Tweet [1]
In this earlier post [2], I began a discussion of what are called Pigovian taxes, taxes to reduce activities where the harm caused by one’s actions are imposed on others, what economists call negative externalities. At the end of that post, I mentioned the importance of Coase in challenging the virtues of Pigovian taxes. Today’s Wall Street Journal brings a perfect example [3] of the intuition implicit in Coase’s challenge:
Breathing common urban air pollution is much more
deadly than previously thought, according to a major study published in
today’s New England Journal of Medicine.Today’s study, which followed 58,600 postmenopausal
women for seven years, found the added risk of cardiovascular death
from living in the most-polluted areas including Cincinnati and
Riverside, Calif. was roughly 150%. Breathing air heavily polluted by
soot from automobiles and power plants may raise the risk of death for
older women at nearly the same rate as smoking cigarettes. The study
focused on the most deadly kind of soot, known as fine particulate
matter, which comes from burning fossil fuels like gasoline, diesel
fuel and coal.
The article goes on to say that previous studies found a much lower risk of cardiovascular death so I don’t know if this study is more reliable. I’m also not sure of the importance of an increase in the risk of death of 150%. The article fails to mention what the base level of risk is, which matters. But let’s take these as facts. Let’s assume that soot kills women. What should we do?
One answer is a tax on the emitters of soot—cars, trucks, and coal-based power plants. Another answer is to have a standard setting a maximum level of emissions. That standard could be placed on vehicles or it could be a standard imposed on an area requiring that a particular area such as a city or a county find ways to make sure that their air reaches a certain level of cleanliness.
For all kinds of reasons [4], politicians prefer standards. If properly set, the standards can mimic the effects of the tax, achieving the same reduction in pollution that a tax would create, though not necessarily at the same cost because the standard often require a particular technology, such as a catalytic converter.
When we find out that a particular kind of pollution is more harmful than we thought, most economists would suggest reducing the maximum level:
Today’s findings may lend more ammunition to those who want the
Environmental Protection Agency to lower the legal limit for fine
particles in the air, said Rogene F. Henderson, a pollution expert who
heads the agency’s outside panel of scientists. The current limit was
set in 1997 at an annual average of 15 micrograms a cubic meter.
But the Pigovians among us [5] will surely argue that this new discovery that the risks of soot are higher than before increases the justification for raising the gasoline tax as a way to reduce cardiovascular death among women.
But thinking about externalities in a Coasian way suggests an alternative approach.
Most economists think that the essence of Coase’s insight about externalities is that in a world of zero transaction costs, externalities take care of themselves. The implication here would be that if transaction costs were zero, and if the harm to women from a higher risk of death was larger than some of the benefits from driving, women could get together and pay drivers to drive less. The payments would cause drivers to internalize the externality—they would now bear a cost from driving in the form of foregone payments so they would now drive the "right" amount, the same amount that they would drive under the correctly set Pigovian tax.
But of course this scenario is absurd. Transaction costs are not zero. This whole payment scheme is immensely impractical. So this whole Coasian argument seems to go out the window.
But the payment scheme is a parody of Coase. Coase did point out that when transaction costs are zero, externalities will take care of themselves. But he recognized that transaction costs are never zero and are often prohibitively high. Coase’s real insight was that because transactions costs are not zero, you should be careful as to how you assign property rights and liability. [6] When you assign those property rights and liability for harm, take into account the costs of compliance.
What I learn from Coase in this example is quite surprising. Assume the facts about soot in the story are true. Do women want a tax on drivers or a change in the standard to force cities to clean up their air?
The answer seems obvious. Of course they do. But the answer is not obvious. If the costs of compliance with those stricter mandates of the EPA are high enough, women would prefer that nothing happens. (And the costs here include the foregone net benefits from the lower amount of driving and other economic activity reduced by the tax or the standard.) Why? Why would women want to endure greater risk? The answer is that the costs of compliance with say, stricter standards may be very high and a serious chunk of those costs will fall on women. Maybe there is a cheaper way for women to avoid the harm from cardiovascular risk, by the creation of a medication that reduces the risk of a heart attack or the risk of dying from a heart attack. Or maybe not. Maybe it’s really hard to come up with such medications or medical treatments. Maybe there’s no way to avoid the damage from the extra risk in which case a stricter standard or an increase in the gas tax might be a good idea.
But the point is that there is no way to argue on purely economic grounds that the increased risk from air pollution justifies a higher tax on gasoline or stricter standards even in a world where politics plays no role in setting those standards or the level of the tax. (And I would argue that assuming that politics will NOT play a role is akin to assuming a world without transaction costs.)
In this essay [7], I give an example of why the technology alternative can easily be preferred to a tax or a standard. I show how those who are harmed in a situation of a negative externality can actually be made better off by ignoring the externality.
I have no idea whether the costs of lowering soot in the air are large or small relative to the costs of improved medical care to deal with cardiovascular risk. But what I learn from Coase is that that is the relevant comparison for deciding on the merits of increased regulation, not simply asserting that there is an externality.