I. Introduction
The strengths and weaknesses of constitutional federalism have been debated for centuries. Similarly, we have had centuries of debate over the extent to which there should be constitutional constraints on the scope of government power more generally.Footnote 1 But one major possible advantage of building decentralization and limited government into a constitution has been largely ignored in the debate so far: its potential for reducing the costs of widespread political ignorance.Footnote 2
The argument of this essay is simple, but has potentially far-reaching implications: Constitutional federalism enables citizens to “vote with their feet,” and foot voters have much stronger incentives to make well-informed decisions than conventional ballot box voters. The same goes for limits on the scope of government that enable citizens to vote with their feet in the private sector.Footnote 3
The informational advantages of foot voting over ballot box voting suggest that decentralized federalism can increase both citizen welfare and democratic accountability relative to policymaking in a centralized unitary state. Since at least the pioneering work of Charles Tiebout,Footnote 4 scholars have analyzed foot voting extensively, but its informational advantages over ballot box voting have largely been ignored.
These advantages are important both for those who believe that political ignorance is instrumentally harmful because it undermines democratic accountability in government, and for those who value democratic control of public policy for its own sake.Footnote 5 Widespread political ignorance undermines each of these potential benefits of democratic government.Footnote 6
In Section II, I briefly elaborate on the theory of rational political ignorance, explaining why most ballot box voters have little incentive to acquire more than minimal political knowledge. In addition, the theory of rational ignorance implies that voters will often make poor use of the usually very limited knowledge that they do possess. The empirical evidence on political knowledge generally supports both of these predictions.
Section III shows that incentives for knowledge acquisition are much stronger when citizens have the option of voting with their feet rather than just at the ballot box. The same holds true for incentives to make rational use of the information that is acquired. Reductions in transportation costs and in identification with state governments in recent decades both increase the likelihood that foot voters will acquire adequate information about their alternatives, and will use that information rationally. These points hold true for the United States but may be less valid in countries where federalism tracks deep ethnic or religious divisions that make it difficult for citizens to migrate from one region to another or to admit that a regional government dominated by an ethnic group other than their own is doing a better job of governance than the one where they currently reside. With these qualifications, the benefits of foot voting strengthen the case for constitutional limits on central governments in order to facilitate decentralization.
Section IV considers some empirical evidence indicating the informational superiority of foot voting over ballot box voting. In particular, I summarize evidence demonstrating that even a severely oppressed and often poorly educated group can acquire sufficient information to engage in effective foot voting. That is exactly what happened with numerous poor African Americans in the Jim Crow era South who acquired sufficient information to realize that conditions were relatively better for blacks in northern states, and also better in some parts of the South compared to others. The resulting migration significantly bettered the condition of African Americans throughout the nation.
Section V suggests that limits on the scope of government have some of the same informational advantages as political decentralization. Indeed, those advantages may be even greater in light of the fact that exit costs are usually lower in private sector markets than in the case of interjurisdictional migration. In Section VI, I discuss some implications of my analysis for constitutional design. The informational benefits of foot voting strengthen the case for constitutional limits on central governments in order to facilitate decentralization. They also strengthen the case for limits on the power of government relative to the private sector.
Finally, Section VII provides a conclusion that summarizes the implications of my analysis and notes some of its limitations.
The argument presented here is comparative. It holds that foot voting has significant informational advantages over ballot box voting, not that foot voting is without difficulty or that it overcomes all information problems completely. However, comparative analysis is important because foot voting and ballot box voting are the most important realistic alternatives facing many societies across a wide range of issues.
II. The Logic of Rational Political Ignorance
Scholars have long recognized that most citizens have little or no political knowledge.Footnote 7 An individual voter has virtually no chance of influencing the outcome of an election—somewhere between 1 in 10 million and 1 in 100 million in the case of a modern U.S. presidential election, depending on the state where one lives.Footnote 8 The chance of casting a decisive vote is somewhat greater in other democracies with smaller populations, but is still extremely small.Footnote 9
As a result, the incentive to accumulate political knowledge is vanishingly small, so long as the only reason for doing so is to cast a “better” vote. Even highly intelligent and perfectly rational citizens could choose to devote little or no effort to the acquisition of political knowledge. The theory of rational ignorance implies that most citizens will acquire little or no political knowledge and also that they will often make poor use of the knowledge that they do acquire. Both political knowledge acquisition and the rational evaluation of that information are classic collective action problems, in which individual citizens have incentives to “free ride” on the efforts of others.Footnote 10
