Citizen or Consumer? Michael Munger Responds

When it comes to voting by citizens in a democracy, there are four essential questions, as I see it, in marrying up the “should” and the “is.”

  1. How do citizens and consumers actually choose?
  2. How should citizens and consumers choose, in an idealized sense?
  3. How can this answer be affected by civic education?
  4. What are the implications of these answers for the scope and nature of the state? And we mean the actual state, the government(s) we live under, not those we can imagine

Having an ideal conception of how citizens and consumers act, and an empirical notion of the extent of the difference, is useful. The problem is that all too often we use the following kind of reasoning: Consumers are mostly selfish, both in fact and in the conception of those who study markets. Citizens should be public-spirited, in the minds of those who study political theory.  Therefore, on this view, the state is better than the market. (I have written elsewhere about the problem of this comparison of markets-as-they-are and polities-as-we-imagine)

The difficulties with this approach are two. First, it is by no means clear that the fact that consumers, and producers, act in ways that benefit themselves means that society is thereby harmed.  Whether you think that system is evil but necessary (Bernard Mandeville) or morally justified and anodyne (Adam Smith), the selfish motivation of market participants is not an obvious strike against markets as a social order.  The institutions matter.  But let us leave this topic to one side, as it is not the focus of our discussion today.

The second difficulty is germane. Suppose that voters are not, as a matter of empirical fact, public-spirited. Rather, voters are also self-interested, though of course their concept of the “self” may be different in public and private settings. The different institutional contexts may have different implications for behavior, but that way that actually works out often means that voters know less and even care less than consumers about getting things right. As I have written elsewhere, every flaw in consumers is worse in voters.

But this is again an empirical objection. It is not intrinsic to the notion of voters that they behave the way I claim they do, now, in the United States. The question before us in this debate[1] is how to evaluate the differences in the nature of market consumption and participation (by voting and otherwise) in the polity.

In a number of writings, I have argued that the Fundamental Human Social Problem (FHSP) is the design, or maintenance, of institutions that make self-interested individual human action not inconsistent with the welfare of the community. This is a complex formulation, and its full exegesis is beyond my scope, and your patience, for today.  But the key point for our purposes is that there are two very general ways of addressing the FHSP.

The first is to take self-interest as given, and (relatively) immutable and durable. We grant autonomy to citizens and consumers because of their essential human dignity, and allow them to pursue their own subjective conceptions of the good, provided the community is not harmed (hence, the double negative in the FHSP). Institutions should be designed, or preserved when they arise as spontaneous orders, to reconcile the conflict of self-interested actions in ways that redound to the benefit of the polity.

The second is to seek to engage in an intersubjective reconceptualization of the “self,” through civic education, so that the individual sees himself as embedded in a social context. I think it is a point in Professor Stoner’s favor, and I want to concede this publicly, that those of us in the “Public Choice” tradition both undervalue this approach as a goal, and underestimate it in terms of effectiveness. Human beings really do care about acting, and being seen authentically to act, as part of a community.

We might call the first approach—recognizing the reductionism inherent in naming—the “neutrally self-interested” view, and the second the “virtue-conscious” view. All societies exhibit aspects of both kinds of behavior, and all societies work to create settings where each of these is nurtured and promoted.

And it is important to point out that the dichotomy I have drawn is false, or at least misleading, because many kinds of action serve both approaches. An example is property rights: we might ask that the state should promote clear title and well-defined property rights because that reduces the transactions costs of commercial activity, promoting economic growth.  But we might also settle on property rights as a means of promoting virtue, arising from fundamental conceptions of dignity, autonomy, and hard work.

But there is no reason to believe the dichotomy at the level of behavior, either. If it is true that citizens as voters do, as Professor Stoner claims, take seriously their obligations that arise from being embedded in a social context, I see no reason to doubt this about consumers. “Self-interest” is not “acting without regard for the interest of others.” If people have intersubjectively reconceptualized themselves, we might anticipate that they carry this same conscience, evaluating their actions in light of Adam Smith’s “impartial spectator,” in the grocery store as well as in the voting booth.

There is plenty of obvious evidence in favor of this claim. People voluntarily recycle, though it is expensive in terms of time. They pay higher prices for “fair trade” products, believing that workers benefit. They choose products that have less negative impacts on the environment, and buy meats and other foods that are “sustainably produced.”

So if the question we are presented with is, “Should citizens act the way that some economists caricature consumers?” then the answer is, “Of course not.” But consumers shouldn’t act that way, either. Fortunately, they don’t.

A more interesting question is whether consumers and citizens are fundamentally different intrinsically. At some level, the answer has to be “yes,” because consumers have better information and are responsible for the outcomes of their choices. My selection of a car, or a smart phone, is affected by the fact that cheap and reliable information is available about alternatives. More importantly, if I “vote” for the new Banana iPhone, that’s what I get, because consumers have direct agency.

