Looking But Not Seeing: The (Ir)relevance of Incentives to Political Ignorance

ABSTRACT

Ilya Somin's Democracy and Political Ignorance represents a missed opportunity to fully examine the implications of public ignorance in modern democracies. Somin persuasively argues that existing levels of public ignorance undermine the main normative accounts of democratic legitimacy, and he demonstrates that neither cognitive shortcuts nor heuristics can provide a quick fix for democracy. However, Somin seeks to find a simple explanation for public ignorance in the conscious, rational choices of voters. He thus commits to the position that voters choose to be ignorant and irrational—and to the simplistic implication that given the right incentives they would choose otherwise. This position is empirically problematic, methodologically flawed, and theoretically redundant. On the more plausible view that ignorance is the inadvertent result of social complexity, it is clear that simply focusing on incentives tells us little about what voters would or would not know under different institutional circumstances.

Keywords:

• Ilya Somin
• political ignorance
• public ignorance
• radical ignorance
• rational ignorance
• rational irrationality

Notes

1. For reasons of space, this claim will have to remain unsubstantiated for now. Suffice it to say that while theorists advocating a largely or fully procedural account of democracy might claim that their conceptions of democracy confer legitimate authority regardless of how informed citizens are, such accounts beg the question of how the procedural process of democracy can be insulated from the technocratic functions of the state without robbing the citizenry of everything but symbolic significance. If citizens actually care about changing the conditions in their societies, as would plausibly seem to be the case, then epistemic issues apply regardless of which democratic standard we use to judge their actions and institutions.

2. Most famously associated with Charles Tiebout's 1956 formal model of competitive federalism and Albert Hirschman's 1970 account of economic decision making, variations of foot voting/exit rights have been embraced in philosophical treatments of individual liberty (Nozick 1974, pt. III; Kukathas 2003), as a key mechanism within various normative theories of democracy (Okin 2002; Hirst 1994; Cohen and Rogers 1995), and even as a central mechanism within existing social democracies (Dowding and John 2008; Warren 2011).

3. Somin provides a possible explanation for this on page 119, on which, alongside citations to Tiebout 1956 and Warren 2011, he suggests that the informational effects of foot voting have been all but ignored. His own approach, based as it is on the theory of rational ignorance, is thus a significant departure from the existing foot-voting literature.

4. There are two near-exceptions to this point. The first is Somin's elaboration of the concept of foot voting in his brief discussion of “private planned communities” (137–39). Compared with “traditional interjurisdictional competition between regional and local governments,” he contends, small private communities have stronger incentives to be responsive to citizens’ demands (because they have no other sources of funding), are able to emulate the structure and operation of private firms in the market, and run a much lower risk of citizens forming “irrational” attachments (138). Put simply, foot voters living in such communities should be more like straightforward consumers, evaluating their possible options in terms of their affordability and value rather than irrelevant or specious political considerations. Yet this conception of foot voting makes too strong a distinction between jurisdictional competition and markets. After all, the competition Somin seeks to realize in practice would be precisely between different conceptions of political community; to the extent that irrationality is likely to overwhelm anything more than a particularly private attachment to one's neighbors, it is difficult to see how foot voting could ever work within politics. In fact, it is more plausible to argue in the other direction, that really-existing competition between political communities (and even between nation-states) already promotes rational movement between jurisdictions on the basis of precisely that kind of reasoning set out in Somin's example. On this view, the pertinent question is therefore not how to incite more competition (unless one wishes to assume, come what may, that more competition is always the answer), but rather why existing competition has not realized the benefits Somin hypothesises.

