Why does competition among brand drugs often fail to reduce prices? In this post, I address that question, which is obviously important in structuring efforts to reduce prescription drug costs. Based on the answer, I then propose a mechanism to increase competition among brand drugs and lower prices. Specifically, I advocate the creation of nonprofit entities to purchase the intellectual property rights to any one of several competing brand drugs and license those rights to multiple generic manufacturers.
Understanding Prescription Drug Competition
In model systems, the profit-maximizing price for a competing monopolist depends on consumer preferences and the prices of competing drugs, and this may result in an equilibrium in which competition causes prices to rise. The key idea is that with competition, manufacturers tend to lose customers who value their drug less than the average, and they maximize profits with a price near what their remaining customers are willing to pay.
This situation contrasts with the usual effect of generic competition where competing drugs are exact equivalents rather than close substitutes, and prices typically fall substantially as more generic manufacturers enter the market. Shortening the period of high prices during brand-only competition would benefit consumers in the short run.
The Value Of Pharmaceutical Intellectual Property Depends On How Many Producers Sell The Same Drug
A way to enhance competition among brand drugs is suggested by considering the value of intellectual property (IP) protections to a brand manufacturer facing competition from a near-substitute drug sold by a monopolist versus a brand manufacturer facing competition from multiple generic manufacturers. If the competitor is a monopolist, the price of the substitute is high and IP rights to the original drug remain valuable; but if the competitor drug is licensed to multiple manufacturers, its price falls substantially, which takes market share from the original brand, decreasing the value of its IP rights. Thus, a credible threat that a near substitute drug might be licensed to multiple manufacturers could significantly decrease the value of IP protections.
A Path Forward: Non-Profit Entities To Purchase And License Brand Drug IP Rights
Suppose a non-profit entity or entities were formed with the announced intention of buying IP rights to any one of a group of close substitute, brand drugs. Each monopolist would face pressure to sell its IP rights at a reasonable mark-up over the value of its IP given generic competition, since if it did not sell its IP but one of its competitors did, it would suffer a bigger loss.
The effect of such a non-profit company would be to enhance competition by converting brand-brand competition to brand-generic competition. This would promote the economic interests of consumers, generic manufacturers, and insurers, who pay the bulk of the cost of prescription drugs. Thus, generic manufacturers and insurers would be reasonable candidates to fund such an endeavor.
The Nature Of The New Entities: Nonprofit And Tax Exempt
A for-profit company would have no incentive to license IP rights to multiple manufacturers since it could make more money by selling a monopoly drug. The project therefore requires a non-profit. Could such a non-profit qualify for tax-exempt status? Under IRS code 501(c)(6), an organization that promotes economic interests of a defined group, such as a trade association for insurers or generic manufacturers, qualifies as tax exempt so long as it doesn’t perform specific services for particular members. Thus, it might be important for the non-profit to be structured in a way that avoided promoting narrow business interests of particular funders.
Previous work has described potential advantages of government buying patents and licensing them broadly, but this entails much greater government intervention and interference with private enterprise than the idea proposed here: focused, non-profit, non-governmental, patent buyouts. The same strategy, however, could be used by governments in countries where government plays a more active role in pharmaceutical markets.
What Drugs Would Be Appropriate Targets?
A first requirement is that the drugs be easily produced by generic manufacturers, i.e. small molecule drugs rather than “biologics.” Drugs for which insurance companies have little bargaining power, resulting in higher margins, would be better candidates than drugs for which insurers can already effectively bargain to lower prices, for instance by threatening to exclude a drug from a formulary or put it on an unfavorable co-pay or co-insurance “tier.” Examples of drugs for which insurers have little bargaining power are those in six “protected classes” for which Medicare Part D requires all drugs be covered: drugs for immune-suppression, cancer, retrovirus infection, depression, seizures, and psychosis.