Voters, by contrast, have limited and inaccurate information about candidates, all of whom have incentives to be vague and resources to portray themselves as honest, as far as a 30-second television spot can reveal. Voters have almost no information about particular policy alternatives, and ambiguous promises from leaders. Of course, it doesn’t matter much anyway because voters have no agency: my choice to “vote” for Candidate A rather than Candidate B has no direct impact on the outcome (unless there is a tie, which happens so rarely that we can discount the possibility).

Jason Brennan has written provocatively but forcefully, in fact, that a moral voter will actually decline to vote if he is not well-informed, precisely because the impact on the society is negative.  Poorly informed consumers may be helped by reviews, or labeling, but if they choose badly they harm only themselves. Poorly informed voters do little harm to themselves, but impose what economists call “negative externalities” by choosing what makes them feel good in the voting booth rather than selecting what is good for society. Ignorance and lack of agency, on this account, leads voters to act in the bad way that consumers are caricatured as acting (but do not in fact act). The surprising conclusion, then is that voters act more like consumers act that consumers do!

The point of my argument is that we cannot make very general claims about consumers and citizens in the abstract. The context, the institutional context, matters. And the relevant institutional context is the constitution.

The core precept of Public Choice is self-interested (though perhaps not selfish, or egoistic) individuals participating in “politics as exchange.” Thomas Hobbes called the unconstituted collection of people the “state of nature.” Buchanan and Tullock (1962) imagine a constitutional “moment” when a group of individuals confront the problem of deciding how to decide. They impose two conditions, or qualifications, on the decision process.

First, the choice must be unanimous, meaning that everyone consents. This is a means of solving the infinite regress problem (how to decide, how to decide how to decide). Unanimity protects each individual, ensuring that no minority, no matter how small, can be subjected to coercion involuntarily.

Second, the choice must be disinterested, or made without complete knowledge of the effects of the rules on the chooser’s welfare, much like the Rawlsian “original position.”

Unanimity answers the question raised by Rousseau: How can a person be both free and yet bound by wills not his own? Because he consented. Anyone who voluntarily signs a contract expects to benefit.  But he would benefit even more if he could cheat while other parties perform as promised. Since this is true for each potential signer, monitoring and enforcement are required, and it is in the interest of all the signers to ensure that the contract is enforced. These might be provided by the “state,” but a formal state apparatus is not logically entailed by the need for enforcement. What is necessary for people to be able to make free choices for mutual benefit is the ability to sign binding contracts. And binding contracts require a threat of enforcement if the contract is breached. As Hobbes said, “Covenants, without the Sword, are but Words, and of no strength to secure a man at all.”

But then is the ex post enforcement—force and forfeiture of property—coercive or voluntarily imposed? If I agreed to be bound, am I bound by wills not my own? There are two states of play, one before the signing of the contract and one after. Oliver Williamson (1979) claims contractual relations undergo a “fundamental transformation.” Before a contract is signed, there is no coercion in the negotiation process. Each party can exit without harm. But after the contract is signed, exit from the agreement may be expensive and going back on the contract triggers punishment. If threats are made in the pre-contract stage, that would be coercion.

The contract would never have been signed unless the there was an ex ante expectation that ex post sanctions would be threatened. All parties to the contract are being coerced voluntarily, as a means of committing to the terms of the contract, which we assume they want to do. As the quote beginning this chapter shows, Rousseau was focused on this problem of reconciling freedom and coercion.  Coercion is justified in three steps:

  1. Citizens agree unanimously to the rules specified in the original contract. Each consents to laws passed under these rules, even if one does not like the outcomes.
  2. This unanimity requirement applies only to those who voluntarily join the contract. I need not agree. But then I must exit the group, because membership (and the consequent enjoyment of the benefits of membership) implies consent.
  3. The transcendent truth embodied in the constitution is the “general will,” or the self-interest of the collective, rightly understood. It would be irrational for me to want something different from that which I should want; if I disagree with the general will, I am mistaken or evil.

Exactly 200 years later, almost to the day, Buchanan and Tullock (1962) argued that Step Three is unnecessary: A voluntary contract gets the group everything it needs. There is still the problem that “membership implies consent,” because it is easy to imagine that a citizen in a state may have no real options to leave. This is a problem with no very good solution. Any group must balance the externality of involuntarily including someone against the costs of negotiating unanimity. Buchanan and Tullock argue that true unanimous consent rule is very expensive in terms of transactions costs, because it encourages strategizing and efforts at hold-up.

