The Competitiveness of Nations in a Global Knowledge-Based Economy

William H. Davis *

Our National Patent Policy

American Economic Review, 38 (2)

May 1948, 235-244.

* President’s Committee on Patent Policy

Our highly developed private enterprise economy demands, or so the economists tell us, a constant flow of technological innovation.  To stimulate that flow we have chosen to rely principally upon patents for inventions - the grant of temporary private monopolies for substantial technological innovations.  Our patents for inventions are federal grants to the first inventor for seventeen years of the exclusive right to make, use, and sell the patented invention.  They are based on the constitutional provision that to promote the progress of science and useful arts Congress may secure to inventors for a limited time the exclusive rights to their discoveries.

The temporary right to exclude others from the use of an invention is an incentive essentially material in its nature, addressed to the cupidity of men.  It brings to bear upon the individual all the impulsions that can arise from economic necessity.  It has been said of this incentive (1) that it encourages invention and research; (2) that it induces an inventor to disclose his invention or discovery, instead of keeping it a trade secret; (3) that it offers inducement to risk capital for development through the trial stage that precedes marketing; and (4) that it attracts capital into new production which might not appear to be profitable if competing producers were free to follow the leader.  Let us look at these incentives one at a time.

1. As to encouraging invention and research, is it not the pressure of economic necessity on the inventor or producer that is the truly effective impulse to technological innovation; is not that the necessity which is proverbially the mother of invention?  Invention is surely an expression of the deep creative impulse of mankind, an impulse that hardly needs another spur.  So far as organized research is concerned, that would remain indispensable to modern competitive industry if the patent system were abolished.

2. But the inventor, like the rest of us, has to live and so we need to make it possible for inventors to earn a livelihood; we must make it at least as profitable to supply the continuing demand for inventions as it is to supply ordinary articles of commerce.  The inventor with his new idea in mind must be able to find a market.  His difficulty is that as Mr. Jefferson was wont to remark, a new idea is something “which an individual may exclusively possess as long as he keeps it to himself; but the moment it is divulged it forces itself into the possession of everyone.”  Consequently, if an inventor has no other choice, no artificial pro-


tection, he will certainly use all his ingenuity to keep his invention secret, or as nearly so as may be.  Thus the choice is always between full disclosure in exchange for the temporary monopoly the patent system offers him, or secret practice to the extent and with all the safeguards that his ingenuity can devise.  All my experience strengthens in me the conviction that this inducement to choice of disclosure in preference to secret practice or carefully guarded know-how is the central point at which the incentive needs to have its maximum effect.  Nor is the margin of choice at this point as deep as a well or as wide as a barn door.  The advantage or disadvantage of swapping disclosure for the temporary monopoly may be a very close question in many cases, even under a system of temporary monopoly grants considerably more efficient than ours is today.  An inefficient patent system, or one that begrudges the temporary monopoly, may very easily defeat its own basic purpose of full and free disclosure.

3. and 4. The effect of temporary private monopolies to induce the venturing of risk capital in the initial development stage and in the stage of commercial production is contingent upon a free enterprise system based on voluntary investment of private capital in a competitive economy.  In such a system these inducements may be important, but only in such a system.  The significance of this has been emphasized very much, to my mind at least, by our recent doings in the field of atomic energy.  In this field the government has been driven to reserve to itself a total monopoly of the production of fissionable material and atomic weapons.  The inducements to risk capital have, therefore, no significance, and so we have done away with the grant of patent monopolies in these fields.  Disclosure has been made mandatory by law; a compensation is by government award.

Since no patents are to be granted for producing fissionable material or atomic weapons, any innovations which security requirements allow the Atomic Energy Commission to publish will become immediately available to everyone.  We will have, in effect, for those published innovations a general licensing system without payment of royalty.  Indeed, the Attorney General has recommended that as a basic policy all government-owned inventions should be made fully and unconditionally available to the public without charge, by public dedication or by royalty-free nonexclusive licensing.