Market features that impact the price that IP rights would likely command would also be important in determining which drugs the new nonprofit entities should target. For example, a manufacturer with a large share of a market would face larger losses from generic competition, and hence would be willing to spend more to buy the IP rights of a competitor to prevent their licensing to generic manufacturers; thus, markets in which multiple companies have comparable market share would likely yield lower IP prices for the sort of non-profit entity proposed here. IP prices would also be lower for drugs for which a company is not planning new drug development in the same class, since it would not be in a brand company’s interest to sell rights to a drug that could compete as a generic with brand drugs it planned to develop in the future. The purchase of IP rights to drugs for which all the close competitors have long remaining patent lifetimes could yield potentially large consumer savings, but such drugs would likely command correspondingly high IP purchase prices.
How Would Buyouts Affect Innovation Incentives?
An effective buyout and licensing program would decrease incentives to develop “me-too” drugs, which provide less consumer value than first-in-class drugs but contribute a great deal to insurance costs. But since first-in-class drugs would also lose value due to generic competition from close substitutes, this proposal would reduce innovation incentives generally.
However, innovation incentives may be excessive from a consumer welfare perspective given the increased demand that comes from widespread insurance. Moreover, because biologic drugs are difficult for generic manufacturers to reproduce cheaply, the buyout scheme would further incentivize development of biologics as compared to small molecule drugs. The potential of this proposal for negative as well as positive effects on consumer welfare from altered innovation incentives deserves further consideration.
How Much Funding From Insurers Or Others Would Be Required?
This depends on many factors such as the number of competing brand manufacturers, market share of the generics after sale, and price of generics compared to surviving brands. As an example, suppose three brand manufacturers produce close substitute drugs, each selling for price p and comprising one-third of a market whose total net present value is M. Suppose further that if one company sells its patent rights to a non-profit which licenses the rights to multiple generic manufacturers, the price of the generic drug is 0.2p and the generic takes 80 percent market share; the remaining brand manufacturers (with 10 percent of the market each) raise their price 50 percent because the customers they retain are those with a higher willingness-to-pay for their drugs. The net present value of the business of each surviving brand manufacturer drops from M/3 to roughly 0.1*1.5*M = 0.15M, so this figure—0.15M—is an estimate of the minimum price a brand manufacturer would accept for its IP.
How much would insurers save? Twenty percent of the market would pay 1.5*p per unit of drug and 80 percent would pay 0.2p, so the average price would be 0.2*1.5*p + 0.8*0.2*p = 0.46*p. If insurers paid 75 percent of drug costs (25 percent co-insurance rate), they would now pay 0.75*0.46*p = 0.35*p, compared to 0.75*p before generic competition. This would yield a saving of 0.4*p per unit of drug. For simplicity, suppose the net present value of their savings over the life of the drug is 0.4*M; this is an estimate for the maximum they would be willing to pay for the IP rights, and it is comfortably larger than the estimated minimum a manufacturer might demand.
Now suppose that only a fraction Y of insurers agree to fund the non-profit venture. The saving to this group (Y*0.4*M) must equal or exceed the minimum price required by a brand manufacturer (0.15*M) for a sale to be feasible; this implies that Y > 0.15/0.4, or about 37 percent of insurers (or insurers covering about 37 percent of consumers), would be required. If there were five competing near substitute drugs (four surviving brand manufacturers), with the same assumptions only about 19 percent of insurers would be required.
While the above numbers are purely hypothetical, the main point is that the magnitude of potential savings makes the scheme feasible even if only a fraction of insurers could be enlisted to back the project.
What Could Government Do To Foster This Type Of Competition?
The proposed strategy could be thwarted if pharmaceutical companies pooled resources to buy brand drug IP that otherwise might be licensed to generic manufacturers. In cases where multiple, near-substitute brand drugs compete, the savings to consumers from generic licensing could easily exceed the loss to any single brand manufacturer, so a purchase price for the IP above the price that any single manufacturer was willing to pay could be feasible.
However, if the combined loss to multiple brand manufacturers exceeded potential savings to groups funding the non-profit, brand manufacturers might find it rational to pool resources to outbid non-profits. This would raise anti-trust issues. From a policy perspective, the Justice Department could clarify if it would oppose such combinations on anti-trust grounds, and Congress could consider legislation to prohibit such activity. As both Republicans and Democrats advocate enhanced competition to bring down drug prices, this type of governmental intervention could have bipartisan support.