Our collective understanding of electoral clientelism has advanced considerably over the past decade or two, thanks to an innovative body of research that has explored the mechanisms sustaining clientelist relationships over time and the behavior of the actors who participate in clientelist exchanges. In Patronage at Work: Public Jobs and Political Services in Argentina, Virginia Oliveros offers a major contribution to this literature by focusing on patronage, defined as the subset of clientelism in which “the good that is exchanged [for political support] is a public sector job” (p. 5). Patronage employees (who she calls “supporters”) are an important and understudied piece of the clientelistic political machines found in Argentina and many other countries. As Oliveros demonstrates, supporters engage in a wide range of political activities on behalf of the incumbent politicians to whom they owe their jobs—by campaigning and mobilizing votes during an election, but also by “doing favors” and distributing material benefits, even outside of the campaign season. From Oliveros’s careful descriptions, which are based on extensive field research and an innovative survey of public employees, we can get a better sense of the complex ecosystem of brokers, supporters, and personal networks that jointly comprise local political machines.

Oliveros asks fundamental questions about how patronage works. What exactly do supporters do to “earn” their jobs? Why do they do it? And most importantly, how do their patrons (the incumbent politicians) keep them from shirking? To work, the arrangement must overcome a commitment problem (p. 20): Supporters need to commit credibly to work on behalf of the incumbent, and once employed they must live up to that commitment. Common approaches suggest that norms of reciprocity or the threat of monitoring and punishment can explain why supporters hold up their end of the bargain. But Oliveros find these explanations wanting on both theoretical and empirical grounds. So, she develops a third explanation: Patronage is “self-enforcing” because the interests of supporters are aligned with those of the politician who hired them. Supporters do not work on behalf of the incumbent because they feel the need to repay a favor, or because they fear being fired by the incumbent if they shirk; they do it because they perceive that their benefit (the job) will probably only last for as long as the incumbent manages to win reelection (chapter 2).

Through four central empirical chapters (3–6), Oliveros methodically builds a case for the “self-enforcing” explanation of patronage. These chapters are exemplary: They are clearly written, present empirical evidence supporting each step in the argument in a logical progression, and reinforce quantitative findings with key insights gleaned from fieldwork.

The core findings are primarily derived from a survey of public employees fielded in three Argentine cities. Oliveros describes a carefully designed instrument with embedded list experiments and other features that allow her to elicit information on sensitive topics (although some interview subjects are surprisingly candid about their participation in the patronage system!). The descriptive results alone are worth the price of admission. For example, data from the list experiments allow Oliveros to estimate that 22% of public employees in the field sites participate in election campaigns, 21% attend rallies, 12% serve as polling monitors on election day (all from chapter 4), and 44% “grant favors” (chapter 5). This is an astounding amount of patronage, and the estimates are supported with interview evidence that depicts the quotidian reality of patronage work, as well as its centrality to the operation of political machines. For example, employees talk about politicians bringing in “their people” at the start of their terms, and about the expectation of campaign work in return (pp. 69–72). Others explain that supporters with law degrees are asked to serve as election monitors because the (nonpartisan) poll workers will often defer to them when a question arises (pp. 94–95). Supporters describe the types of favors they are likely to do for people (pp. 106–12). In one notable exchange, a non-Peronist employee who was not asked to serve as a monitor for her Peronist boss, perhaps because she was not perceived to be sufficiently trustworthy, contributed to the cause anyway by delivering lunch to her Peronist colleagues who were serving as monitors (p. 148). Evidence like this, cited throughout the book, serves to illustrate the routinized, common-knowledge nature of political work.

These chapters offer convincing evidence in support of three propositions derived from Oliveros’s theory: (1) patronage jobs are disproportionately distributed to perceived supporters, often through informal personal networks; (2) supporters are more likely to provide both electoral support and “political services” (p. 29), which also tend to flow through personal networks; and (3) supporters tend to believe that their job is contingent on the “political success of the incumbent” (p. 130). While other theories of clientelism might also predict the first two points, the third is more distinctive: Oliveros refers to it as the “main empirical implication of the theory” (p. 130), and extensive evidence supporting it helps to distinguish her explanation from the alternatives. For example, she shows that supporters who express more concern about what would happen to them if the incumbent lost the next election are more likely to serve as election monitors (p. 145), while those who have either job tenure or comparable private-sector employment options are less likely to participate in political activities.

The theory of self-enforcing patronage has many important implications. For one, we should expect patronage to be a problem in any country that has less-than-perfect civil service institutions. In a cross-national comparative chapter, Oliveros offers evidence to suggest that patronage is widespread, even in a country like Chile that has a reputation for good governance and a professional civil service (chapter 7). Relatedly, the self-enforcing nature of patronage will make it extremely difficult to overcome because the patronage workers are essentially free to the incumbent, and they do not have to be monitored (p. 104). As Oliveros writes: “If clientelistic arrangements can be self-sustaining without punishment, … clientelism becomes less costly, more difficult to detect, and even more difficult to curb” (p. 202).

And of course, patronage has important implications for the quality of democratic governance, which Oliveros discusses at the end of the book (chapter 8). She rightly argues that patronage violates basic democratic principles because it misuses state resources and tilts the electoral playing field toward the incumbent. Yet much more could be said on this topic. Future research could explore the difference between “political favors” and constituent service (p. 110), or ask how patronage helps to perpetuate inequality between clients and patrons (which is an important but often overlooked aspect of clientelism). And while Oliveros’s focus on public employees is appropriate and valuable—indeed, this is the main contribution of the book—one wonders about the linkages between supporters and voters. The book suggests that voters are tied to the political machine through personal networks, which once again raises the possibility of a norms-based logic of clientelism. But we also know that voters value good governance in addition to personalistic favors (p. 29), which raises questions about the limits of patronage, and the conditions under which voters might choose to punish politicians for using it. These are just a few of the additional questions and ideas that might inspire future readers who take on this excellent and thought-provoking book.