How can we explain extreme variations in presidential strength in a context where the constitutional and electoral rules remain largely unchanged? In his book on the rise and fall of President Alberto Fujimori in Peru, Barry Levitt seeks to measure explicitly a long-held view among scholars of Latin America: The rules of the game in politics only matter so much in practice. Levitt challenges a purely neoinstitutionalist account of presidentialism by identifying and explaining broad swings in the constraints on executive power in Peru. The puzzle is that these changes occur during a 20-year period when the formal rules remained relatively stable.

Rather than rely exclusively on what John M. Carey would call “parchment institutions” (“Parchment, Equilibria, and Insitutions,” Comparative Political Studies 33 [September 2000]: 735–61), Levitt argues for incorporating other, “meta-institutional” variables to explain how successive Peruvian presidents sought and wielded power in a comparatively short period of time. Whereas the presidencies of Alan García (1985–90) and Alejandro Toledo (2001–6) experienced significant pushback when either was perceived as overstepping its bounds, Fujimori's 10-year rule (1990–2000) was much less constrained by the legislature or other agents of resistance. Levitt demonstrates successfully that constitutional changes in the early 1990s do not fully explain this pendulum shift from executive control to hyperpresidentialism and then back again. He argues, instead, that shifts in the adherence to constitutional norms by elites, as well as changes in the nature of party organization, affect the level of presidential power. Where both meta-institutions are weak, then the executive will be stronger. On the other hand, “(much) stronger constitutionalism and (modestly) stronger parties made the executive weaker” (p. 208). Where there is both the desire and the capacity to constrain the executive, then the president will be weaker than he or she prefers.

Levitt makes several contributions to our understanding of executive-legislative relations and democratic politics more generally. First, he advances a formal conceptualization of a particular type of representative institution: the electoral movement. For many party scholars, neither the minimalist definitions of political parties, such as that advanced in Anthony Downs's (1957) An Economic Theory of Democracy, nor more organizationally complex renderings, such as Angelo Panebianco's (1988) Political Parties: Organization and Power, line up neatly with the personalist vehicles and ad hoc coalitions that regularly contest elections in countries like Peru. Levitt provides us with some much-needed leverage to distinguish the multiple electoral organizations that compete in places where parties are weakly institutionalized and party systems are highly volatile.

Second, the author successfully incorporates new meta-institutional variables into the study of presidentialism. It makes intuitive sense that organized and disciplined parties in Congress will be better able to counteract a power-seeking president than less organized and highly undisciplined electoral movements. I am also persuaded that a desire, on the part of political elites, to adhere to constitutional rules can be a powerful check on the manipulations and machinations of the executive branch. Neither of these variables sits comfortably with the formal rules that new institutionalists typically use to explain powerful presidents. Yet they undoubtedly affect the nature of presidential power in practice.

Given the novelty of the argument, it is disappointing that Power in the Balance never fully operationalizes the independent variables used to explain hyperpresidentialism. This is particularly the case with constitutionalism. The author initially defines constitutionalism as “the extent to which the rules that govern executive power and legislative-executive relations actually channel and constrain the political behavior of flesh-and-blood politicians” (p. 5). He rightly differentiates between the rule of law on paper and the rule of law in practice. Later on, the book adds to this definition, first, the norms and expectations that surround constitutionally prescribed behavior (p. 44) and, second, the “acceptance” on the part of politicians of the actual behaviors of their brethren (p. 90). Each of these addenda reflects dimensions that go beyond the de facto rule of law in a country. The first taps into what we might call the culture of the rule of law in a country. The second reflects an attitudinal dimension emanating from the individual. This conceptual complexity is a welcome exercise that invites the kind of qualitative work that Levitt undertakes. Still, the reader is never sure which definition—the minimalist or the more complex—is under investigation in the text. I would have benefitted immensely from a list of theoretically justified empirical markers that we should expect to see where, for example, there is acceptance of flagrant violations to the rule of law or where constitutional norms are effectively in operation.

Without these observable indicators, it is difficult to know what evidence of the additional dimensions of rule of law should look like. It is equally difficult to know with certainty that a decline in constitutionalism preceded and therefore caused hyperpresidentialism, rather than the other way around. At one point, the text hints at reverse causality when it asserts that Fujimori “denigrated constitutional norms from the bully pulpit of executive office” (p. 65). Given the rather fluid commitment to constitutionalism in Peru's history, which the author briefly but effectively highlights in Chapter 2, it seems possible that the emergence of hyperpresidentialism could have provoked yet another turn away from adherence to rule of law. The narrative that connects changes in constitutionalism to the decline and growth of presidentialism leaves these theoretical and conceptual questions unresolved.

The author is more successful at demonstrating the constraining capacities of an organized political party, both as a part of the opposition in the legislature and when its leader is in power. Levitt capably demonstrates that President García was much more constrained by his party, the Alianza Popular Revolucionaria Americana (APRA), than Fujimori was by the electoral movements that supported him. Unfettered by an “organizational counterweight to his power,” Fujimori was free to make and break laws as he saw fit (p. 156).

Here, too, operationalization of party organization should be more explicit. The organizational prowess of a party is difficult to measure, as Maurice Duverger pointed out long ago (Political Parties: Their Organization and Activity in the Modern State, 1954). The dimensions of party organization that might matter for constraining the executive are multiple. This book would have benefited from a clear list of the organizational dimensions at work in the argument. Without such a list, the classification of parties and electoral movements at times seems post hoc in nature. They appear to be defined by virtue of their constraining capacities. This is problematic theoretically. It is also problematic empirically. Several Peruvian parties, such as Acción Popular, underwent organizational changes during the period under consideration. Should these changes matter for the party's constraining capacities? We cannot know unless we are sure about the organizational dimensions that are important for the causal argument.

These issues aside, Levitt offers key insights into an understanding of executive-legislative relations in countries where the rules in and of themselves cannot completely explain governance. In these countries—and perhaps everywhere—existing norms and the nature of the representative institutions also matter when it comes to balancing power between the different branches of government.