The fact that we have, almost inevitably, been led to a system of government awards in the excluded fields of atomic energy and toward a policy of public dedication of all patentable inventions owned by the government discloses at once the potentialities and the limitations of any such system of general compulsory licensing as that recommended by the TNEC seven or eight years ago.  That proposed substitution of


compensations fixed by an administrative or judicial agency of government for temporary monopoly is appropriate enough wherever the government is prepared to supply the risk capital necessary for the development of a particular art or has acquired inventions as a result of research or development already paid for.  As an over-all policy it is consistent with an economy of government ownership, but it is inconsistent with an economy of private enterprise.

Without attempting here to follow the significance of this fact into many fields of application of compulsory licensing which have been suggested, I confine myself to pointing out that the idea of a royalty fixed by the government necessarily entails some sort of obligation to enforce the exclusive right against those who do not voluntarily pay royalty; that standards of reasonable royalty are practically unavailable to a government agency; and that such a system would be devoid of that separation of the wheat from the chaff among patented inventions that characterizes the realistic operation of private grants of temporary monopolies.  I may add, however, that my rather extensive experience during World War II in the administration of governmental controls in areas where there were no commonly accepted standards of measurement and no established methods of enforcement leaves me with a deep conviction that the practical administration of any system of general compulsory licensing lies utterly beyond the limitations of democratic government.

However that may be, it seems quite clear that in the critical economic adventure that lies immediately ahead of us we are committed to our present kind of patent system which aims as a general policy to secure to inventors for limited times the right to exclude others from the use of their inventions or discoveries.  That does not mean, in my understanding of it, that compulsory licensing need be wholly eliminated.  We already have it in more than one area of the patent system.  Since 1910 there has been compulsory licensing for use by or for the government, by act of Congress.  And compulsory licensing is recognized by the Supreme Court as an appropriate means to correct a monopolistic situation built up in violation of the antitrust laws.  Furthermore, the federal courts have always had and have often exercised the discretion to withhold injunction, after adjudging validity and infringement, where public health or safety is involved or where very special circumstances make injunction an excessive hardship for the defendant without substantial benefit to the plaintiff.  Nor have we yet heard the last word on the subject of compulsory licensing as a remedy for unjustified nonuse.  Since one great object of the patent privilege is to bring inventions into actual use and put their benefits within the reach of others, it would seem not unreasonable to declare by act of


Congress that prolonged and unjustified failure to use or license is an abuse of the patent right, and to make such abuse a defense to a suit for injunction and a ground for declaratory judgment proceedings in the federal courts.

If, as I suspect, we are committed to the granting of temporary monopolies for the promotion of technological innovation for the immediate future, the important thing seems to be to make the best of this resource, as we must now make the best of all our productive resources.  Obviously this choice of incentives entails a cost, since private monopoly is in general inconsistent with free enterprise.  But any scheme of promotion entails some risk and faint heart ne’er won fair lady.  But courage is a virtue only when directed by prudence, so the real question is whether the particular scheme is worth its cost.  This question has given rise to much discussion of the virtues and the vices of our patent system.  That discussion has yielded, I think, very little fruit.  It is a very ancient saying that criticism comes easier than craftsmanship, and that is peculiarly true of our patent system at the present moment.

One reason is that we have no basis for factual comparison with any alternative.  While the grant of temporary private monopolies has been adopted in every industrialized country in the world, no alternative incentive has ever been tried anywhere except recently in Russia.  We do not know anything about the results in Russia and if we did, that knowledge would be without much significance for an economy of private enterprise.  Another reason is that our patent system falls so far short in these days of effective administration that it is practically impossible to assess what might be, in a modern industrial economy, the true virtues and the true vices of an effectively administered system.  In the absence of any alternative experience, it is extremely difficult to get any factual knowledge on which to base such an assessment anyway, and the present administration of our system is so encumbered with delays and frustrations that its every virtuous aspect and every vicious aspect are entangled in and distorted by these inefficiencies - so much so that I think we waste time in theoretical discussion of virtues and vices under present conditions.  The most we can do, I believe, is to try to define standards for an effective patent system and then devote our energies to bringing our system up to those standards, as nearly as may be.  And this ought to be done now, when we need all our resources.  What I am about to say about such standards and the possibility of approximating them in practice is a somewhat condensed repetition of thoughts expressed more at length than I have time for here as part of a symposium on patent laws published in the current issue of Law and Contemporary Problems at Duke University.

