The question of who exercises their right to vote, and who stays home, is central to political science, yet there is much we still do not understand. Two recent books shed new light through insightful yet distinct analyses. Despite much hand-wringing in the popular press, turnout is not down, both say; but neither is it what we might call good. Both are driven by survey and other quantitative data (sometimes the same data sets), yet they begin with diverging central questions. This makes for a fascinating comparison, leading to different—though not incompatible—answers.

In The American Nonvoter, Lyn Ragsdale and Jerrold Rusk look explicitly at those who do not vote, a clever perspective shift that illuminates gaps in our thinking. E. E. Schattschneider (The Semisovereign People, 1960: 102) famously referred to voter apathy as a “sickness” and to nonvoters as the “soft underbelly” of American democracy. Separating the purely ignorant from those who collect political information but do not vote for other reasons is critical in diagnosing that sickness. Rational abstention might be safe to ignore as a choice (although, if due to widespread apathy, it might point to a real problem with democratic governance [see, e.g., David T. Risser, “The Moral Problem of Nonvoting,” Journal of Social Philosophy 34(3), 2003: 348–363). The thwarting of an earnest desire to express through the vote “political voice” (the useful term from Sidney Verba, Kay Schlozman, and Henry Brady [Voice and Equality, 1995]), however, requires remedy.

Ragsdale and Rusk explicate four prior schools of thought on voting: psychological engagement, personal resources, social connectedness, and mobilization (Chapters 1 and 3). The conventional wisdom is that voters (versus nonvoters) have higher levels of education, age, and income, greater residential stability, social connections, and psychological engagement with politics, and are more likely to be mobilized in campaigning. Nonvoters are younger, poorer, more politically apathetic, and less knowledgeable, educated, mobilized, and engaged. True as far as it goes, Ragsdale and Rusk say, but demographics are not destiny. Comparing voters and nonvoters supports the conventional wisdom (Chapter 4), but does not necessarily explain nonvoters’ motivations. There are plenty of nonvoters with some college education and awareness of issues. (“[P]olitics,” they write on p. 92, “is often hard to ignore.”)

The missing piece, the authors argue, is uncertainty, which is always present to some extent in the “national campaign context” (p. 11), but can be high or low in any given year. (Campaigns are idiosyncratic, but Ragsdale and Rusk hypothesize an uber-context at the national level pervading state and local campaigns, even in non-presidential years.) Four indicators measure national uncertainty (p. 59): economic volatility (“sharp increases or decreases in . . . growth”); large technology changes (“a new form of political communication”); dramatic national events (including the U.S. involvement in international conflicts); and a federal expansion of voting rights (which “shock[s] . . . the political status quo”). The devil, of course, is in the details. What counts as “large” in terms of technological changes or major international/national conflicts, all of which dominate news lately? (For a work hinging on the exact definition of uncertainty, the book could offer more by way of clear criteria.)

Uncertainty falls along a continuum, and “the higher the level of uncertainty in the national campaign context, the lower the level of nonvoting” (p. 24). Theirs is not entirely a Downsian model, but nonvoting, Ragsdale and Rusk believe, is often rational. Uncertainty interacts with the factors identified by previous schools of thought (demographics, etc.) in important ways, such that “There is no uniform answer to the question, ‘Who does not vote?’” (p. 95). Why individuals who could cast ballots do not “lies in the mix among the campaign context [“external uncertainty”] and candidate preferences [“internal uncertainty”], as well as individual backgrounds, mobility, political interest, and campaign contact” (p. 125). The question is twofold: is national uncertainty high or low, and can individuals find a candidate whom they trust to do a good job in the face of that “external uncertainty?” Voting is only rational if citizens can find a candidate who satisfies this “internal uncertainty.” (What this implies, interestingly, is that “people decide for whom to vote before they decide whether to vote”; p. 93).

Specifically, nonvoters come in four flavors (Chapter 4): the politically ignorant (the category that comes closest to the conventional wisdom); the indifferent; the dissatisfied; and those experiencing personal hardship (operationalized, unfortunately, as being unemployed [p. 115], which leaves something to be desired; why not poverty? Politics easily falls by the wayside in the daily struggle for survival). This is a fresh take, and it has the great advantage of ringing true with common experience, as well as fitting the data (which the authors parse carefully, in particular time periods and an overall historical spread). It is an impressive array of data and analysis, although their explanations of methods and findings sometimes feel rushed or not entirely accessible for, say, undergraduates or a more general public. It is particularly useful that they offer a South versus non-South analysis, rather than assuming homogenous effects across regions.

