Whether democratic institutions deliver accountable, responsive, and stable governments is a core question in political science. G. Bingham Powell has been an integral contributor to this research for decades and is clearly the leading scholar on the question of “ideological congruence”—the degree to which government preferences reflect the preferences of citizens. In 2000, he produced what I (and I imagine the overwhelming majority of my colleagues) consider the defining work on the subject with his book, Elections as Instruments of Democracy: Majoritarian and Proportional Visions, which presented a compelling case that there is a critical trade-off in constitutional design: clarity of choice and accountability versus diversity of choice and representational inclusivity. Majoritarian designs tend to deliver the first, whereas proportional designs tend to deliver the second. That is, sacrificing clarity and accountability seems to be the price for designing institutions that regularly deliver ideological congruence. In his most recent contribution, Ideological Representation: Achieved and Astray, Powell asks the related, yet distinct, question within the context of parliamentary governance: When, and under which circumstances, does congruence break down?

The “when” portion of this question is central, because it clarifies the contribution of this book relative to Powell’s impressive previous scholarship. Here, the theoretical arc is that voter preferences are translated into government compositions in three successive stages. First, voters’ preferences are aggregated into parties’ vote shares. Second, these vote shares are converted into parties’ legislative seat shares. Third, legislatures select governments, composed of one or more parties, which are ultimately charged with making (or, at the very least, proposing) policy. Each of these stages presents an opportunity for slippage in congruence—inefficiencies in the conversion of preferences from one point to the next in the chain of ideological representation. Powell’s goal is to determine at which stage congruence goes astray most often, why it goes astray, and whether deviations in earlier stages can be rectified in later stages. The central finding is that congruence failures occur in similar proportions at each stage, but are almost never corrected. Further, deviations, particularly in the first stage, tend not to simply persist but also to compound through the chain, growing ever larger.

The book proceeds through the three stages of ideological congruence in turn. The first stage focuses on aspects of the party system: the menu of options before voters in terms of both the number of alternatives and their ideological spread, as well as the tendency of voters to “defect” from their nearest alternative (for reasons of accountability, tactical voting, and so on). The second stage focuses on the rules governing the translation of votes into seats, paying particular attention to “distortive” rules (those that increase disproportionality in the votes–seats conversion) and the conditions under which median-voted parties do not become the median party in the legislature. The third focuses on aspects of the bargaining environment that may lead to the formation of governments that are ideologically dissimilar from the median voter (particularly legislative polarization) or governments in which authority is highly concentrated in a single party (such as a large plurality party).

If there is a critique of the book to be made, it is perhaps most apparent here, in this third stage. From time to time, Powell chooses to consider the observable characteristics of the objects being translated, rather than the process through which translations occur. Note that this is rare in the Stage 1 analysis, vanishingly rare in the Stage 2 analysis, but more common in the Stage 3 analysis. For example, a fair amount of discussion is devoted to differences in congruence when the median and plurality parties are one and the same and when they are distinct, as well as to differences in congruence under large and small plurality parties. Of course, this is in keeping with the analysis at earlier stages of the book, and these analyses are informative and valuable. However, more consideration of the institutional factors that may privilege the plurality party relative to the median party when they are distinct or large relative to small plurality parties (e.g., the nature of recognition rules for government formation) would have been a welcome addition. In the grand scheme of the book, however, this is a very small quibble, particularly given the degree to which focusing on these observable characteristics makes the book both more accessible and more focused on its mission: exploring the aggregative nature of congruence failures.

The book makes an important contribution and is to be commended. Although Powell is careful to note that congruence “seldom fails disastrously,” there is a lot of descriptive, theoretical, and normative value in understanding “how and why the path to congruence can go astray” (p. 204). Simply noting that democratic institutions are imperfect is insufficient; we want to know how and why they may underperform our idealist hopes. Powell does an excellent job of empirically mapping process and performance, as well as explaining the theoretical mechanisms that underlie them.

The central argument, that congruence slippage happens cumulatively and in distinct stages, is communicated clearly and elegantly. The analysis is rigorous but also efficient: figures are favored over tables where appropriate, and simple models are favored over complex models wherever possible. This combination of clear argument and assessment makes for an enjoyable and brisk read, but the 200+ pages are still quite dense—filled to the brim with analysis. As such, although I recommend the book for advanced undergraduate seminars, it is likely a bit beyond less advanced undergraduate students. Of course, it should be assigned widely in graduate comparative seminars.

Finally, the book makes two ancillary but, to my reading, very important contributions in the later chapters that I urge my colleagues not to miss. Powell devotes an entire chapter to the exploration of minority governments, and this work is sorely needed. Minority governments are common (about one in three in Powell’s sample) yet understudied occurrences, typically treated as either equivalent to majorities in form and function (they are neither) or considered oddities or perturbances to large-scale analysis. Minority governance deserves the careful consideration that Powell gives it, and the discipline sorely needs this description, analysis, and discussion. Second, through the concluding chapters, Powell presents a very thoughtful, normatively oriented discussion of the findings of the book and the trade-offs that they imply (complete with data brought to bear, of course). I found this section a much more rewarding read than typical conclusions and a fitting end to a well-written book.

Given the depth and range of Powell’s contributions over the years, it is tempting to conclude this review with sweeping statements about the articles and books he has written, the students he has mentored, and the institutions he has shaped, but this is not the venue for those statements. Instead, I conclude by writing that this book lives up to the expectations set by Powell’s previous contributions—a standard to which we should all aspire.