I have been working through reports from the mid 1940s on government ownership of patents. There are a number of arguments against government exploitation of patents, but these arguments clearly failed. What arguments won out? Here’s one, from the National Patent Planning Commission:
Since the Government can grant only revocable nonexclusive licenses under patents which it owns, it is for all practical purposes restricted to (a) publication, (b) the procurement of a patent which is in terms dedicated to the public and (c) the acquisition of a patent under which nonexclusive licenses may subsequently be granted. It often happens, however, particularly in new fields, that what is available for exploitation by everyone is undertaken by no one.
The sentence I have highlighted is a simple assertion. There are multiple claims. (1) If something is available to everyone, then no one will “exploit” it; (2) especially in “new fields”; and (3) this happens “often.” We have to get at, as well, the use of “exploitation” rather than “use” or “practice.”
There’s a fundamental problem with trying to evaluate this assertion, because every invention that’s never patented and never used, or patented and offered non-exclusively and never used might be offered as “evidence” of the truth of the statement. Yet, just because a given invention is freely available to all does not mean that this availability is the cause of no one using the invention. That’s just an assertion.
We can point to “new areas” where an invention available to all is indeed “exploited.” We might mention insulin, developed at the University of Toronto in 1921, patented, and made available on a non-exclusive basis. Exploited. Or consider John von Neuman developing the digital computer in 1951, and deliberately not patented–and exploited by many. Or consider Jonas Salk developing a polio vaccine in 1955, available to all, and exploited immediately.
The Harbridge House report in 1968 studied the years 1957 and 1961, and found that nearly all Department of Agriculture inventions developed within the Department and made available non-exclusively were “exploited.” Of course, given the nature of the original assertion by the National Patent Planning Commission, citing instances in which an invention has been made available to all and has been exploited cannot disprove the assertion. These could just be “exceptions that prove the rule.” But there may well not be any rule here. Steven Johnson, in Where Good Ideas Come From, presents scores of instances of what he calls “non-market, networked” innovation–inventions available to all, being developed through a network of individual efforts, to become useful to all. Johnson argues that the number of monopoly-based innovations (stuff that gets exploited) is tiny compared to the number of networked, non-market innovations.
For the assertion to be meaningful, someone must compile evidence for its general truth. I don’t find that evidence. I find lots of contrary evidence. It is as if the assertion is persuasive primarily because it comes without evidence. At least, there sure doesn’t seem to be support for the “often” or the “new fields.” We are left with the claim that what’s available to all won’t be “exploited” by anyone. Richard White, in Railroaded, points out that train engineers in the 19th century modified their engines based on the particulars of the rail routes they ran. They shared their modifications in a kind of open hardware commons. Since they ran their own routes, sharing was apparently not a block to exploitation. The open source movement in software developed in much the same fashion as the technology-sharing of the railroad engineers.
Patent enforcement has long been denied to surgery techniques–yet it appears we have no dearth of new surgery techniques being practiced. Some university medical schools in the 1950s argued that medical patents were simply unethical. Apparently they did not consider availability to all fatal to the use of new methods of saving lives. Share the code and develop applications that work better and work together. No problem with exploitation. No patent monopoly needed.
And then there’s the Cohen-Boyer gene splicing patents. Stanford’s Neils Reimers observed later that the patents were unnecessary to the use of the Cohen-Boyer techniques–everyone was going to use them, patents or not. For having the patents, Stanford harvested $260m or so, but the exploitation then happened despite the patents and the cost they added to the use of the technology. The same thing appears to be case with CRISPR technology today. Same with a bunch of lab medicine diagnostic tests–anyone with any competence can practice the methods, patent or no. There is no particular need for a “commercial” product version–and a patented commercial version often gets in the way of better local practice.
It may be that there are instances in which, but for a monopoly patent position, some thing is never to be used because it requires costly development into a commercial product first, and no one will take on the development without a patent monopoly. Please, someone, give examples–any example. Even one.
Let’s continue with the assertion made by the National Patent Planning Commission:
There undoubtedly are Government-owned patents which should be made available to the public in commercial form but which, because they call for a substantial capital investment, private manufacturers have been unwilling to commercialize under a nonexclusive license.
