“Do Inventors Value Secrecy in Patenting? Evidence from the American Inventor’s Patenting Act of 1999,” S. Graham & D. Hegde (2013)

The patent system has many ridiculous properties for us economists to grouse about (Boldrin and Levine have a well-known book on the topic, but I think you’ll find James Bessen’s tome the best). The problem of disclosure, whereby patentholders are meant to exchange a description of how their design works in exchange for the limited monopoly that is a patent, is particularly strange. First, it is not obvious at all that disclosure matters as a method for spreading information – as Petra Moser and coauthors have shown, it is very easy to be obtuse enough in some types of patents that a well-trained outsider will find it impossible to construct the original device, and further, in many cases simply seeing the patented device is sufficient to understand how it works. Second, since patents are not disclosed immediately upon application, there is quite a bit of scope for the dreaded “submarine patent”: I see someone infringing, but I don’t accuse them of anything until they do a bunch of work building up the industry, then at the height I pop up and sue them for all they are worth.

Now, a 1999 law, the AIPA, had a provision meant to restrict submarines somewhat. Eighteen months after applying, your patent application is made public; note that it can take many years for the actual patent to be granted. There are exceptions here, but basically, you can only request the application be kept secret longer if you never want foreign patent protection, and, further, you can also request an earlier disclosure of the app, which the authors do not examine specifically in the current draft. A number of thinkers (among them both Samuelson and Friedman!) were worried that this requirement would be harmful to small inventors, who often lack the legal ability to sue early infringers, particularly when the infringers are foreign firms.

We have data to settle the issue now (Stuart Graham, an academic, is the Chief Economist at the US patent and trademark office; if you want evidence of how bad the economic reasoning behind our IP policy is, note that Graham is the first ever Chief Economist appointed at the PTO!). Graham and Hegde collect data, harmonized in an EU database, of all US patent applications since the policy went into place, and code them by whether they also applied for a foreign patent, the technology class, and whether the applicant was a large firm, a small firm or solo inventor, or a foreign firm. Overall, less than 8% of applicants requested secrecy; of those who could have requested secrecy since they never file overseas, about 15% request secrecy. Small inventors preferences are fairly similar to large firms. Of the patents where secrecy was requested by small inventors, the patents kept secret appear to be, if anything, less important inventions: they receive fewer onward citations, have fewer claims, and are granted in a shorter amount of time (earlier papers suggested that breakthrough inventions tend to involve a lot of finetuning in their claims, hence longer waits between application and grant). In none of the technology classes are more than a quarter of applications kept secret. Unmentioned in the paper is a more recent fact: the percentage of applicants requesting secrecy continues to fall every year.

Given all this evidence against the importance of secrecy, it is perhaps no surprise that there is a currently a bill in Congress that would remove these disclosure requirements. What can you do?

2013 working paper (No IDEAS version). If you want some other cool recent empirical work fighting bogus ideas about innovation (bogus, yes, but which nonetheless carry great weight in policy discussions!), check out the great work by Paul Heald at the U of Illinois law school concerning the question of “necessary property”. An argument by the content industries about why we should retroactively extend copyright (where, for sure, a 2013 law cannot affect incentives to create in 1920) is that IP without an owner will either be overexploited (Mickey in porno films) or not updated (no quality audiobooks of classics, say). Heald shows that music which falls out of copyright is no more or less likely to appear in modern films, and that bestselling books in the public domain (those written 1913-1922) are much more likely to have high-quality audiobook versions for sale than bestselling books still under copyright (those written in the following decade). Heald’s results put the lie to the argument that “content needs an owner to be exploited optimally”, but you don’t even need his research to know this: even if content needed an owner for efficient exploitation, what reason do we have to think that the previous copyright holder is the most efficient one? Why not, say, rotate the rights to the early Mickey Mouse films randomly among preservationists and film firms? Indeed, why not auction off the retroactive extension? (But of course, you know why we don’t do these things: because the executives and congressmen who supported the CTEA care and understand not a whit about the welfare analysis, but quite a bit about lobbying from their “I will never take a lobbying job” hack of an ex-colleague.)

“Patent Alchemy: The Market for Technology in U.S. History,” N. Lamoreaux, K. Sokoloff & D. Sutthiphisall (2012)

It may appear that the world of innovation looks very different today than it used to. Large in-house R&D outfits – the Bell Labs of the past – are being replaced by small firms who sell the results of their research on to producers. Venture capital funding of research appears more and more important, both for providing capital to inventors and to linking the inventors up with potential buyers. Patent trolls hound the innocent, suing them for patent violations they weren’t even aware of. The speed with which patents are evaluated has slowed to a crawl, and the number of patents being granted continues to grow. Many patents are merely defensive, acquired solely to keep someone else from acquiring them.

