Criminal politics, the complex interplay between criminal organisations and political institutions, is said to be increasingly pervasive and pernicious across the globe. Perhaps unsurprisingly, the study of criminal politics is growing in prominence. In recent years, more and more efforts have been made to understand the interactions of political and criminal actors, to conceptualise the often-blurred lines between state and non-state actors and between licit and illicit activities, and to identify how criminal politics affects governance, democracy and the provision of public goods, not least of public security.

Criminal Politics and Botched Development in Contemporary Latin America is a fine example of the growing body of literature on criminal politics. Andreas Feldmann and Juan Pablo Luna present an intriguing account of how criminal activity affects politics, state institutions and development in Latin America. The authors advance an ambitious argument about the interrelated activities of politicians, organised crime actors and state agents in Paraguay, Peru, Uruguay and Chile. Unfortunately, the argument's empirical underpinnings fall somewhat short of its conceptual richness and theoretical sophistication. Nevertheless, the book presents an invaluable contribution to the study of criminal politics, both in Latin America and elsewhere. It represents an invitation to further explore criminal politics and test the book's assumptions, hypotheses and inferences.

Feldmann and Luna construct an elegant and convincing theoretical framework. According to the authors, criminal politics results from the spatially bounded interactions of the different manifestations of the drug industry (production, domestic or international trafficking, money laundering) and the state (politicians and state agents).

On the criminal side of criminal politics, drug trafficking organisations’ influence in a given country depends on the competitive advantage of drug trafficking, the existence of other criminal actors and the linkages with the legal economy. The greater the size of the illegal economy vis-à-vis the legal economy, the more important is criminal politics. Drug trafficking organisations tend to prefer predictability and relative peace to generalised and extreme violence, which can be bad for business; thus, they seek out a modus vivendi with politicians and state actors.

On the political side of criminal politics, politicians can choose to either ‘contest’, ‘seek rents’ from or ‘restrain’ criminal activity in a specific place or of a particular type. Because drug trafficking is highly lucrative, it can lead politicians to collude with rather than combat drug-related activities. The implementation of politicians’ strategies depends on state agents, such as law enforcement and the judiciary. Because state agents have more or less leeway, depending on whether there are principal–agent problems created by ‘institutional fragmentation’ and ‘functional autonomy’, politicians’ preferred course of action might not be implemented.

Depending on the combination of politicians’ and state agents’ responses to criminal actions, six possible ‘ideal types’ of criminal politics emerge: when there is rent seeking, either a protection racket or Janus-faced corruption; when there is contestation, either iron fist policies or lip service lacking enforcement; or, when there is forbearance, either opportunistic corruption or standoff and nonconsequential state activities. Each of these arrangements can be applied to different functional or spatial manifestations of the drug industry. Furthermore, these are dynamic arrangements that can change over time with the occurrence of reconfigurations of either the political or the criminal side of criminal politics.

To illustrate their theoretical argument, Feldmann and Luna investigate criminal politics in Paraguay, Peru, Chile and Uruguay. These four cases provide both variation on the dependent variable (criminal politics) and independent variables (institutional capacity, political structure, drug industry). They also display variance in terms of the outcomes of criminal politics (violence, corruption, institutional and developmental impacts). To study the four cases, Feldmann and Luna draw on primary research (interviews with state officials, politicians and actors involved in drug trafficking) and secondary sources (scholarly research and media reports). The authors describe differences and similarities in historical development and present-day configuration of drug production, drug trafficking and money laundering in Paraguay, Peru, Chile and Uruguay. This description of the drug industry is followed by an analysis of politicians’ interactions with criminal actors and different drug-related activities in each of the four countries.

Based on their case studies, the authors classify a country's criminal politics as protection racket, Janus-faced corruption, iron fist, lip service, opportunistic corruption, or standoff. While they characterise the criminal politics of Paraguay as protection racket across the board, they identify substantial variation in the configuration of criminal politics across functional and spatial areas in Chile and Uruguay. Because Peruvian state capacity is highly circumscribed and the political system fragmented, there are uniformly limited interventions and pervasive corruption. The book's central finding is that, depending on the socio-political configurations of the state – principally the state's capacity and the fragmentation of the political system – different configurations of the drug industry result in different manifestations of criminal politics. For instance, a centralised, legitimate and capable Uruguayan state leads to different criminal politics from the weak, patrimonial and uncoordinated Peruvian state.

Criminal Politics and Botched Development in Contemporary Latin America has many strengths, but also some weaknesses. The book is at its best when laying out the nature and logic of criminal politics. While it promises to shed light on an important social phenomenon, the brief empirical section does not always deliver on that promise. On one level, this is by nature: the illicit activities and collusion involved in criminal politics occur out of sight, ‘render[ing] even an accurate description of them exceedingly challenging’ (p. 11). On another level, this is by design: Feldmann and Luna readily concede that their approach is ‘exploratory’ (p. 4), and that they engage in only ‘preliminary mapping’ of criminal politics (p. 34). As such, their findings and conclusions are of ‘tentative nature’ (p. 71). Rather than the last word on criminal politics, then, Feldmann and Luna's book opens up multiple and exciting lines of inquiry. Their conclusion extends an invitation to continue the study of criminal politics: ‘We hope the analytical framework and the preliminary evidence … will help scholars embrace the challenges and theoretical promise of researching criminal politics’ (p. 71).