Recent elections around the world have challenged the conventional wisdom in political science that incumbency provides politicians with an insurmountable electoral advantage. Instead, in elections from South Korea to South Africa, the 2024 electoral cycle was the most damaging on record, with a clear “throw the bums out” message that led incumbents to lose vote share at a scale unparalleled in more than a century of elections (Burn-Murdoch Reference Burn-Murdoch2024).
Luis Schiumerini’s Incumbency Bias thus comes at a particularly good time, providing a unified theory for understanding both when incumbency is a blessing for politicians and when it is a curse. Building on cases drawn from subnational executive offices in Argentina, Brazil, and Chile, Schiumerini painstakingly develops and tests his theory, seeking to understand the causes of incumbency bias, under what conditions it helps or hurts incumbents, when bias is personal and when it is partisan, and the consequences for democracy.
Incumbency Bias has many virtues, but perhaps greatest of all is that it is a methodological tour de force. It will serve as a useful model for graduate students in comparative political science on how to structure and present the findings from a sound multilevel research design that includes regression discontinuity designs, natural experiments, differences-in-difference approaches, and survey experiments. Each chapter takes a neatly defined question and tackles it in great detail. Rather than a sequence of unconnected papers, though, Schiumerini goes to great lengths to tie his chapter findings back to the book’s central theoretical contribution, namely that incumbency bias, understood as the systematic difference in the electoral success of electoral contenders solely based on their institutional status as incumbents or challengers, “emerges and varies because of how institutions condition the alignment between citizen expectations and incumbents’ capacity to deliver on these expectations.”
One reason that the conventional wisdom of a positive incumbency bias has been so pervasive is that incumbency clearly confers benefits: incumbents have access to resources and power, are able to provide services to constituents, and have important symbolic power relative to challengers. But more recent research has cast doubt on just how important these advantages may be. In weak party systems or high corruption environments, for example, scholarship has shown that there may be an incumbency disadvantage, since voters have good reason to fear that if reelected, incumbents may either shirk from providing good governance or actively engage in rent-seeking behavior (Klašnja and Titiunik Reference Klašnja and Titiunik2017).
Schiumerini argues that incumbency bias, whether of the devil you know or the throw the bums out variety, emerges because voters engage in “bounded accountability.” Because of information constraints, voters focus primarily on incumbent performance, without considering the challenges that subnational politicians may face in delivering on voter expectations. The innovative twist to his argument is that there is a frequent mismatch between voter expectations of performance and incumbents’ capacity to meet these expectations. Schiumerini’s theoretical innovation is to recognize that this mismatch can explain both positive and negative incumbency bias: bias in either direction results from the interaction between citizen expectations, the capacity of the office held, and the information environment, which is heavily conditioned by the strength of programmatic political parties. He finesses this claim by incorporating institutions. Political institutions determine the policy scope of officeholders’ mandates, that is, the range of policies for which the incumbent is responsible (Niedzwiecki Reference Niedzwiecki2018). Fiscal institutions, particularly dependence on transfers from the central government, shape officeholders’ capacity to deliver. Because institutions are stable over time, they help to explain enduring differences in the returns to office across countries and across levels of government. The institutional origins of incumbency bias distinguish it from an alternative explanation, differences in candidate quality, and lead to systematic biases for and against incumbents in the aggregate across classes of politicians, such as Brazilian mayors or Argentine governors.
Incumbency Bias uses the variation in political, fiscal, and federal institutions across Argentina, Chile and Brazil to develop a series of powerful analyses of incumbency (dis)advantage in mayoral and gubernatorial elections. Fiscal institutions, for example, are very constraining for Brazilian mayors, who are highly dependent on transfers from the federal government. Argentine governors, by contrast, face virtually no constraints on spending. The impacts of the fiscal constraint on incumbent bias vary depending on voters’ understanding of officeholder’s policy scope: Brazilian mayors have few resources but voters expect much of them, leading to negative incumbent bias, while Chilean mayors receive relatively robust transfers but have such narrow policy scope that voters do not expect much from them, contributing to positive bias.
Brazilian municipal elections are the empirical backbone of this book, providing what Schiumerini calls the “strongest possible test” of his arguments. To tackle the challenge, Chapters 3 and 4 evaluate, respectively, six and four electoral cycles in Brazil, thus encompassing between 22,280 and 33,420 individual mayoral contests. The analysis of this massive data set is carried out in a variety of creative ways. For example, Schiumerini uses the fiscal shock generated by the path-breaking Fiscal Responsibility Law of 2000 as a natural experiment to assess how the new fiscal constraints affect mayoral capacity, finding that negative incumbency bias only appeared in municipalities running deficits, namely the municipalities where the law had a binding effect on mayoral spending. Another test looks at the effect of a change in commodity prices on incumbency bias. Schiumerini builds on Zucco and Campello’s work on commodity shocks to argue that exogenous price shocks may change the direction of incumbency bias, with citizens using the shocks to reassess incumbents’ competence (Campello and Zucco Reference Campello and Zucco2020). Among other convincing results, he finds that a one standard deviation positive shock from the median in municipal commodity prices contributes to a five percentage point increase in mayors’ incumbency advantage.
Later chapters evaluate gubernatorial elections in Argentina and Brazil, mayoral elections in Chile, and citizen reasoning about incumbents. Chapter 5 examines gubernatorial elections in Argentina and Brazil. Although both countries are known for strongmen governors with considerable autonomy and power—the notorious coronéis, for example—Schiumerini finds that voters are strongly influenced by fiscal shocks and their effects on public goods spending in Argentina, but not in Brazil. Chapter 6 develops a natural experiment around a change in term limits to investigate the relative importance of personal versus partisan sources of incumbency bias in Chile, finding that the benefits of positive shocks accrue more to personal candidacies than to coalitions. Chapter 7 dives into the microfoundations of how citizens process information, using two distinct survey experiments in Brazil and Argentina to evaluate the plausibility of the book’s theory of bounded accountability.
Because Incumbent Bias is focused on aggregate patterns of incumbent bias in a huge number of elections, rather than on individual candidate strategies or campaigns, real-life politicians and their choices are largely hidden from view. Among the various technical presentations of the multiple models, experiments, and surveys, there are few mentions of living, breathing mayors or governors, or about the blood and guts of a real campaign. The reader wonders how the many colorful characters who have populated subnational offices in the region—Jânio Quadros, Paulo Maluf, and Carlos Menem come to mind—fit into Schiumerini’s arguments. Because of the broad aggregate focus, there is also little mention of the networks that sustain incumbents and shape how subnational executives strategize about their careers beyond subnational office, such as the implications of mayoral office for promotion into executive cabinets, legislative seats, or higher executive posts. Although Schiumerini does offer insightful quotes from his multiple interviews of politicians and political consultants, the absence of a broader descriptive dimension may restrict the pedagogical appeal of Incumbent Bias to all but the most methodologically-oriented students.
This does nothing, however, to diminish the impressive contribution that Incumbent Bias represents, both to the study of Latin American politics and to the study of incumbency bias more broadly. The unevenness of competition between incumbents and challengers is of great normative concern to scholars of the region’s democracies. By demonstrating that this bias varies in systematic ways and uncovering some of these aggregate biases’ institutional determinants, Schiumerini has greatly expanded the possible scope of future research on this topic. The book’s empirical exploration of his unified theory of incumbent bias, capable of simultaneously explaining both its positive and negative expressions, provides future scholars with a valuable exemplar of finely-crafted applied empirical research. The book also promises to spark valuable dialogue with adjacent literatures on patronage, clientelism, and effective public goods provision.
