I pondered earlier why an economist would use the term market failure. A fellow wrote in to tell me that he uses it as a short-hand to describe a "situation in which we would expect the free market to fail to meet the Pareto optimality criterion." He allows as how a better term might be "intractable situation" since it may not be possible to achieve an optimal solution. He is still of the opinion that regulation can come closer to the optimum in most cases of market failure.
I'm less convinced than him for the same reason that people will drive their car in a situation where the total cost is cheaper to use mass transit. A car is effectively a sunk cost. You can sell your current car, but that is almost always followed by the purchase of a new car. Well, a government is a sunk cost. Markets have a problem with sunk costs, because over time a commodity ends up being sold at the marginal cost (the cost to produce the last item sold). But the last item sold doesn't account for the sunk cost.
In the market for services, government services are over-provided.
Traditionally, not all of a car's price is considered a sunk cost. Only that which cannot be recovered is "sunk". And yet, in most parts of North America, you need a car to do everything you'd like to do. In some very small parts of America, you can reasonably do without a car: villages where you can walk, or cities which have mass transit. Thus, deciding to live anywhere else requires that you sink the cost of owning a car.
I fear that I'm being too succinct, and waving my hands at things you may not understand. Please send email if I've outrun you.
posted at: 15:03 | path: /economics | permanent link to this entry