Two potential areas for rapid innovation are telecommunication and medicine. I am very concerned that these areas are regulated by the FCC and the FDA, respectively, two agencies that were established in different eras. My fear is that these agencies are culturally incapable of adapting to the environment that scientific advances have created. I have proposals for sidestepping each agency.
I must admit that I have no scientific expertise in either telecommunications technology or biochemistry. I may be misjudging the most likely paths for innovation in these areas. Even so, it still may help to create new institutions for allocating spectrum and rewarding medical research that are free from the shackles of FCC and FDA tradition. We need agencies that will look at these problems with fresh eyes.
The Spectrum Arbitration Board
All responsibility for regulating spectrum should be transferred from the FCC to a new agency, which I call a spectrum arbitration board. The board will adopt a common-law approach to spectrum, rather than a command-and-control approach.
The command-and-control model, in which the FCC dictates how particular bands of spectrum may be used and by whom they may be used, seems to be based on a view that different uses of spectrum must be separated in order to avoid the problem of “interference.” However, there are other solutions to this problem.
For example, there is a communication method called “spread-spectrum” that does not create interference. The spread-spectrum signals only affect receivers that are programmed to understand them; to other receivers, spread-spectrum signals appear as noise that is readily filtered out.
Another important concept is cognitive radio. That means a transmitter that can adjust automatically to find an open frequency with which to communicate with a similarly-programmed receiver.
Even if these technologies do not take over telecommunications, it is still the case that spectrum is allocated very inefficiently in the United States today. For example, in spite of the fact that fewer than 10 percent of households still receive television signals over the air, large swaths of spectrum are reserved for broadcast television.
If spectrum were used efficiently, it is possible that Americans could all have access to low-cost high-speed wireless Internet connections everywhere. In addition, the so-called “Internet of things” could emerge more rapidly.
Unlike the FCC, the arbitration board would presume that any spectrum could be used for any purpose. The board would set ground rules for users to deal with one another to resolve potential conflicts. These ground rules might specify which user has priority until an agreement can be reached. The ground rules might set expectations for how negotiations ought to be conducted and resolved. If parties are unable to resolve disputes, then the board would rule on that specific dispute.
Many economists have argued that spectrum should be governed by property rights, comparable to the property rights in land. If the arbitration board determines that this is indeed the case, its job will consist of defining such property rights and determining who owns them. For example, the owner of the television license for channel 6 in Philadelphia might be said to own that slice of spectrum in that region. It could then use, rent, or sell any portion of its spectrum property.
However, there are proponents of the newer technologies who doubt the need for strong property rights in spectrum. As an analogy, consider the problem of avoiding airplane collisions. The property-rights solution would be to allocate specific swaths of sky to different airlines. Instead, we use rules, technical requirements, and air traffic control systems. Perhaps the arbitration board will determine that the best approach to dealing with spectrum is something similar.
An Agency to Offer Prize-Grants for Medical Research
Medical research has entered a new world, created by the genomics revolution. However, as Peter Huber has pointed out, the institutional structure for creating and protecting property rights has not kept up.1 Drug developers now can use genetic information to identify individuals who need or are likely to respond to a drug. This makes standard testing protocols, which call for broad-based, blind trials, inappropriate. It also blurs the distinction between basic and applied research.
Huber and others focus on the FDA approval process as an impediment to progress. Their thinking is that if the FDA were faster and more flexible about granting approval for medical innovations, then patients would benefit sooner.
I am looking at the problem more broadly than that. To me, it appears that 21st-century medical research is difficult to reconcile with the patent system, and the FDA approval process merely compounds that more fundamental problem.
The current system rewards companies that come up with widely-applicable drugs, devices, or medical tests that can be patented. It is not suited to rewarding incremental additions to knowledge or the development of therapies tailored to narrow populations.
What we want to reward is the development of practical knowledge. However, some knowledge is not patent-able. Consider, for example, the discovery of “off-label” uses of a drug, which rewards the original developers of the drug but not those who discover the new use. Another example would be a protocol for combining drugs to treat certain cancers in certain patients but which does not involve the development of any new molecule.
We could broaden the patent system to allow protocols and new uses for drugs to be patented. However, the objective of policy should be to increase the amount of knowledge in the public domain, a goal which is not served by a thicket of patents. Broadening the patent system would serve to reward patent lawyers rather than researchers.
Instead, we need to do a better job of aligning the institutions that surround biomedical research with the path that medical discovery is likely to take going forward, as we seek diagnostic tools and treatments based on genetic information and biochemistry. What I propose here is a hybrid of a research grant and a prize, which could be awarded to either for-profit or nonprofit entities.
The prize-grant would be a contract between a research organization and a funding institution. The funding institution could be either a government agency or a private foundation. The National Institutes of Health would be the natural home for a prize-grant agency within the U.S. government.
The process would work as follows. First, the research organization proposes a study to test for a particular result. This might be the identification of a biomarker for a disease, or a demonstration that a new diagnostic tool is more accurate than existing methods, or the creation of a treatment targeting a specific population.