Some critics reject the rational ignorance theory on the ground that it allegedly also predicts that citizens will choose not to vote.Footnote 11 However, as Derek Parfit has demonstrated theoretically, and Aaron Edlin et al. have supported with empirical evidence,Footnote 12 the decision to vote is rational so long as the voter perceives a significant difference between candidates and cares even slightly about the welfare of fellow citizens, as well as his own. For example, if an American voter believes that the victory of the “right” candidate will give to each of his 300 million fellow citizens an average of $5,000 in net benefits, then it is rational to vote even if the chance that one's vote will be decisive is less than 1 in 100 million. The enormous benefit of casting a decisive vote outweighs the very low probability that it will happen. This remains true even if the potential voter values the welfare of fellow citizens who are strangers to him far less than he values his own welfare (e.g., valuing a $1,000 benefit to a fellow citizen as equivalent to $1 for himself).Footnote 13 In a 2006 publication, I extended the Parfit and Edlin analysis to show that, for most citizens, it will usually be irrational to acquire significant amounts of political information for voting purposes, even though it is rational to engage in voting itself.Footnote 14
A. Low levels of political knowledge
We cannot know for certain that the theory of rational ignorance is correct. But the available evidence strongly supports it. There is little doubt, for example, that political knowledge levels are extremely low, and have been so at least since the start of modern survey research in the United States in the 1930s.Footnote 15
It is impossible to summarize the extensive evidence of widespread ignorance here. However, it is important to note that it covers a wide range of issues. The majority of citizens are often ignorant of the facts of specific policies, even very important ones. For example, around the time of the 2004 election, 70 percent of Americans were unaware of the recent passage of President George W. Bush's prescription drug bill, the largest new government program in decades.Footnote 16 Citizens are often also ignorant about the general “rules of the game” of politics, such as determining which officials are responsible for which issues.Footnote 17 And studies have long demonstrated that most citizens do not understand the basics of political ideologies such as liberalism and conservatism.Footnote 18
Absent the rational ignorance hypothesis, it is difficult to explain the fact that political knowledge levels have remained roughly stable at very low levels for decades, despite massive increases in education levels and in the availability of information through the media and now the Internet.Footnote 19 The rational ignorance theory is also confirmed by data indicating that the strongest predictor of political knowledge—more important even than education—is the survey respondent's level of interest in politics.Footnote 20 This supports the prediction that those who acquire political information will do so primarily for reasons other than a desire to become better voters. Instead, they view political information as a consumption good, much as dedicated sports fans will acquire knowledge about teams and players despite the fact that they cannot affect the outcome of games.Footnote 21
B. Illogical use of knowledgeFootnote 22
The theory of rational ignorance suggests not only that most citizens will acquire little political knowledge, but that they will often make poor use of the information they do possess. The key consideration is that the theory of rational ignorance does not predict that voters will choose not to acquire any information at all. Rather, it predicts that they will acquire very little or no information for purposes of voting. Footnote 23
Some voters, however, will learn political information for other reasons. Obviously, scholars, politicians, political activists, journalists, and others have professional reasons for being informed about political developments. However, such professional consumers of political information are only a tiny fraction of the population. Far more common are those who acquire political knowledge because they find it interesting.Footnote 24 There are not enough such people to eliminate widespread political ignorance, but they do nonetheless form by far the largest bloc of relatively well-informed voters.
Citizens with a strong interest in politics often function like “fans” cheering on their preferred ideology or political party, rather than as rational assessors of information.Footnote 25 They evaluate information in a highly biased manner that tends to confirm rather than objectively test their preexisting views.
This prediction is supported by studies showing that people tend to use new information to reinforce their preexisting views on political issues, while discounting evidence that runs counter to those views.Footnote 26 Although some scholars view such bias as irrational behavior,Footnote 27 it is perfectly rational if the goal is not to get at the “truth” of a given issue in order to be a better voter, but to enjoy the psychic benefits of being a political “fan.” Rationally ignorant voters may limit not only the amount of information they acquire but also “how rationally they process the information they do have.”Footnote 28 To put it a different way, such citizens' mode of processing information may be rational for purposes of psychic gratification, but irrational for purposes of improving the quality of their votes. Pursuing the former at the expense of the latter is itself perfectly rational, since the chance that any one vote will be decisive to an electoral outcome is infinitesimally small.
A recent study shows that the most knowledgeable voters tend to be more biased in their evaluation of new evidence than those with less preexisting political information.Footnote 29 Similarly, large numbers of relatively knowledgeable citizens believe ridiculous conspiracy theories about political events, such as claims that the 9/11 terrorist attacks were an “inside job.”Footnote 30
If those who acquire political knowledge do so in order to cast “better” votes, such findings would be difficult to explain. But if, as the rational ignorance hypothesis implies, the main goal is to enjoy psychic benefits similar to those available to sports fans, then the greater bias of the more politically knowledgeable is perfectly rational. The fact that they acquired more knowledge in the past suggests that they value the “fan” experience more than those who acquired less; thus, it is not at all surprising that they tend to be more biased in their evaluation of new data.