Ragsdale and Rusk are thoughtful and engaging writers. Their work is thorough and often persuasive, particularly in discussing specific types of nonvoters and reasons for not voting. “If the candidates are unattractive—‘tweedle-dum, tweedle-dee’ candidates who do not seem distinct from each other, or the ‘lesser-of-two-evils’ candidates who are similarly bad at addressing the national campaign context—then people will be less likely to vote,” they write (p. 93). Certainly we can recognize the theoretical sense, as well as real-life data, behind this idea. The concepts underlying their theory are nicely explained, although some will quibble with certain operationalizing moves. And scholars in the “mobilization” and “resources” camps could be forgiven for thinking their models are somewhat undersold. Yet overall, this work offers an exciting addition to our understandings, one that raises questions and leads to new directions in research.

Jan Leighley and Jonathan Nagler offer a more traditional approach, rooted firmly in the idea that “demographics matter,” but with a few terrific twists. Who Votes Now (many decades in the making) is less ambitious in scope, and less lively in tone, than Ragsdale and Rusk’s book, but more careful in execution and explanation. It does not seek to provide a new theory but updates and adds to the seminal Who Votes (Raymond Wolfinger and Steven Rosenstone, 1980). Not surprisingly, it has become an instant classic in the “resources” school. “The demographics of turnout,” they write, “are inherently important to questions of democracy and representation” (p. xv).

Leighley and Nagler’s first and perhaps most important move is to assert (later shown empirically, Chapter 6) that preferred policies differ between voters and nonvoters; in this, these authors evidence a departure from the Wolfinger-Rosenstone model. While partisan or even candidate preferences may not differ, policy preferences do, and “[P]olicy makers cater more to the wishes of voters than nonvoters. . . . [W]ho votes matters for the most basic outcome of politics: who gets what” (pp. 2–3). In particular, they are concerned with the “income gap” (more accurately “wealth gap”) between voters and nonvoters; this is highly salient, because even “over a period of increasing economic inequality, the income bias of voters has remained the same” (p. 3). Voters are not representative of nonvoters economically or in redistributive stance. Increasing economic inequality can exacerbate unequal turnout, leading to alienation (and, one could add based on 2016, deep anger) when many feel that the political system is failing them (p. 9).

Another departure from the original demographics model bears mention; if you firmly believe easier voting (same-day registration, mail-in ballots) will solve these problems, Leighley and Nagler will break your heart. They find modest effects, but no silver-bullet solution; reforms too often disappoint, have unintended consequences, or, even when successful, perversely increase inequality while increasing turnout (p. 10; and Chapter 4). Yet they do find a surefire way to increase voting: “[W]hen candidates offer distinct choices, eligible voters are more likely to turn out” (p. 3; and Chapter 5). (On this, Ragsdale and Rusk likely would agree, in the interest of decreasing “internal uncertainty.”) As Leighley and Nagler write: “The blur of modern campaign politics often seems to overlook, or even forget, that elections are choices not just between candidates or parties, but also between issue positions” (p. 139). Better campaigns and clearer elite/media discussion of policy choices, in addition to better voting laws, are needed.

Throughout, Leighley and Nagler are careful to explain each step, often also recounting difficulties involved in operationalization or analysis. They are cautious with language, sometimes overly (see, e.g., the discussion of “perceived policy alienation” on p. 125). While I appreciated such attention to detail, it may make for difficult reading for those not as enthralled with the nuances of survey wording or previous literature. The authors also include, with each chapter, an appendix for the data nerds among us, although this too might turn off some readers, like my students, who, I fear, run at the sight of Greek letters. The many well-crafted charts, however, should help with accessibility.

On the whole, those interested in voting are in for a treat with either (but preferably both) of these books. Also worth noting is that both attest to the continued importance of American National Election Study (ANES) data and are excellent models of collaborative thinking. Graduate students take note: a terrific project for enterprising young scholars (with lots of time) would be combining these theories. How might “engagement,” “interest,” “mobilization,” and “uncertainty” relate? Or these models could become stepping-stones toward greater complexity. Thoughtful digging would likely find that certain factors (uncertainty, issues, engagement, mobilization) matter more for certain groups (demographic, but not only) in producing turnout—knowing for whom and how this works would be a wonderful contribution. These works give the most complete picture to date of who does and does not vote and why, but there is plenty more to be done.