Lamoreaux et al, building on earlier work by Lamoreaux and Sokoloff as well as Tom Nicholas’ interesting recent research, point out that none of the above is strange. The rise of in-house R&D is a phenomenon that doesn’t show up in great number in America until well into the twentieth century, only becoming dominant after the Second World War. Around the turn of the century, most innovation was done by small, independent inventors, or by small research firms like Edison’s outfit. A series of intermediaries, principally but not always patent lawyers, served both to file the proper paperwork and to link inventors with potential buyers; the authors provide a bunch of juicy historical stories, derived from lawyer diaries during this period, on exactly how such transactions took place. Railroads were frequently being hounded by patent trolls who tried to catch them unaware, and traveling patentbuyers crossed the Midwest and South suing farmers for using unlicensed barbed wire or milk buckets. Patents took an average of three years to be processed by the early 1900s, and the patenting rate was near an all time high. Firms regularly bought patents just so their competitors wouldn’t have them.

This is all to say that, to the extent we are worried about certain aspects of the patent system today, looking to history may be a useful place to begin. “Submarine patents”, acquired by trolls and kept unused until a particularly juicy potential violator has started to earn large profits, don’t appear to have been too prominent at the turn of the century – given how lucrative this business appears, perhaps an investigation of why they only appear in the present would be worthwhile. The role of a patent as a saleable piece of knowledge, allowing non-producers to do useful research and then sell that research to a firm who finds it useful, surely has some role, as Arrow pointed out in his famous 1962 essay. When patents instead simply add transaction costs or result in thickets, discouraging activity by true innovators, something has gone awry. And when something goes wrong in the world, it is rarely the case that history can offer us no useful guidance.

2012 working paper (No IDEAS version). Prof. Sokoloff passed away from cancer at a young age in 2007, so this may become his final published paper – it incorporates a great number of ideas he worked on throughout his career, so that would be a fitting tribute.

“Recruiting for Ideas: How Firms Exploit the Prior Inventions of New Hires,” J. Singh & A. Agrawal (2011)

Firms poach engineers and researchers from each other all the time. One important reason to do so is to gain access to the individual’s knowledge. A strain of theory going back to Becker, however, suggests that if, after the poaching, the knowledge remains embodied solely in the new employer, it will be difficult for the firm to profit: surely the new employee will have an enormous amount of bargaining power over wages if she actually possesses unique and valuable information. (As part of my own current research project, I learned recently that Charles Martin Hall, co-inventor of the Hall-Heroult process for aluminum smelting, was able to gather a fortune of around $300 million after he brought his idea to the company that would become Alcoa.)

In a resource-based view of the firm, then, you may hope to not only access a new employer’s knowledge, but also spread it to other employees at your firm. By doing this, you limit the wage bargaining power of the new hire, and hence can scrape off some rents. Singh and Agrawal break open the patent database to investigate this. First, use name and industry data to try to match patentees who have an individual patent with one firm at time t, and then another patent at a separate firm some time later; such an employee has “moved”. We can’t simply check whether the receiving firm cites this new employee’s old patents more often, as there is an obvious endogeneity problem. First, firms may recruit good scientists more aggressively. Second, they may recruit more aggressively in technology fields where they are already planning to do work in the future. This suggests that matching plus diff-in-diff may work. Match every patent to another patent held by an inventor who never switches firms, attempting to find a second patent with very similar citation behavior, inventor age, inventor experience, technology class, etc. Using our matched sample, check how much the propensity to cite the mover’s patent changes compares to the propensity to the cite the stayer’s patent. That is, let Joe move to General Electric. Joe had a patent while working at Intel. GE researchers were citing that Intel patent once per year before Joe moved. They were citing a “matched” patent 1 times per year. After the move, they cite the Intel patent 2 times per year, and the “matched” patent 1.1 times per year. The diff-in-diff then suggests that moving increases the propensity to cite the Intel patent at GE by (2-1)-(1.1-1)=.9 citations per year, where the first difference helps account for the first type of endogeneity we discussed above, and the second difference for the second type of endogeneity.