C. Information shortcutsFootnote 31
Many scholars have argued that voter ignorance is not a significant problem because citizens can offset their ignorance through the effective use of information shortcuts.Footnote 32 For example, voters who know little else about a candidate can still determine a lot of useful information about him or her simply by knowing his or her party affiliation.Footnote 33 Voters who know little about the details of public policy can use “retrospective voting” to punish the party in power for poor performance if things generally seem to be going badly under its rule.Footnote 34 Elsewhere, I have criticized shortcut theories on the ground that citizens must have a significant preexisting base of knowledge to use them effectively.Footnote 35 For example, retrospective voting can only be effective if citizens know which political leaders are responsible for which issues, and whether or not their policies were more effective than available alternatives. Otherwise, they might end up “punishing” incumbents for policy issues that they have no control over or for enacting policies that actually produced better results than those of their opponents would have.Footnote 36 In recent years, some scholars who were previously highly enthusiastic about the utility of shortcuts have partially conceded the validity of such criticisms.Footnote 37
Moreover, even if information shortcuts can in theory provide an adequate substitute for more detailed knowledge, the theory underlying shortcuts implicitly assumes that voters will (1) choose the right shortcuts and (2) use those shortcuts in a logical manner to assess opposing candidates and parties. However, if rationally ignorant voters economize not only on the acquisition of information but also on the degree to which they assess it in a logical manner, these assumptions are unlikely to hold true. Empirically, voters often pick poor shortcuts or fail to use them logically. Even the most sophisticated and highly rational voters may rely on shortcuts that have little relevance to political candidates' likely performance in office. For example, a recent study of elections for the presidency of the American Economics Association (AEA) shows that the relative physical attractiveness of the rival candidates is a powerful predictor of which candidate prevails in the voting.Footnote 38 The AEA electorate consists of academic economists who are presumably knowledgeable about the functions of the AEA—and arguably more committed to rational, maximizing behavior than is the average voter in ordinary elections. If AEA voters nonetheless rely on dubious information shortcuts, it is likely that voters in other elections also do so.
Recent evidence suggests that even the most knowledgeable might systematically pick ideological shortcuts that mislead more than they inform. A study of the accuracy of predictions by experts in politics and international relations finds that their predictions of political events are generally no more accurate than would be produced by random chance.Footnote 39 Of greater interest for present purposes is the finding that the most inaccurate experts are those who tend to make their predictions on the basis of broad generalizations—that is, experts who rely the most on ideological shortcuts.Footnote 40
Such behavior is consistent with the prediction of rational ignorance theory that citizens who acquire and evaluate political information do so primarily for reasons other than becoming better voters. Biased and illogical evaluation of political information is perfectly rational behavior for people whose goal in acquiring information is something other than truth-seeking.
None of this suggests that information shortcuts are completely useless. Indeed, in my view, they often have some benefits. However, for present purposes, the only necessary inference is that information shortcuts fall far short of fully offsetting the detrimental effects of rational political ignorance. To the extent that this is true, foot voting is likely to have important informational advantages over ballot box voting.
III. Foot Voting Versus Ballot Box Voting
Foot voting provides much stronger incentives than ballot box voting for both information acquisition and rational information use. People voting with their feet are largely free of the collective action problems that lead to rational ignorance in the political process.
A. Information acquisition
As we have seen, one of the main causes of political ignorance is the fact that it is “rational.” Because even an extremely well-informed voter has virtually no chance of actually influencing electoral outcomes, he or she has little incentive to become informed in the first place, at least if the only purpose of doing so is to cast a “correct” vote. By contrast, a person “voting with her feet” by choosing a state or locality in which to live is in a wholly different situation from the ballot box voter. If a “foot voter” can acquire information about superior economic conditions, public policies, or other advantages in another state, he or she can move to that state and benefit from them even if all other citizens do nothing. This creates a much stronger incentive for foot voters to acquire relevant information about conditions in different jurisdictions than for ballot box voters to acquire information about public policy. Unlike in the case of ballot box voters, information acquisition by foot voters is largely exempt from collective action problems.
In most cases, foot voters also don't need to acquire as much information as ballot box voters in order to be adequately informed. Unlike a ballot box voter, a foot voter need not connect his judgment of relative conditions in various states to specific elected officials and their policies.Footnote 41 If voters don't realize which officials are responsible for which issues or don't separate out the impact of public policy from that of other social conditions, they may end up punishing or rewarding incumbent officeholders for outcomes over which they have no control.Footnote 42 The officeholders themselves can try to take credit for positive developments that were not really caused by their policies. Presidents, for example, try to take credit for any economic prosperity that occurs during their term in office, even if they did little or nothing to cause it.
By contrast, foot voters don't need comparably detailed knowledge. It is enough for them to know that conditions are better in one state than another, and then be able to act on this knowledge by moving. So long as public officials themselves know that their policies can affect social conditions in ways that attract foot voters, they will have an incentive to implement better policies in order to appeal to potential migrants. Not only does foot voting create a stronger incentive to acquire knowledge than ballot box voting, it also usually requires less knowledge to implement effectively.
B. Information use
In addition to providing superior incentives for information acquisition relative to ballot box voting, foot voting also improves incentives for rational information use. Part of the reason for this is the same as that which underlies foot voters' superior incentive to acquire information: the absence of a collective action problem. But there are also other reasons to expect foot voters to make better use of the information they acquire than ballot box voters do.
As noted above, people have a strong tendency to process political information in a highly biased way that tends to confirm their preexisting ideologies and prejudices. This is true of both ordinary voters and political activists and experts. By contrast, most modern Americans lack the same kind of commitment to their states that many have to their ideologies and partisan affiliations. Over the last hundred years, citizen identification with state and local governments has largely faded away in most parts of the country, replaced by a sense of national identity as Americans.Footnote 43 For that reason, people are likely to be more objective in analyzing information bearing on their decisions about where to live than their decisions about who to vote for. The latter decisions implicate strong partisan and ideological commitments, and sometimes also ethnic or religious ones. The former—at least in the modern United States—usually do not.