The assertion takes the form of a matter of “undoubtedness.” There’s nothing here of an assertion that there are such government-owned inventions, or that the Commission knows exactly which ones they are (and that there are a lot of them)–implied in that “often” of the first part of the assertion. Here too we have a change in the assertion, from “exploitation” of inventions to developing inventions “in commercial form.” Perhaps that’s the sense of the original assertion: that often, in new fields, what is available to all is not something that companies will spend money to develop into something to sell. Follow the sense of this claim. If everyone can practice an invention–can make it for themselves or have some local “gadgeteer” or shop fabricate what they need–then what’s the point of trying to create a “commercial” version to sell as a product? One first would have to eliminate use of any independent version of the invention in favor of a mass-produced product version to be sold commercially.
The role for a patent becomes apparent–suppress one form of exploitation in favor of another. “Wouldn’t it be nice if everyone was forced to stop doing something for themselves, or relying on a local workshop to make things for them, and instead had to buy from us in whatever form we decided was the most profitable to manufacture?” “Wouldn’t it be nice if the results of research in the area of this invention were forced to come to us for our review and approval, so that no-one could develop a better product based on openly available technology?” One gets the drift of how a patent might be used, if only the government were to allow it to private persons, not the inventors, to give effect to the full force of its monopoly power.
The Commission asserts as a general truth that companies often spend large sums to create new products. There’s nothing to indicate, however, whether companies spend large sums regardless of the starting patent position. It may well be that a company intent on creating a new product must spend an even greater sum to design around a patent that blocks research and development–or to barge ahead, infringe, and aim to invalidate the patent or force a cross-license deal, or figure that it’s better to have a share of the market now and be forced to pay a royalty than to have no product and no market position, all because of a nasty little piece of paper.
The Commission’s assertion may be particularly true for big corporations who can’t imagine doing anything unless it makes them huge profits. For their efforts, they assign highly paid planners and managers, spend money on engineers and marketing experts, and bring in attorneys and risk managers to limit the downside of the effort. At Xerox PARC, inventions were classified into “strategic” and “non-strategic.” If an invention was declared strategic, then Xerox formed a new division with senior managers to turn the invention into a commercial product. Inventors prayed that their inventions would not be declared strategic, since next to nothing that went the strategic route ever came to anything–but there would be a great expenditure. The amazing outputs from PARC that did become exploited–the graphical user interface, the mouse, ethernet, and postscript–were non-strategic inventions that were left outside corporate commercialization efforts and thus were able to find applications.
Thus, even where it is true that large companies spend huge amounts to develop new products, it is not therefore true that those same new products would cost so much to develop by other methods, such as in the hands of a small company, or a startup, or by a collaboration among companies.
A patent monopoly can be used to prevent competitors to develop their own products based on an invention. But a patent monopoly may also prevent local users from practicing an invention–users who don’t need a commercial product version because they can develop their own, at much, much less cost than a company might incur for the product version.
Furthermore, a patent monopoly, exploited to prevent competition from companies and local users alike, can antagonize people who otherwise might support a company’s efforts. Antagonized people can work to undermine a monopoly position–by developing a competing standard, by filing patents to block development of the monopoly invention, by refusing to collaborate or supply parts, by opposing the company’s political lobbying efforts, by hiring away talent. A monopoly isolates a company. It’s a lonely, expensive place to be.
Let’s make the Commission’s assertion sharper: What is available to all to use is less attractive to some investors to be “exploited” as a commercial product. That much would appear to be true. But that is not the argument that is being asserted: the argument is that often in new fields inventions don’t get used because there’s no patent with a monopoly owner. But that argument does not appear to be true. It just sounds good, in a way, as an assertion, and there is this strange double sense in which it may well be true. If one has to compete with open use of an invention, then one had better build a much better version–better features, better quality, readily available, standardized. But even then, the argument isn’t that one has to have a patent monopoly to do it–it’s just that one has to be better than local practice. It’s another thing to say, there’s no motivation to be better than local practice unless one can prevent everyone else, including local practice, from improving. That’s a strange argument. But that appears to be what is being asserted. More:
Accordingly, it seems evident that the Government has been handicapped in its effort to further the promotion and development of some of its inventions to the point where they are available to the public in the form of a commercial product.
The “accordingly” appears to be a transition to a non-sequitur, and a weak one at that–“it seems evident.” There’s no confidence in the assertion, just a statement of seeming. Again, as stated, the assertion may well be true: it indeed might seem that the Government is handicapped by its patent policy. That much may well have been true, that things seem to be a way, if assertions are to be believed. But if the assertion is wrong, then seeming isn’t what we should be basing policy on.
