Michael Perelman: Steal This Idea

Michael Perelman is a fairly prolific left-wing economist. I noticed several of his books, then looked him up in the library, finding this one: Steal This Idea: Intellectual Property Rights and the Corporate Confiscation of Creativity (2002, Palgrave Macmillan; paperback, 2004, Palgrave Macmillan). I have my own critique of various issues in intellectual property, and consider it to be an important issue. This is a useful book, but Perelman makes a critical mistake in treating intellectual property as a single issue. It is, rather, an artificially agglomerated set of laws, each with its own issues. (Richard Stallman has lectured us to death on this point. The lack of any reference to Stallman, or more generally to free software, is a shortcoming in the book.) But while Perelman talks about intellectual property, for the most part he means patents. And his critique of patents does go beyond the usual norm to consider such issues as the negative impact of patents on scientific research and the contribution they make to a culture of secrecy and litigiousness.

The book has lots of case examples, especially on patents. One general point is the role of intellectual property in increasing inequality both within the US and throughout the world. This is one case where it may be appropriate to generalize about intellectual property, since even such innocuous properties as trademarks, backed by sufficient advertising and an appropriate culture, have precisely that effect.

The selected quotes are mostly self-explanatory.

Some general statements (p. 3):

Today, matters are completely different. Intellectual property covers just about everything. The system is riddled with overlapping claims. The contemporary system of intellectual property, rather than spreading information, creates a pervasive atmosphere of secrecy. Litigation is becoming far more important than creativity. In fact, I will show that intellectual property rights threaten to stifle creativity. Taking a historical view, we can compare the system of intellectual property to a stimulant that may well have promoted economic and cultural progress in an earlier period, but now threatens to exhaust creative activity.

Even in the arts, intellectual property rights offer very little to the mass of creative artists. In contrast, intellectual property rights grant enormous powers to corporations that distribute music or run movie studios. These corporations typically wield their power to the disadvantage of the artists, as well as society at large.

In science, intellectual property rights encourage secrecy and wasteful duplication of effort. They hold back economic progress by fostering inefficient monopolies. They encourage costly litigation that dissipates an unimaginable amount of time and resources. Over and above these problems, intellectual property rights pervert the entire scientific process by undermining the traditional incentives to engage in the basic scientific research essential to developing future improvements in technology.

Actually, Perelman doesn't have much to say about the arts case. The implication that copyrights benefit corporations at the expense of artists is certainly true, but that doesn't mean that artists don't benefit from copyrights, or that simply eliminating copyrights would benefit artists.

On developed vs. developing countries (p. 6):

Internationally, a regime of intellectual property rights condemns the poorest countries of the world to an even more disadvantaged future. For example, the United Nations reports that in 1993, just 10 countries accounted for 84 percent of global research and development expenditures. These same countries controlled 95 percent of patents registered in the United States during the past two decades. The rich, industrialized countries now hold 97 percent of all patents worldwide. Compounding the inequity, more than 80 percent of patents granted in developing countries belong to residents of industrial countries. No doubt this situation has worsened in the intervening years.

(p. 34):

Texas Instruments struck first. Typically license fees ran about 1 percent of revenues. In 1987, Texas Instruments raised its royalties on chips to 5 percent. The company filed a suit against one Korean and eight Japanese semiconductor companies, accusing them of infringing semiconductor patents. The settlements yielded the company more than $600 million in payments, according to a 1990 report. The company became so aggressive in seeking royalties that by 1992 it earned $391 million in royalties, compared to an operating income of only $274 million.

Other companies are even more successful. For example, IBM's annual report announced that the firm had earned more than $1.5 billion in income in 2000 from its intellectual property portfolio.

