It is often said that intellectuals have a hard time dealing with new ideas. Unfortunately, for purposes of public debate, open-source government funding of drug development is a new idea, and people in policy positions seem to be having a very hard time understanding it. So, I will try to write this post in a way that even a policy wonk can figure it out.
The basic idea of government funded research should not be hard to grasp since the government already funds a large share of biomedical research. The National Institutes of Health gets over $40 billion a year in federal funding, with the Biomedical Advanced Research and Development Agency (BARDA) and other government agencies getting several billion more. This puts the government’s total spending in the $45 to $50 billion range, compared to a bit over $90 billion from the industry.[1] So the idea that the government would fund research really should not be that strange.
Most of the public funding does go to more basic research, but there are plenty of instances where the government has actually funded the development of new drugs and also done clinical testing. But under the current system, most of the later stage funding does come from the industry and is funded through patent monopoly pricing. Relying on open-source government funded research for later stage development and testing would be a major change.
The Outlines of a System of Government Funded Research
To my view the best way for the government to support the development of new drugs is through long-term contracts (10-12 years), which would be awarded through competitive bids for research in specific areas, like cancer or heart disease. The plan would be that the contracts would be relatively large, with the idea that the winners would be comparable to prime contractors for the military. (I describe this system in somewhat more detail in chapter 5 of Rigged [it’s free].)
Major military contractors, like Lockheed or Boeing, typically contract out to many smaller companies in specific areas. This is a good model. Most of the major innovations in the development of new drugs have come from start-ups, who are often bought out by major pharmaceutical companies like Pfizer or Merck. The winners of prime contracts under this system would be foolish not to look to award contracts to innovative start-ups, to ensure that they have something to show for their work.
One condition that would apply to both prime contractors and any subcontractors is that all research findings would have to be fully open, meaning that they are posted on the Internet as soon as practical. This would apply both to pre-clinical research and the results of clinical trials. The posting of trial results would mean that researchers around the world would be able independently analyze the data and assess the effectiveness and risks of drugs and vaccines for different populations.
I have had many people ask me what would be the incentive for the companies that win contracts to actually innovate as opposed to just spinning their wheels. Since they presumably would want to renew their contracts when they expire, that should provide substantial incentive for them to have something to show.
Also, the researchers would presumably want to actually do something with their time rather than just looking to collect a paycheck. I have also suggested having a large pot of money (e.g. $200 million a year) to pay out as prizes, similar to a Nobel Prize. If a researcher, or group of researchers, have a major breakthrough that will radically improve the treatment of heart attack victims, why not give them $10 million? But this prize would be on top of their ordinary pay, not a replacement.
People have often raised the problem of political influence determining the awarding of contracts. There is always a risk of political interference, which of course arises under the current system as well. One advantage of this system is that the full public posting of results should at the least make blatantly political decisions difficult, if not impossible. If a company had received $30 billion to research lung cancer over a ten-year period, and had nothing to show in terms of new drugs or major innovations, it would be hard to justify another long-term contract to the same company.
All of the drugs developed through this system (the funding would include carrying new drugs through the FDA approval process) would be available as generics from the day they are approved. This would mean that new drugs that may sell for thousands of dollars, or even tens of thousands of dollars, under the patent monopoly system, are likely to sell for ten or twenty dollars. Drugs are rarely expensive to manufacture and distribute, under a system of government-financed open-source research they would also be cheap to buy.
Government Funding in the Pandemic
The pandemic provided a great opportunity to experiment with open-source government funding. While we did the government funding part, with the U.S. government alone putting up $10 billion through Operation Warp Speed, we did not get the open-source part. The government effectively paid for much or all of the research, but still gave private companies patent monopolies.
Ideally, we would have negotiated an international pact, where all countries would contribute to research, based on their size and relative wealth, and all findings would be fully open to researchers around the world. I’ve heard people object that it’s difficult to negotiate these sorts of deals, and we needed to act in a hurry. Some people have also insisted that China and Russia never would have agreed to such a deal.
