Patent Cure-All?

A patent cure-all?

CLINICAL depression is a depressingly common ailment. In America alone, there are around 11m sufferers. Yet although medical experts reckon that 80-90% of them could be helped, particularly if they were to take drugs such as Prozac, only a fraction of them seek treatment. One reason for this is that the price of such drugs is high: a month' s supply of Prozac costs over $60.

Many new drugs are expensive because they are patented. By awarding a patent, the government gives the company that invented the drug a property right over its product. This means that the drug cannot be copied by rival firms, unless the owner of the patent chooses to license it to them.

The rationale for patenting is simple: in effect, it grants patent- holders a monopoly and allows them to reap higher profits than they would in a competitive market. The prospect of such fat profits is what encourages firms to innovate in the first place. In the drug industry, for instance, patents make it worthwhile for companies to invest the millions of dollars needed to develop and test new treatments.

But patents have a cost to society. To see why, consider the chart. The marginal cost of producing drugs after they have been developed is usually low. Were the market truly competitive, therefore, the price of most drugs would also be low (line Pc on the chart). The amount of drugs sold in this market would then be Qc. The entire area above the marginal-cost line and under the demand curve up to Qc represents the value that consumers get from the drug (the ''consumer surplus' ' in economists' jargon).

However, when a patent is awarded to it, a drug company will maximise its profits by charging a higher price, Pm. Its aim is to maximise the size of rectangle A on the chart, which represents the extra profit that it reaps from being a monopolist. As a result, the consumer surplus is now the much smaller area B. Area C goes neither to consumers nor to the company: it is simply wasted.

The challenge is to come up with a means of encouraging innovation that avoids this. One solution is for governments to subsidise research, and then let the market decide the appropriate price for a drug by distributing the formula widely. In America, around half of health- care research is publicly funded. The problem with this is that governments are notoriously bad at identifying what research is worth paying for, making public subsidies an even more inefficient way of fostering innovation.

Is there a better way? In an intriguing new paper*, Michael Kremer, an economist at the Massachusetts Institute of Technology, suggests that there is. He proposes a half-way house between patents and public subsidies that would help society to capture more of the social value of drugs, but without the problems inherent in public subsidies. His idea is that governments should offer to buy patents from the drug companies that hold them and then make them publicly available so that anyone can produce the drugs in question.

But what should the government pay for a patent? To answer this, Mr Kremer proposes that it hold an auction at which rival drug companies would be invited to bid for a patented drug that another firm had invented. Such a competitive auction should establish a price for a patent that would be equivalent to the ''monopoly profits'' that firms would get from owning the patented drug.

The government would then offer to buy the patent at a premium to this auction value. This premium would reflect the extra value that society gets from the drug being available cheaply and it would encourage more private research into new drugs. Consider the chart again: the auction price would show how big area A is. The perfect premium would add in B and C.

In theory, such a scheme combines the best of both patents and state subsidies. Like the direct funding of research, it avoids the inefficiencies of monopoly pricing and removes incentives for companies to duplicate research. And like the existing patent procedure, it means that the direction of research is still determined by the private sector.

Hardly a panacea

There are, however, several pitfalls with the idea. First, drug firms might collude with one another in auctions, deliberately bidding up prices so that the government ends up paying more than patents are really worth. In order to encourage drug companies to put in bids that accurately reflect a patent's value to them, Mr Kremer suggests that once in a while the government should not buy the patent itself, but instead allow it to go to the highest private bidder. Alternatively, it could offer to buy the patent at a premium to the third- or fifth- highest bid, rather than the highest one. This would mean that the patent-holding firm would have to bribe more than one other firm to push bids up.

A second tricky area is defining the scope of a patent. A clever firm might, for example, invent a drug that was only effective if used together with another one, which has a separate patent. For Mr Kremer' s system to work, a government would have to buy both patents.

There is also a danger that if the premium that a government pays for a patent over the auction price is greater than areas B plus C in the chart, it will end up subsidising drugs by more than their social value. In that case, the gain from the new system might not outweigh the costs involved in raising the extra tax revenues to pay for the patents. Although it is difficult to estimate just how much such a scheme would cost, for some drugs the likely bill would run into billions of dollars.

If nothing else, Mr Kremer's proposal should stimulate more research into the issues surrounding the economic costs and benefits of patents in industries in which they are widely used as a means of encouraging innovation, but where the result forces up the prices of the product and so lowers consumption. More than most other industries, the drug business fits this description.

* ''A Mechanism for Encouraging Innovation'', by Michael Kremer. HIID discussion paper no. 533, May 1996.

A patent cure-all? (alternative to pharmaceutical patents)(Finance and Economics: Economic Focus)(Column)., Vol. 339, The Economist, 06-15-1996, pp 75(1).