(p. 36):

From the standpoint of exports alone, this emphasis on intellectual property proved highly successful. In 1947, intellectual property comprised just under 10 percent of all U.S. exports. By 1986, the figure had grown to more than 37 percent. By the early 1990s, the best estimate was that intellectual property accounted for well over 50 percent of exports from the United States. In 1999, U.S. exports in the form of royalties and licensing revenue alone exceeded $37 billion -- topping aircraft exports, at $29 billion, and telecommunications equipment. Moreover, the trade surplus in intellectual property -- the exports minus imports -- is running at about $25 billion annually, and growing. As already noted, IBM alone enjoyed worldwide licensing revenues that exceeded $1.5 billion, according to its annual report for 2000. These figures exclude payments for physical goods, such as computer chips, which also embody intellectual property.

On radio patent conflicts, which the US Navy had worked to manage during WWI (pp. 51-52):

Once World War I was over, the litigation recommenced. Between 1900 and 1941, a total of 1,567 infringement suits entangled 684 different radio patents. These patent suits extracted a heavy price in terms of technological development. Reviewing the history of the British radio industry, one writer observed with a notable understatement that radio manufacturers in Britain wasted "a lot of ingenuity" during the 1920s devising circuit arrangements that reduced the royalties that would otherwise have to be paid to British Marconi. Although radio tube technology advanced in the process, this progress would have been much greater had researchers not directed so much energy to working around existing patents.

(p. 60):

Cadtrack was headed toward bankruptcy in 1983 when the head of licensing at IBM called to discuss the possible licensing of a patent for moving a cursor on a screen. In 1985, [Cadtrack CEO Eugene] Emmerich went to the board of his company and suggested that the company get out of the production business. The company then laid off all of its employees and concentrated on collecting revenues from its patent. By 1997, when the patent finally expired, he had signed deals worth about $50 million with 400 companies.

Emmerich was proud that only one company refused to take Cadtrack's license -- Commodore. He boasted, "So we took them to court and got a permanent injunction barring sales of their computers in the U.S. When that happened, their creditors called in their loans and they went bankrupt. That little patent of ours put Commodore out of business."

The latter paragraph helps explain why IBM brought up the issue of the patent in the first place. While it may seem to have added to IBM's costs, thereby cutting into profits, the net effect was to drive one of IBM's major PC competitors out of the market. Perelman doesn't go into this sort of strategy, but IBM has played this game before. In fact, their initial success in mainframes was at least in part based on their licensing of early computer patents: while IBM's competitors were busy suing each other over patents, IBM advanced to dominate the market.

(pp. 83-84):

The researchers who developed the transistor were not doing pure science. A 1931 paper had already laid out the basics of a quantum mechanical model of a solid semiconductor. Their work was not merely applied, either. It was something in between. The history of the transistor also illustrates how important accidents can determine intellectual property rights. A team at Purdue University was within weeks of discovering the transistor.

If AT&T had been free to use its intellectual property rights in the transistor the way contemporary firms can and do, modern technology would be far less advanced than it is today. Merges and Nelson have written:

Because of an antitrust consent decree, AT&T was foreclosed from the commercial transistor business. . . . [As a result] AT&T had every incentive to encourage other companies to advance transistor technology, because of the value of better transistors to the phone system. AT&T quickly entered into a large number of license agreements at low royalty rates. Many companies ultimately contributed to the advance of transistor technology because the pioneers patents were freely licensed instead of being used to block access.

Because of government intervention, intellectual property rights did not limit the revolutionary potential of the transistor as they might have. Richard C. Levin, economics professor and later president of Yale University, speculated some time ago that the computer industry might not have developed if AT&T had not been forced to license the transistor to all comers.

(p. 102):

In this world, academic careers rest on the ability to land corporate or government (mostly military) contracts. Researchers can either work at the behest of corporate "donors" or attempt to become independent by seeking out profitable discoveries either to patent or to use as the basis for their own firms.

The legal system is bending over backward to accommodate such practices. Today, when a biologist can patent a sequence of genetic material or a mathematician can patent an algorithm, money rather than the acclaim of colleagues becomes the coin of the realm. Researchers, who once worked in the open to win recognition from their peers, now shroud their research in secrecy in the hope of striking it rich.

