By Stan Liebowitz
Intellectual property is normally defined as the set of products protected under laws associated with copyright, patent, trademark, industrial design, and trade secrets. The U.S. Constitution expressly allows for intellectual property protection, albeit for a limited time, in the form of protection of “writings and discoveries” in order to promote “science and useful arts.” This article focuses on the two most important categories: copyright and patent law.
Copyright, which covers the expression of ideas (e.g., through words or music), currently lasts for the rest of the author’s life plus seventy years (or ninety-five years after publication if the product is a “work-for-hire”). But the protection is very narrow. If someone else should, by a remarkable coincidence, write exactly the same song or story as you without ever coming into contact with your work, your prior copyright does not prevent him from selling his work. Copyright currently exists on a work without any effort on the part of the author to attain copyright and without any requirement of quality or originality.
Patents, in contrast, last for twenty years and apply to inventions. The protection, although shorter, is broader than that of copyright. If someone else independently creates a duplicate of your invention after you have patented yours, your patent can make his invention worthless since he will not have the legal right to sell his version. This may be true even if his invention is slightly different from yours. For this reason, being the first to patent a valuable idea is very important, and “patent races,” as competitors vie to be first, can be a wasteful use of resources. Unlike copyright, getting the legal patent from the patent office requires spending resources, and before a patent is granted, the ideas that are to be patented must pass several legal hurdles regarding their originality and quality.
Although expression and invention must be transformed into physical embodiments before they can have market value, they can also exist, and indeed must originally exist, in the creator’s mind. As such, traditional laws of property, which require physicality, do not apply. Traditional laws of economics, such as the assumption of scarcity, also seem not to apply because individual expressions and ideas cannot be used up.
Economists have a term for goods that cannot be used up—“nonrivalrous consumption” (sometimes known as “public goods”)—and these goods require a different form of analysis than more typical economic goods. In particular, the prices individual consumers are willing to pay are summed to arrive at an overall demand for the intellectual product, a process known as the vertical addition of demands. This differs from the more traditional summation of quantities each individual demands at various prices—horizontal addition—to derive the market demand for rivalrous goods.
There is no practical mechanism for the ideal production of nonrivalrous goods, as Kenneth Arrow concluded in his classic 1962 article. Harold Demsetz properly noted, though, in his classic 1969 critique, that efficiency does not require perfection, and so markets might still produce these goods efficiently. The difficulty of ideally producing nonrivalrous goods is contained in the well-known trade-off between production and consumption, to which we now turn.
The Consumption-Production Trade-off
The fact that ideas and expressions do not get used up allows for an unusual result in terms of the “efficient,” or ideal, level of consumption. I will illustrate the example with a creative expression, but it applies equally well to ideas that are patented.
Since my listening to a song does not reduce your ability to listen to the same song, efficient consumption of that song, once it has been produced, is to allow everyone who has a positive value of the song that is greater than the cost of transmitting the song (often assumed to be zero) to consume the song.
This is a quite remarkable result. Typical goods, such as apples, are scarce, meaning that there are fewer in existence than the number that potential consumers would wish to eat if apples were freely available; thus, some rationing mechanism, such as price, must be used to determine who gets the apples. Efficient consumption requires that consumers who value the apples more get them and consumers who value them less go without. The correct allocation of apples is important in achieving efficiency.
The ability of one unit of a nonrivalrous good to provide for the entire set of users turns usual rules of consumption efficiency on their head. Everyone, today and in every future generation to come, without limit, can listen to Beethoven’s Ninth Symphony. There is no allocation problem to be solved. Thus, efficient consumption of Beethoven’s Ninth, if transmittal costs are zero, requires that everyone who places a positive value on hearing the symphony be allowed to listen to it.
This would seem to imply that everyone should be allowed to consume all the products that are normally copyrighted, and everyone should be allowed to use the ideas that are normally protected by patent. But there is a fly in this ointment, which is why I used the term “trade-off” in this section’s heading.
The requirement that all potential consumers be allowed to consume the intellectual products puts some serious restrictions on the price(s) that can be charged for the product. If consumers have differing values for the good, but the producer (creator) can charge each potential consumer a price slightly below the maximum price that consumers are willing to pay, then all potential consumers would consume the product and efficiency would be achieved, a result known as perfect price discrimination. But if, more realistically, the producer is unable to charge different prices to different potential customers, then no matter what price the producer picks, some potential consumers will be priced out of the market—unless the producer picks a price of zero.