What do we find? It is true that, after a move, the average patent held by a mover is cited more often at the receiving firm, especially in the first couple years after a move. Unfortunately, about half of new patents which cite the new employee’s old patent after she moves are made by the new employee herself, and another fifteen percent or so are made by previous patent collaborators of the poached employee. What’s worse, if you examine these citations by year, even five years after the move, citations to the pre-move patent are still highly likely to come from the poached employee. That is, to the extent that the poached employee had some special knowledge, the firm appears to have simply bought that knowledge embodied in the new employee, rather than gained access to useful techniques that quickly spread through the firm.

Three quick comments. First, applied econometrician friends: is there any reason these days to do diff-in-diff linearly rather than using the nonparametric “changes-in-changes” of Athey and Imbens 2006, which allows recovery of the entire distribution of effects of treatment on the treated? Second, we learn from this paper that the mean poached research employee doesn’t see her knowledge spread through the new firm, which immediately suggests the question of whether there are certain circumstances in which such knowledge spreads. Third, this same exercise could be done using all patents held by the moving employee’s old firm – I may be buying access to general techniques owned by the employee’s old firm rather than the specific knowledge represented in that employee’s own pre-move patents. I wonder if there’s any difference.

Final Management Science version (IDEAS version). Big thumbs up to Jasjit Singh for putting final published versions of his papers up on his site.

“Returns to Scale in Research & Development: What Does the Schumpeterian Hypothesis Imply?,” F. Fisher & P. Temin (1973)

Schumpeter famously argued for the economic importance of market power. Even though large firms cause static inefficiency, they had dynamic benefits in that large firms demand more invention since they can extract more revenue from each new product. Further, they supply more invention, Schumpeter hypothesized, since the rate of invention has increasing returns to scale in the number of inventors, and in the number of other employees at the firm. (Axioms A and B). The second part of that statement may be for many reasons; for instance, if the output of a research project could be many potential products, a larger firm has the ability to capitalize on many of those new projects, whereas a small firm might have more limited complementary capabilities. Often, this hypothesis has been tested by checking whether larger firms are more research intensive, meaning that larger firms have a higher percentage of their workforce doing research (Hypothesis 1). Alternatively, a direct reading of Schumpeter is that a 1% increase in the non-research staff of a firm leads to a more than 1% increase in total R&D output of a firm, where output is just the number of research workers times each worker’s average output as a function of firm size (Hypothesis 2).

And here is where theory comes into play. Are axioms A and B necessary or sufficient for either hypothesis 1 or 2? If they don’t imply hypothesis 1, then the idea of testing the Schumpeterian axioms about increasing returns to scale by examining researcher employment is wrong-headed. If they don’t imply hypothesis 2, then Schumpeter’s qualitative argument is incomplete in the first place. Fisher and Temin (that’s Franklin Fisher and Peter Temin, two guys who, it goes without saying, have had quite some careers since they wrote this paper in the early 70s!) show that, in fact, for both hypotheses the axioms are neither necessary nor sufficient.

An even more basic problem wasn’t noticed by Fisher and Temin, but instead was pointed out by Carlos Rodriguez in a 1979 comment. If Axiom 1 holds, and the average product per researcher is increasing in the number of researchers, then marginal product always exceeds average product. If market equilibrium means I pay all research workers their marginal product, then I will be making a loss if I operate at the “optimal” quantity. Hence I will hire no research workers at all. So step one to interpreting Schumpeter, then, is to restate his two axioms. A weaker condition might be that if the number of research and the number of nonresearch workers increase at the same rate, then average product per research worker is increasing. This is implied by Axioms A and B, but doesn’t rely on always-increasing average product per research worker (Axiom C). This is good for checking our two hypotheses, since anything that would have been implied by Axioms A and B is still implied by our more theoretically-grounded axiom C.

So what does our axiom imply about the link between research staff size and firm size? Unsurprisingly, nothing at all! Surely the optimal quantity of research workers depends on the marginal product of more research workers as firm size grows, and not on the average product of those workers. Let’s prove it. Let F(R,S) is the average product per research worker as a function of R, the number of researchers, and S, the number of other employees at the firm. I hire research workers as long as their marginal product exceeds the researcher wage rate. The marginal product of total research output is the derivative of R*F(R,S) with respect to R, or F+R*dF/dR. As S increases, this marginal product goes up if and only if dF/dS+R*dF^2/dRdS>0. That is, I hire more research workers in equilibrium if my non-research staff is bigger according to a function that depends on the second derivative of the average output per researcher. But my axioms had only to do with the first derivative! Further, if dF/dS+R*dF^2/dRdS>0, then larger firms have a larger absolute number of scientists than smaller firms, but this implication is completely independent of the Schumpeterian axioms. What’s worse, even that stronger assumption involving the second derivative does not imply anything about the share of research workers on the staff.

