TCS Daily


By James V. DeLong - July 29, 2003 12:00 AM

The current hysterical assault on industries that deal in intellectual property, primarily pharmaceuticals and entertainment, seems utterly baffling. These industries spew out extraordinary floods of worthy things: life-saving, life-enhancing drugs; breathtaking movies; music for every possible taste. And both are bitterly demonized, as if they cheat us by asking that money be paid in return for their wares.


Much of the demonization comes from the left, particularly academia, powered by intellectuals' habitual disdain for the market and lust for a government run by themselves. But disaffected intellectuals are always with us, and the deeper problem is that this disdain has found a lever to wield that has been created by the real culprits -- the economics profession.


This lever is a syllogism that I have now heard to the screaming point: It starts with the proposition, "Economics teaches us that in a competitive market prices equal marginal cost -- the extra costs incurred by producing an additional unit." It goes on to note that the marginal cost of an additional pill, or an additional copy of a movie or song, is close to zero. Therefore, the argument concludes triumphantly, "economics teaches us" that such products should be priced at zero. Any other condition demonstrates that undue "market power" exists, and is immoral. If one demurs to the logic, on the ground that it costs about $800 million dollars to produce the first pill, or $100 million to produce the first print of the movie and this initial investment -- not just the marginal cost of the second pill or the second print -- must be recovered from somewhere, the perpetrators of this logic usually shrug and talk about the need for new business models.


If you doubt the power of the "price should equal marginal cost" mantra, or the moral component that has been infused into it, check out speeches by officials of the Antitrust Division, such as William Kolasky, former Deputy AG: "An economist would say that a market is perfectly competitive when firms price their output at marginal cost." Or a statement by Lawrence Summers, former Secretary of the Treasury and current President of Harvard: "[T]he most basic condition for economic efficiency [is] that price equal marginal cost." Or go look at almost any textbook, complete with diagrams.


To be fair to these two gentlemen, they immediately move on to note that of course this principle creates problems in an investment-intensive context because it does not allow for the recovery of capital cost. They have no answer for this dilemma, however, and they leave an impression that such situations are aberrational, perhaps temporary, and not a real issue, though Summers notes that this segment of the economy is growing.


The net result is to leave two misconceptions to run amok. First, that the fundamental theorem has any meaningful application outside of the classroom, and, second, that investment-heavy industries are exceptional cases when in fact they constitute almost the entire economy.


In explaining these misconceptions, it is helpful to focus on specific situations. Suppose I have a widget factory, routinely selling my widgets for $100 each. Someone comes to my factory and says "I'll give you $10 if you add one more widget to today's production run for me." I calculate that it would cost $9 in materials and labor to make the widget. Economic theory teaches that, judging solely by the optimal use of resources at that frozen moment of time, I should make the deal. If I do not, then welfare will be reduced because the buyer will not receive his utility, and the resources will either go unused or be put to an inferior use.


But this really says nothing about the price of widgets generally. Does it mean all my other widgets should have been sold for $10? No -- when I was thinking of investing in a widget factory my marginal cost calculation had to encompass the investment cost, so it was then nearer $100, not $9, whereas the theorem refers only to this precise day, even this hour. Does today's marginal cost say anything about my future prices? Not really; I need to recapture depreciation and interest, so $9 will not be enough.


So should I make the deal? Maybe. It depends on the overall effect on my business. Will my other customers find out? Will this guy pay the regular price if I refuse? How much will the competition charge? Will I lose good will from others? Do I save transaction costs by refusing to bargain, so that a special price cut will cost me more in the long run? In the real world, "price" is not just this immediate deal but its ramifications across time and space, and comes from complicated business calculations.


"Marginal cost" has a similar time dimension -- it all depends on how a long a period one is looking at and what costs one must include. The axiom "prices must equal marginal cost" does not tell you whether the relevant time dimension is a decade, a year, or an hour, which makes it into a meaningless statement. So to set up an identity between marginal cost and price, without a tight specification of the assumptions about time, or to assume that short-term marginal cost is the ticket, produces nonsense.


To drive the point home, look at real world examples.