Next, the funding organization would offer a specific amount that the research organization will receive if the study produces a positive result. If the study produces a negative result, then the research organization receives nothing. Either way, the results of the study are placed in the public domain.
The contract would also specify a method for third-party verification of the result. The cost of verification would be paid for out of an escrow account funded in advance by the research team, although the third party would be chosen by the funding institution. In some cases, the verification might be done concurrently with the study, perhaps by the third party auditing the study to determine whether it is conducted in a manner that produces reliable results.
A research organization would have to decide whether or not to proceed with a project, based on its view of the likelihood of success, the cost of the project, and the value that the funding institution is willing to assign to a successful result. The researchers and their backers, whether a drug company, university, or venture capitalist, would have to make the decision whether to accept the contract and undertake the costs of the research.
The prize-grant contract would differ from a plain research grant in the following ways:
- Profit-seeking companies would be just as eligible as non-profit research institutes to receive prize-grants. The studies contribute to basic research.
- The burden of assessing the probability of success would fall on the research organization, not on the funding institution. Those doing the research would in effect be betting on its success.
- Those doing the research would have the incentive to use funds wisely and for their intended purpose. The cost of waste, including excess overhead, would be borne by the research organization, not by the funding institution.
- The funding institution would have to screen applications to ensure that the research will be conducted safely and ethically. However, it would neither have to “qualify” the researchers based on their scientific credentials and beliefs nor assess the probability of success. A team headed by an oddball who is scorned by mainstream scientists could obtain a contract, risking the team’s resources that the oddball’s idea will pan out.
- The funding institution would have to assign a value to a successful result before the research is undertaken, so that this value can be specified in the contract. This would be the most administratively difficult part of the process, and each research organization would have to submit an application fee to cover the cost to the funding institution of assessing the value of the intended result of the research.
The prize-grant would differ from an ordinary prize in the following ways:
- The criteria for winning the prize would typically be first suggested by the researchers, with funding institutions then assigning a value for the prize, prior to the research.
- Prizes often would be for incremental achievements, not just for spectacular accomplishments.
- Large pharmaceutical companies and other private firms would be just as eligible as non-profit researchers to receive prize-grants.
We can think of the current intellectual-property regime in medical research as a prize-grant approach in which the prize is a patent. However, prize-grants differ from patents in the following ways:
- The prize for a successful result is specified by the funding institution. With a patent, the value of the prize is determined in part by patient demand but also by the purchasing rules of insurance companies and governments, by legal jousting, and by gaming of the system.
- Useful research that does not result in a patentable product gets rewarded under prize-grants, whereas under the patent system such research does not get rewarded.
- Regardless of the outcome of the research undertaken in pursuit of a prize-grant, findings would be immediately placed in the public domain. In contrast, patents set a term of monopoly on the use of information, during which the prices of patented products can be set far above production cost.
It is possible that existing drug companies would be unable to adapt to a regime in which they are paid in cash rather than in patents, and in which they are rewarded for providing incremental knowledge rather than discovering blockbuster drugs. It could be that new profit-seeking research institutions would have to emerge in order for the price-grant system to work.
One way to understand the prize-grant approach is to recognize that pharmaceutical research conducted by private firms is a public good. Such research adds to the general stock of knowledge. If there were no public-goods aspect to pharmaceutical research, then we would not need patents. If there were no patents, then the prices of drugs would be driven down to production cost, which is much lower than the prices charged for patented drugs. On the other hand, without patents, private firms would not have the incentive to discover and test new drugs.
Patents have always been a problematic way to promote innovation. They raise prices of products far above marginal cost. They impose legal costs involved in obtaining, attacking, and defending patents. They provide an artificially high incentive to develop substitute products that devalue the patented invention. They create an artificial disincentive to develop complementary products, because the high price of the patented product limits its market penetration, adversely affecting would-be product complements. Most important, there is a lot of important knowledge to be obtained that cannot be shoehorned into a patentable molecule or device.
In pharmaceuticals, the challenges with using the patent system are increasing. As Huber has pointed out, the nature of molecular medicine is changing. The system of rigid, blind clinical trials needs to be replaced by a regime of focused trials in which researchers learn and adapt as they go.
This modern approach to drug development combines biochemisty and genetics. It tends to blur the distinction between basic research and applied research. It used to be that basic research looked at the biochemical mechanism of disease and applied research then tested cures (or sometimes the other way around — cures were found, and basic research then tried to understand the mechanism of the cure). With personalized medicine, the testing process itself yields important information about the biochemical mechanism of disease.
The human trials that used to be considered applied research are now vital in their contribution to basic research. However, the structure of this research needs to be tailored toward discovering new information for the public domain, not on fitting a rigid FDA process for approving a patentable drug. The prize-grant approach could provide a better alternative.
Notes
1 Peter Huber, The Cure in the Code: How 20th Century Law is Undermining 21st Century Medicine (2013).
The opinions expressed here are solely those of the author and do not necessarily reflect the views of the Cato Institute. This essay was prepared as part of a special Cato online forum on reviving economic growth.