Some scholars claim that the decline of identification with state governments is an argument against federalism,Footnote 44 because citizens no longer have a sense of “community” that is linked to state government. However, citizens' lack of commitment to their states and localities facilitates effective foot voting, and to that extent actually strengthens the case for devolution of power away from the center.
Obviously, this point applies with much lesser force to countries where regional governments are the focus of ethnic or ideological loyalties. For example, French Canadian nationalists may be strongly attached to Quebec and reluctant to move to other provinces even if the latter have better policies. Even in federal systems of this type, however, foot voters may still be more rational in their evaluation of information than ballot box voters if their ethnic group is in the majority in more than one jurisdiction. For instance, Switzerland has multiple French-, German-, and Italian-speaking cantons. While German-speaking Swiss may be reluctant to migrate to a French-speaking canton, the same might not hold true for movement between different German-speaking jurisdictions. Moreover, even in the relatively rare cases where foot voters' biases in favor of their home jurisdiction are as powerful as those of ballot box voters in favor of their ideology or party, the former will still have stronger incentives to try to overcome their biases because of the absence of a collective action problem.
C. The role of interjurisdictional competition
Interjurisdictional competition also improves the acquisition of information by foot voters. States and localities seek to attract new residents and businesses as sources of tax revenue. Therefore, state and local governments have strong incentives to establish policies that will appeal to potential immigrants and convince current residents to stay.Footnote 45 The power of the competitive pressure comes from governments' constant need to attract additional revenue to finance expenditures that can pay off key interest groups and increase political leaders' reelection chances. Interstate and interlocality competition for residents facilitates the creation of public policies that advance the interests of the majority, even in the absence of informed ballot box voting.Footnote 46
In addition, competition gives both state governments and private organizations incentives to disseminate information about the advantages of living in one jurisdiction as opposed to others. While the same is true of competitors for political office, information disseminated by competitors in the foot voting market is arguably less likely to be inaccurate or misleading than political advertising. Since foot voters have strong incentives to examine information more closely than ballot box voters, competitors in the former market are less likely to get away with deceptive or overly simplistic claims than those in the latter. Political rhetoric and advertising routinely employ misleading or deceptive rhetoric and claims,Footnote 47 some of which are quite effective. By contrast, as I discuss in Section IV, evidence suggests that competitors in the market for foot voters generally disseminate relatively accurate information to their “consumers” even in cases where the latter are extremely poor and ill-educated.
D. Implications for group migration
The informational advantages of foot voting over ballot box voting potentially extend to group migration, as well as migration by individuals and families. Historically, religious and ideological groups have sometimes chosen to migrate together in order to establish a community where they could live in accordance with their principles. Well-known examples from American history include the Pilgrims' migration from Europe to establish their colony at Plymouth, and the migration of the Mormons to Utah.Footnote 48 Such group migration differs from moving decisions by individuals or families because it requires coordination across a larger number of people, and is often undertaken for the purposes of establishing a community based on a specific religion or ideology.
Group migration within a federal system need not involve the establishment of an entirely new state or province, as happened with the Mormons. It could simply mean moving from one existing state to another whose policies are more hospitable to the group's purposes. For example, Mennonite religious groups in Canada and the United States migrated west without any intention of establishing their own state or province.Footnote 49
Some aspects of group governance may raise informational problems similar to those that arise from ballot box voting. In a large group with a democratic governance structure, individual members might have little incentive to acquire information on which to base their votes. They could be rationally ignorant for much the same reasons as most voters in the political system are. However, group migration does generally include an individually decisive decision on the part of each member to join the group in the first place, and often an additional decision to choose to migrate along with the other members instead of staying behind. For these reasons, participants in group migrations probably have stronger incentives to acquire and rationally evaluate relevant information than do ballot box voters.
E. The problem of moving costs
The most obvious drawback of foot voting relative to ballot box voting is the problem of moving costs. People who migrate from one jurisdiction to another must pay the cost of transporting themselves and their possessions, as well as assume the burden of finding new jobs and social ties. In some cases, these costs will prevent foot voting even in situations where another jurisdiction might be more attractive to the potential migrant than her current home.
This essay is not a comprehensive evaluation of the costs and benefits of foot voting. It focuses on its advantages with respect to information-gathering. Thus, I will not fully consider the issue of moving costs. Nonetheless, a few brief remarks are in order. First, moving costs are not so great as to preclude interjurisdictional mobility for millions of people. A recent Pew survey finds that 63 percent of Americans have moved at least once in their lives, and 43 percent have made at least one interstate move.Footnote 50 Contrary to claims that foot voting is an option primarily for the affluent, census data finds that households with an income under $5,000 per year are actually twice as likely to make interstate moves as the population as a whole.Footnote 51 As I discuss in Section IV, historically poor and oppressed populations have often taken advantage of foot voting opportunities. Increases in societal wealth and improvements in transportation technology have made migration cheaper than ever before. While moving costs continue to be a shortcoming of foot voting, they fall far short of vitiating its informational advantages. In Section V below, I discuss how private planned communities and interjurisdictional competition that does not require physical mobility can reduce the impact of moving costs still further.
