Why economists need to think about freedom

Should we restrict fake news and other political speech designed to create discord and undermine democratic institutions? Should we impose a soda tax to combat the obesity problem? Should we restrict the ability of parents to pass on their wealth to their children in order to level the playing field and insure equality of opportunity for all?

All of these questions propose some limitation on freedom in order pursue desirable consequences. Some of these questions pit one aspect of freedom (free speech, freedom to transfer wealth) against another (democracy, equality of opportunity).

Taking political speech as an example, one argument against restricting even noxious speech is that no one can be trusted to police the boundary between acceptable and unacceptable speech. If the government were given this power, it could eventually use it to silence its opponents. This is a good argument, but it is important to think a step further about why it would be bad if a government were to silence its opponents.

Why is freedom of speech important in the first place? One reason is that a free market for ideas will lead to the truth, which is essential for selecting wise policies. This is an instrumental justification for this freedom in terms of other types of goods (e.g., wise policies and the benefits they produce). If one thinks purely instrumentally in this way, the question of where to draw the boundary around permissible speech is just the question of balancing the positive effects of speech on, say, policies and discoveries, against the negative effects of noxious speech on the same, taking the difficulty of policing the boundary into account.

But the instrumental justification is incomplete: Being free to express one’s views is part of what makes for a good life and a good society. Free expression is one of the ends for which we should be striving. Freedom also has an inherent value, apart from its other consequences.

An evaluation of the many policies impacting freedom ought to consider both the instrumental and the inherent costs and benefits of freedom. In assessing the costs of freedom, it is important to note that freedoms themselves can come into conflict. Free speech can be used to persuade people to restrict other freedoms. If I am free to advantage my child, this might restrict opportunities for yours. In general, the enjoyment of freedom is not zero-sum, but at the other extreme, it is not purely win-win either. Not everyone can make the same decision; if I make it, you cannot. Perhaps we can share the decision, for example, by voting, but then I have less control than if the decision were all mine. Freedom is, in some respects, scarce. In other respects, we may have opportunities to expand freedom for all.

Economists on both the right and the left have long been interested in freedom, and freedom is part of the rhetoric of economics, but freedom is not part of the formal mathematical apparatus of mainstream economics. Amartya Sen and other social choice theorists have made notable formal contributions, but this work is not part of the mainstream economics curriculum that most students encounter when they study economics. So, work remains to be done.

Freedom may seem like a vague concept that it is difficult to reason about. In the remainder of this post, I will give the reader a sense of style of reasoning that is formalised in my recent paper and in related work.

As I have said, freedom is scarce. To see this, consider an election. The voters are Ann, Bob, and Carol, and the candidates are Clinton and Trump. Not everyone can control the outcome, so how is control to be shared? We can either run the election by majority voting or random dictatorship. Random dictatorship is a voting method in which each voter is selected to be dictator with equal probability, and the dictator can choose the winner. Which of these two methods maximises the control that citizens collectively have over the outcome?

Look at the question from Ann’s point of view. Suppose Bob and Carol vote independently and are each equally likely to prefer Clinton or Trump. Under random dictatorship, Ann’s vote will decide the outcome whenever she is selected dictator, which happens 1/3 of the time. Under majority voting, Ann’s vote decides the outcome whenever Bob and Carol disagree, which happens 1/2 of the time. So, if we measure control by the chance that your vote is decisive, a measure known as Banzhaf power*, Ann has more power under majority voting, and the same goes for Bob and Carol.

The point I would like to emphasise is that this way of thinking is not restricted to elections. Whenever a group of people, large or small, must share control, we can think about efficiently and equitably allocating that control.

Let us consider a different kind of example. Suppose that Joe has owned his home for 30 years. A developer offers to buy Joe’s property for $300,000 so that she can build a shopping mall. Joe agrees. Now imagine instead that Joe is forcibly removed from his property so that the shopping mall can be built. He is given $300,000 in compensation. Is the second situation equivalent to the first?

In terms of outcomes, the situations are equivalent. In both cases, the developer pays $300,000, Joe receives $300,000, Joe’s house is torn down, and the shopping mall is built. But they are not equivalent in all respects. In the first situation, Joe has control that he does not have in the second. He can say no. The right to say no, even to an offer that one would accept, is valuable.

Perhaps Joe would secretly be willing to take $300,000 for his property but pretends in negotiations that he values it more highly in order to receive a higher price. The pretence precludes a mutually profitable exchange. When we think about whether this would justify the seizure of Joe’s property, we should not think only about whether the property is more valuable in Joe’s hands or in the developer’s, but also about what sort or rights Joe ought to have over his property, in this case, his home.

Economics focuses on allocating goods to those who value them most, and evaluates policies from the perspective of how well they satisfy people’s preferences. These considerations are important, but they are not the whole story. Our lives are not only valuable insofar as we are happy or not, insofar as we get what we want or not; they are also valuable insofar as we are creative agents who freely decide what to do. Questions of how we allocate rights and freedoms are on a par with questions of how well people’s wants are satisfied. Questions of public policy are inevitably questions of practical ethics, and if economics is to engage with practical ethics, it must employ a broad ethical vocabulary that includes not just preference satisfaction and well-being, but also fairness and freedom of choice.