Elephantstrunk

This Crooked Timber post reminds me of a pet theory, under-googled and probably unoriginal where correct, that the virtues of markets get badly misunderstood.

It goes as follows:
Markets are intrinsically inefficient. For there to be competition there is very likely to be excess capacity and significant effort devoted to suboptimal solutions. It seems therefore that for a given definition of efficiency, the market solution will not only not be efficient but will be so in a way that a real, live, human planner might well be able to exploit. That strongly contradicts what is supposed to obtain in a perfectly competitive market.

Instead the advantage of markets come from their adaptability and ability to learn which is directly related to their capacity to support allegedly suboptimal solutions and to adjust their verdict upon them. While at a given moment the market based solution might well be significantly suboptimal, over time effective and relatively undisruptive adaptation and learning will win out. Essentially I think that in the story of the Hare and the Tortoise, the market is the Tortoise. Hindsight based evidence comes from the rapid boom and bust of the Soviet and Nazi economies.

For this reason the problem with government services is often that they are undercapitalised and over optimised. This has particularly bad consequences when market reforms mix with state funding. What is clearly missing from school choice is excess capacity, the NHS costs less than US state medical care but attempts fill the role of both state and most US private health expenditure.

In general I think accepting such a theory implies that many criticiisms of state-control are wrong headed and therefore in many instances also wrong. However in the case of the Crooked Timber post I think the argument works against the state solution.

The problem is that one size does not fit all and the state would be prevented from making flexible and case sensitive decisions on the matter. One commenter asked why 75% and not another number. It’s a good question but already gives away too much. Why is one number correct for all companies? How can a steel mill and Microsoft fit into the same scheme? Highly cyclical businesses would suffer.

This is in part because the idea seems to assume the existence of some kind of economic constant at the level of firm profitability which I am sure does not exist. In essence the impossibility theorem can only be the start of the argument. It doesn’t say what actually is going on inside firms and certainly doesn’t give you the information needed to fix an optimal rate of such a tax although it might allow you to argue that there are reasons why a tax might actually be useful. Many taxes are economically inexplicable and people are happy to argue for tax cuts from premises which in fact suggest that the world would be a better place if there was no tax at all.