On the first point, it really should not have taken very long to negotiate a pact. We are throwing money around all over the place. No one thinks that we are getting things exactly right. Some industries and individuals are getting compensated in ways that they probably don’t need/deserve and undoubtedly some people are being left behind. If the United States or some other country chips in 20 percent too much or too little, it would be chump change relative to the costs of the pandemics and various rescue packages being put forward.
As far as including China and Russia in a deal, I have no idea whether they would be anxious to join if given the opportunity. They have joined in many other international agreements, so there would be no prima facie reason to assume that they would not want to be parties to this sort of arrangement. We also can’t know for certain that they could be counted on to contribute their agreed upon share and to make results fully open, but as practical matter both countries have been reasonably good about adhering to other agreements to which they are a party.
It would have been enormously advantageous if China and Russia would have been included in a pact with open-source research. Ideally, if all the successful vaccines were fully open, including their production processes, manufacturers anywhere in the world that had the ability to produce these vaccines could have done so.
And, they could have begun to ramp up production even before the vaccines had been determined to be safe and effective. The cost of manufacturing 1 billion vaccines that turned out not to be approved is trivial, in both lives and money, compared to the cost of waiting to have 1 billion vaccines become available so that they can be administered.
While we can never know how much quicker we could have learned about the effectiveness of vaccines, and arranged their distribution, in an open-source world, we can have some idea just from what we have learned over the last several months.
For example, Pfizer discovered that it is possible to get six shots out of a standard shipping vial, not the five they originally believed. This means getting 20 percent more vaccines. If this knowledge was available sooner, it might have meant hundreds of thousands or even millions of additional vaccines.
There is also evidence, based on data from Israel, that the Pfizer vaccine is highly effective after just one shot. It is possible that if all the clinical trial data was fully public that this result could have been discovered sooner. This would have allowed countries to adopt a one-shot strategy with the second shot coming after most people had received their initial shot.
Pfizer also discovered that its vaccine can be kept in a normal freezer for up to two weeks, instead of requiring super-cold storage. If this was known sooner, it would have greatly facilitated the storage and distribution of the vaccine.
And, Pfizer reported last month that it had discovered a way to alter its production process to nearly double its output of vaccines. It’s hard to believe that if engineers all over the world were familiar with Pfizer’s production process not one would have been able to realize this potential improvement more quickly.
We also got an interesting lesson about incentives with reports that Astra Zeneca cherry picked the data it presented to the Food and Drug Administration (FDA) to gain approval of its vaccines. While Astra Zeneca denies the charge, this problem would not exist in an open-source world.
First, the company that developed a vaccine or drug would have no special incentive to see it approved if it was not safe and effective. While it would like to have something to show when a contract came up, the risk of having a drug approved that later turned out not to be safe or effective would likely far exceed any potential benefit.
More importantly, the clinical trials would not be under its control. The data from these trials would be fully public so that any researcher anywhere in the world could analyze it independently. If the FDA or some other regulatory agency misread the data and made the wrong call, it is virtually certain that independent researchers would be able to recognize the mistake and call public attention to it.
We would not have stories like Purdue Pharma misleading physicians about the addictiveness of its opioids. Under a system of open-source research they would have neither the incentive nor the opportunity.
If People Act the Way Economic Theory Predicts, We Should Want Open-Source Research
Economists always want to look at the incentive structures that we create with our policies. While patent monopolies do create incentives to develop drugs and vaccines, they also create incentives to keep as much research as possible secret, so as not to benefit competitors. The huge markups allowed by patent monopolies also create an enormous incentive for drug companies to lie about the safety and effectiveness of their products.
Open-source, publicly funded, research radically alters the structure of incentives. The incentive is to try to have research results spread as widely and quickly as possible in the hope that others can build on them. And, there is no incentive or opportunity to push drugs that are unsafe or ineffective. This is a big deal.
[1] The figure on private R&D spending comes from the National Income and Product Account, Table 5.6.5, Line 9.
