Saving a life usually grabs more attention than preventing a death. Perhaps that is why the achievements of drugs—antibiotics, anticholesterol treatments, and countless others—have overshadowed those of simple vaccines. Yet the vaccines for polio and smallpox have been two of the greatest public-health triumphs of the past 100 years. From a developing country's point of view, immunization still promises the most cost-effective public-health strategy for maladies ranging from measles to polio to tetanus. It will be a long time before the world's most devastating infectious diseases are eradicated, as smallpox has been, thus making vaccination against them unnecessary. In the meantime, preventing their occurrence through vaccination is vastly cheaper than caring for their victims, but the economics of drug development often don't line up with the economics of health care provision in poor nations.

Sometimes, the result is that no vaccine gets developed. Two million people, mostly in the developing world, die of malaria annually, but only in the past five years has there been a concerted global effort to develop a vaccine against it. In other cases, vaccines do exist but are too expensive for the markets in question. For example, though a vaccine for the pneumococcal disease—a bacterial form of pneumonia that claims more lives than malaria—actually does exist, the $200 cost of a four-dose treatment is beyond the reach of most people in the developing world, where total annual health care expenditures come to less than $10 a head. For this reason, vaccines for diseases prevalent in these countries are a risky, and often a loss-making, proposition.

The need for subsidies from private donors and governments alike will continue to grow as the price of vaccines goes on rising

Manufacturers may well be able to justify to their shareholders low returns on some vaccine projects for developing countries. But if a project risks substantial losses or, by diverting finite resources, prevents a company from making and marketing more profitable products, it is likely to shy away.¹ Hence the need for subsidies from private donors and governments—a need that will only grow as the price of vaccines goes on rising around the world.

These and other kinds of entities recently committed well over $1 billion to distributing existing vaccines to the world's poorest people and to hastening the development of the most needed vaccines. Coordinating this initiative is the Global Alliance for Vaccines and Immunization (GAVI), composed of governments, international agencies, private foundations, and drug companies.² GAVI is assessing how alternative finance mechanisms could best allow the governments of developing nations to promote the development, production, and distribution of vaccines by paying for them in a different way (Exhibit 1), not just by paying more.

McKinsey has helped GAVI to mine one seam of opportunity. In certain areas, funding entities such as individual governments or international organizations can help cut the financial risks borne by the developers and marketers of vaccines, and at a lower cost than the company itself would have to assume. Sometimes, these entities would advance funding; in other cases, they would have to commit themselves only to purchasing a vaccine once it was produced. As a result, drug companies would be more willing to take on what had previously been commercially unpromising vaccine projects and might even be able to reduce prices per unit. The shifting of costs to the party that is in the best position to bear them would leave a greater sum of value to be divided between buyer and seller. In the language of investment banking, this would be an "arbitrage opportunity."

A more than risky business

At present, governments in the developing world, as well as international agencies such as the United Nations Children's Fund (UNICEF), purchase vaccines much as private citizens might buy over-the-counter drugs. If Chiron, GlaxoSmithKline, or Merck should develop a particular vaccine, governments and nongovernmental organizations may buy large quantities of it or none at all. The vaccine maker bears all of the up-front risks, while governments and public-health organizations make no commitments until the moment of purchase.

The costs of developing and testing vaccines that never reach the market must be factored in to the price of all vaccines that do`

These risks often account for a substantial portion of the vaccine company's costs. The risk that a proposed vaccine won't prove itself in clinical trials, for instance, is substantial; only about two of every nine vaccines awaiting clinical trials actually become marketable products. The costs of the trials of the other seven vaccines—not to mention their research costs up to that point—must be factored in to the price of the two successful ones. In this way, the risk of clinical trials adds to a vaccine's price. Other such risks exist as well.

At present, few of the risks are shouldered by national governments and international agencies. Yet the capabilities of these entities, and the public-health interests they are supposed to serve, make it both appropriate and economical for them to assume some part of a manufacturer's risk in marketing a vaccine and, when the right conditions are present, in developing it as well. These are the arbitrage opportunities that McKinsey and the World Bank are evaluating for GAVI.

Delivering a meningococcal vaccine for Africa

The cost of the different risks varies markedly from one vaccine to the next. A vaccine for meningococcal meningitis, for example, exists today, but it protects only against a strain found in the Western world, not the strain that has caused epidemics in a belt running from Senegal to Ethiopia. The disease, fatal in 5 to 10 percent of cases, can cause permanent disabilities such as deafness and seizures. Since 1996, there have been over 300,000 cases, mainly affecting young children.

