« Women and wealth | Main | Markets in everything: rentable goat edition »
A prize for the Edwards plan?
I am awaiting details but this proposal from John Edwards is not entirely crazy.
Mr. Edwards said he wanted to discourage pharmaceutical companies from obtaining long-term patents on medicines for specific ailments like Alzheimer’s and cancer. Instead, an upfront cash prize would be made available to serve as an incentive for research on such drugs.
I worry, however, that the prizes will be far too small. Since the social value of breakthrough medicines greatly exceeds the private profit, prizes of tens of billions of dollars would not be unreasonable. In fact, optimal prizes must increase the profits of US drug companies. Can a Democrat like Edwards sell that? And who will decide how the prizes are handed out? Can the US government award billions of dollars in prizes without significant rent seeking?
Partly for these reasons, I would much prefer a patent buyout as suggested by Michael Kremer (Kremer's paper can also be found in Entrepreneurial Economics.) Let's at least have a few experiments to buyout say 5 years of the time remaining on some important patents.
Do keep in mind that the problem of expensive drugs is overblown - a typical new drug will go off patent in 12 years anyway. The real issue is how best to increase the incentive to develop new and important pharmaceuticals.
Posted by Alex Tabarrok on June 15, 2007 at 07:20 AM in Economics | Permalink
Comments
What is crazy, given such data points as "Reagan and AIDS" and "Medicare and Viagra," is to think that Edwards' proposed "expert panel" would be immune from political interference.
Posted by: KipEsquire at Jun 15, 2007 8:18:01 AM
Stiglitz has been promoting the prize idea, for example at http://www.cptech.org/ip/health/js_new_scientist_06.pdf (PDF).
Posted by: tom s. at Jun 15, 2007 8:25:21 AM
So who lives in Iowa or New Hampshire... that would go to an Edwards event prepared to ask him a question that might get this issue in play? A softball question might be, do you agree with economists who say we should triple the current Pharma research, and if so, how would you make that happen?
Meanwhile, shouldn't Americans be ENCOURAGED to resent the burden of that decade of high prices... when Germans and Canadians are free riding? How do we get them to help pull the wagon?
Is there any public policy question that is more likely to dramatically impact your family?
Posted by: dave meleney at Jun 15, 2007 8:32:44 AM
As Michele Boldrin and David K. Levine point out in their book Against Intellectual Monopoly, there are enough appropriation mechanisms on the market to enable pharmaceutical firms to earn back their cost of capital invested in drug R&D and to make competitive rents.
Why not get rid of the patent regime-junta altogether by abolishing patents?
Just as slave owners should not have been bought out, so too drug patent owners shouldn't be. They are reaping ill-gotten monopoly rents, just as slave owners were.
See also Gary P. Pisano, "Pharmaceutical Biotechnology," in Benn Steil et al., eds., _Technological Innovation and Economic Performance_ (Princeton U. Press, 2002).
Posted by: Bill Stepp at Jun 15, 2007 8:53:13 AM
In theory, an upfront cash payment would have the same incentive effect as a 17-year patent, if the upfront cash payment is set as the discounted present value of the stream of future revenues that would flow to the company under the patent system. Of course, this would not be particularly easy to calculate. Alex is right to worry that, under Edwards's (or anyone else's plan), the upfront payment might be set too low to create sufficient incentives. Meanwhile, as another commentator notes, because it is not always easy to predict the social value of a particular pharmaceutical at the time it is first introduced, the government might choose to exercise Eminent Domain to acquire property rights in the drug after the fact of production. The use of Eminent Domain presumably would not reduce the incentives to innovate as long as the compensation level is set equal to or higher than the net present value of future benefit streams under the pre-existing patent.
Posted by: Dan Cole at Jun 15, 2007 9:14:04 AM
When would the award be granted? After FDA approval?
One thing to consider is that the awards have to be large enough to pay for all of those expensive clinical trials that do not pan out. Blockbuster drugs can be worth tens of billions of dollars.