If I contract for coercion, if I voluntarily agree to have someone else punish me for violating my promise, the resulting use of force is no longer coercion, at least not in the usual sense. In Homer’s Odyssey, Odysseus “contracted” with his men to bind him to the mast, so that he could resist the songs of the seductive Sirens. The solution shows the power of enforceable contracts: Odysseus has to find a way to order his men to ignore his orders. The ropes literally bind Odysseus to do what he wants himself to want to do, rather than what he knows he will want to do later when the song of the Sirens attracts him. Odysseus must be able to enter voluntarily into this contract to be coerced, or he would not have been truly free.

But if you were on a passing ship, you would see Odysseus struggling with the ropes and begging to be set free. The crew (apparently) disobeys his command and busily binds him even tighter. Clearly Odysseus wants to escape but is being held against his will. Should you try to help? Who would you help? Odysseus, who involuntarily now wants to be set free? Or would you help the men trying to carry out the orders Odysseus gave back when he controlled his will, before the call of the Sirens robbed him of the power to make choices? Whose “will” would you validate if you intervene?

If a constitution is an agreement, and people become members by consenting to the contract, then the members of that group are agreeing to be coerced because they expect to be better off as a cooperating group than they were as non-cooperating individuals. Binding Odysseus to the mast, and then binding him even more tightly when he changes his mind (as he knew he would when he gave the orders to prevent his escape) is not coercion in the usual sense. This means that agreeing to suffer coercion at the hands of a group is potentially better for each individual in the group, compared to collections of individuals who cannot enforce agreements. The ability to enter into agreements, and to agree to be punished if the agreement is violated, is the essence of the constitution of groups.

Thomas Hobbes was quite correct that the ability to sign binding contracts is necessary for human survival and group flourishing. But all he really established was that some kind of governance could be justified, because almost any system would improve the welfare of citizens over an anarchic state of nature. He does not say how would the group select among all the many kinds of governance structures, some private and some involving direct state action, that might be constituted? Hobbes showed that almost any viable constituted group is better than his “state of nature.” For a given group, however, the Hobbesian argument can provide no guidance about which constitution to select.

The pure contractarian answer extends the Hobbesian idea, but remains agnostic about what a particular collection of individuals would, or should, choose. The “best” agreement is whatever the parties to the particular exchange situation choose to agree on, knowing everything they know about local conditions and their own needs. The individuals constituting the group will use their moral intuitions in choosing the rules, and the rules become norms that guide moral intuitions. To be free, the group has to be free to choose its contract, and then be free to enforce that contract.  Maybe there is a state, maybe there isn’t. All that is necessary for politics to be important is that there is a group, and a contract.

After the group is constituted, of course, many people will look for ways to cheat, just as Odysseus did (and knew he would). After the agreement, people will try to escape the punishment they promised to accept. If they are caught, they will protest that the punishment is against their will, because they would prefer that everyone else is bound by the promise but that they can escape. But the consent, the unanimous consent, to the original contract means that the coercion was voluntarily agreed to. One way of understanding constitutions is that they can, in some circumstances, allow groups to solve collective action problems that otherwise would prevent the capture of substantial gains from cooperation.

Of course, if I did not agree to the contract, even if I am the only one who did not agree, then the coercion is not voluntary because I did not consent. This is why the qualification of unanimity, even if it is hypothetical, is central to constitutive arguments. If I do not consent, I am not bound by wills not my own, and I am not free.

Public Choice scholars argue that societies must have limits on the scope, and limits on expansion of the size, of government. We can imagine a state that will carry out many activities we favor, but that state does not exist in isolation. Those people, or others who will replace those people, will act badly. For that reason, the best feasible state is minimal, and humble. The question of whether we can reliably use civic education to cause people to act as if they were constrained, or perhaps just prudent, is an important one that Public Choice scholars have given short shrift. But the alternative assumption such that restrictions are not necessary is likewise naive, and dangerous.

The real answer, perhaps, is humility. The fetishizing of markets as “perfect” and the state as flawed is every bit as mistaken as the reverse. Consumers and citizens are, after all, the same people. The same flaws in information and motivation that operate in one setting will pass through into the other in a kind of institutional osmosis. It is unlikely that consumers and citizens will act in the same way, of course, because the institutional setting is different.  But they are the same people.  If it is possible to perfect citizens, it is possible to perfect consumers.  I agree with Prof. Stoner, and depart from many of my Public Choice colleagues, that it is worth trying.

But a system that assumes the perfection of either consumers or citizens is headed toward disaster.

[1] Debate held on March 26, 2015, with James R. Stoner, Jr., of Louisiana State University at Louisiana Tech University in Ruston, Louisiana, sponsored by the Waggonner Center for Civic Engagement and Public Policy.