Unfortunately, an analysis of costs suggests that the expense of developing and testing a vaccine for the African strain would make the resulting product too pricey for its intended market. Clearly, such a vaccine would be a losing commercial proposition for vaccine makers. But this analysis assumes that the manufacturer bears virtually all of the risk, which, as Exhibit 2 shows, accounts for a substantial share of the cost. At the top of the exhibit are the costs that a company would incur in providing an approved mass-produced drug to a known market. Because these costs don't include any of the risk, they account for only 50 percent of the final price in the developing world.³ If this risk were assumed by governments or international health agencies—at little or no additional cost to them—their aid dollars would go twice as far.

Making demand more certain

One of the biggest risks facing vaccine manufacturers is uncertain demand; even after a vaccine has been discovered, refined, tested, and produced, they still don't know if anyone will ever buy it. Demand risk is the gap between the ordinary market price of a vaccine and the price if demand were known at all times and production were set accordingly to meet it. From 14 to 18 percent of the market price of a meningococcal vaccine, for example, would be attributable to demand risk. To some extent, of course, the number of doses sold is a function of price, which reflects not only the producer's costs but also factors such as competitive products and changes in manufacturing capacity. But in the case of health care in developing countries, significant uncertainties persist at almost any price—indeed, even when a vaccine is offered for free.

The almost universally endorsed hepatitis B vaccine, for instance, is available at very low cost, yet take-up rates in developing countries vary enormously. While 69 percent of the countries of the Middle East have offered immunization to their people, just 39 percent of the former Communist countries of Europe have done so, as have only 12 percent of the countries in sub-Saharan Africa. For a variety of reasons, some governments are choosing not to include the vaccine in their national health programs, but even governments that want to inoculate their populations often find this difficult. In some cases, the resources of the local health care systems are limited; in others, the delivery systems can't handle perishable medical supplies.

If vaccine buyers were to oblige themselves to purchase a certain number of units of a vaccine, the demand risk of the company that produced it would plummet. At present, manufacturers know that donor countries and not-for-profit organizations will continue to buy vaccines but have little idea which vaccines they will buy, and in what quantities. The need is not for extra expenditures but for a commitment of funds in advance to particular vaccine programs. This approach not only would increase the odds that drug companies will invest in developing needed vaccines but would also lower their ultimate cost; for one thing, manufacturing facilities built specifically for particular vaccines would be much less likely to lie idle. Such arrangements—though unusual or impossible when the market comprises thousands of individual decision makers—is the norm when the market consists of a handful of governments or perhaps even a single all-purpose agency.

Some national governments, particularly impoverished ones, have trouble committing themselves to purchasing products that don't yet exist. International agencies, though, can do more. GAVI, through its Vaccine Fund, has committed itself to purchasing upward of 300 million doses of various existing vaccines until 2004. Besides securing a lower price, GAVI is signaling its intention to purchase vaccine makers' future products when they are developed. But firmer commitments will probably be needed. To spur the development of a meningococcal vaccine, it may be necessary to reverse the effects of years of market uncertainty by offering a demand guarantee before the product is even developed.

Another aspect of demand risk is the possibility that vaccines, once purchased, won't actually reach the people for whom they were intended. Education, advocacy programs, and stronger distribution systems can reduce this kind of risk, as well as increase demand, thus making investment more attractive to vaccine makers. Governments can help reassure health care workers and the general population about the value of vaccines and educate people in their use. The educational arms of international agencies can support these efforts more credibly and effectively than can the manufacturers. GAVI's partners and the Vaccine Fund are also working with governments to assure the soundness of their distribution systems before the arrival of a vaccine.

The R&D lottery

Another risk that funding agencies might target is research and development, which in the case of our hypothetical meningococcal vaccine represents about a fifth of total costs. Here, governments have a cost advantage, especially in the early stages of research. To a private company, early-stage R&D for vaccines may be an unattractive proposition, since there is a high risk that it won't be scientifically fruitful or that the results will be too basic to be patentable. Indeed, that is why much basic research relies on government funding. At this stage, discoveries—even if they are potentially valuable—often enter the public domain. Individual vaccine makers can soak up the results of such research and then go ahead to develop the final vaccine, for which they themselves will hold the patent. Governments and public-health agencies don't have to guess which vaccine maker will ultimately come up with a viable and marketable product. To them, the important thing is getting the vaccine they need regardless of which company happens to produce it. Consequently, they are the logical parties to underwrite research in the early stages.