I think that prizes would be excellent for drugs with high social value but low market value. Many avenues of pharma R&D are simply not pursued because the payoff would not be high enough. However, for drugs where market forces alone fund the research, a market value patent buyout also seems productive. Why not a hybrid approach?
An aside: the US government will tell you that the government owns all patents, it just gives the patent assignee the exclusive right to use that patent. Therefore, the government does not “need” eminent domain to practice your patent, they just need to pay you for it.
Posted by: Brent at Jun 15, 2007 9:28:52 AM
The irony of the drug patent debate is that, of all IP, drugs are probably the area where some form of government monopoly is most needed! There was a paper a few years back that looked at what percentage of inventions in given fields would have been invented anyway (first-mover advantage, blind luck led to the invention, etc.); pharma was very low on that list.
That said, the deadweight loss from patents is quite large (and the "big rectangles" that those with monopoly power create through inefficiency often outweigh the loss from the Harberger triangle). There's no question that some type of buyout would be much better.
Posted by: cure at Jun 15, 2007 9:29:34 AM
The fastest and easiest way to lower drug costs is to force Europeans and Canadians to pay their share of development costs. Let them make one time payouts to the drug companies, since their medical systems are already socialized.
Posted by: Matt at Jun 15, 2007 9:31:46 AM
How about a first place ribbon?
I love how you assume John Edwards understands incentives at all. He is too busy playing the class warfare game.
An upfront cash prize? This must be a joke. I have a chemistry set in my basement. I will get to work right away.
Posted by: rick at Jun 15, 2007 9:38:27 AM
Bill, I disagree with your comparison. On a philosophical level, slave owners were claiming property rights over human beings that they were fundamentally not entitled to because the slaves were stolen originally and because sentient life-forms are capable of owning property and are therefore not rightfully considered property by others. On the other hand, pharmaceutical companies develop life-saving drugs and then assert a right to own the non-sentient item that they have created. These are completely different scenarios.
On an economic level, monopolies are not always bad in the sense that a monopoly on a cancer cure is preferable to not having a cancer cure. While theoretically it is possible to replace monopoly rents with a cash prize and end up with the same incentive structure, I have a few concerns:
First, the necessary amount of the cash prize to replace monopoly rents may be prohibitively high. Second, the government would likely use the patent to cure disease less efficiently than private pharmaceutical companies because government has the incentive to maximize transaction costs. Third, if the prize was not set high enough and the exchange of patent rights for the cash prize was made mandatory, this sort of system could dramatically decrease the incentive to develop life-saving drugs.
Posted by: Paul R. Dorasil at Jun 15, 2007 10:25:32 AM
Ezra Klein at
http://ezraklein.typepad.com/blog/
apparently has some more details. To quote:
Spoke to some folks at Edwards HQ on their proposal for prize money as a spur for pharmaceutical development. This was muddled in their fact sheet and much of the initial reporting, but this program is not, in any way, a replacement for the current system of patents. It does not, in any way, change the way patents are awarded, or how long they last, or who can apply for them. Rather, it creates a separate and parallel track, a pilot program of sorts, wherein a committee would identify diseases and conditions that would benefit from alternative incentives for innovation, and offer prize money as the reward. An example -- not given to me by anyone at the campaign -- would be prize money for malaria treatments, as drugs for conditions affecting mainly third-world residents don't find a lot of urgency in the R&D departments of publicly-traded pharmaceutical companies.
Posted by: spencer at Jun 15, 2007 10:27:23 AM
From a net-present-value accounting point of view, patents are a very reasonable way of assesssing the value of pharmaceutical innovations.
Eventually, there comes a point in time where the net present value of the revenue already made from the patent equals the net present value of the future revenue stream of the patent. At this point in time, what the public has already paid (the past monopoly profits) is equal in value to what the patent holder loses from the expiry of the patent (future monopoly profits), and the expiration seems fair.
If (for lack of a better model), we say that the patent's revenue stream is a constant-income perpetuity, then a 17-year patent corresponds to a ~4% discount rate in this model.