The grant of temporary private monopolies as an inducement to sub-


stantial technological innovations, which is the essence of our patent system, implies a contract in which the price paid by the inventor is the disclosure of his invention and the price paid by the government is the assurance that for seventeen years no one will be permitted to make use of the invention without the patentee’s consent.  This calls for a full and frank disclosure from the patentee and for good faith and diligence from the government.  That patents shall be granted only for substantial innovations which the inventor has in fact originated and fully disclosed and that the protection by government of the patentee’s exclusive right shall be prompt, sure, and reasonably inexpensive, are thus requirements inherent in the very nature of the system.

The effectiveness of any incentive depends upon its appeal to the individual to whom the incentive is offered.  The nature of the incentive needs to be such as to arouse in him some eagerness to accept it in preference to other alternatives.  In the case of inventors this means in preference to the alternative of keeping his innovation secret, or as secret as he can.  To my mind, the growing emphasis on undisclosed know-how and the broadly expanded resort to complex patent licensing agreements both reflect, in the realistic world of American industry, dissatisfaction with the delays and frustrations of our patent system as now administered.

That within the century there have been significant changes in the character of our industrial society and particularly in the relation of the individual inventor to the progress of applied science and useful arts is clear.  It is a commonplace remark that we have passed from a stable society to an adaptive one, and nowhere is this more apparent than in the accelerated flow of technological innovations.  This increase in the tempo of technological improvement and its effect upon the environment of the patent system has correspondingly increased the importance of clear definition, simple and prompt procedure, and surety of enforcement.  But, unfortunately, in this period we have suffered retrogression rather than progress.  It would be easy to be pessimistic about this, but I do not think we can afford to be.  On the contrary, I think we now need to treat this situation as one offering an opportunity for correction that could, at this critical moment in our industrial life, quite significantly promote the progress of science and useful arts.

The primary functions of the Patent Office are to afford a publicly accessible general repository of industrial technology, and to act promptly in granting patents for real inventions and in refusing patents for anything less.  It must be admitted that the Patent Office is markedly deficient in performing these functions.

That there should be available in the Patent Office a greatly im-


proved repository of industrial technology will be very clear to anyone who takes the trouble to study the Patent Office classification system and the equipment of the Patent Office library.  Every Commissioner of Patents for years has been aware of these deficiencies but has never yet had adequate appropriations to correct them.  The time has now come, I believe, when decisive action in this direction must be undertaken.  Although the problem lies within the Patent Office, it is closely related to a more general problem of creating a publicly-accessible repository of the whole body of scientific knowledge.  This relation has been recognized.  In all recent legislation dealing with governmental research and development there has been evident a Congressional desire to make scientific and technological knowledge readily available to the public.  It is a large problem.  It warrants, I believe, immediate and careful consideration by a well-chosen group of experts, either under the auspices of the executive branch of government, or under the joint auspices of the Executive and the Congress.

As to prompt issuance of patents by the Patent Office, the situation, which has long been deteriorating, has now become one of critical emergency.  The backlog of the Patent Office in the summer of 1946 was two and a half times as great as in 1942 when it was already much too great.  There were over 100,000 applications and amendments awaiting action by the examiners.  The average delay in Patent Office actions was more than a year, and the average time for passing an application through the Office was six years.  This cannot be permitted to continue very much longer without a breakdown.  It calls for emergency action including increased appropriations.  Its permanent correction requires more than emergency action.  Basic reforms in Patent Office equipment and procedure are needed.  Abandonment of the idea that the Patent Office must be self-sustaining is called for.  Better tools for the Patent Office examiners are an inescapable necessity.

One significant legislative proposal that has met with pretty general approval is the so-called “twenty-year” law that no patent shall in any event have a term longer than seventeen years and all patents shall expire at a date not more than twenty years from the date when the application was filed, with limited discretion in the Commissioner of Patents to make allowance of not more than two years from the filing date for unavoidable delays not chargeable to the applicant.  The enactment of this bill could be expected to put pressure on the applicant and the Patent Office, and even on the appropriation committees of the Congress, to speed up the issuance of patents.