A zero price, alas, provides the producer with no revenue. If producers receive no revenue, then there is little reason to believe that production will occur. This, then, is the problem brought about if one attempts to achieve efficient consumption: there is nothing for potential consumers to consume since nothing will be produced at a price of zero. Reducing consumptive efficiency is the cost involved in allowing for increased creation of ideas. As government increases the production of creative ideas by giving producers of these ideas more and more control over the production of the embodiments of these ideas, the consumption of these embodiments becomes less and less efficient.
The application of this trade-off occurs when the competitive model is grafted onto these ideas. If anyone can make a copy (embodiment) of an idea or expression without the permission of the original creator, the price of embodiments would be expected to drop to zero (plus the transmission/reproduction costs), leaving nothing for the creator.
Therefore, if markets are to be used to provide producers with a pecuniary incentive to create intellectual products, creators must be given some degree of control over the use of their products, prohibiting others from copying their ideas or expressions. This is where copyrights and patents come in.
This prohibition on copying is generally referred to as a “monopoly,” although for many intellectual products the term is not economically correct. As Edmund Kitch (2000) correctly pointed out, providing property rights does not confer economic monopoly—which would imply that consumers have only a small number of alternative products that are not very good substitutes. In reality, the “monopoly” conferred by copyright is no greater than the monopoly that each worker has on his or her efforts, or that each firm has on products bearing its name. The monopoly created by patent law would generally be somewhat stronger than for copyright law because it is a realistic possibility that others would have independently created the same idea, but the patent eliminates the use of such independent creations. Nevertheless, competition is still possible between the patent holder and other ideas or technologies not limited by the particular patent.
Intellectual property protection, then, can be seen to create two countervailing results. First, it provides authors and inventors the wherewithal to receive remuneration for their activities, which has the beneficial impact of increasing the production of expressions and ideas. On the other hand, copyright and patent laws allow the owners of the intellectual properties to charge positive prices for their use, restricting the usage and consumption of these ideas below their ideal levels.
Several simplifications in the above story weaken its generality. First, it is not clear that competition will remove all revenues from creators. Although profits will be competed to zero in the long run, the long run does not happen instantaneously. Because the creator is usually the first to market, that temporal advantage would allow the creator to generate some revenue above the cost of reproduction.
Second, the extent to which copyright owners require remuneration to create their artistic works is not clear. The claim that creative production requires remuneration of the producers is fully consistent with the usual market principles. Adam Smith’s famous maxim that production does not come from the “benevolence” of butchers, bakers, or brewers, but instead derives from their self-interested behavior, certainly has a plethora of empirical evidence to support it.
Nevertheless, it has often been argued that artistic creation is often undertaken for reasons having little to do with pecuniary rewards. The idea that artists require payment seems antithetical to various romantic notions of “art.” (Inventors, on the other hand, are usually thought to be less interested in “creation for creation’s sake” and, as such, are usually assumed to require compensation.) Even if one does not subscribe to the romantic view of art, however, the fame from creating successful works can bring its own rewards and often the ability to generate additional revenues in other markets (such as concerts for musicians).
The Optimal Term
One of the most important aspects of intellectual property is its limited term. In principle, economic efficiency would require that the length of protection be the minimum necessary to provide the author/creator with the incentive to create the product. (An additional problem with patents is the proper “scope” of the patent.) In this way, the restriction on consumption would be minimized, and yet the creator would receive sufficient remuneration to produce the product. This would require a different term for each intellectual product. Composers (or recording artists) who could make money by performing at concerts, for example, would not need as long a copyright term as composers who do not concertize.
Although attractive in theory, such an approach does not appear practical and has not been used. The question, then, becomes what the efficient terms for patent and copyright might be—5 years, 50 years, 150 years? This has led to much debate, but no consensus.
The problem with determining the optimal term of intellectual property protection is that no one knows any of the key pieces of information needed to determine the optimal term. These include how much incentive is required to induce creators to create, the size of the harm from reduced consumption during the term of the intellectual property law, the size of the revenues generated during the term of protection that can be used to pay the creator, and future interest rates. No one has these facts, and the difficulty in learning them is such that we may never be in a position to determine the optimal term with any precision.
Alternatives to Intellectual Property
Because of the imperfections of market-based intellectual property systems, various alternatives have been proposed. In most of these alternative systems, the government funds creators. The advantage of such a system is that competition would drive the price of the intellectual products down to their transmission/reproduction costs, and no consumers would be denied the product because the price was above the transmission/reproduction costs. Of course, the funding of these products through tax revenues (see taxation) causes its own set of inefficiencies that might otherwise appear to be hidden from view, and there is no reason to believe that the inefficiencies from the tax code will be less than the inefficiencies from having too few users of intellectual products.