The moral is the same one you were probably taught you first day of economics class: using reasoning about averages to talk about equilibrium behavior, so dependent on marginals, can lead you astray very quickly!

1971 working paper; the final version was published in JPE 1973 (IDEAS). Related to the comment by Rodriguez, Fisher and Temin point out here that the problem with increasing returns to scale does not ruin their general intuition, for the reasons I stated above. What about the empirics of Schumpeter’s prediction? Broadly, there is not much support for a link between firm size and research intensity, though the literature on this is quite contentious. Perhaps I will cover it in another post.

“Diffusing New Technology Without Dissipating Rents: Some Historical Case Studies of Knowledge Sharing,” J. Bessen & A. Nuvolari (2012)

The most fundamental fact in the economic history of the world is that, from the dawn on mankind until the middle of the 19th century in a small corner of Europe, the material living standards of the average human varied within a very small range: perhaps the wealthiest places, ever, were five times richer than regions on the edge of subsistence. The end of this Malthusian world is generally credited to changes following the Industrial Revolution. The Industrial Revolution is sometimes credited to changes in the nature of invention in England and Holland in the 1700s. If you believe those claims, then understanding what spurred invention from that point to the present is of singular importance.

A traditional story, going back to North and others, is that property rights were very important here. England had patents. England had well-enforced contracts for labor and capital. But, at least as far as patents are concerned, recent evidence suggests they couldn’t have been too critical. Moser showed that only 10% or so of important inventions in the mid-1800s were ever patented in the UK. Bob Allen, who we’ve met before on this site, has inspired a large literature on collective invention, or periods of open-source style sharing of information among industry leaders during critical phases of tinkering with new techniques.

Why would you share, though? Doesn’t this simply dissipate your rents? If you publicize knowledge of a productive process for which you are earning some rent, imitators can just come in and replicate that technology, competing away your profit. And yet, and yet, this doesn’t appear to happen in many historical circumstances. Bessen (he of Bessen and Maskin 2009, one of my favorite recent theoretical papers on innovation) and Nuvolari examine three nineteenth century industries, American steel, Cornish steam engines and New England power weavers. They show that periods of open sharing on invention, free transfer of technology to rivals, industry newsletters detailing new techniques, etc. can predominate for periods a decade and longer. In all three cases, patents are unimportant in this initial stage, though (at least outside of Cornwall) quite frequently used later in the development of the industry. Further, many of the important cost reducing microinventions in these industries came precisely during the period of collective invention.

The paper has no model, but very simply, here is what is going on. Consider a fast growing industry where some factors important for entry are in fixed supply; for example, the engineer Alexander Holley personally helped design eight of the first nine American mills using Bessemer’s technology. Assume all inventions are cost reducing. Holding sales price and demand constant, cost reductions increase industry profit. Sharing your invention ensures that you will not be frozen out of sharing by others. Trying to rely only on your own inventions to gain a cost advantage is not as useful as in standard Bertrand, since the fixed factors for entry in a new industry mean you can’t expand fast enough to meet market demand even if you had the cost advantage. There is little worry about free riding since the inventions are natural by-products of day-to-day problem solving rather than the result of concentrated effort: early product improvement is often an engineering problem, not a scientific one. Why would I assume sales price is roughly constant? Imagine an industry where the new technology is replacing something already being produced by a competitive industry (link steel rail ties replaced iron ties). The early Bessemer-produced ties in America were exactly this story, initially being a tiny fraction of the rail tie market, so the market price for ties was being determined by the older vintage of technology.

Open source invention is nothing unusual, nor is it something new. It has long coexisted with the type of invention for which patents may (only may!) be more suitable vectors for development. Policies that gunk up these periods of collective invention can be really damaging. I will discuss some new research in coming weeks about a common policy that appears to provide exactly this sort of gunk: the strict enforcement of non-compete agreements in certain states.

2012 Working Paper (IDEAS version)

“Profiting from Technological Innovation,” D. Teece (1986)

Teece’s 1986 article in Research Policy is surprisingly little known among economists given that it has been cited something like 10,000 times. I want to give an interpretation of the article similar to that of Sid Winter in his article written on the 20th anniversary of the original.