A more direct attack, aimed at a reduction of the overall load on the Patent Office, has been proposed.  It is well known that a great many patent applications are filed, particularly by industrial corporations but


not only by them, in cases where the owner of the innovation does not really want or expect to exercise a monopoly.  These are the so-called “defensive” patent applications.  Unfortunately, the applicant does not get that defensive protection unless the patent issues.  The result is that industry and the Patent Office are burdened with prosecution of these applications to final allowance, even though no one expects that the issued patent will, or intends that it shall, become the basis of a monopoly.  It has been proposed that this burden be lifted from the Patent Office and from industry by authorizing the applicant to abandon the application after a first action by the Patent Office and after complying with requirements of the Commissioner as to form, and that an application so abandoned be published with the same evidentiary effect now attributed to the filing of a patent application which eventuates in an issued patent.  The proposal is worth very careful consideration.  It may be a feasible way to reduce very substantially the load on the Patent Office.

It is an anomaly in our patent system that appeals from the decision of the Board of Appeals of the Patent Office may follow either one of two courses, as the dissatisfied applicant may choose.  He may either appeal to the Court of Customs and Patent Appeals, on the record made in the Patent Office, or bring a bill in equity in a federal district court.  It has been suggested that the appeal to the Court of Customs and Patent Appeals, an administrative procedure, should be eliminated.  Statistical studies and informed opinion indicate that such change could be expected to result in a significant reduction in the number of appeals.  It would also contribute to that degree of clarity and uniformity in the actions of the Patent Office examiners and the Patent Office Board of Appeals which must be achieved to break the intolerable bottleneck that the Board of Appeals now is.

In the field of judicial interpretation and administration of the patent laws, considerable thought has been given to the delays, uncertainties, and costliness of patent litigation and to the very considerable gap that seems to exist between the Patent Office and the courts as to what is patentable subject matter.  The principal suggestion now current in this field is the creation of a single court of patent appeals to take over, subject to review by the Supreme Court on certiorari, the final jurisdiction in patent litigation that is now divided among the ten Circuit Courts of Appeals and the Court of Appeals of the District of Columbia.

This suggestion has been under discussion for many years.  There are two schools of thought.  One regards the suggestion as a wise and necessary thing to eliminate conflicts, uncertainties, and multiplication of litigation on particular patents that now prevail.  The other fears


that separation of the judicial process in patent cases from the general body of our federal system would tend to isolate the court from those contacts with human and commercial problems which keep the judicial vision broad, and would lead to technical and narrow attitudes and judgments destructive of what should be a living and dynamic system.

Among many suggestions as to how these two points of view might be reconciled, perhaps the most carefully considered one is that the proposed single court of patent appeals should have not more than two permanent judges, one of whom would be chief judge, and that the other judges should be designated from time to time, for temporary service, by the Chief Justice of the United States from the federal judiciary, the temporary judges being in the majority in every case.  In this way it is hoped to secure the uniformity, finality, and reduction of litigation sought for, and at the same time to retain the beneficent effects of the broader judicial experience to which the whole body of federal judges is continuously exposed.

There is another proposal that suggests a somewhat different approach to many of the problems that grow out of our multiple courts of appeal.  That suggestion is that the final judgment of a federal court holding a patent or claim thereof invalid should have the effect to revoke the patent, subject to the right of its owner to have it reissued after due examination in the Patent Office.  This suggestion implies a sort of continuing responsibility of the Patent Office, called into being, however, only by a final judgment of a federal court.  It seems worthy of careful consideration, which should include, of course, its possible effect upon the already highly developed law of reissues.

A great deal has been said and written about the question whether, and to what extent, and how and why, the monopoly granted under patents for inventions is in conflict with the antitrust laws.  My belief is that there is no necessary conflict between these two long-established concepts of American life; that, on the contrary, if properly administered, they complement and support one another.  Yet it cannot be denied that patents have been made use of in the building up of combinations and conspiracies in restraint of trade which have violated the antitrust laws, and that the patent privilege has sometimes been projected beyond its legitimate scope to interfere with the free flow of commerce in unpatented articles and materials.

