Feed on

Yesterday the New York Times ran a piece describing Professor Bill Baumol’s “cost disease” problem in economics. Very simply, the theory says that some sectors that are labor intensive would be expected to have its costs increase faster than less labor intensive sectors because productivity improvements are harder to come by when lots of human labor is required. So, sectors like the arts, education and health care are susceptible to larger cost increases than we see elsewhere.

Let’s abstract from the fact that expenditures and costs are not the same thing because some (much) of the increase in “cost” in these areas come from the fact that we are a lot richer, and dedicate a very small share of our budgets to necessities today as compared to in the past – leaving us with more disposable income to spend in these other “productivity stagnating” areas. So, let’s assume that all of the increased expenditures are in fact the areas where costs themselves are increasing. (To understand the distinction, clearly Americans are spending a lot more in aggregate in iPods today than they did 5 years ago, even as the price (cost) of an iPod is much lower today than it was 5 years ago – such increased spending hardly constitutes a crisis).

Baumol’s canonical illustration of the cost disease is that it still takes 4 people the same amount of time and effort to perform a string quartet rendition of Mozart today as it did back in the 18th century. In education the argument would be that a professor could teach only so many students. In health care, a surgeon can only operate on one patient at a time, and doctors can see only one patient at a time. But such a narrow view abstracts from two hugely important factors. First, it ignores the wide range of substitutability that is available today that was not available in the past.  Take the case of a common stomach problem – ulcers. If the good in question is “having my stomach cut open by a surgeon” then perhaps there are limitations to what technology can do to replace human labor or to make human labor more productive (I would still vigorously dispute this as people having an extreme lack of imagination – recall the claims of the US Patent officer at the end of the 19th century that everything useful that can be created has already been created). Indeed, Baumol argues that even if we get a robot to perform the surgery, we still require human monitoring of the robot.This ignores the improvements in quality and reliability we are likely to get from this, and also ignores the possibility that robots can operate more quickly than humans. Even so, Baumol would argue that such technological improvements are merely transitory, giving us one time shifts down in costs, and the higher trajectory of cost increases would remain in that sector. That may be so, but then that makes the assumption that no additional one time shifts are possible or likely. I simply do not find that plausible. For example, suppose some extractable resource exhibits diminishing returns as you extract it – you would expect its price to increase as you extract more of it. But if a large number of identical lodes are available, each with similar cost structures, one would not expect to see a general rise in the cost of the resource over time. Similarly, there is no reason to believe that regular improvements in technology cannot be applied even to the most labor intensive areas of medicine. Wasn’t the US Patent Office famous for arguing in the 1890s that all of the good inventions that could be created have been created? To think that we cannot improve productivity dramatically even in labor intensive areas is to suffer from an extreme lack of imagination.

However, the good in question in most medical cases is not something like, “having my stomach cut open by a surgeon,” but rather something to the effect of , “a healed stomach.” If what we care about is having a healthy stomach then there is no reason to believe that there are limits to how we can achieve this at lower cost. In fact, in the case of an ulcer, whereas in the past your only options were to deal with it, or to have expensive and dangerous invasive surgery, today many ulcers can be treated with a … pill. And these are pills whose marginal cost of production is very nearly zero. Is there any reason to believe that many diseases and ailments could not be treated this way?

But even pointing to advances in medicine to treat disease suffers from an extreme lack of imagination. In the case of ulcers, more research and better understanding of human physiology and psychology may lead us to nutritional or lifestyle adjustments that can treat the ulcers or even to prevent them from occuring in the first place. In that vein, do we not now have a wide variety of nearly costless vaccines to prevent the emergence of many other diseases from hitting us? And are there not gene therapies and modifications that could be on the horizon? And no, there is no guarantee that these will emerge. But so long as men are tolerably free and society is reasonably tolerant of entrepreneurial risk taking it is absurd to think that entrepreneurs would not find this cost disease as an opportunity rather than a curse. It is precisely the presence of high costs which incentivizes the kinds of things I am talking about today. Government provision or government subsidization in the name of some mythical “market failure” will only inhibit the process, and ultimately make us poorer than we otherwise would be.