Schumpeter famously argued that “perfect” competition is, in fact, not so, as the lack of rents given no incentive for firms to spend on R&D, and since growth is so much important for welfare than static inefficiency, we ought be more forgiving of market power. Ken Arrow, in a well-known article from the 1962 NBER Invention volume, maintains that Schumpeter’s logic is incomplete, and that with patent licensing, monopolies can make things worse. Consider a good with marginal cost 2 and demand such that Q=6-p. In the competitive market, price is 2, quantity is 4, and industry profits are zero. With a monopoly, price is 4, quantity is 2, and industry profits are 4. An innovator invents a technique that lowers marginal cost for the good to 1. In the competitive market, he can license this good to all producers, accruing licensing profits of 1×4=4. In the monopoly market, the monopoly with marginal cost of 1 would optimally sell 2.5 units at 3.5 each, earning 2.5×2.5=6.25. Therefore, the invention increases monopoly profit by 6.25-4=2.25, and the inventor can earn no more than 2.25 by licensing to the monopolist. It seems, then, that whether monopoly or perfect competition leads to more invention depends, at least in part, on the ability of inventors to license without being appropriated.

Teece takes that logic a step further. As most inventions can be appropriated, either by direct imitation, or by inventing around the relevant patent, inventions will only pay off for the inventor if she owns the best complementary assets. Consider the case of EMI’s CAT scanner and Searle’s Nutrasweet. The CAT scanner was both invented and commercialized by EMI, leading to a Nobel for one of EMI’s engineers. Nonetheless, EMI would be out of business within a few years, while competitors made bundles of money from similar scanners. Nutrasweet, on the other hand, was enormously profitable for Searle. Why the difference?

The difference is access, through contracting or ownership, to complementary assets. EMI’s imitators had much better medical technology manufacturing and distribution technology than EMI itself. Searle, on the other hand, took deliberate steps to protect itself once its patent ran out, by establishing a strong brand during the patent period, by limiting outside manufacturing (since those contract manufacturers are potential future competitors), and by doing R&D on a product that is difficult to imitate without violating patent; for one, other alternative sugars would need to go through their own FDA approval, which takes years. Teece’s article also provides a second reason why large firms spend more on R&D. It’s not just that they will have market power in the product’s market, but also that they are more likely to own complementary assets.

Final Research Policy version (IDEAS version). A site note: this is the 300th article we’ve discussed on this site. I would love to see more focused research blogs. There are a few (e.g., the NEP-HIS blog with a weekly post on economic history), but that’s it. I’d be glad to share my experience from this blog with anyone interested. For one, the potential audience for discussions of new research is huge – at least half of the readers of this site are non-academics, but instead represent the curious, people working in the tech sector, undergraduate students, etc.

“Innovation: The History of a Category,” B. Godin (2008)

What is innovation? What, indeed, is invention? I am confident that the average economist could not answer these questions. Is invention merely a novel process or idea? A novel process or idea for a given person? A new way of combining real resources like capital and labor? A new process which allows more of something to be created using a given amount of real resources? Does the new process need to be used, or embodied in technology, or is the idea enough?

None of these definitions seem satisfactory. A poem is a “new idea”, but we wouldn’t call it an invention. Novelty for a given person without technological embodiment, as a definition, doesn’t seem to distinguish between diffusion and simple learning. The idea of technology as a Solow residual means that merely using different mixtures of capital and labor to make the same product doesn’t qualify, and further the Solow residual includes things like Bowles-style adaptations to a more cooperative or trusting culture, which we generally don’t think of as innovation. Was Schumpeter correct that invention is a mere act of creativity “without importance to economic analysis”, or does the sequential nature of ideas mean that even non-embodied ideas are economically important?

In an interesting “genealogy of an idea”, Benoit Godin examines the history of how the terms invention and innovation were used in the Western World. The term invention goes back to Cicero, who listed the development of new argumentative concepts as one of the five tools of rhetoric. From the 15th to 19th centuries, invention was used occasionally to mean novel thoughts, but also novel recombinations (as in painting) or simple imitation (such as the patents given to importers in 18th century England).

It is really quite late in the game – well into the twentieth century – that something like “innovation is the invention, embodiment and diffusion of a commercial product” begins to be accepted as a definition. Part of this involves the shift from the individual inventor, the lone genius, to commercial firm R&D, as well as a recognition that simultaneous discovery and ex-post construction of credit meant that the lone genius inventor probably never existed. The terms discovery and invention began to separate. Science policy began to focus much more on the quantifiable, inventions as discoveries embodied in products or countable as patents. The word innovation became identified with an economic sense rather than an artistic sense which it previously possessed.’

