In contemporary politics, there is no shortage of pundits and scholars identifying frustrated majorities (and governing minorities) as the root cause of our most recent “crisis of democracy.” In Democracy in America (2020), Benjamin Page and Martin Gilens, for example, make the case that the solution to America’s latest democratic crisis is to empower majorities so that public policy better reflects the public will. Seth Hill thinks differently. Frustrated majorities arise because political candidates are attempting to win popular elections by securing the most votes. They are not ignoring voters or are constrained by institutional design; they are simply responding to voter intensity in ways that increase the probability that they will be elected.

In his ambitious new book Frustrated Majorities, Hill sets out to explain why majorities in the American political system frequently lose to more committed minorities. At first glance, this is a story we know well. On issues like gun control and abortion, popular majorities lose to minority factions. These are issues where intensity of opinion, and not just direction of opinion, matters. No reader will be surprised by this observation. “Frustrated majorities” is perhaps the defining descriptor of the American political system. James Madison intentionally designed the US Constitution to frustrate majority factions driven largely by passion, rather than reason and, as a result, easily duped by demagogues and “pretended patriots.”

What is missing from popular and scholarly laments, according to Hill, is an explanation for why politicians appeal to committed minorities, rather than less committed majorities, as a viable (and perhaps even optimal) electoral strategy. Using game theory, Hill develops a model, based on what he coins “intensity theory,” for how this works. Candidates want to win the most votes, they know the preferences of voters on issues, but remain uncertain about the intensity of public attitudes. Within this context, intensity is revealed by the costs voters are willing to pay to achieve their policy goals. Intensity matters because candidates need to know which potential voters will vote according to their policy preferences, rather than partisan heuristics, candidate image, or some other consideration.

Candidates gauge intensity from the signals voters send through costly actions (e.g., emails, personal contacts, and campaign contributions). According to the model, voters intentionally and strategically signal their intensity through their actions. Candidates, in turn, signal their alignment with voters’ policy preferences via their policy statements, positioning, and platforms. Across issues, candidates assess the weighted intensity of voter preferences on any given set of policies to decide whether to support an intense minority or a less committed majority. The key here is that majorities can be frustrated in the absence of Madisonian institutional constraints or by interest group or money-based distortions of the popular will. Neither is necessary for majority frustration; you need only a plurality-based election system. There is one more piece to the puzzle: the conditional responsiveness to intensity, Hill shows, leaves all voters better off in a utilitarian sense when candidates are responsive to intense minorities rather than apathetic majorities.

The model here strikes us as well constructed in terms of its internal logic but questionable in terms of its applicability. With apologies to Jerry Lee Lewis, “there is a whole lotta signaling going on.” First, the model assumes candidates know where voters stand on the issues but not the intensity of their preferences. Candidate knowledge of voter preferences is, at best, knowledge that is laden with uncertainty and, at worst, misinformed. Candidates often mistake voter preferences, systematically overestimating certain types of preferences and underestimating others. This may be because candidates infer policy preferences from the same costly signals that they use to estimate intensity where voters “preach to the choir” (David Broockman and Timothy Ryan, “Preaching to the Choir: Americans Prefer Communicating to Copartisan Elected Officials,” American Journal of Political Science, 114, 2016). This has consequences for the types of policies that emerge from the political system and how closely those policies align with voter preferences (David Broockman and Christopher Skovron, “Bias in Perceptions of Public Opinion among Political Elites,” American Political Science Review, 112 2018). Hill is aware of this literature and cites it but largely discounts it.

Perhaps more problematic are issues where there is no majority opinion or no meaningful opinion at all, where constituents are uncertain or cross-pressured, and where opinions shift over the course of a campaign and elected officials must estimate what opinion informed by political messaging might look like after a political campaign (R. Douglas Arnold, The Logic of Congressional Action, 1990).