Even in the later stages of R&D, the risk is still high that no commercial application will result. Acknowledging this reality, the International AIDS Vaccine Initiative (IAVI), along with the private sector, has invested in the development of an AIDS vaccine. But a company that uses whatever intellectual property results from these ventures must agree to supply the resulting vaccine to developing countries at an affordable, "costs-plus" price. If the company should fail to do so, control of the intellectual property reverts to IAVI, which can then license another manufacturer to use that same intellectual property for vaccines that it offers in all markets, not just in developing countries.

Another kind of risk is associated with trials of efficacy—the last stage in testing a proposed vaccine before health authorities decide whether it may be marketed. Failure at this stage imposes a heavy cost on vaccine makers because it puts at risk all of the R&D that has been completed and paid for until then. As regulations become more stringent and the public grows less tolerant not only of harmful but also of merely uncomfortable side effects, the risk represented by these trials has risen.

Governments, however, are in a privileged position to deal with the mounting costs of such a trial: through their public-health programs and relationships, they can recruit participants more easily than can vaccine companies new to a country, and the trial can be monitored by their existing networks of health workers and hospitals. Vaccine makers on their own would have to replicate all of this for the sake of a single clinical trial.

Such interventions are most useful when an existing vaccine needs merely to be approved for use in a less developed country. In this case, trials of efficacy are the single largest part of the incremental cost of getting the vaccine introduced there. By helping to mount them, a government can markedly reduce the price it eventually pays for each dose of a vaccine.

Other risks, other solutions

The opportunity costs of many vaccines are high because the resources used to create them could be spent on the next Viagra

Even if public-private collaborations overcame the risks posed by demand and R&D, the risk that a new vaccine might not be manufactured would remain. A vaccine's direct production costs, typically less than 10 percent of the final price, aren't large. But vaccine manufacturers think that the opportunity cost is much higher, since their capital and talent could be dedicated to products with higher returns—perhaps a vaccine for the developed world or, more likely, the next Prozac or Viagra. In this case, it is hard to argue that buyers are better suited to bear production costs than are sellers. Yet as a practical matter, if buyers don't do so, existing vaccines may not reach those who need them. A public-sector grant or low-cost loan, for instance, might be needed to coax highly profitable drug companies to produce a vaccine for yellow fever, which is almost nonexistent in industrialized countries.⁴

Some countries, too, face the risk of product-liability litigation, which may well ensue from adverse reactions to vaccines. This risk can be significant—it adds up to 10 percent to the price of a pneumococcal or DTP⁵ vaccine, for example. In 1985, makers of the DTP vaccine were facing $4 billion in claims filed in US courts. In view of the possibility that the production of the vaccine might be halted, during the following year the government created its own insurance scheme—a low vaccine-sales tax whose proceeds go into a fund to compensate people who suffer adverse reactions. The legislation caps compensation and limits the right to file suit. Many European countries have gone even further in this direction.

One risk for pharmaceuticals companies is more illusory than real: the sale in rich areas of cheap vaccines intended for poor areas. Pharmaceuticals are indeed sold in this way, but, unlike them, vaccines are biological products that must be kept under constant refrigeration, which can be assured only by channels that are controlled directly by manufacturers and governments. The widening gap between the price of the vaccine for Hib⁶ in Latin America and in the United States since the vaccine's introduction, in 1994, shows that the problem of gray-market sales isn't real. Moreover, in developed countries that have both public and private health care systems, global pharmaceuticals companies continue to charge widely divergent prices for the very same vaccines, much as they do in the developing world: in Brazil and India, for example, these companies sell the hepatitis B vaccine to the public health sector for far less than the amount that they can charge the private sector (Exhibit 3).

Arbitraging risk is just one mechanism for getting more vaccines more quickly to more of those who need them. Governments and health care organizations are quite often better suited to the task of reducing the risk-associated costs of vaccines than are their producers. But identifying the risks that actually matter and the best way of containing them is going to require tight collaboration among governments, manufacturers, and international bodies dealing with health care questions. Such collaborations will eventually promote a keener understanding of the economic issues and market failures that continue to prevent vaccines from reaching the people of the developing world.