Even the economic definition that would eventually be adopted is not the only one that could have developed. Schumpeter is often recognized as the father of economic studies of technological change, but his definition of innovation includes many concepts no longer covered by that term. For Schumpeter, innovation was tightly linked to creative destruction, or the dynamic ability of economic change to remake the commercial sphere. The opening of new commercial markets, for example, was an important part of innovation, whereas pure science was not.

http://www.csiic.ca/PDF/IntellectualNo1.pdf (2008 Working Paper – this is still unpublished, as far as I can tell).

“Did AMD Spur Intel to Innovate More?,” R. Goettler & B. Gordon (2011)

The relation between competition and innovation is theoretically ambiguous. On the one hand, as Schumpeter pointed out, having market power allows you to recover rents from new product sales, so you might expect monopolies to innovate more. On the other hand, innovation is costly, so without competitive pressure, you may simply rest on your laurels and keep selling your old product.

Goettler and Gordon, in a recent JPE, use the Intel/AMD microprocessor competition to investigate this issue. Innovation is easy to measure here – we simply look at the processor speed at the frontier for each firm, and avoid any messy issues about the difference between patented inventions and “actual” inventions. We can also track for over a decade the price differences in each firm’s top chips, the speed differences, and the response. The market is also for all practical purposes a duopoly with very little attempted entry. Computers possess another interesting property, in that they are durable goods. Past products compete with future sales. You may wish to keep prices high when you have market power this period in order not to cannibalize future sales if you expect a good innovation to appear next period for which you can charge even higher prices. Many sectors of the economy involve durable goods, of course.

The authors use a simple model to estimate consumer preferences in a structural model with spillovers (it is harder to push the frontier than to catch up). They find that, if Intel had a monopoly, innovation would have been 4% faster, but consumer surplus would have been 4% lower due to the higher prices charged by Intel, which is the standard Schumpeterian tradeoff. They find consumer surplus is maximized in a world where Intel has some anticompetitive power, though not monopoly power. The reason is that monopoly firms in durable goods markets still need to innovate because of competition with their old products, whereas duopolists can only earn rents to cover R&D costs if the two firms are selling different technologies. There are a number of interesting comparative statics as well. If spillovers are nonexistent, then the two firms race until one has a sufficiently large technological lead, at which point the other firm gives up, and no more innovation takes place, while if spillovers are large, the returns to each firm from doing R&D are low. In both cases, monopolists in a durable goods market innovate more. If spillovers are of an intermediate level, then duopolists will innovate more. As the authors note, “such variation might be one reason cross-industry studies have difficulty identifying robust relationships.”

The estimation involves some technical difficulties which may interest the Pakes-style IO readers. I am not an IO guy myself, so perhaps a reader can comment as to the more general style of this sort of paper. While I find the theory interesting, and am impressed by the difficulty of the empirical estimation, what exactly is the value of this sort of estimation? We know from theory the important qualitative tradeoffs. The style of estimation here can really only be done ex-post – the methods here could not be used, for example, to identify contemporaneously whether a anticompetitive behavior in a particularly durable goods industry is harmful for social welfare. I don’t mean to single this paper out, as this comment applies to a huge number of IO articles.

www.columbia.edu/~brg2114/files/dynduo.pdf (Final working paper)

“Directed Technical Change,” D. Acemoglu (2002)

If I increase the supply of something, its price should go down. And if I decrease the supply, its price should rise. Some markets do not seem to follow this pattern, however, with skilled labor in the US since 1970 being a famous example. As the percentage of college-educated workers has risen the U.S., the premium paid to the college educated has also risen. How can this be? One hypothesis is skill-biased technical change: the innovation that has occurred over the past few decades, computers included, has been complementary with the skills of educated workers. When might we expect innovation to complement certain factors?

An old and incorrect answer, previously discussed on this site, is that innovation will replace “expensive” factors of production. If labor is dear, for instance, firms will try to invent machines to replace labor. This intuition is wrong: in competitive markets, all factors are paid their marginal products, so saying labor is “dear” is just like saying labor is productive. And you might imagine we’d want to develop innovations that are complementary to our most productive factors!

Daron Acemoglu has a nice paper from a few years back – already very highly cited – dealing with these issues. Take a good produced using two factors with a CES production function; that is, the way in which factors are substituted for one another does not depend on how much of each factor we are already using. Let each factor have its marginal productivity improve by technology multipliers A1 and A2, and let innovations (which increase A1 or A2) be developed in any structure where the amount of new innovation responds in the natural way to the social value created by improving the technology multiplier. Acemoglu uses a monopoly innovator, but broader assumptions here about how social value is captured will not change the basic point.