Second, despite the signaling noted earlier, the same candidates who know how voters align on the issues are assumed to be less knowledgeable about voter intensity. It is not clear why this would be the case. One might fairly argue that candidates would be better judges of intensity of opinion, given the information readily available to them via costly actions, and less certain of majority preferences. This would not hurt the model—we would still have an explanation for frustrated majorities based solely on electoral considerations—but it would fundamentally undermine the substantive conclusions that frustrated majorities pose little or no problem for democracy (and may even be a good thing).

In any event, to the extent that candidates rely on polling and not just personal observation, they should have access to information about the direction and intensity of voter preferences. Hill is, of course, correct that polling is laden with uncertainty. Pollsters are well aware of this uncertainty and attempt to gauge how changes in question wording, question order, and context alter survey response (John Zaller and Stanley Feldman, “A Simple Theory of the Survey Response: Answering Questions versus Revealing Preferences,” American Journal of Political Science, 36, 1992). Message testing, used to inform campaigns about how opinions are likely to shift over the course of a campaign, further illustrate pollsters’ understanding of opinion fluidity. Despite the challenges, pollsters also attempt to gauge the salience and intensity of political issues. These efforts are imperfect but often revealing. Hill seems to recognize this, using intensity measures (strongly agree–strongly disagree) in the first of his empirical chapters demonstrating the applicability of the model within the context of embryonic stem cell research. In other chapters, he uses measures of “costly actions,” a measure of intensity easily (and often) gauged via survey research, or “most important problem” items to gauge salience.

If we are interested in empirical generalization, there are even more perplexing measurement challenges. Can we measure intensity in a way that allows us to predict when majorities will be frustrated by committed minorities, rather than provide post-hoc explanations? If we measure intensity via costly actions, do actions ever become too costly, crossing the line from conventional to unconventional participation and undermining their own cause? What happens when we move from unconventional protests to political violence? Do costly political actions ever become counterproductive (Omar Wasow, “Agenda Seeding: How 1960s Black Protests Moved Elites, Public Opinion and Voting,” American Political Science Review, 114, 2020)? Even if we remain within the realm of conventional participation, how do we measure the cost of an action? Is the cost of a political contribution equal to the cost of an email? And what should we conclude about those voters (and nonvoters) who cannot pay the cost?

There are other issues. One reason why intensity matters more in American politics is that the aggregation of opinion is imperfect across congressional districts. In a system with proportional representation, majorities would still be frustrated, but the frustration would presumably occur less often. Related, frustrated majorities can occur even when a majority of elected representatives share the policy preferences of a political majority. According to Jake Grumbach’s (2022) new book Laboratories against Democracy, interest groups strategically search for states and localities where they can more easily and intentionally frustrate majorities. Overall, institutions still matter, and much of the frustration in American politics is baked into the institutional cake.

Finally, a minor but we think important point: Hill blurs the distinction between opinion intensity and opinion salience. In the public opinion literature, these are closely related but conceptually distinct. Voters can feel strongly about an issue (abortion), but it may be less important than other issues they might consider (the economy) when casting a ballot.

Overall, this ambitious book is well worth reading. The model is carefully constructed and tightly argued, and the subsequent empirical chapters provide supportive evidence. Hill does an excellent job assuring the book is accessible for less technically inclined readers, moving his proofs to the appendix and leaving the text for conceptual description of his model. The book is well written, informed, and appropriately provocative. Most students of American politics know well that intensity matters, so the central argument is not new, but Hill pushes the observation into new and challenging territory. Strategically, candidates might rationally appeal to committed minorities not to undermine democracy but because they want to win the most votes. Paired with Democracy in America, Laboratories against Democracy, or some similar text, lamenting the decline of majority rule or policy responsiveness would make for an interesting set of readings in advanced undergraduate or graduate-level courses. Our guess is that most readers will not buy entirely into Hill’s conclusions, but those conclusions will spur a lot of thought about whether frustrated majorities are a feature or a bug in a democratic political system. Challenging our preconceptions is what a good book does, and on this count Seth Hill’s Frustrated Majorities unquestionably succeeds.