Michelle Taylor-Robinson's book argues that since the poor are typically less capable than the rich of holding their elected representatives to account—due to the costs of monitoring and sanctioning—they rarely achieve effective political representation. The extent to which the poor nevertheless “count” in national politics depends on formal and informal institutions (electoral systems, nomination procedures, and forms of clientilism) that either mitigate or aggravate the political biases that the author identifies. One of her main points is that clientilistic representation—which is typically regarded as sham representation, inferior to its programmatic counterpart—is often the only form of representation for which the poor can hope.

The first two chapters introduce the research problem and the main theoretical arguments. Chapter 3 examines the confidence that the poor have in national legislatures in nine countries in Latin America (Argentina, Brazil, Chile, Colombia, the Dominican Republic, Mexico, Peru, Uruguay, and Venezuela). Chapters 4 through 7 are devoted to a case study of Honduras, and especially its National Congress, which is a legislative institution on which the author has worked extensively in the past. Chapter 8, the concluding chapter, investigates the policy implications of her argument by comparing the adoption and subsequent fate of conditional cash transfer programs in Honduras and Brazil.

This is in many ways a book in the best traditions of comparative politics, examining the manner in which institutions, socioeconomic conditions, and political behavior interact. The author's willingness to engage with all these issues and theoretical literatures is commendable, making the book relevant to a wide readership: to students and scholars of political institutions, comparative political economy, inequality, and representation, and to those with a special interest in Latin American politics and society. In the course of her investigation of the interaction of political institutions, socioeconomic conditions, and political behavior, the author moves among several levels of analysis: individuals, parties, institutions, regions, and countries. The scope of the argument is really quite impressive.

It also makes things complicated, and I did not always find the author's arguments easy to follow. For example, the book claims that institutions do not merely create different incentives for politicians, party elites, and citizens influencing the strategic interaction among these three types of agents; institutions also shape the identities and preferences of elected representatives (pp. 14–15). But it never explains how to combine these two different theories of institutions. The theoretical analysis in Chapter 2 is more or less exclusively rationalistic, deriving testable hypotheses from principal–agent theory and semiformal game-theoretic propositions (pp. 50–52). The empirical chapters, however, and particularly the two chapters on the Honduran Congress, are primarily concerned with the “roles” that politicians play (or see themselves as playing).

It would have been helpful, I think, if the book had stated more clearly how incentives and identities combine to shape the political behavior of elected representatives. One way to do this would have been to examine empirically how political careers develop over time. How does it become a part of the identity of a politician to represent the constituents of a poor district, as opposed to defending the interests of a national party or protecting the integrity of the legislature as an institution (the other motivations that the author considers)? Which types of politicians do parties and voters select, and why? And how do different kinds of politicians respond to electoral victories and defeats? The cross-sectional research design in Chapter 6 and Chapter 7 is not well suited to answering these kinds of questions.

The manner in which the theoretical claims of the book are formulated and tested is also less clear than it could have been. One of the author's main points—and I think it is a correct one—is that formal and informal institutions interact: “A key argument of this study is that it is not one but a combination of institutions that influences the sanctioning capacity of poor and rich people…. If we only considered one type of institution and the incentives it created (say electoral institutions) we would have an incomplete picture” (p. 8). This argument appears to have the form “the effect of x ₁ on y depends on x ₂,” but Chapter 2 nevertheless examines electoral systems, nomination procedures, and forms of clientilism separately, and the discussion of how the effects of one of these factors (or sets of factors) might be conditional on the others is limited. Moreover, the comparative analysis of nine Latin American countries in Chapter 3 does not allow the author to differentiate between the effects of electoral institutions and the effects of clientilism, and the evidence seems to indicate that these variables are highly correlated (see the table on p. 63), which suggests that they may be causally related in some way that is not accounted for (although the author does discuss some of these problems on pp. 69–71).

The claim that clientilistic representation is the only form of representation that the poor can hope for—and that clientilistic representation is better than no representation at all—is a provocative idea, with far-reaching implications. The question is how far-reaching. It occurred to me, as a European and a student of Europe, that it would have been interesting to compare the results of the author's study of Latin America to evidence from the recent European past. Was clientilism the only effective way for the poor to get political representation in Western Europe around 1960, when the average GDP per capita in Western Europe was approximately at the same level as the average GDP per capita in the larger Latin American countries today? The answer, I think, is no. One possible explanation is that the distribution of wealth and incomes was more equal in 1950s and 1960s Western Europe than it is in contemporary Latin America. Another explanation might be that the state formation process in Western Europe generated more robust bureaucratic institutions. There is a lot for students of comparative politics and political economy to think about here, thanks to Taylor-Robinson's thought-provoking and timely book.