The social value of innovations in each factor are increasing in the price paid to the factor and the total quantity of that factor used. If one factor is, say, skilled labor, then my incentive to create innovations improving the productivity of skilled labor depends both on how much skilled labor will be used, and on how productive the marginal skilled labor already is (since my invention is a multiplier on the existing marginal product). Imagine now that I increase the relative supply of skilled labor, exogenously. Will I see more or less skilled labor-augmenting invention? On the one hand, there is more skilled labor, so I can sell my innovation to a bigger market, but on the other hand this extra labor has a lower marginal product, so there is less productivity to enhance. Which effect dominates? With CES production, there is a simple rule. If the two factors are gross substitutes, an increase in the relative supply of a factor will increase the incentive to develop innovations augmenting that factor, and vice versa for gross complements. That is, with gross substitutes, an increase in the supply of one factor will not affect the relative factor prices (read: relative marginal products) very much, so the effect of an increased amount of that factor which I can augment dominates the effect of lower marginal product on that factor.

In the short run, before innovations can be created, the now more abundant factor sees its rent (wage) decline. This is the usual substitution effect. But what about in the long run, after technology is created? Here we need to model explicitly the monopolists who create inventions. It turns out that if the elasticity of substitution between factors is sufficiently high, an exogenous increase in the relative supply of one factor will increase the rent received by that factor. That is, the long run factor demand curves will slope up! This is because when the factors are gross substitutes (the elasticity of substitution is at least 1), innovation will be directed toward the now more abundant factor. The higher the elasticity, the more innovation. At some point, there is so much productivity-enhancing innovation directed toward the more abundant factor that even though the marginal units of this factor were relatively unproductive without the innovation, and hence received a lower wage, the response by innovators will be high enough that the now-more-abundant factor is paid even more than it was before the exogenous supply increase. A quick aside: theoretically, the increased elasticity (though not the sign change) of long-run response vis-a-vis short-run response is well known. It is called the Le Chatelier Principle and comes to economics via Paul Samuelson. Milgrom and Roberts have a lovely paper on why Le Chatelier works. The three theorems in this paper are proof positive of the usefulness of monotone comparative statics. Topkis is used to prove a result in two lines that must have taken pages to prove, and in less generality, with earlier techniques.

Consider again the concrete example of skilled labor since 1970. Goods are produced with skilled and unskilled labor. The supply of skilled labor increases, due to the GI Bill and other exogenous factors. This causes the skill premium to fall initially. If the elasticity of substitution is above 2, the long run wage premium to skilled labor will increased due to the effect of incentives to develop technologies augmenting the now larger base of skilled labor. This is one explanation for why you may have seen skill-biased technological change after the 1960s, and why there may have been enough of it to raise the skill premium. (Note that the elasticity of substitution itself is fixed in this model, but you might imagine that certain types of innovations may affect this factor.)

Those interested in Acemoglu’s work may enjoy an empirical paper by a PhD student on the job market this year, Walker Hanlon, applying Acemoglu’s result to the context of the Cotton Crisis, the shift in Britain from using US to using Indian cotton during the US Civil War. He has some nice data showing that even though Indian cotton became relatively abundant, there was a great amount of invention dealing with gins and other techniques for handling idiosyncratic issues in the Indian supply, and that the elasticity of substitution between US and Indian cotton was high enough that, indeed, the relative price of Indian cotton to US cotton rose by the end of the Civil War despite the relative abundance of the Indian cotton.

Final REStud version, Oct 2002 (IDEAS)

“Patent Reform: Aligning Reward and Contribution,” C. Shapiro (2007)

Carl Shapiro, in addition to being a bigshot in the academic study of invention, is also a member of Obama’s Council of Economic Advisers. I’m not sure how much of a role he had in advising on the Leahy-Smith patent reform act that was passed last year, but many of the reforms seem to come directly from this NBER Working Paper, so I imagine his role was a big one.

