December 2004

The Sickness of the U.S. Vaccine Market

Richard Manning
Pfizer

I wrote two papers about ten years ago. The first studied the childhood vaccine market, in particular the DT and DPT vaccines. Both protect against diphtheria and tetanus, but the DPT vaccine has an additional protection against the pertussis virus. The prices of both vaccines move, essentially, in tandem until the early 1980s, when litigation began that focused largely on the pertussis component of the DPT vaccine. A simple linear regression of DPT price on four explanatory variables shows that the liability component is by far the strongest and most consistent explanatory factor in the increase in DPT price. The price skyrockets by thousands of percent while the number of manufacturers falls from about five to two. Also, we attempt to roughly calculate how well the tort system compensates injured people. Using the change in price due to liability, we factor in congressional testimony on frequency of injury and average damage award, finding a five to eight dollar implied cost per dollar of compensation. The increase in price for the vaccine due to liability represents a kind of insurance premium consumers are forced to pay. In the second paper, we show that tying this insurance to vaccines decreases demand for them and is, therefore, inefficient.

Christopher Snyder
George Washington University

I am interested in the differential profitability of vaccines versus drugs. There seem to be three groups of factors that may characterize this relationship: policy factors, such as how the government procures vaccines; inherent factors, such as technical issues of production; and lastly, wedge factors. These wedge factors are the ones that separate social benefits from a firm's private incentives. For instance, consumer myopia affects the decision to consume vaccines versus drugs because the vaccine consumer does not already have the disease. Also, there is a free-rider problem because vaccine consumption reduces transmission rates, a public benefit to private consumption. Some let others get vaccinated, reducing the costs of potential side effects but reaping the benefits of decreased transmission. Drug consumption, however, almost exclusively benefits only those who consume them.

There is also a durable good problem. Because vaccinations do not need to be consumed repeatedly, there is very little opportunity for firms to discriminate on price by selling to high demanders first and then lowering the price to others. Another distinct problem is the heterogeneity of consumers. A simple example here is a population in which a large portion has small chance of contracting the disease and a small portion has a large chance of contracting the disease. In this case it is very difficult for a firm to get the pricing such that, all other things being equal, a vaccine is at least as profitable as a drug. Michael Kremer and I use HIV/AIDS data to show that a profit-maximizing price for a vaccine would only sell to the very highest-risk consumers. We also look at other diseases for which there is no vaccine and find a similar result. However, you would expect the opposite result with the flu vaccine. It has to be consumed every year, which mitigates the durable good problem, and the distribution of risk is much more homogeneous. Its shortage can probably be explained by other factors.

Scott Gottlieb
AEI

I would like to give some top-line thoughts first. There is not much venture capital money flowing into vaccine development either in biotech companies or in big pharmaceutical companies. One reason is that the specialized help and facilities necessary for vaccine production can also be used in the higher-margin business of biologics. Also, innovation can be discouraged by a hesitant Federal Drug Administration; for example, FluMist was not approved for use by the age group for whom it might be most attractive, children aged one through five.

On the recent flu vaccine shortage, I think the British overreacted. Rather than getting rid of all of it, perhaps the vaccines could have been re-filtered or better investigation done on the point of contamination. Still, why was the market dependent on only two manufacturers? The cost of vaccine production is very high. Because these vaccines are made in chicken eggs, manufacturers have to maintain their own chicken farm, which means start-up costs of $300 to $400 million. Also, it means they have to sell tens of thousands of units in order to recover their costs. Newer technologies such as cell-based manufacturing would reduce fixed costs and would, therefore, reduce the amount a manufacturer would have to sell to cover costs. This technology would also reduce production time by a month or two reducing costs incurred by setting contracts so far in advance.

Cell-based manufacturing would also make it possible to develop vaccines to particularly virulent strains of the flu, strains which cannot be developed in chicken eggs because they kill the embryo. The worry with cell-based manufacturing is that some of the genetic material from the embryo will transfer to the vaccine and then to the patient causing damage. The FDA is overly cautious after Vioxx and has not approved this technology. A better use of government funding might be to study this production technique rather than spend money trying to develop new vaccines. Finally, a large marketplace disruption comes from the government's monopsony purchasing power. Because it wants a uniform product, there is little or no differentiation in vaccines; companies cannot charge more for a specialized form of the vaccine. Instead, they compete by lowering costs.

John E. Calfee
AEI

I think what we need is a "normal" market for vaccines. There are certain elements that you would expect to find: market research followed by development, manufacturing, regulatory oversight, and then marketing and selling this product in an environment which responds to fluctuations in demand with reasonable limits on liability awards. Some vaccines coming out now, such as ones for the human papillomavirus, meningitis, and rotavirus, see fairly normal market conditions, but the market for childhood vaccines as well as for the flu vaccine is not a normal one. Inefficiencies include, for example, the government as a dominant payer, essentially setting prices due to their market power. Weak intellectual property rights discourage competition. Unpredictable swings in demand make manufacturing difficult, especially given the technology required to produce flu vaccines. I will echo Scott in making the distinction between the government funding research which replicates what the market could do better and more fundamental research. Also, tort reform would help protect manufacturers from the leakage recently observed in the Vaccine Compensation Fund. Finally, the single greatest fix in the market would be increasing the freedom and dynamics of pricing. Subsidizing consumer purchases instead of bulk purchasing would greatly improve price adjustment to market incentives.

AEI research assistant Michael Petrino prepared this summary.