IV. The Power of Foot Voting under Adverse Conditions: African Americans in the Jim Crow Era South
To my knowledge, there has not yet been a study that empirically documents the informational advantages of voting with your feet over ballot box voting. It is difficult to construct a research design that gets at the issue directly. However, there is telling historical evidence of impressive information acquisition by foot voters even under extremely adverse circumstances. The case of African Americans in the Jim Crow era South (roughly 1880–1960) is a particularly noteworthy example. During that time, southern state governments adopted a wide variety of laws discriminating against and oppressing their black populations; this extensive system of racial oppression was collectively known as “Jim Crow.”
If information acquisition for foot voting could be effective under the severely adverse conditions endured by southern blacks in the Jim Crow era, it is likely to be at least equally effective in other, less extreme circumstances. Moreover, the limited available evidence suggests that black southern foot voters were better-informed than the (on average) wealthier and more educated southern white ballot box voters of the same era.
A. Southern black migration during the Jim Crow era
African Americans in the Jim Crow era South, most of them poorly educated and many illiterate, were able to learn enough information about the existence of relatively better conditions in other states to set off a massive migration to the North and also to parts of the South that were relatively less oppressive than others.Footnote 52 Between about 1880 and 1920, over one million southern-born African Americans migrated to the North or the West.Footnote 53 By 1920, these migrants accounted for some 10 percent of the total black population of the United States, which then stood at 10.4 million.Footnote 54 There was an even larger black migration from South to North in the years immediately following World War II.Footnote 55 The earlier migration, however, is of special interest for present purposes, because during this period southern blacks were even more severely disadvantaged than during the later one and would have found it more difficult to acquire information about migration opportunities.
In addition to migration from the South to other parts of the country, there was also extensive African American population movement within the South itself.Footnote 56 Intraregional migration was often driven simply by the search for economic opportunity, but also by differences among southern political jurisdictions in the degree to which they oppressed the local black population.Footnote 57
Southern blacks in the early twentieth century labored under severe disadvantages that one might expect to prevent effective foot voting. Most were extremely ill-educated, in part as a deliberate result of state government policy. As late as 1940, only 5.4 percent of southern blacks over the age of twenty-five were high school graduates, compared to 26.1 percent of contemporary American whites.Footnote 58 Even those southern blacks who did have access to education nearly always attended inferior segregated schools that were deliberately structured to provide only very limited education for black students.Footnote 59 In light of these problems, the success of so many African American migrants in acquiring the knowledge they needed is strong evidence in support of the informational advantages of foot voting.
B. Information acquisition by southern black migrants
Southern black workers relied on a variety of information sources to facilitate migration decisions. One important resource was the information provided by relatives and acquaintances already living in the North or in more tolerant southern jurisdictions.Footnote 60 Many black migrants were “armed with firsthand reports from trusted friends and relatives” about conditions in the North.Footnote 61 Ballot box voters cannot easily rely on comparably knowledgeable and trustworthy information sources.Footnote 62
Other information was provided by the contemporary black media, which actively encouraged migration.Footnote 63 But most scholars put special emphasis on the information-spreading activities of “emigrant agents” employed by businesses seeking to recruit African American workers.Footnote 64 The agents provided valuable information to African Americans considering moving to the North, and sometimes also helped arrange transportation for them. While agents had obvious incentives to exaggerate the benefits of moving,Footnote 65 these were to some extent kept in check by information provided by migrants who had already made it to the North,Footnote 66 and by the likelihood that workers who were deceived about the opportunities available to them might move back to the South themselves and tell others to disbelieve the agents. These mechanisms provided a check on deception by emigrant agents of a kind that is not usually available in the case of political rhetoric used to persuade ballot box voters.
In addition to successfully acquiring information about job opportunities, many black migrants also chose to move in part because they came to realize that northern state governments and social mores were less hostile to blacks than those in the South. Although economic opportunity was a key factor in motivating migration, the desire to escape racial repression was also important. Migrants themselves often cited lynching, racial discrimination, and other hostile government policies as important factors in their decision to leave the South.Footnote 67
Some contemporary African American leaders recognized the potential of foot voting as a tool for mitigating their people's oppression, and urged southern blacks to consider migrating to the North. As early as 1886, Frederick Douglass—the most prominent African American leader of the nineteenth century—argued that “diffusion is the true policy for the colored people of the South,” that as many blacks as possible should be encouraged to move to “parts of the country where their civil and political rights are better protected than at present they can be at the South,” and that “[a] million of dollars devoted to this purpose [of assisting black migration out of the South] would do more for the colored people of the South than the same amount expended in any other way.”Footnote 68 A 1917 NAACP publication claimed that migration north was “the most effective protest against Southern lynching, lawlessness, and general deviltry.”Footnote 69
C. Effective use of knowledge
Most scholars agree that black migration to the North during the Jim Crow era was generally effective in achieving the migrants' goals. Although the North was far from free of racism, most migrants were able to better their lot significantly, both economically and from the standpoint of protecting their civil and political rights.Footnote 70 These results suggest that the migrants made effective use of the knowledge that they acquired, and generally chose their destinations wisely.
