of parties to extend their political control into state bureaucracies and the judiciary through legislative delegation, which can generate a corruption dynamic that links the legislature to other branches; and the lack of financial accountability of parties to such institutions as auditing agencies that monitor finances and ensure transparency. These mechanisms together generate higher demands for cash resources in party systems, along with lower ability of other institutions to countervail corrupt parties. Theoretical questions can be raised about the demand-side mechanisms, which imply a “satiability” of individuals' corruption demands in weak party systems that deserves further comparative exploration. For Yadav, the proceeds of corruption are often presumed to be used for political rather than personal effect, which begs questions about the way that corruption is best conceptualized.

In developing this argument, Yadav offers extensive description of an intricate research design and methodology: a two-country qualitative comparison that includes process tracing of specific legislative cases, subnational sampling, interviews, and survey research, all complemented by cross-sectional time series of 64 countries. The book is a model of transparency, with extended explanations of the research design incorporated into the body of the text; while many such issues could be relegated to appendices at little analytical cost, the level and clarity of documentation on research design and strategy sets an example for others to follow.

This is impressive work, and the reader is likely to find that most queries are resolved or addressed within the book, given Yadav's attention to detail. In fact, some of the effects uncovered here might be found to be even stronger if unresolved questions could be tested directly. For instance, there is some analytical slippage between early chapters on legislative corruption and the large-n study that uses several broad international corruption statistics (such as Transparency International, among others) that are presumably somewhat more heavily weighted toward corruption within state bureaucracies. The book's theoretical linkage between party power in the legislature and delegation within the bureaucracy offers only a partial bridge here, yet the robust findings link indicators of legislative party strength to the less legislature-specific international corruption statistics. This is empirically compelling and can contribute to theory building as more direct cross-national tests of legislative corruption are developed.

There are opportunities for building on this fine work. This includes the prospect of examining whether there are within-case variations in corruption levels over time (which appears to be true in statistics for India); testing this could either shore up or challenge the comparative-static proposition that legislative institutions are the principal causes of corruption. Yadav's qualitative comparison does not capture these variations and it is unclear how the time series analysis as conducted contributes such a time-variant perspective. The concept of corruption remains tricky as well, and the book will not resolve these conceptual debates. The author distinguishes clearly between lobbying and corruption, but lines grow fuzzier in the specifics: illegal bribes versus influence, for example, or campaign contributions from business that coincide with generally pro-business policy versus preferred policy for certain sectors versus particularistic benefits to specific firms. Finally, as a smaller issue, many cited sources do not appear in the reference list; this is unfortunate, because Yadav's extensive research will likely leave many readers seeking other cited works.

These questions detract little from the book's contribution. Its multistep causal argument—carefully argued and thoroughly documented—will leave scholars with much to chew upon. This is true for those seeking to understand and explain corruption, or state performance broadly, as well as those examining the governance impacts and economic effects of legislative institutions, parties, and interest groups.