Free to Move: An Interview with Ilya Somin, Part Two

This is part two of a two-part interview with Ilya Somin (George Mason Law School) about his new book, Free to Move: Foot Voting, Migration, and Political Freedom (Oxford University Press, 2020). Part One appears here.

JB: How might you apply the thesis of your book to the internet? I have argued that we need many different kinds of social media companies with many different kinds of content moderation rules in order to protect freedom of expression. But the central objection to what I've called "social media federalism" is the concern that Facebook and Instagram create network effects both for participants and for the advertisers who want to reach end users. Moreover, Facebook's ability to gather enormous amounts of data has given it super-competitive abilities to make predictions and attract advertisers, and thus make enormous profits. How do your arguments about foot voting and political freedom apply to social media?

Ilya Somin: I don’t have your expertise on internet issues, so I may have to leave the heavy lifting on this issue to others who know more. But I am skeptical that Facebook and Instagram have such monopoly power that they could forestall competition if the latter had content moderation policies that many users preferred – and otherwise provided comparable quality of service. I’m old enough to remember when many people claimed that Microsoft had established a lock on the operating system market and could use it to monopolize the browser market, as well. Those predictions were quickly falsified. The idea of unbreakable social media monopolies strikes me as comparably dubious. At the very least, putting up with the flaws of Facebook and Instagram and relying on competition to curb them strikes me as preferable to having the federal government (or, worse still, an international body like the UN), establish moderation rules, as advocated by various people on both the right and the left.

JB:  How does your account of political freedom help us understand the current massed protests in the United States? Speech, assembly, petition and protest are often thought of as central to political freedom-- because people exercise this freedom by changing minds and by demonstrating that they and others like them disagree with the status quo. These aspects of political freedom tend to be collective rather than individual. Massed protests are exercises of political freedom, but they require collective participation. Political revolution is seen as an important act of political freedom, but revolution requires collective action rather than individual action. Political revolution can be violent, but it doesn't have to be. It could occur within the forms of politics, as we saw in the United States during the civil rights revolution and the New Deal. A critic might charge that emphasizing foot voting neglects this conception of political freedom. What do you think?

Ilya Somin: Protests, marches, and revolutions can all potentially help improve government policy (and also, of course, make it worse). But they provide little in the way of real meaningful political freedom to the vast majority of people, other than those few who have a substantial likelihood of making a decisive difference to the outcome.

If having a tiny chance of exercising influence was enough to have meaningful political freedom, then we would not need democracy or protests. An absolute monarchy where the king occasionally listens to your views on policy, then rolls a million-sided die and implements your views if he rolls a 1, would give you the same level of freedom of choice! Ditto for a dictator who occasionally chooses one of her million subjects at random and then follows her policy preferences for the next year.

Moreover, each of these means of influence where individuals have little chance of affecting the outcome suffers from the same problems of rational ignorance as voting does. Protestors – like ballot box voters -  often don’t actually know much about the issues at stake, and do a poor job of evaluating the evidence they do know, because they have little incentive to get it right. Foot voters, by contrast, do better, because they know their decisions are actually likely to make a difference.

Protesting and other means of exercising “voice” over public policy by mechanisms that go beyond voting also run into the problem that the ability to engage in them effectively his highly unequal. Only a minority of Americans (about 25%) engage in them at all, and only a much smaller number have more than a minimal chance of making a difference.

Moreover, such influence is a zero-sum game. Increasing my leverage over government decisions necessarily reduces yours and everyone else’s, and vice versa.  Even if we somehow manage to make political influence completely equal, we still have the problem that then each person would only have an infinitesimal chance of making a difference.

By contrast, foot voting can be made available to a much wider range of people, there need not be any zero-sum game, and each individual’s decisions have a high chance of mattering.

I’m not arguing that protesting is always a bad thing, anymore than I am saying that in the case of ballot-box voting. Both can, at times, accomplish real good, and both are preferable to authoritarianism. But neither offers the kind of meaningful political freedom that foot voting does, and both often fall prey to the dangers of rational political ignorance.

JB: Albert O. Hirschman's famous typology of exit, voice, and loyalty, which you discuss in the book, suggests that there are tradeoffs between exit and voice-- between individual and collective forms of political freedom. That is, it may be that focusing too much on foot voting undermines the remaining community's ability to exercise collective political freedoms. We discussed these collective freedoms in the last question.

But there is another way of making Hirschman's point. It has to do with the idea that one of the purposes of government is to invest in public goods and maintain the quality of those public goods. Some public goods require large amounts of public investment-- to start them up, and then to maintain them over time. The claim is that exit will drain the willingness of the public to invest in public goods in the first place, and then to maintain them, because they predict that they will have to pay a larger share of expenses. That will be especially so if the people who remain are less able to pay than those who leave.

I was thinking about this aspect of foot voting in the context of the coronavirus pandemic. Many affluent people have fled New York for other parts of the country, leaving less affluent New Yorkers to deal with the mess and the dangers. On the other hand, migration patterns caused by the coronavirus will probably shake up old ways of doing things and create new opportunities. How should we think about the advantages and disadvantages of foot voting for public goods?

Ilya Somin:  Hirschman famously argued that exit rights destroy incentives for people to improve institutions in their existing communities, by exercising “voice.” Why invest in public goods (or other public services) if you can just leave? Then, those left behind might suffer.