As the theory of foot voting under competitive federalism would predict,Footnote 71 the resulting migration not only benefited the migrants themselves but also forced racist southern state governments to “grant … African-Americans greater educational opportunities and greater protection in their property and person” in an effort to get them to stay and continue to provide labor for white-owned farms and businesses.Footnote 72 For example, fear of losing black labor was one of the motives that led southern state governments to finally make some belated efforts to crack down on the lynching of African Americans in the 1920s.Footnote 73 Lynching was cited by many migrants as an important cause of their decision to move.Footnote 74 In a related dramatic example, interjurisdictional competition for the labor of migrating black coal miners led to successful lobbying by coal companies for a reduction in school segregation in early 1900s West Virginia.Footnote 75 As Douglass had predicted in 1886, “the condition of those [southern blacks] who must remain will be better because of those who go.”Footnote 76
Obviously, the ability of southern blacks to vote with their feet did not come close to fully mitigating the baneful effects of Jim Crow.Footnote 77 Foot voting was an improvement over preexisting conditions, not a panacea. It did, however, provide important informational benefits and a measure of political empowerment to a widely despised and poorly educated minority.
Although exact comparisons are difficult, it seems likely that potential southern black migrants of the Jim Crow era were able to learn considerably more about relative conditions in different jurisdictions than most modern voters have learned about the basics of our political system. At the very least, large numbers of poor and ill-educated southern blacks learned enough to understand that relatively more favorable employment opportunities and public policies awaited them in other jurisdictions, a realization that contrasts with the inability of most modern citizens to acquire sufficient knowledge to engage in effective retrospective voting.Footnote 78
If foot voting could provide powerful informational advantages in the exceptionally adverse conditions of the Jim Crow era South, there is strong reason to expect that it is more effective under modern conditions, where education levels are much higher, information costs are lower, and no large group is as thoroughly oppressed as poor southern blacks were a century ago. People in less dire circumstances than early twentieth century southern blacks can acquire information more easily.
D. Comparison with contemporary southern white ballot box voters
In considering Jim Crow era black migration as a case of foot voting, it is difficult to make a direct comparison to ballot box voting. Most southern blacks during that period were, of course, denied the right to vote, so they did not have the opportunity to address through ballot box voting the same issues that many sought to resolve through foot voting. However, southern whites of the same period did address racial issues at the ballot box, and it is worth comparing their apparent knowledge levels with those of black foot voters.
White southerners had far higher average income and education levels than African Americans. As of 1940, 24.6 percent of southern white adults over the age of twenty-five had high school diplomas, compared to just 5.4 percent of southern blacks.Footnote 79 And this difference in quantity of education coexisted with a massive difference in quality. Income and education are both highly correlated with political knowledge levels.Footnote 80 Despite these comparative disadvantages, southern black foot voters seem to have acquired fairly accurate information about migration opportunities, and seem to have made effective use of their knowledge of which jurisdictions had policies more favorable to blacks. By contrast, southern white ballot box voters were apparently ignorant of important basic facts relevant to Jim Crow era racial policies.
Widespread southern white support for Jim Crow policies in the early twentieth century was in part based on purely normative disagreement with racial egalitarianism. However, white support for many such policies was also in part the result of gross ignorance on factual matters and failure to make rational use of political information.
To take one of the most notorious examples, for decades large proportions of white southern voters seem to have accepted the blatantly false claim that many, if not most, black men were out to rape white women.Footnote 81 This widely accepted myth was the principal rationale justifying the southern states' policy of permitting the lynching of numerous blacks accused (often falsely) of the rape or murder of whites.Footnote 82 Although scholars and civil rights advocates demonstrated the falsity of such claims as early as the 1890s,Footnote 83 most of the white southern electorate apparently remained unaware of this fact, or unwilling to consider it.Footnote 84
Southern white voters were also, for decades, unable to recognize that the exclusion of the region's large African American population from much of the educational and economic system was an important contributing factor to the region's underdevelopment—a point obvious to most economists. Although the relative economic backwardness of the South was a major issue in regional politics throughout the late nineteenth and early twentieth centuries, few white southerners urged desegregation as a method for promoting economic development until well after World War II, and even then such views were mostly advanced by business leaders and other elites rather than by ordinary voters.Footnote 85
In the absence of suitable survey data from the period, it is difficult to say whether these white southern views on racial issues were driven by ignorance per se, or by failure to rationally evaluate the information voters did know. Most likely, a combination of both was at work. Either way, the result is consistent with the rational ignorance hypothesis, and also contrasts with the more effective acquisition and use of information on racial issues by black foot voters.
