Who Really Creates Externalities?

And some solutions

A person bumped into my backpack today. I was standing in a line, my backpack extending into a space where people walked. I inconvenienced someone, and they inconvenienced me. I could have stood further out of the path, they could have avoided me, or we could both act. What is optimal? Who should change their behavior?

This gets after a much deeper question of externalities, which we shall begin by defining. An externality is when some action affects those who were not involved in contracting for it. This can be both positive or negative. A negative externality would be something like pollution, which affects many people who cannot cost-effectively redress the harm done to them. It can be thought of as whenever a contractual agreement would occur, if all property rights were perfectly defined and the cost of transactions were 0, but cannot due to some imperfection of the two assumptions. This framing of transaction costs was introduced by Ronald Coase’s “The Problem of Social Cost” in 1960. 

The obvious solution is to place a tax upon negative externalities, and give a subsidy to the positive ones. If someone is emitting some noxious gas, placing a tax on its emission would surely bring the amount emitted in line with what is socially optimal. We cannot, however, simply assume that this is the optimal mechanism. Suppose it were actually cheaper for everyone to buy filters for their homes than to stop emitting. Or, if the example seems too implausible, suppose it were actually cheaper to capture carbon from the air than tax its emission. If property rights were perfectly defined and transaction costs were 0, the government would simply define a target enforceable by fines on someone, and people would find the optimal way to stop it, but we do not live in such a world. Who starts out with the right to pollute or not pollute does not matter (except so far as it influences who has more money) in a frictionless world, but in a world with friction, it is of extreme importance.

In order to get to improve the world, the government should not engage in blind reaction against those who seem to be creating the externality. To create an externality, you need someone there to be harmed too. If someone moves underneath the flight path of airplanes coming into an airport, who has created the externality really? The airport, or the person who moved? Rather, the government should first define what its goals are, and then use market mechanisms such as prediction markets to assess what the most cost-effective way of achieving that goal is. This seems to me one of the more exciting reasons to use prediction markets — it can only be an improvement over normal cost-benefit analyses. (Indeed, the creation of such analyses can be done by government employees who then bet parts of their salary on whether or not their analysis is accurate).

As for me — I should have moved out of the common area. :)