There can hardly be any doubt that such abuse or misuse of patent rights is increased by the current obesity of our system of issuing patents and of interpreting and enforcing patent rights.  The inducement and the opportunity for such abuse might be quite different if we had a patent system in which patents were promptly issued only for patentable inventions and could be adjudicated and enforced quickly


and at minimum expense.  Suppose, for instance, that the average patent application could be processed by the Patent Office in six months from filing, that a suit for infringement could be brought to decision in a District Court within six months from filing the complaint, and that another six months would cover any appeal.  The picture would certainly change its whole aspect.

In the meantime, we can and we should segregate in our thinking abuses which may arise out of defects in our system of issuing and enforcing patents or out of defects in the patent grant itself from those abuses which result from license agreements, contracts, and combinations entered into by patent owners.

It is as true of owners of patents as it is true of the owners of other property that they are forbidden by the antitrust laws to enter into any contract, combination, or conspiracy in restraint of trade, or to monopolize or attempt to monopolize any part of trade or commerce beyond that limited monopoly secured to them by existing patents.  Within this area of combination encroaching upon the territory forbidden by the antitrust laws, the patent laws afford no protection.  Such conduct can be prevented by enforcement of the antitrust laws, and calls for no change either in the antitrust laws or in the patent law.

Even though it may thus be assumed that the antitrust laws afford adequate safeguards against bilateral contracts, combinations, or conspiracies in restraint of trade where patents are involved, yet the fact remains that a patent monopoly may give the owner a unilateral power, either without any specific contract or by means of patent licenses which are in substantial effect unilateral, to interfere with the free flow of commerce in things which lie outside the patent monopoly.

Thus, throughout the history of the patent law, questions have arisen as to whether and to what extent the patent owner may use his monopoly to restrict a licensee or purchaser in the manufacture, use, or sale of unpatented things.  The result has been to build up judicial interpretation by a considerable body of law within this field.  Particularly in recent years the Supreme Court has developed a line of cases condemning unilateral uses of patent monopolies to restrain or interfere with the free flow of commerce in unpatented articles, materials, or industrial practices; and as part of this more recent development the Court has evolved the judge-made rule that a patentee who has so abused or misused his monopolistic rights is debarred from exercising those rights until the improper use has been ended.

This case-by-case judicial evolution of restrictions upon the activities of patent owners has been contemporaneous with growing concern about our patent system from two opposed points of view. At the one extreme are those who conclude that free competition calls for abolition of


the patent laws, and at the other are those who regard these restrictions with terrified alarm as destructive encroachments upon the patent monopoly.  Between these extremes, it would, I suggest, be possible for mature minds to find agreement on at least two principles: (1) that it is desirable, and consonant with the mores of the nation, to foster competition in the production of and among patentable inventions, and (2) that any restriction upon that exclusive right to make, use, and sell which characterizes the patent monopoly reduces the incentive of the grant and so tends to discourage disclosure and reduction to practice and to encourage secret trade practices in preference to disclosure, so that all excessive or unnecessary restrictions are dangerous in themselves.

It is clear that dividing lines in this field cannot accurately be drawn merely on legal considerations.  Their location depends rather on the answers to practical economic questions in our complex and vigorous industrial structure.  Unfortunately, there is a marked deficiency of real knowledge of the factors that determine the answers to these practical questions.  The tendency has been rather to resort to emotionally inspired guesses, or even dogmatic assertions, both in and out of court.

Out of this situation arises a very serious question whether we should continue to rely upon the gradual, step-by-step development of these lines of demarcation by judicial process or make a carefully studied legislative attempt to define more explicitly and with more assurance of stability those activities of patent owners in the borderland of dispute that are acceptable and those that are forbidden.  It would, of course, be possible to conduct this sort of rule making partly by legislative action and partly by a duly authorized and instructed administrative tribunal.  The Clayton Act and the Federal Trade Commission Act are suggestive precedents.