There are, of course, some important distinctions between Jim Crow era white ballot box voting on racial issues and black foot voting. Racial issues were obviously of greater importance to African Americans than to whites, and the former therefore probably had stronger incentives to be informed about them. Moreover, the knowledge necessary for effective foot voting is in some respects simpler than that necessary for ballot box voting.Footnote 86
However, the similarities between the two cases are still strong enough to make the comparison meaningful, even if imperfect. Race and its associated economic underdevelopment were arguably the most important political issues in the Jim Crow era South, and whites had almost as great a political stake in them as blacks did. The one-party system and other political institutions of the pre–Civil Rights Movement South were organized around the objective of maintaining white supremacy.Footnote 87 Moreover, to the extent that whites did have less interest in racial issues than blacks, this factor is at least partially offset by their higher income and education levels.
V. Foot Voting in the Private Sector
The informational benefits of foot voting make the case not only for federalism as an alternative to centralization, but for the market and civil society as an alternative to government. In many situations, the private sector may be an even better mechanism for foot voting than federalism is. Voting with your feet against a product in the market usually has much lower moving costs than doing so against a regional or local government. One can switch to a different product or firm without changing one's residence. The same point holds true for most civil society organizations. In this respect, the informational argument for foot voting has implications for the balance between the government and the private sector, as well as for the relationship of one level of government to another.
Private-sector foot voting is already a reality for many services traditionally performed by local government in the United States. As of 2004, over 52 million Americans lived in private planned communities such as condominium associations.Footnote 88 These organizations routinely provide security, trash removal, environmental protection, zoning rules, and other services that are usually the responsibility of the state.Footnote 89 Similar enterprises have proven popular in Europe, Latin America, and parts of Asia.Footnote 90
Competing private planned communities have significant advantages over traditional interjurisdictional competition between regional and local governments.Footnote 91 A single metropolitan area can contain many more private common-interest communities than government bodies. This makes it easier for each potential resident to find the community that best fits his needs, and also cuts down on potential moving costs by reducing the distance most movers would have to travel.
Unlike state and local governments, which are often subsidized by higher-level governments,Footnote 92 most private planned communities are exclusively dependent on residents for their revenues. This increases their incentive to compete for residents and meet their demands. Should private communities fail to do so, the property values of current owners are likely to fall, and they cannot use tax revenue collected in other areas to make up the difference. By contrast, state and local government officials usually have a much weaker stake in attracting migrants and incentivizing current residents to stay.
Finally, potentially irrational attachments to a state or locality might inhibit decisions to move out of a political jurisdiction, especially in cases where that jurisdiction is closely associated with an ethnic or religious group with which the decision-maker feels a strong sense of identity.Footnote 93 By contrast, few people have strong emotional or ideological attachments to a private planned community. This factor might make foot-voting decisions involving private planned communities more rational, on average, than those where potential movers choose between competing governmental jurisdictions.
The relative advantages of the private sector over government might be even more significant for the production of goods and services not tied to particular physical locations. In such cases, people can vote with their feet without actually moving at all, thereby eliminating moving costs from the equation. Here, too, the informational advantages of foot voting suggest that private provision has an important advantage over government.
In recent years, the Swiss economist Bruno Frey has argued that regional and local governments can take on some of the characteristics of private-sector firms, breaking the link between territory and jurisdiction.Footnote 94 Frey claims that various government bodies specializing in different issue areas could have overlapping jurisdictions, and that individual citizens could change government service providers without a physical move. It is too early to give a definitive verdict on these proposals. If Frey's theories turn out to be viable, they could provide a blueprint for ensuring that foot voting will often be as effective in the public sector as in the private. Something resembling Frey's proposal already exists in the field of commercial transactions in the United States, where businesses and others are often able to choose for themselves which state's law will govern their dealings with each other, often without making a physical move.Footnote 95
A complete comparison of private planned communities and political bodies is outside the scope of this essay. My aim here is to note a potentially important and underanalyzed advantage of private communities over ones controlled by political bodies. I do not assert that this advantage necessarily outweighs all competing considerations. To the extent that the informational benefits of foot voting are even greater in the case of private-sector institutions than political jurisdictions, they argue for increasing the authority of the former relative to the latter.
VI. Implications for Constitutional Design
Widespread political ignorance and irrationality strengthen the case for constitutional limits on the powers of central governments, and also for constitutional constraints on the size and scope of government power.
As I discussed above, foot voting has major informational advantages over ballot box voting. It reduces incentives for both ignorance and irrationality. Even if this conclusion is accepted, however, it is possible that the benefits of foot voting can be left to legislatures to balance as they see fit. Since determining the size and degree of centralization of government involves many complex trade-offs, it is possible that legislatures will be in a better position to balance the relevant considerations than constitutional drafters or courts exercising the power of judicial review.
Despite this concern, there is reason to believe that ordinary legislative activity will undervalue the informational benefits of both decentralization and limited government. Perhaps the famous “political safeguards of federalism” would make constitutional limits on central government power unnecessary. Some scholars argue that the political power of regional governments is sufficient to prevent excessive centralization, because the regions can use their clout to prevent it, and voters will punish overcentralization at the polls.Footnote 96
Unfortunately, the very political ignorance that makes decentralization and limited government desirable also reduces the chance of achieving them through the ordinary legislative process. Few voters have a solid understanding of federalism, and fewer still are likely to be aware of the interconnection between limits on federal government power and “foot voting.”Footnote 97 For these reasons, they are unlikely to punish elected officials who promote overcentralization.