Most academic economists working on IP-related issues think, for a variety of reasons, that IP is currently far stronger than the optimal level. Indeed, many would prefer a world with no patents and copyrights at all to the current system. But let’s take the simplest possible reform: if the social benefit granted by a patent exceeds the social value created by the invention, we ought limit the strength of the patent. You might wonder, how is it even possible for the patentholder to gain more than the social value of his invention? A standard monopolist with a patent still creates consumer surplus and some deadweight loss – that is, social value not captured by the inventor – unless the monopolist is perfectly price discriminating. Shapiro, drawing on a number of earlier papers, gives three nice examples where return to the patentholder exceeds social value. Unless otherwise noted, we assume there is zero deadweight loss created by the patent; if there is deadweight loss, the reason for weakening the patent is even stronger.

First, we know from Loury (1979) and Tandon (1983) that if a patent gives the first firm to invent the full social value of his invention, there will be too much effort expended trying to win that prize; when each firm is deciding whether to expend more effort on R&D, they do not take into account that their increased effort lowers the probability of winning for the other firm. Tandon shows that this “patent race” effect is particularly strong for inventions that are relatively cheap to produce, such as those that are close to obvious. One way to fix this problem somewhat is to allow a second firm who independently invents at roughly the same time as the first firm to invent to sell the product without needing a license. That is, if a product is easy to invent, and two firms expend a lot of effort on it in an attempt to win the patent race, the second firm’s effort is not a total social waste since it may lead to a second independent invention, turning the eventual monopoly (with high deadweight loss) into a duopoly (with lower deadweight loss). Many economists and legal scholars have proposed allowing an independent inventor exception, but Congress has thus far shown no interest in taking up this idea. This is perhaps no surprise: Congress refused to pass the Public Domain Enhancement Act a few years back, an IP-related law that is as big a free lunch as you will ever see.

Second, probabilistic patents are often not challenged. Imagine a patent that, if challenged in court, has a 30% chance of being upheld as valid; many such weak patents exist. Assume that is totally free to challenge the patent, meaning there are no legal or transaction costs. Shapiro shows the following example, drawing on a paper of his with Joseph Farrell. Let a patent with probability .3 of being upheld when challenged be licensed to an oligopolistic downstream industry. The patent adds $10 of value to the products of all downstream inventors, so if the license royalty is greater than $3, the patentholder is earning more than the expected value of his patent. Imagine a royalty of $6. If I challenge the patent in court and my rival does not, then when I win the challenge, I and my rival in the downstream product market are both able to use the invention without paying any license fee, hence our costs are the same, and hence winning the challenge does not earn me any more profits due to competition with my rival. If I lose the challenge, then my rival pays a royalty of only $6, whereas I will have to pay $10 for each unit where I infringe, and hence I will be at a disadvantage in the downstream market. Therefore, neither firm will challenge the patent in equilibrium, and the inventor will earn more than his true social contribution.

Third, hold-up, particularly in the form of the “patent ambush,” can lead to excess returns. Imagine I can sell my product with noninfringing design A at a price of 100 dollars, or with infringing design B, for which I will need to license a previous invention, at a price of 120 dollars. The patent thus increases the value of my product by $20. If I Nash bargain with the inventor, we will split the gains from using his invention in my product, and therefore I will pay $10 to use the invention, and earn $110 per unit by producing design B. This intuition is very different if I first make investments, then learn about the patent. Imagine A and B both require 40 dollars of fixed cost per unit, each, to design. If I don’t know about the patent, I will design product B, and plan to earn 80 dollars per unit. The patentholder will then come to me and tell me I need a license or he will sue for infringement. Once the fixed cost of B is sunk, the surplus from obtaining a license is 20+40=60 dollars, since not obtaining a license means I will need to produce A, which costs another 40 dollars and sells for 20 dollars than design B. So a Nash bargaining outcome is that I pay 30 dollars for the license and produce B. That is, the patentholder can use holdup to extract extra rents after I have made specific investments.

One way to fix the last two problems is to allow informal post-grant challenges to patents, perhaps by third parties. This makes weak patents in important industries less likely to cause hold-up after specific investment, and also limits the ability of third parties to take advantage of the reluctance of licensees to challenge once license terms have been established. The new patent reform does vastly increase the scope for post-grant review.

What’s too bad about the 2011 patent reform is that the types of examples provided by Shapiro above are only the most clean-cut, overwhelmingly obvious ways to improve the efficiency of the patent system. They don’t even pretend to approach what would be necessary for an optimal IP regime. Aside from a handful of congressmen, (Zoe Lofgren and Ron Wyden on the democratic side, or Jason Chaffetz on the Republican side, among them) Congress is filled with IP maximalists. For the sake of social welfare, it’s too bad.

May 2007 NBER Working Paper (IDEAS)