In their book, Gregory Koger and Matthew J. Lebo develop their notion of “strategic party government” (SPG). They define this as refinement of how to conceptualize the role of parties and legislative behavior in the U.S. Congress. Their motivating question and organizing themes are straightforward. First, Koger and Lebo argue that scholars too often conflate ideology and party in roll call behavior. Second, they contend that individual legislators often must choose between pursuing a course that effectively balances constituent interests and party interests. From here, their SPG approach expands to incorporate explicitly the electoral implications associated with legislative partisanship. The competing incentives faced by members are encompassed in what Koger and Lebo call “Madison’s Joke” (pp. 16–17). Parties are not monolithic entities, but rather collections of individuals with diverse sets of preferences. Legislators face competing and simultaneous pressures from both their constituents and their legislative parties. At times these two forces may work in tandem, but at other times they may conflict, forcing a legislator to choose one side at the expense of the other. These pressures can be manifest even within heterogeneous parties, in part because one side may be cajoled into unity largely to defeat the other party.

This argument is not controversial, per se. In many respects, the arguments parallel those of Frances Lee (2009) in her book Beyond Ideology. Furthermore, that legislators’ decisions should manifest in meaningful terms with respect to electoral fortunes is also highly likely. The real test of the argument, then, comes when one seeks to aggregate the individual-level behavior to assess the impacts at the party or electoral level. That is an ambitious undertaking which requires a carefully crafted and methodologically sophisticated approach to assess accurately. Such analyses are doubly tricky because they require the authors to find their hypothesized effects not only across multiple levels of analysis, but across an array of dependent variables as well. This is especially apparent as one moves from micro-foundations evaluated via experimental analyses, to analyses of long time series of highly aggregated measures such as election returns, win rates, and party unity scores. This marks one of the strengths of the present book. Koger and Lebo start with an experimental approach to establish how voters punish party unity in elected officials, a crucial first principle. Next, they shift to analyses of the electoral penalty for increased party unity, along with macro-level patterns of unity to assess how majority-party unity scores are influenced by majority size, as well as by minority partisanship. In the closing analytical chapter, they analyze legislative action on the Central America Free Trade Agreement through the lens of the SPG theory.

Much of the SPG theory rests on the notion that members face countervailing pressures to support their party or their constituencies. As the authors note, however, notions of party unity or party support are often inaccurately portrayed as ideological voting. While conceptually distinct, separating partisan from ideological voting proves difficult. Yet failure to establish this distinction negates a key plank of the SPG theory. Koger and Lebo’s experimental results show that respondents do indeed differentiate between legislators’ ideological and partisan voting records. The key implication of this is that ideological voting, even extreme ideological voting, was viewed more positively than high levels of party voting. In other words, being a “party hack” plays more poorly with voters than being a committed ideologue.

Having established that voters view party unity in a relatively negative light provides a launching point from which to investigate members’ roll call behavior, partisan agendas, and the interconnected nature of the parties’ electoral fortunes across time. The authors’ primary aim is to identify how party unity and a party’s electoral fortunes move across time. The most notable—and highly robust—finding is that higher levels of party unity lead to a statistically significant reduction in vote share. That effect is moderated somewhat, however, for the majority party. As the majority party succeeds in passing legislation, the resulting enhancement in party reputation provides them with positive electoral benefits. Those benefits, however, do not offset the losses incurred by “whipping” the rank and file to secure the level of party unity needed to pass key agenda items. The nature of the dynamics observed in party unity across time exhibit some interesting regularities. Interestingly, the process tends toward an equilibrium level of unity for both the minority and majority parties. High unity at time t results in electoral losses, which leads to lower levels of unity at time t+1. Given this dynamic, the parties adjust their respective strategies and, as a result, both sides adjust their party unity accordingly. As one of the parties increases (or reduces) unity, the other party follows suit.

Interestingly, the majority party drives much of this process, including aspects of the minority party’s strategy of opposition. The minority party is rather insulated from the costs associated with party unity, and certainly is not poised to share in the benefits associated with passing the majority’s agendas. As such, the minority often pursues a rational “best response” strategy of obstruction, making it notably more painful and costly for the majority party to pass its agenda. The majority party is then left to save face and pass something so they can point to some type of tangible legislative success.

Again, these findings might not be all that surprising, as in many respects they are often intuitive. What this response overlooks, though, is how difficult it can be to address such questions empirically. In this regard, Koger and Lebo provide a clear picture of how difficult it is methodologically to tease out the causal elements of interest needed to assess their hypotheses. As such, they bring an important element of sophistication and precision to addressing a series of questions many individuals tend to gloss over. That said, the high level of aggregation across time seems to treat the effects as directly comparable. Given the diversity of the prevailing party systems and political cleavages across time, this result is impressive on the one hand, but is likely too blunt on the other. For instance, does a change in party unity during the era of the strong Speaker in the Gilded Age mean the same thing as it would after the overthrow of Joseph Cannon and during the Ryan speakership? A minor point, but this would be interesting for the authors to discuss more.

The other contribution Koger and Lebo make comes in their assessments of what their findings mean for legislative politics in a polarized system. The answer they leave us with is a distinctly pessimistic one. While citizens and legislative scholars understood the incentives for members to “position take,” what we did not have before was a clear metric for how certain types of legislative strategies paid dividends and imposed costs. Koger and Lebo show us why parties rarely seem to undertake legislative initiatives to fix pressing problems or to pursue “good policy” outcomes that might foster bipartisanship. In the polarized arena legislative actions are often meant to serve partisan electoral ends, rather than govern in the common interest, and policy benefits are secondary concerns.