As I explain in the book, Hirschman’s argument has several limitations and flaws. It only applies in the special case where 1) the people who choose to exit would otherwise be able to force beneficial reforms in local policy that others could not push through in their absence;  and (2) those people nonetheless prefer to exit rather than exercise voice de­spite the likelihood that their use of voice will be successful, and 3) those left behind lack exit options of their own. Thus, the argument does not apply to the many people who have too little political influence to effectively use voice to improve the provision of public goods (or other services).  It also does not apply in many situations where the better solution is not to force the relatively influential people to stay but to improve exit options for others.

Even when the Hirschman argument is applicable, it overlooks the ways in which strong exit rights actually foster better provision of public services rather than undermine it. Exit rights enable people to more effectively “sort” themselves into those communities that best fit their needs and preferences. With more options, more people can find a community that is a good “fit”  - perhaps even one they truly love. The stronger your affinity for a community, the more likely you are to invest in maintaining and improving it.

As I describe in the book, social science research bears out this point. For example, political participation is actually better in small suburbs that are subject to strong exit options, than bigger jurisdictions that are more costly to leave. It’s also worth noting that Hirschman-like reasoning was previously used to justify arranged marriages and laws that make it hard to divorce. If you have little or no choice about who to marry, you are, by this reasoning, more likely to invest in maintaining the one relationship you are allowed to have! Yet, as noted in the book, greater freedom of choice in marriage actually increases partners’ investments in the relationship.

Finally, exit rights strengthen investment in public goods because of the competitive pressure they put on institutions. Local governments – and also private planned communities – know they are likely to lose taxpaying residents and investors if they provide poor public services. That incentivizes them provide better ones. Social science research finds that this dynamic actually works well in the policy area that apparently inspired Hirschman to develop his theory: school choice. Far from leading to the degradation of public schools, school choice programs  that enable students to use vouchers to transfer to private schools actually  lead to improvements in public education in their area, because public schools are forced to compete to retain students. By contrast, simply increasing funding for traditional public schools without increasing choice does little or nothing to improve education. I cite relevant studies in Chapter 6 of the book, where I address the Hirschman argument in more detail.

In addition to these practical considerations, there is also a moral reason why it is wrong to force people to stay in a jurisdiction, so that other residents thereof can benefit from their labor (or in this case, their tax payments and political clout). Such reasoning is inimical to the idea that individuals own their own labor, and that it is not the property of the government or other members of the community. This is a problem with the oft-expressed view that international migration should be curtailed to curb “brain drain” or to ensure that people fulfill their supposed duty “stay home and fix their own countries” (both discussed in Chapter 5 of the book). But it is also a flaw in the Hirschman rationale for constraining internal foot voting, at least in so far as it is seen as justifying coercing people into staying, as opposed to merely exhorting them to do so.

JB: What is the role of family members and children in your story? Children don't themselves exercise political freedom when their parents vote with their feet. And spouses and dependent family may lack effective choice-- unless we postulate that the decision is made equally by all members of the family who are affected. In other words, is your account of political freedom really the freedom of heads of households? Or should it be the collective freedom of members of families who are equally involved in the decisive choice? If all family members are not equally involved in the choice, can we say that they are politically free under your theory?

Ilya Somin: It’s true that spouses often constrain each other’s foot voting choices. But spouses usually have far more ability to influence each other’s decisions about where to live or what private-sector services to use than they do of influencing any outcome in the political process. My wife has vastly more influence over my choices – and I over hers – than either of us has over government policy, even though we are both in professions where we likely have much more political influence than the average voter.

Moreover, the dating and marriage market – itself an important form of foot voting! – allows people to select spouses whose preferences on such matters are relatively similar to their own. At least as a general rule, people tend to marry partners with similar views. Certainly, there is, on average, much more agreement among spouses than between either of them and the electorate as a whole, which is generally much more diverse in its preferences.

Influence within the family is rarely perfectly equal. But the same is true of influence in even the most democratic political system. Indeed, those systems typically feature far greater inequalities than families do. A relatively small elite of activists, politicians, and others have vastly greater influence than the average voter, in many cases by an order of magnitude or even more.

Of course, in a patriarchal society, wives often have less clout relative to their husbands. But in such societies, women also have less political influence than men. And whereas government policy (in a democratic society) is likely to be dominated by the views of the median voter and the political elite (who, in this scenario, are likely to be highly sexist), individual women making foot voting decisions can try to seek out relatively egalitarian spouses and relatively egalitarian jurisdictions in a federal system. In Chapter 2 of the book, I describe how exactly these sorts of dynamics helped play a role in increasing women’s rights in the nineteenth century, as Western states began to adopt relatively more egalitarian policies in order to attract women.

Children, of course have far less influence over family decisions than adult spouses do. But even here, they probably have more influence over their families than they do over the political process. After all, in the vast majority of jurisdictions, those under the age of 18 are not allowed to vote, and have little if any ability to participate in politics in other ways. In Chapter 2, I explain why children’s political clout probably won’t increase very much if we adopt proposals to lower the voting age to 16, as some jurisdictions have done (and which I have a degree of sympathy for).  Even if we go as far as lowering the voting age to six, as British political scientist David Runciman has proposed, the new child voters would still face the same limitations as their adult counterparts: the low significance of any individual vote, and the resulting incentives for rational ignorance.

 I address the issue of families and children in more detail in Chapter 2 of the book.