An airliner in Chicago boarding for a trip to San Francisco has some seats empty. Just before the gate closes, a traveler offers the agent $20 to get on. The cost of the extra fuel burned is less than this, and all other costs are, as of that frozen moment in time, fixed, so the airline would indeed make money by taking the deal. But this hardly means that "economics teaches us" that $20 is the correct price in general for the Chicago-San Francisco run. It does not even teach us that the agent should make this one deal, since much depends on calculations of the effect on the overall price structure. Such as, would all passengers then wait at the gate until the last minute, when bargains would be available, and what would this mean for airline pricing and operations?


One can replay analogous scenarios in many other contexts. An automobile assembly line that is running at less than full capacity. A customer entering a stationery store to buy a pen that is priced at $2.00 when it cost about 20 cents to make. Telecommunications services -- the marginal cost of sending a bit around the world a dozen times is close to zero.


In fact, reverse the question. Try to think of an industry in modern America in which marginal cost pricing is workable, in which it would actually promote long-term efficiency. Further, since the dynamic long term is necessarily the sum of the static short-terms, the common distinction between short term "static efficiency" and long-term "dynamic efficiency" does not work either, except under particular limited assumptions wherein the economy is composed of end-of-season or going-out-of-business sales. In the real industrial world, marginal cost is simply the lower bound on price, not the ideal. A company will sometimes sell at this level, but its overall price level must be sufficient to cover every cost, and to conceive of marginal cost pricing as the ideal is demented.


This tension between marginal cost theory and the real world has always caused discomfort. Generations of antitrust lawyers have regarded themselves as the guardians of marginal cost pricing, and view real world departures from it as temporary aberrations in need of correction. This results in blackly humorous interactions between them and businessmen who know, inarticulately, that the standard is impossible to meet, but must pretend to be with the program. They cannot confess that they actually spend all their time scheming how to acquire market power so they will not get caught in the death spiral of marginal cost pricing.


The tension has been tolerable as long as companies could find roads to market power that eluded the technicalities of antitrust doctrine and the attention of a tabloid-ized press. Primarily, heavy investment costs create barriers to entry which, combined with a knowledge of game theory and some conscious parallelism, allow industrial America to function.


But the IP industries have been stripped of this protection. The heavy costs in the pharmaceutical business are in the initial research and the regulatory costs, not is setting up a plant to stamp out pills. The entertainment industry is beleaguered by the new reality that duplicating and transmitting bits takes no capital investment, once the basic computer and telecom networks are in place, so the cost barrier has vanished completely.


Both these industries have been reduced to relying on the protections of the legal system, which means the political system, and that is creating a problem, because the marginal cost mantra has undermined their moral position. People really do feel outraged at the idea of full cost pricing.


It is time for the economists to stop misleading a gullible public. They cannot claim ignorance. Their predecessors of the late 19th and early 20th centuries were skeptical of the Sherman Act and subsequent antitrust laws precisely because of the difficulties of accounting for fixed costs in competitive analysis. Ronald Coase wrote on "The Marginal Cost Controversy" in 1946, in an essay incorporated into The Firm, The Market, and The Law. Professor George Bittlingmayer of Kansas has examined the instabilities in pricing created by fixed costs, using core theory, a concept developed by Professor Lester Telser of the University of Chicago. Holman Jenkins, Wall Street Journal columnist, gets into the issue from time to time.


But in the context of pharmaceuticals and intellectual property, the concern is over the moral impact, not the long-term economics. The current situation, wherein sectors of the economy that are both vital and incredibly productive are systematically delegitimized by moral opprobrium, is not tolerable. It is demonstrating the dark side of Lord Keynes famous dictum:

[T]he ideas of economists and political philosophers, both when they are right and when they are wrong, are more powerful than is commonly understood. . . . . Practical men, who believe themselves to be quite exempt from any intellectual influences, are usually the slaves of some defunct economist. Madmen in authority, who hear voices in the air, are distilling their frenzy from some academic scribbler of a few years back.

The academic scribblers need to scribble some new ideas.

The author is Senior Fellow & Director, Center for the Study of Digital Property, Progress & Freedom Foundation, Washington, D.C. Mr. DeLong is an attorney. His ideas are his own, and do not necessarily represent those of PFF.

TCS Daily Archives