Posted by: Cyrus at Jun 15, 2007 10:33:56 AM
Bill Stepp's point that we should abolish patents without recompense "just as slave owners were not bought out" misses the fact that the British did, indeed, buy out slave owners. And, in doing so, avoided a war that costs us more than half a million lives. More interesting, I think, is Dave Meleney's point about those free-riding foreigners. Basically, what we have now is a system where the development costs of drugs are borne by American patients, while many foreign patients only bear the marginal costs of drug production. If we allowed for the reimportation of drugs, wouldn't this result in one or both of two things? (1) Drug companies would stop selling drugs in foreign countries where the laws mandate that the drugs be sold at prices that do not allow a recovery for R&D costs and/or (2) the price of drugs in foreign countries would increase. In both cases, it seems likely that drug prices in the US would drop.
Posted by: M.D. Fatwa at Jun 15, 2007 10:58:40 AM
How about this variation on a patent buyout: The pharma company must submit a price that they would accept for a patent buyout. They must pay a yearly fee based on that price.
If a drug is great, gov't or private charity would have an incentive to raise the funds to buy out the patent. What keeps the drug company from setting insane buyout prices? If they do then they incurr needlessly high fees that would be a waste of money if no one wants to buy out the drug. The fees could be devoted to outcomes based research to see the best ways to use drugs that already exist.
Posted by: Boonton at Jun 15, 2007 12:20:23 PM
How about this variation on a patent buyout: The pharma company must submit a price that they would accept for a patent buyout. They must pay a yearly fee based on that price.
If a drug is great, gov't or private charity would have an incentive to raise the funds to buy out the patent. What keeps the drug company from setting insane buyout prices? If they do then they incurr needlessly high fees that would be a waste of money if no one wants to buy out the drug. The fees could be devoted to outcomes based research to see the best ways to use drugs that already exist.
Posted by: Boonton at Jun 15, 2007 12:20:24 PM
(1) Drug companies would stop selling drugs in foreign countries where the laws mandate that the drugs be sold at prices that do not allow a recovery for R&D costs and/or (2) the price of drugs in foreign countries would increase."
Or 3) the foreign countries break the patent and allow a third party to make the drug thus continuing to pay marginal cost and continuing to fail to pay for R&D costs.
Posted by: Sebastian Holsclaw at Jun 15, 2007 12:26:51 PM
Since the patent system, and the resulting profits to drug companies, has a proven track record for making new big drugs, why not just leave it alone? If any additional incentive makes sense, and given the free-riding Canadians and Europeans and others, then why not just eliminate all taxes on profits earned by drug companies in the United States?
The place for prizes is where the current patent system does not have any incentive effect, as in nutrition research or drugs with small profit potential. Competing prize givers is another necessity. Why not incentives in the tax system for such prize setups?
Incentives work in complicated contexts. The capital market will smoke out irrationality in corporate management, and correct it, often in spite of ferocious resistance. The incentives working on politicians are somewhat different, where a Darwinian natural selection on successful dissembling is the main survival trait. Do we really want decisions on the direction of drug research determined professional dissemblers working Caplan's irrational voters?
Posted by: cliff styles at Jun 15, 2007 12:33:40 PM
Paul,
The taxation that supports the State--"the biggest mass murder, armed robber, and enslaver in all of human history," as Rothbard put it, is a form of involuntary servitude, just as slavery is. So patents, which are government-granted monopolies, depend on a form of slavery. Patents (and intellectual property generally) are neither property, nor consistent with liberty.
As Boldrin and Levine point out in their chap. on pharma patents (google David K. Levine; their book can be downloaded there--one of these days Cambridge will publish it), drug patents aid and abet rent seeking and the creation of "me too" drugs, but do little if anything to stimulate innovation in this area.
Italy, which didn't have drug patents before 1976, had
a thriving drug industry. So did Germany and the Netherlands in the pre-patent era. The UK, which did have patents, had little drug innovation compared to its continental competition.