The argument pushes on. What seems evident? That the Government can’t “further the promotion and development of some of its inventions.” We might ask just what the phrase “further the promotion and development” means? Clearly, “further the promotion and development” does not mean “the Government cannot develop its own inventions.” It must mean something more like “the Government cannot get others to develop its own inventions.” But it’s more than just develop–“the Government cannot get others to develop its own inventions into commercial products.” Now, again, it may well be true that once the Government owns a patent, it is really difficult to get just anyone to develop the underlying invention into a commercial product–especially if everyone can use the invention without needing a commercial product version. But here the depth of focus has to be government ownership of a monopoly right, which if the government licenses non-exclusively, royalty-free, the invention is essentially in the public domain–perhaps the only difference is that the government might refuse to grant a license to foreign companies seeking to sell product here competing with domestic companies.
But then the argument is that a robust public domain works against the development of commercial products for profit. That’s a different thing altogether, a kind of enclosure argument. It’s not an argument about how trade secrets limit the “progress” of the useful arts–something that patents on inventions were developed to address; rather, it’s an argument that if everyone publishes their inventions and doesn’t hold trade secrets, why nothing at all will happen. Companies won’t do anything new without a monopoly. We have seen that in general this assertion is simply untrue. Monopolies might be sweet, but they are not essential.
So just what does the National Patent Planning Commission argue here? That the government (“often”) has a difficult time getting its inventions made into commercial products (“exploited”) unless it can hand a patent monopoly to a favored single company. In the early aircraft industry in the United States, the federal government had to intervene because so many companies had patented bits and pieces of aircraft technology that no one could build an airplane in the United States, and the progress of the industry had shifted to Europe. The government told the companies to cross-license or face government action. If the federal government holds patents on bits and pieces of new technology and chooses favorites to receive patent monopolies on those bits and pieces, the government helps to create gridlock in an industry. This is especially true where an exclusive license forbids sublicensing royalty-free and non-exclusive (that is, contributing the licensed bit or piece to a standard or cross-licensing it to prevent a patent fight with a competitor).
If the argument is that the government has a difficult time getting anyone to develop a commercial product based on a government-owned invention, that may indeed be accurate. But that accuracy does not mean that a commercial product is necessary. Perhaps the problem is government ownership. Perhaps the problem is the nature of the licensing. Perhaps the problem is that the government offers nothing along side the license–no services, no production of part of the invention, no readiness to purchase product produced. Why would anyone choose the government as a business partner? Any of these things might be more likely than that there is a kind of invention that cannot be used by the public but for development into a commercial product, and that the expense to develop the product–and the profits to obtained from sales so spare–that a company will not act without a monopoly running nearly two decades. And that’s assuming that the government licenses its patented inventions immediately. What happens if the government delays for five years, cannot find a monopoly partner? Is the monopoly argument still good if it’s only for twelve years? Eight? Five? At what point does the monopoly argument fail for lack enough monopoly time? And if eight years is enough, then why permit a monopoly license to go longer than eight years?
We might say, then, that what may have been true is that there might have been some government-owned inventions for which only a monopoly deal would lead to private investment to create a commercial product, and if no commercial product is created, the public could not possibly benefit from the invention. It’s just–name one such product. I can name a whole class–those pharmaceutical products that require expensive testing and formulation. That’s not all pharmaceutical products, but at least some. But what else? Hello?
So it is evident in a seeming sort of way that there must be some government-owned patents that would do better if the government held them for their monopoly value and conveyed that monopoly to some favored single company, which then would spend a great deal of money to make a product, and would in some years of monopoly pricing without competition justify that expenditure with sufficient profits to cover costs and to be more than the profits from other products that the company might have otherwise developed. The argument, then, is that the government’s monopoly positions might induce companies to develop the government-owned inventions instead of their own inventions or inventions available from others. There’s nothing in this argument that suggests that the government’s inventions are better than any others that a company might develop. Rather, the argument is that by offering a patent monopoly, the government can make its inventions more attractive as a profit-making venture than other inventions, including those in the public domain or readily shared or the basis of a standard.
But by the time we are at this argument, we have lost the idea that a huge amount of money must be devoted to “development” of the invention into a “commercial product.” Now the argument is simply that the profits for doing so make development more attractive than other opportunities. Somehow that fails as public policy unless the government’s inventions are pretty darned essential to public well-being. Again, what exactly is the National Patent Commission talking about here in general terms, with wishy-washy assertions about what seems, often, undoubtedly, to be the case? Actually, virtually nothing.
Part 2 is here.