An even more serious problem with this system is caused by the government’s need to decide how much revenue to pay to creators. Even a well-intentioned government would have great difficulty determining the optimal size of the pot to be shared by various authors/creators. Even though the market-based system of intellectual property is imperfect in generating revenues, it is far more likely to track the direction and size of the “optimal” market than will even the best-intentioned efforts of the government. The reason is simple: without a market to provide guidance, it is virtually impossible to divine even a rough approximation of the prices and quantities that would be found in an ideal market.
A final problem with this solution is determining which authors and which inventions should be most richly rewarded. For material such as books and music, it might not be too difficult to get some measure of relative sales of reproductions, and from that to determine the relative payments to be made to authors since one can examine the market shares of the sales of reproductions. Determining the relative market value of inventions would be much more difficult if the government were to just grant rewards for inventive activity.
One interesting idea to overcome this difficulty in valuing patents was suggested by Michael Kremer (1998), who proposed that the government purchase patents and then put them in the public domain. In his model the government determines the value of the patent by holding an auction. The high bidder occasionally gets the patent (to keep bidders honest), but most of the time the government takes the patent, and the current patent holder can refuse the price if he thinks it is too low. This plan solves the underconsumption problem and allows follow-on innovations to occur more easily. Unfortunately, it introduces its own distortion through the use of taxes to pay for this scheme. Further, it is susceptible to gaming whereby the original patent owner pays others to overbid for his patents, and it is susceptible to the government overpaying when the patent owner has negative inside information about the value of the patent.
Finally, all of this assumes that a government-run intellectual property system is well intentioned and insulated from political considerations. Given our understanding of government regulation of markets (see public choice), it seems fair to say that politics will certainly enter these decisions and that the bureaucrats in charge of these agencies will be influenced by the various parties involved. Giving away other people’s money is not a job that mere mortals are likely to accomplish without a good deal of political influence, if not outright corruption.
The reliance on property rights comes to the fore in the current controversy surrounding file sharing. Several prominent legal academics, following the lead of Lawrence Lessig (2001), have suggested that traditional copyright protection of music (and movies) be discarded and replaced with an organization along the lines of ASCAP or BMI, which would collect monies from income taxes or taxes in markets such as ISPs or blank CDs and distribute the monies to creators. This model, most fully developed by William Fisher (2004), is still a variant of the government provision discussed above, with all of its limitations. It is possible, however, that file sharing cannot be controlled and that the market for sound recordings and movies might essentially vanish. In that case, some alternative is probably better than nothing, although government provision can, in principle, create losses greater than the vanishing of the market.
The specter of digital rights management (DRM), which would allow copyright owners to monitor and control use of their works through software built into a work, has also created a stir. A traditional defense against charges of copyright infringement is “fair use,” which is intended to allow actions such as copying part of another work in an academic analysis or criticism. Many legal academics have decried the possible loss of fair use that would occur if the software required payment for uses that were previously deemed “fair.” The concern over fair use appears to be overstated because one of the purposes of fair use, to allow usage when the costs of transacting exceed the value of the transaction, no longer holds under DRM. Further, if someone wishes to use a snippet of someone else’s work in an academic article, that would still be allowed, although the words might have to be hand copied instead of electronically cut and pasted.
Another recent controversy surrounds the Sonny Bono Copyright Term Extension Act of 1998, which retroactively increased copyright protection from fifty years to seventy years after the death of the author. The act was challenged as being unconstitutional. A group of seventeen famous economists (Akerlof et al. 2002) wrote a brief in that case criticizing the efficiency of the act. Part of their criticism had to do with the retroactive increase in copyright for old works. After all, increasing the term of copyright for already created works cannot increase the number of already created works, although it would increase the harm from having some consumers priced out of the market. The other part of their criticism had to do with the extension of the copyright term. In their view, the present value of the additional revenues fifty-plus years down the road was too small to have an impact on the production of new creations, and thus served no purpose. Stan Liebowitz and Stephen Margolis (2005) criticized this brief on several grounds. First, they argued that one reason to allow current copyright holders to control stewardship over already created works is to reduce externalities, for example, by preventing overuse of certain copyrighted characters that lowers their social value the way overuse of the same house style can lower the value of a neighborhood. Second, even a small increase in expected revenues can have a relatively large impact on the number of new creations (relative to the present value of future deadweight losses), as long as the elasticity of supply is not zero, a factor the seventeen economists ignored. Third, a high percentage of best-sellers, which are responsible for a majority of trade sales, remain in print for more than sixty years, thus indicating that the copyright extension would be expected to have an impact on incentives to create.