(p. 133):

Dean Baker of the Center for Economics and Policy Research made a few rough calculations concerning the costs of intellectual property in the pharmaceutical industry. Presently, people in the United States spend close to $100 billion a year on prescription drugs. In the absence of patent protection, Baker estimated that the cost of these drugs would fall to less than $25 billion -- a savings of more than $500 a year for every household in the country. By contrast, the proponents of deregulation in the airline, trucking, and telecommunications industries put the gains from each of these policies in the neighborhood of $10 billion to $20 billion annually.

Yes, but what about the great medical advances that arise out of the efforts of these companies? Dean Baker observed:

According to its own data, the pharmaceutical industry funds only 43 percent of medical research in the United States. The federal government funds close to a third of all medical research, primarily through the National Institutes of Health. Universities, private foundations and charities account for the rest. These other methods of funding research have a proven track record. This research has produced a long list of major medical breakthroughs, including the discovery of penicillin, the polio vaccine and AZT (though not its use as an AIDS treatment). In just the past two months, NIH researchers developed a vaccine that will prevent the transmission of AIDS through breast-feeding, and a use for aspirin for people undergoing heart surgery. The industry is presently spending approximately $20 billion a year on research. Some portion of this spending, probably in the neighborhood of one-third, is devoted to researching copycat drugs. But in the absence of the patent and the amount of research spending that would have to be picked up in the absence of patent protection comes to approximately $13.3 [b]illion a year. This amount is approximately equal to what state and federal governments could expect to save on Medicare and Medicaid payments for prescription drugs in the absence of patent protection. . . . In other words, this would allow the patented price of drugs to fall to a free market price that on average would be less than 25 percent (and in many cases less than 5 percent) of the patent-protected price.

The quote actually says $13.3 million, but the math and logic argue for billion. Cost is actually only one issue here. Public funding of pharmaceutical development would also entail public testing, which would make it harder to hide dangerous complications -- indeed, it would pretty much eliminate the motivation to cover them up. That in turn would limit liabilities, a big expense for the industry in its own right. Public research would also put more emphasis on cures and vaccines, which are economically less profitable to the industry than long-term palliatives.

(p. 156):

Federal officials have not challenged the industry in this respect. In fact, they have not even bothered to keep track of the products, including drugs, that have profited from federally funded research. A 1995 study done at MIT found that of the 14 new drugs the industry identified as the most medically significant in the preceding 25 years, 11 had their roots in studies paid for by the government. In 1999, a preliminary report by the inspector general's office of the Department of Health and Human Services found that as many as 22 percent of discoveries financed by the federal health institutes were not reported by universities, as is required. More than 2,000 inventions developed with government money were reported to the health institutes last year, but officials told the New York Times that they had no idea which, if any, companies had licensed those inventions, or how they were being used.

(p. 159):

Like most forms of public investment, public health has suffered from terrible neglect in the United States. In the words of Laurie Garrett, author of a panoramic study of the decline in public health: "It took centuries to build a public health system, and less than two decades to bring it down. Once the envy of the world, America's public health infrastructure was, at the end of the twentieth century, indeed in a shambles.

The full consequences of erosion of the public health system will not be felt until the nation faces an emergency, such as the rapid outbreak of a dangerous epidemic for which the system is not prepared. The anthrax scare of 2001 should have brought home this point.

(p. 160):

The history of tetraethyl lead, the poisonous gasoline additive that has since been banned, brings together a number of threads in this book, including the point I just mentioned about the rationality of preventing rather than curing illness. Intellectual property rights were a central factor in the initial development of this lethal product. The early research on gas additives actually favored alcohol, which could be made from agricultural waste products. The opportunity to gain a monopoly through patent rights was the main advantage of lead-based additives.

(pp. 177-178):

But information, a major constituent of intellectual property rights, is not scarce. As Kenneth Arrow recently noted, "Patents and copyrights are social innovations designed to create artificial scarcities where none exist naturally." In spite of the efforts to make information artificially scarce, economists realize that information differs from scarce goods, such as detergents or canned soups.