Which brings me to the second point. You simply cannot discuss the issue of “cost-disease” in an institutional vacuum. Clearly, as we have grown richer, we have wanted to spend our extra income on a whole host of other things, and some of those are sectors that are labor intensive and would seem to suffer from the same “structural” problems that education and health care are argued to be suffering from. But why no cost diseases there? Is it merely a coincidence that the two sectors with the fastest cost growth also happen to be among the two areas with the most government, union and third party involvement (I say among because the financial sector has historically always been the one most heavily influenced by government)? And the “solutions” to the cost disease problem, as outlined in the aforementioned New York Times piece seems to be more government involvement – either via subsidies (hint to students, what happens to costs when demand is subsidized) direct provision or some other convoluted mechanism.

Third, much of what one would consider a reflection of a “cost-disease” is nothing of the sort. For something to exhibit this “cost-disease” feature, we must hold constant everything we can about the product and ask the question, is the cost of providing the same relative product increasing faster than the cost of providing other products whose characteristics and qualities are the same. But it is hard to do that in the health care sector. Or is it? What would it cost to provide 1963 quality health care to patients today? And how would that compare to what it would cost to provide alternative goods? I am not sure you’d get the same answers that Baumol and other state lovers are getting. Even so, much of what you would call cost increases today I would characterize as … cost decreases.

How can this be? Well, if you were stricken with cancer 100 years ago, your choice was to suffer and die or to undergo some medieval experimental treatment. I’d pin the costs of treating cancer then at near infinity. State lovers seem to be pinning this cost at zero – since no human resources are used to treat such an ailment. As our know-how and income have increased, we are now able to eliminate some forms of cancer, and mitigate the suffering and prolong the lives of folks with some kinds of cancers. Even if these come at high costs, I would argue that it would be incorrect to call these advancements an increase in costs of delivering health care. Or to push the discussion to the top level – would you point to higher family incomes today as compared to 100 years ago and condemn all of our spending on new products as being a result of an intractable cost-disease problem?

That economists have thought up the “cost-disease” theory to explain patterns in the data does not mean that the paradigm is correct, nor does it mean that if it is correct, that there is anything obviously to be done about it. For example, if my theory of typewriters leads me to posit a theory that typewriter prices and quantities are a function of these strange things called supply and demand, and for some reason the demand for typewriters plummets as new technologies develop … does it follow that something should be done in the typewriter sector? Even if typewriters were vital products for all Americans? No. But, the cost disease paradigm seems to be gaining in popularity, much like the “externalities” paradigm has, both inside and outside of economics, to justify government intervention into the affairs of individuals and firms. I dispute here that the cost-disease issue is even happening, or if it is relevant. But even if you do not agree with one iota of what I argue above, it still does not follow that government intervention in the health and education markets is any more justified than if there were no such thing as the cost disease. There are two reasons I would argue this. The first I link to above – all of the usual public choice arguments apply. But second, if in fact you wish to argue that the cost disease matters, is not the government itself an institution that is perfectly characterized by this same exact problem? And how could you justify the intervention in the cost disease problem by an institution suffering from that very same problem? Would you argue with a straight face that they have the knowledge and ability to overcome the cost disease problem in education and health care, but for some reason do not apply it to their very own institution? Would you argue that getting consumers more education and health care at reasonable costs is more important than getting better government services at more reasonable costs? On what grounds?

I’ll repeat myself: if a society is reasonably free and tolerates entrepreneurial risk-taking – the presence of the cost-disease presents an enormous opportunity and not a problem for economic agents. And more important than any assumption of rationality in economics is that “no cash be left on the table.” It is highly unlikely that entrepreneurs would leave such enormous piles there for the taking, but take it aggressively everywhere else.

Here is an excellent article analyzing the cost disease problem in the arts sector. It starts with a discussion of how even Baumol’s canonical example of the string quartet today still requires the same 40 minutes to perform that it did 200 years ago may not even be a correct illustration of the cost disease.

Leave a Reply