Many observers worry that rising economic inequality may undermine a core principle that lies at the heart of democratic governance: the responsiveness of legislators to their constituencies, regardless of their social position. Indeed, a large wave of recent scholarship has documented substantive political inequities in the responsiveness of policy makers to the interests of the rich in North America and Europe. The book by Christopher Witko, Jana Morgan, Nathan Kelly, and Peter Enns falls squarely within this stream of scholarship and examines plausible mechanisms underpinning unequal representation in the US Congress.
Let me get the main conclusion out of the way: This is an excellent and enjoyable book. It is likely to become required reading on this topic by academics, journalists, and interested political observers. It uses a nice blend of sophisticated data analysis and well-written, detailed case studies, making the book an engaging read for both academic and lay audiences. The latter, especially, will appreciate the clear and comprehensible writing style. The authors do an admirable job explaining technical concepts in “plain English” with minimal use of jargon. Academics will appreciate the book’s conceptual innovations, the careful data collection, and detailed case knowledge used in crafting several case studies.
Chapter 2 sets the stage by emphasizing the issue of nondecisions (Peter Bachrach and Morton Baratz, “Decisions and Nondecisions: An Analytical Framework,” American Political Science Review, 1963). For an area of scholarship that tends to focus on observable outcomes, such as recorded roll-call votes of legislators, it is an important reminder that studying legislative decisions can easily miss the fact that what matters most to some citizens might not even be on the legislative menu. The authors therefore focus on the changing agenda of Congress and individual legislators as captured by legislative speech in the House and Senate between 1995 and 2016. It is important to note that this conceptualization deviates from the notion of “agenda-setting” typically used in sequential (formal) models of legislative decision making in that it views the agenda as being continuously shaped and evolving both within and between Congresses. A potent example of this view can be found in chapter 7, a case study of the genesis of minimum wage legislation passed in the 110th Congress. The authors document legislators’ attempts to put the spotlight on the issue of minimum wage increases in the 109th Congress (generally by adding amendments). While these attempts might seem futile in terms of legislative outcomes, they did substantially shift the agenda by increasing speech concerning the minimum wage and forcing opposing legislators to at least acknowledge the issue in their speeches (pp. 243ff).
With a quantitative measure of Congress’s agenda in hand, the authors raise the question how it is shaped by different societal groups. They distinguish two sources of group power: structural and kinetic. Structural power refers to a group’s (socio)economic position, while kinetic power refers to the resources a group deploys toward shaping the agenda (such as money or information). Structural power has a material as well as an ideational component: Business groups are powerful actors not just because of their economic might but also due to policy makers’ perception that they are central to the performance of the economy (to an extent that, say, labor unions are not). This conceptualization is valuable and moves beyond simplistic capitalist-worker dichotomies, allowing the authors to capture differences between the structural power of finance and manufacturing. But note that in terms of measurement it remains a construct (Abraham Kaplan, The Conduct of Inquiry: Methodology for Behavioral Science, 1964, p. 55), that is, it is captured using an agreed-upon meaning rather than through a directly observable quantity (think “bureaucracy” vs. “labor income”). This is not a weakness of the book, but it points toward useful future work fleshing out the operationalization of these key concepts (and their application beyond the American political system).
Chapter 3 provides a detailed overview of the data. The authors machine code all speeches recorded in the Congressional Record between 1995 and 2016. Speech on any given issue (e.g., “inequality” or “the deficit”) can then be quantified through simple counts of words (or text fragments). This is done carefully and explained step-by-step in the text. Chapter 3 also contains captivating descriptive illustrations of Congress’s unequal attention to issues relevant to the interests of the rich versus middle- and low-income citizens. The authors are transparent about limits of their implementation (e.g., the difficulty of ascertaining positive vs. negative connotations of words in context). I am convinced that the created database will be of great interest to many researchers and will provide increasingly fertile ground for future exploration as technology (e.g., sentiment analysis) develops.
The second key ingredient of the authors’ quantitative analysis is a measure of campaign donations (a key example of kinetic power) compiled from Federal Election Commission records and matched to individual legislators. In chapter 4, the authors show that corporate donations increase the amount of speech concerning issues central to the rich, such as debt and the deficit, while donations from labor unions increase speech regarding issues such as inequality and wages. The authors’ results are summarized compactly in table 4.1 (p. 119) that reports the sign of the coefficients and their associated p-values (full tables are in an appendix). While the direction of effects is in line with the authors’ expectations, they note that (contrary to expectations) some relationships “just miss … statistical significance” (p. 120) as is the case for the relationship between corporate donations and speech regarding the deficit. Note that this analysis uses all speeches made during the period under study. Analyzing a population (not a sample) changes the meaning of the reported significance tests. Now, the targets of inference are not the members of Congress in the dataset, but rather the possible population of members and their behavior resulting from a hypothetical continued application of the current regime of unequal influence. I raise this point not to make a sales pitch for the Bayesian inferential paradigm, which only conditions on observed data (see chapter 2 of Simon Jackman, Bayesian Analysis for the Social Sciences, 2009, for a spirited discussion). Rather, lack of significance (alone) might not be the most useful tool to quantitatively assess the influence of unequal kinetic power. It would have been nice to see an investigation of how well the model fits the data, for example using simple cross-validation (leaving out legislators or whole Congresses).
Chapters 5 through 7 provide three detailed case studies of financial deregulation and reregulation and of the long and winding road to minimum wage increase legislation in the 110th Congress.
The three chapters are detailed and they insightfully “trace” the process of influence during legislative decision making drawing on a variety of secondary sources as well as the authors’ quantitative data.
Summa summarum, this is a commendable book combining excellent scholarship with engaging writing. I happily recommend it to you, and I will assign it to my students.