These scarcities, however, serve no social purpose whatsoever. In fact, using the market to exclude people from access to information is self-defeating. It does not increase the supply of information. It only spreads ignorance. Nor does my consumption of information detract from the access of anybody else; it may even add to the pool of social information, possibly creating an advantage for others. As a result, fields of research are very different from agricultural fields. While exclusivity is imperative in the farmer's field, it makes no sense whatsoever in science. After all, the more information that I gather, the more potential information is available to you.

For example, if you let me read your book or use your computer program, you may benefit from sharing the fruits of my experience. In fact, unlike so-called rivalrous goods, which can be used up, the more that people partake of the supply of information, the greater the total stock of information becomes. In short, using information can spawn more and better information. For instance, as a scientist learns more about her field, she has more to share with others. While scientists might compete with each other for the priority of a finding, the discovery of one enriches all. [ . . . ]

I cannot emphasize this point enough: The concept of scarcity is absolutely irrelevant to information. The more the law restricts people's access to information, the less information will be available.

One might also point out that the classic theory of markets, per Adam Smith, assumes perfect information. As such, efforts to limit information only serve to subvert market efficiency.

(p. 182):

Most economists make the case for awarding intellectual property rights to the "owners" of information by applying one side of the logic of public goods. They accept that in competitive markets prices fall toward marginal costs and the marginal cost of information is zero. At a zero price, firms would not have an incentive to produce information because they could not make a profit for their efforts.

Such economists conclude that the solution is to treat the information as intellectual property, thereby converting a public good into a monopoly. In making this case, they ignore the other half of the logic of public goods; namely, the central proposition of economic theory, which maintains that efficiency is maximized when goods sell for their marginal costs. Habitually caught within the narrow confines of their economic models, these economists content that the monopoly is required to provide the incentive to create information.

I'm struck here by the either-or logic: that the only alternatives are free information and monopoly. Since monopolies are generally, and properly, understood as inefficient, economists should go out of their way to devise methods that provide a marketable value for information without locking it up in a monopoly. Such methods are possible; e.g., mechanical licensing for radio performance of music. Alternatively, one could devise systems to promote the creation of free information, putting a value on its creation as opposed to its marginal cost. That so many economists hasten to support monopoly just goes to show that their fundamental instinct is to rally behind the status quo.

(p. 187):

[Paul] Samuelson laid the framework of treating such goods as quasi-public goods by insisting that goods can be more or less rivalrous, falling along a continuum. For example, if a software program costs a few cents to reproduce, it is not entirely non-rivalrous, even though it has much more in common with a public than a private good. In this respect, Samuelson showed that it should be treated as if it were a public good.

While Samuelson was correct to insist on the inefficiencies caused by treating public goods as private goods, he missed a larger dimension of the problem: namely, that the privatization of public goods can distort the nature of the goods themselves, or even the way that they are produced. For example, the scrambling of television signals creates an inefficiency that harms society, but the damage arising from this practice may seem minimal.

I'll write more about this in the future, but briefly I don't see any justification for patent protection, even if it could be modified to limit the worst abuses of monopoly grants. There may be a few minor instances where privately funded research would be abandoned without the promise of a patent payoff (e.g., in pharmaceuticals), but those cases could easily be remedied with a little public funding, and the public information sharing and the ability to build on each other's advances would be a positive advance that patents currently preclude. In most other cases there is no value whatsoever. Indeed, patents are often no more than an artificial means of legally enforced inequity.

Other intellectual property issues are more vexed, and need to be sorted out case by case. But the one thing they do have in common is that they are all cases of creating property by legal fiat as opposed to by possession or obligation. As such, there is no necessary reason that they exist. So such cases need to justify themselves as serving some public good as well as private benefit. Mostly, those cases come down to how people get paid for work. Copyrights, for instance, help to support artists, and they have the advantage of doing so in an unpolitical way, but they are not the only way to motivate artists.

posted 2007-09-26