This might not be a problem if central and regional governments had other incentives that would lead them to avoid excessive centralization. In fact, however, both regional and federal governments often have strong political incentives to concentrate power at the center. Central governments have incentives to expand their power in order to capture more revenue and use it to buy political support; subnational governments have incentives to lobby for central-government grants and to use the central government as a cartel enforcer that suppresses competition among them.Footnote 98 Strikingly, subnational governments in most federal systems get the vast majority of their funds from central-government grants.Footnote 99 This occurs despite the fact that dependence on central-government grants severely reduces regions' incentives to compete for foot voters in order to attract tax revenue, and increases the central government's ability to use grants to suppress regional policy diversity.Footnote 100
The political reality that overexpansion of central-government power often advances the interests of regional governments undermines claims that “political safeguards” are enough to ensure an optimal level of decentralization. Virtually all such arguments rely on the political power of regional governments to serve as a check on the center. But if regional governments actually help promote centralization, their influence in the national legislature becomes a liability for federalism rather than an asset.
Moreover, ordinary political processes often cannot be relied on to prevent government from growing unduly at the expense of the private sector. Political ignorance may prevent voters from being able to effectively monitor government interventions that benefit narrow interest groups at the expense of the general public. Most of the items in government budgets are ones that the majority of voters are probably not even aware of.Footnote 101 Even when voters are aware of the existence of a given program, “rational irrationality” will often prevent them from making effective use of the information they possess. Economist Bryan Caplan's recent research indicates that public opinion is distorted by “antimarket bias” and “antiforeign bias,” which lead the majority of voters to systematically overestimate the effectiveness of government interventions in the economy, and of protectionism and restrictions on immigration.
For these reasons, widespread voter ignorance and irrationality are likely to prevent the political process from producing the appropriate level of decentralization and limits on government needed to restrict the harm. This suggests that constitutional restraints on centralization and the growth of government are needed. How strict should those constraints be? Unfortunately, analysis of the dangers of political ignorance does not, in and of itself, provide an answer to this question. Obviously, political ignorance is not the only factor that must be considered in determining the optimum level of constitutional constraints on government power. A wide range of other considerations—some of which vary from one society to another—must be weighed. However, our analysis does suggest that the need to combat the effects of political ignorance justifies stronger constitutional constraints on centralization and the growth of government than we might otherwise wish to impose.
VII. Conclusion
The informational advantages of foot voting over ballot box voting have important implications for normative theories of federalism. Perhaps the most significant is the way in which they reinforce the case for political decentralization. The more policy issues are under the control of regional or local governments as opposed to the national government, the greater the range of policy choices over which citizens can exercise leverage through foot voting and the more they can make use of its informational benefits.
In addition to strengthening the case for decentralization more generally, the informational benefits of foot voting also bolster the argument for competitive as opposed to cooperative federalism.Footnote 102 The greater the incentive for regional governments to compete with each other for citizens, taxpayers, and businesses, the greater the likely effectiveness of foot voting as a tool for imposing democratic accountability on government. This consideration strengthens the argument for policies associated with competitive federalism, such as limiting central-government subsidies to regional governments, so that the latter have stronger incentives to compete.Footnote 103
Further, the ability of even a severely oppressed minority such as Jim Crow era blacks to acquire the knowledge necessary for effective foot voting suggests the need for a partial rethinking of the traditional view that such groups necessarily benefit from political centralization.Footnote 104 While central-government intervention to protect minority groups is often desirable, this potential advantage of centralization should be weighed against the disadvantages of eliminating foot voting. To the extent that oppressed minority groups often have lower income and education levels and therefore lower political knowledge levels than others, the relative informational advantages of foot voting for them may be even greater than for other citizens. Such benefits of decentralization are even more important in periods when the central government has little or no interest in alleviating the plight of oppressed regional minorities—as was certainly true of the United States during much of the Jim Crow era.Footnote 105 The gains for oppressed groups from foot voting within a federal system imply that there may be even greater foot-voting benefits from international migration, a possibility I discussed in another recent essay.Footnote 106 The differences in quality between regional governments within one society are generally much smaller than those between nations. Foot voting through international migration is the best hope for many of the most oppressed people in the world.
At the same time, it is essential to recognize that the argument of this essay is limited in scope. The informational benefits of foot voting are likely to vary from issue to issue, from nation to nation, and perhaps also from group to group. Obviously, foot voting cannot be used by people who are unable to leave a particular area, or by those who seek to protect immobile assets such as land. Examples include people with very high moving costs because they cannot find employment for their special skills outside a given locality, and those precluded from moving by serious health problems. Similarly, foot voting may not be effective for “network industries” that must operate in every part of a nation simultaneously in order to operate anywhere.Footnote 107
Finally, democracy and political ignorance are far from the only issues that must be taken into account in determining the degree of decentralization that a society should have.Footnote 108 Various other considerations may in some situations outweigh the advantages of foot voting. The argument advanced here is not intended to be a comprehensive theory of federalism or of the appropriate role of government in society. It does, however, raise an important consideration that is too often ignored.
