Tuesday, June 13, 2006

Booze and Hisses for Monopoly Power

In the wake of recent court decisions that promise to create greater competition in the alcohol industry (specifically, by reducing the power of monopolistic distributors), the usual suspects are sounding the alarm: these rulings could lead to lower prices and wider alcohol distribution! (Link via To the People.)

Let’s take a moment to ponder the profound silliness of this argument. Suppose you have a product that is already sold in a competitive market. And suppose this product’s use is associated with serious externalities (or perhaps internalities) that need to be controlled. What solutions jump to mind? Here’s one:
1. Internalize the externalities by imposing greater punishments on the most dangerous uses of the product.
Here’s another:
2. Impose a tax to raise prices and reduce consumption of the product.
But would anyone seriously propose this one?
3. Implement regulations that create monopoly power in the industry, so that sellers can charge higher prices and get larger profits.
This proposal would, I hope, be immediately recognized as absurd. Even if you think raising prices to reduce consumption is a good idea, creating monopoly power is the most haphazard way you could go about it. Regulations that create monopoly power will assuredly raise prices, but not by any predictable or controllable degree. The magnitude of the price increase will depend on the number of producers who remain in the market, the overall demand for the product, the overall demand in the market, the heterogeneity of products within the market, consumers’ willingness to substitute among the different products, and a variety of other factors. These factors are not under the control of the legislature, probably unknown to them, and liable to change over time.

Between approaches 1 and 2, I’m partial to 1. Why? With most products, alcohol included, externalities are associated with particular types of use – e.g., heavy drinking, drinking while driving, and so on. But a price increase will discourage all types of use, not just the most damaging ones. And it might even have the greatest impact on the least damaging uses; there’s good evidence, for example, that heavy drinkers are less responsive than moderate drinkers to price hikes.

But if you’re nonetheless convinced that raising the price is the right way to go, approach 2 (the tax) is clearly superior to 3. At least that way you can choose the tax to correspond to the best available estimates of the external cost of the activity in question. In addition, the tax revenues can be used to compensate victims, to reduce the burden of other taxes, or to fund public services. Monopolistic regulations, on the other hand, cannot be used to create a specific price increment, and their benefits go to the remaining firms in the form of higher profits.

I’m sure you won’t be surprised to find out that it’s alcohol distributors, whose profits are most inflated by the existing regulations, who are backing position 3. Of course, in this case they’re trying to defend existing monopolistic regulations instead of creating new ones, but the logic is the same.


Colin said...

Professor Whitman,

Putting aside the issue of externalities and internalities, isn't real wealth lost simply due to the dead weight loss associated with monopolies.

Essentially, more competition means more alcohol is produced and therefore, breaking up a monopoly would create more wealth, right?

Glen Whitman said...

Colin -- yes, that would be the usual analysis. Less competition can sometimes be good because larger firms can exploit economies of scale, but if so, that will happen on its own without gov't assistance. So getting rid of regulations that encourage monopoly should, in general, create more wealth by eliminating underproduction.

However, when negative externalities are present, that will tend to create overproduction, which reduces wealth. So the underproduction caused by monopoly and the overproduction caused by externalities could (partially) offset each other. That is the kernel of truth behind the argument I'm arguing against in the post.