Somin's second illustration of foot voting concerns the migration of black Americans to Northern states to escape Jim Crow laws. Specifically, Southern blacks sought to move not only to take advantage of the better economic opportunities available in the North, but also to escape the racist persecution they were subject to in the South (129). Despite being in many cases extremely poorly educated, and with access to far less information than modern voters, individual migrants were able to learn of the better conditions and possibilities by way of the media and word of mouth (130). As a result, Southern state governments came under pressure to improve conditions for remaining blacks, who were a significant part of the labor force (131–132). This example clearly demonstrates the value of the principle of the freedom of movement, and of the importance of exit as a fail-safe against illiberal government power. However, the example of migration away from the Jim Crow laws does not take us very far in considering how the ability to exit would moderate or improve government performance today. The issue here is what led racist Southern governments to accede to the belief that racism is morally abhorrent. As Somin readily accepts (132), migration as such did not do much to dent racism against remaining blacks (though it did clearly help those who migrated). Rather, it was the concerted political pressure exerted first on, and then by, the federal government as a result of the Civil Rights movement in the 1950s and 1960s. Black migration very plausibly contributed to this pressure (as blacks in Northern states had greater political freedom), but the mechanism in question was clearly the newly dominant belief that abhorrent racist practices had no place in American politics. If this example is to tell us anything about contemporary politics, we therefore need to address the question of how decisions to exit in the face of much less morally clear-cut problems (such as the questions of how to address poverty or joblessness, neither of which have obvious answers to be found in the Declaration of Independence), would generate the necessary beliefs either in central or local government to solve those problems.

5. See Bennett and Friedman 2008 for a detailed inquiry into whether Caplan's theory can function without the attribution to voters of knowingly false opinions.

6. Though published after Democracy and Political Ignorance, The Pew Research Center's 2014 survey From ISIS to Unemployment: What Do Americans Know? repeats the findings of both the 2007 study cited by Bennett and Friedman and intervening studies: “On nine of the 12 questions included in the survey, only half or fewer get the answers correct” (Pew Research Center 2014, 2). In complementary evidence from the United Kingdom, the Hansard Society found that self-reported interest in politics among British voters varied from between a high-water mark of merely 42 percent of voters self-reporting as not very or not at all interested politics in 2011 to 58 percent self-reporting in the same way in 2012 and 2013, before falling to 50 percent (within 1 percent of the long-run mean of 49 percent) in 2014 (Hansard Society 2014, 34). Across the same years, 56 percent, 58 percent, and 50 percent of respondents, respectively, reported knowing either nothing at all or “[n]ot very much” about politics (ibid., 36). To paraphrase Bennett and Friedman 2008, 12, it is implausible that individuals keen on getting political information for personal gratification would either be as ignorant about politics as the Pew research suggests or as uninterested as the Hansard research finds. Again, Somin might ask in reply: What about those voters who are knowledgeable and interested in politics? Perhaps these are the rationally irrational voters; the former may simply be rationally ignorant. Putting aside the ad-hoc nature of this (hypothetical) claim (discussed above), it is once again belied by the evidence. Building on Larry Bartels's 1996 finding that voters are underinformed relative to their value preferences (a finding that reinforces the points made above), research in Britain at the turn of the century shows that not only do voters more accurately identify the party matching their values the more knowledge they possess (Heath et al. 2002), but that the actual content of these values is changed by the level of knowledge voters possess (Andersen et al. 2001; Heath and Tilley 2003). Put differently, whereas the theory of rational irrationality predicts that values should guide knowledge acquisition, the evidence suggests that knowledge guides values. This by no means provides a settled understanding of just how voters understand their choices (Heath and Tilley 2003, 23–24), but it does demonstrate the importance of engaging with the empirical research in this area, and it suggests the unreliability of accounts, such as Somin's, which assert psychological causalities without such engagement.

7. Somin does not elaborate on whether the form of decentralized state he advocates would be one approximating the night-watchman state or whether it would simply entail more comprehensive federalism. In practice, however, it makes little difference. Whether public goods are provided through public or private organizations, individuals face the same task of identifying who has the better understanding of how to provide public goods in the most effective and least costly way possible.

8. This claim is supported by an overview of the causes and individual effects of internal migration (Greenwood 1997), which demonstrates that employment has a complicated, and only partially understood, relationship with other variables affecting migration rates.