post hoc, ergo time to talk

Saturday, July 4, 2009

Happy Independence Day!

Tuesday, June 30, 2009

There's no problem a good Pigouvian tax can't fix (or is there?)

Another option regulators have for limiting negative externalities like ozone-depleting emissions is to impose what's called a Pigouvian tax on the goods associated with the externality. The tax works by increasing the cost of supplying any given quantity of the good, which, ceteris paribus, will lead to a reduction in the quantity of the good that ends up being consumed, reducing the externality by extension. (Note that, in this respect, it does not matter whom the tax is collected from, i.e. the economic incidence of the tax does not depend on its statutory incidence. In other words, the same people will be paying the tax, via higher prices, regardless of who has to actually give money to the government.)

The government's goal is again to get firms to internalize the external costs their production imposes on society. The ideal Pigouvian tax is equal to the per unit external cost of producing the good that is being taxed. With the government charging the firms for the external costs they impose, the firms are forced to take the consequences of their production into account. Facing this additional cost, they will produce less.
Unfortunately, there are significant informational difficulties in establishing the ideal level of taxation. Measuring external cost is difficult, just like determining the efficient level of production. If the government gets the tax level wrong, it is imperative that it admit the mistake and fix it. The next guess may not be much better than the first, though, and, legislatively, it is relatively difficult to tinker with taxes too much. If the tax is set at the appropriate level however, the efficient level of production will be achieved without the government having to issue mandates like it did under stern orders and without creating the potential for a cartel to emerge (since the last unit produced will have the cost the firm as much as it benefits the firm). So really, there is no problem a good Pigouvian tax can't fix. The problem is finding a good Pigouvian tax.

Seriously, though, there are a couple other issues with Pigouvian taxes that make them something less than perfect, even if set at the level appropriate to reach efficient production. For one, the benefits of the Pigouvian tax are not received in proportion to the injury suffered from the externality it regulates.

Assume the government collects tax revenue and spends it on projects that benefit its entire citizenry. Now let's say that there are still steel plants in Cleveland, that they pollute, and that the federal government has imposed a Pigouvian tax on steel. Suppose revenue collected from this tax is spent nationwide on programs to plant more trees in order to address carbon dioxide emissions. The people who live near the steel plants suffer the most, as they have to breath dirty air. People in, say, Berkley, CA, are virtually unaffected by steel production in Cleveland (ignoring, for a moment, the long-term, worldwide climate change implications). However, through this taxation mechanism, Cleveland steel plants are essentially paying for the pollution they create in Cleveland by buying trees for people in Berkley.

The people most affected by the externality receive a disproportionately small portion of the benefits from its regulation. They do experience a reduction in pollution relative to the case of no regulation, but they are still not being fully compensated for the pollution they continue to experience.

Another problem with Pigouvian taxes arises when they are imposed in environments in which the costs of bargaining between those creating externalities and those affected by them are low, but I'll talk about that later.

Monday, June 29, 2009

The problem with stern orders

One thing to keep in mind regarding regulations of emissions is that the point is not to drive emissions to zero. That would be incredibly inefficient. The goal of a cap and trade system (or any similar system) is to get people (and firms) to internalize the costs their behavior imposes on society, leading society to the right amount of a emissions. When we drive our cars, or when steel plants fire up their furnaces, we create pollution, which affects other people. Our decisions, however, about how much to drive, or a firm's decision about how much steel to produce, are generally based on the costs we face, not on how these behaviors will help or hurt other people. Regulation, like cap and trade, forces us to take the consequences of our actions into account by imposing additional monetary costs on behaviors that hurt others.

A government that uses stern orders limits emissions by dictating levels of production to firms whose products are associated with pollution. The government sets the level of production that will lead to the efficient amount of emissions (the level at which the marginal cost of reducing emissions is equal to the marginal benefit of doing so).

This sounds simple. The government should just figure out what the right level of production is and tell firms to produce it, and the pollution problem will be solved. There are a few problems, however. Let's assume markets are competitive. It is hard to figure out what the right level of production is. The government would need to know all kinds of information about costs and benefits for individual firms in order for it to make stern orders work. All it knows is the market price and quantity of the goods in question. It could ask firms for the other information, but they would all have pretty strong incentive to lie to the government in order to encourage it to set terms that would be favorable to firms. Also, at least in America, firms generally do not take kindly to being told how much of their products they are allowed to produce.
(left - what government needs to know; right - what government knows)

None of these, however, is the big problem with stern orders. The big problem with regulation via stern orders is that it can actually end up helping the polluters (and hurting consumers). In order to understand this, you need to know a little bit about how monopolies work. You can follow the link for a more in-depth explanation, but, in short, a firm that is a monopoly is able to earn higher profits by restricting the available quantity of the goods it sells, which allows it to charge higher prices. A firm that faces a competitive market cannot do this because its customers will simply buy their goods from another firm that is charging the market price.

If a government regulates a competitive market with stern orders, it may help the firms in that market become a cartel, which behaves like a monopoly. In a cartel, multiple firms cooperatively limit the available quantities of goods in order to charge higher prices. The regulations require firms to reduce the quantities they are producing. Cartels, however, are difficult to maintain because each individual member has incentive to cheat by producing more and lowering its price to beat the other members of the cartel. Cartels can try to punish members that cheat (those that engage in "chiseling"), but the measures necessary to do this are often more painful to the cartel than to the cheater. The government regulations, in this case, also provide a strong enforcement mechanism for a potential cartel (the government can do things, like imposing fines or requiring shutdowns, perhaps, that a true cartel cannot do).

A cartel won't necessarily form under stern orders, but it's possible. If one does form, there will be less product available on the market, and consumers will have to pay higher prices for it than they would in a competitive situation. Consumers also experience a loss of welfare, since additional units could be produced the benefits of which would exceed the costs. Thus, this government-enhanced cartel leads to inefficiency.

Hopefully, stern orders don't look so great anymore. Once I've explained the other options, it should be clear that they are actually he worst choice all the time.

Friday, June 26, 2009

The right amount of pollution?

Today, I wished I had my old job. Nothing against my current job, but today would have been a great day to be on the Hill. Today, the House of Representatives passed the Waxman-Markee climate bill, officially known as the American Clean Energy and Security Act, which would, among other things, address the threat of global warming by, for the first time, imposing a "cap and trade" system of regulating emissions.

The vote was remarkably close (219-212), considering the Democrats' sizable majority in the House. One reason for the small margin was that the bill failed to satisfy several of the more liberal members of the Democratic caucus. Their complaints cover topics like too much spending on the wrong types of programs, going too easy on polluters, too much coal, and others.

Some people, liberal and conservative alike, don't like the cap and trade mechanism. Given the alternative means of regulation, however, cap and trade is probably the best option.

A government that wants to regulate emissions has five basic choices:
  1. Stern orders - dictate what level of production the industries in question may engage in
  2. Pigouvian taxes - impose a tax on emissions (via taxes on the goods whose production leads to pollution)
  3. Suits in common law courts - sue polluters
  4. Cap and trade - allocate alienable rights to pollute and allow firms to buy and sell them
  5. Do nothing - take no direct action, but allow polluters and those affected by pollution to bargain in order to reduce pollution to the appropriate level
This could get long fast, and since the Senate is likely to take its time with this bill, I think I'll break up my explanation into multiple posts over a couple days.

Straight from the horse's mouth

This is really cool. Treasury Secretary Timothy Geithner answers questions posed to him by readers of the Huffington Post. It might be a little bit technical, but overall I don't think it's too hard to understand.

I tried to embed it, but I don't think it worked.