To M.D. I would point out that slavery in Britain was dying before the buyout, and that the War Between the States was not fought to destroy slavery, but to impose Lincoln's dictatorial view that the South didn't have a right to secede. Which of course it did. Slavery would have died a natural death, as economically it was on the ropes. See _King Cotton and His Retainers_.
To Cyrus, see Boldrin and Levine on this. They point out that no pharma firm would use a discount rate less than a realistic hurdle rate. Indeed, they'd use a conservative rate of at least their cost of capital, which averages around 10% in the pharma industry, according to Ibbotson. It would probably be higher to build in a margin of safety.
Three B-school profs wrote a study (I don't have the cite, see Boldrin & Levine) showing the benefits of drug patents outweighing the costs at any discount rate under (I think) 5%. But no one in his right mind would use a rate this low. Why on earth three B school profs would is beyond me.
Posted by: Bill Stepp at Jun 15, 2007 1:17:02 PM
Italy's thriving drug industry, like India's, is largely in manufacturing techniques, not research in new therapies. That is very useful, but not what we are talking about.
Posted by: Sebastian Holsclaw at Jun 15, 2007 2:01:42 PM
Indeed, I don't think anyone would seriously claim that drug discovery in the last 20 years has ever been equalled by any non-patent regime. Pharmacology has undergone a revolution and it was almost certainly caused by drug patents.
Posted by: Boonton at Jun 15, 2007 2:27:28 PM
MD Fatwa suggests that had Lincoln compensated slave owners the Civil War would have been averted. There is much reason to be skeptical. In fact, compensation for emancipation was often discussed prior to the outbreak of the War. A compensation provision was even included in the penultimate draft of the Emancipation Proclamation included a compensation provision for slave-owners who did not join the rebellion. After the South took up arms, however, compensation was off the table.
Posted by: dancole at Jun 15, 2007 2:28:05 PM
So if the gubmint awards a prize for a new drug, does that IP then theoretically belong to the gubmint?
Wonder what happens to those myriad class action suits then...
Posted by: fustercluck at Jun 15, 2007 3:07:43 PM
Abolitionists would have argued, rightly, why should non-slave owners have to pay slave owners to free their slaves? Doesn't that imply slavery was moral and the property rights claims of slave owners were true? If that was the case then why shouldn't slavery be permitted?
The south defined itself by slavery. At the founding it was predicted that slavery was economically on its way out but instead it surged and slave owners insisted on having it spread even to free states that didn't want it (see Dred Scott). It's only speculation to think slavery couldn't have been adapted to the industrial revolution that happened after the Civil War.
Posted by: Boonton at Jun 15, 2007 3:12:54 PM
Lincoln's dictatorial view that the South didn't have a right to secede
It's almost as if he thought the people in the South were his slaves.
Posted by: Barbar at Jun 15, 2007 4:30:21 PM
The current patent regime encourages extremely high health care costs which represents the most pressing fiscal crisis facing the U.S. It also encourages money wasted on advertising and copycat research by company X wanting to develop the same drug (or a different drug with the same effect) as company Y.
The solution, whatever it may be, needs to perform the task of redistributing the rewards of research from the pharma companies to the consmers.
I would propose:
A: Cutting the patent period from 12 years to 5, cutting the incentive for copycat research, since companies would only be able to profit from the fruits of said research for 2 years. Once competition to produce the same drug is eliminated, advertising expenditures should fall drastically.
B. Requiring all companies to put their products in the public domain after the 5 year period
C. If this patent period produces excess profits for the pharma companies, tax it and put the money in a health care subsidy fund. If it produces insufficient profits (i.e. incentives) subsidize it from an independent health care fund. Similarly, if a 5 year patent does not sufficiently reduce the incentives for copycat drugs, cut it to 4 years and (if necessary) increase subsidies.
The trick is to redirect competition into a race to produce better drugs, not a race to market the same drug, while at the same time minimizing rent seeking on the part of the government.
Posted by: BC at Jun 15, 2007 4:30:45 PM
