Having people pay for most medicines themselves would alleviate many of the abuses that pharma companies commit. The average wholesale price (AWP) scandal discussed in Chapter 3, in which drug companies posted phony list prices that enabled them to take taxpayers to the cleaners, would never have happened. Neither Medicare, nor Medicaid, nor any private carrier would have relied on the phony numbers published in a manual when deciding how much to pay. Prescription drugs would have sold at real retail prices, just like pain relievers, cold remedies, and other nonprescription items. Drug prices would have a natural limit too: consumers’ willingness to pay. Manufacturers couldn’t charge whatever they want.
A switch from third-party payment to first-party payment would not be a cure-all, though. Pharma companies could still charge monopoly prices for branded drugs under patent because they would still enjoy marketing exclusivity. And even after a drug has gone generic, there are pricing problems that should be dealt with. Any serious plan to control spending on prescription drugs must address both of these problems. We address each in turn.
The first problem requires patent reform, which requires an act of Congress. That makes it exceedingly unlikely to happen. The pharma sector is adept at blocking any proposed legislation that would take money out of its pockets. Consider the fate met by recent efforts to legislate alternatives to patent monopolies. In 2005, Sen. Bernie Sanders filed the Medical Innovation Prize Act. It embodied an idea that has been floating around in academic circles for decades: using prizes to stimulate research and development of new drugs. The bill went nowhere. The similar bills that Senator Sanders introduced every two years thereafter also died. And, when Congress did enact a patent reform law in 2011, it made no provision for prizes. It did not even include a pilot program.
The idea was dead on arrival in Congress because the pharma sector has too much at stake. Citing a white paper produced by the Center for Economic and Policy Research, the Huffington Post reported that “Americans spent $250 billion on prescription drugs in 2009, but would have spent just $25 billion had all drugs been sold as generics, meaning without patent protections.”1 The difference, $225 billion, was the revenue that the elimination of patents would have cost name-brand pharmaceutical companies in just one year. No wonder the U.S. Chamber of Commerce, the lobby of the corporate sector, also pooh-poohs the idea of eliminating patent monopolies.2 It is overwhelmingly likely that existing arrangements will remain in place, in which event “the excess cost of prescription drugs attributable to patent protection [will] likely . . . exceed 2 percent of GDP before the end of the next decade.”3
A well-designed prize regime would eliminate the need for drug monopolies, thereby reducing prescription drug prices to competitive levels. It would also create incentives for innovation, including incentives to develop orphan drugs for small populations, and could also be tailored to encourage drug companies to test the efficacy of old drugs. Finally, if funded openly by the federal government, a prize regime would place the costs of drug innovation on-budget, where they would be borne by all taxpayers—rather than just by the consumers who happen to need drugs (and their insurers).
Because Congress is unlikely to replace patents with prizes, we are reluctant to spend much time discussing how a prize system might be designed. The matter is a complicated one that would take many pages to address. A workable prize regime must answer (at least) the following questions:
These are the same questions that the patent system answers, albeit haphazardly and with many unwanted side effects.
To give readers a sense of both the advantages a prize system could offer and the difficulty of designing one, in the discussion below, we use a regime that ties rewards to innovators’ documented research costs. A prize system that was actually adopted would have to be more complicated than our example, and there might be a better system that would not tie prizes to costs at all.4
Before getting into the nuts and bolts, a couple of background points must be made. First, a reward system based on prizes will not work perfectly. Nothing does. A prize system just has to work better than the existing system of patent monopolies, and there are good reasons to think that one would.
Second, a prize system will require public funding and will therefore have to involve Congress. In theory, Congress could wreak as much havoc with a prize system as it has with the patent system. That seems unlikely, however. Under the current system, drug companies’ returns on research and development (R&D) come from diverse sources: the budgets for Medicare, Medicaid, the U.S. Department of Veterans Affairs, and other public programs, and from purchases of prescription drugs that are paid for by insurers or consumers. Under a prize system, those returns would all come from one place. Because the process of connecting those returns to innovation would be more transparent, legislators who sought to make the prize formula unduly favorable to the industry would have to expect negative press coverage. The issue of funding any pro-industry increase in prize amounts or other changes would also invite scrutiny.
Finally, a prize system would not do away with patents entirely. Pharmaceutical companies operate internationally, and other countries deal with them in different ways. It may therefore be important for them to obtain patents in the United States so as to be able to register those patents elsewhere or to protect their interests in other ways. Our proposal would let drug manufacturers do that. It would only deny them monopolies on domestic sales of new medications.
Normally, a prize system offers a cash payment for an invention that meets a detailed set of success criteria. The payment can be a lump sum or a variable amount tied to subsequent developments. In the case of a pharmaceutical, the amount paid could be tied to things like the number of units sold at a specified price postdevelopment or the documented increase in life expectancy after several years of use. Details like these would have to be worked out.
A different strategy for incentivizing drug makers would tie the size of the prize to actual, documented R&D costs (including conducting clinical trials) that are incurred. Companies hoping to obtain U.S. Food and Drug Administration (FDA) approval for new drugs would submit confidential periodic filings showing all of the drugs they are studying and the associated R&D costs. When FDA approval for a particular drug was granted, a company would apply for a prize and submit an accompanying final statement of its research costs for the new medication. This final statement would be audited and made available to the public. The company would then receive a check from the U.S. government, the size of which would be a predetermined multiple of the approved research costs.5 The same approach would govern pharma companies’ applications for approval of old drugs and orphan drugs.
This proposal has several important strengths. Transparency is one. Because final cost statements will be audited and available for public inspection, R&D outlays would be on the table for everyone to see. The prize multiplier would also have been set in advance through a public process. Everyone would know why the government cut the check that it did.
The prize regime would also create clear incentives. During the process of setting multipliers, administrators would receive input from a variety of sources, as they do now when regulating public utilities. The expected value of new drugs, the cost of capital, and other considerations that influence the economic rationality of risk taking would all be considered, and the industry would have its say. Once the multipliers were fixed, innovators would know what they stood to gain before investing in R&D. Incentives could also be retuned. If experience showed that certain types of drugs were needed but not being pursued, R&D could be encouraged by making multipliers larger.
Conservatives may object to our proposal on the ground that regulators are unlikely to size prizes correctly. We agree there is a risk of error that, given the well-known phenomenon of industry capture, would likely result in prizes that are too large. But there is a serious public goods problem at the heart of pharmaceutical innovation that must be handled in some way. The problem is that, unless they can gain by developing new products, pharma companies will lack incentives to invest in research. But garnering rewards would be difficult if competitors who bore no R&D costs were free to duplicate new inventions and sell them at lower prices than inventors would be able to charge.
Currently, the patent regime allows inventors to capture profit by using the power of the government to prevent competitors from duplicating their creations and undercutting their prices. (The Unapproved Drugs Initiative and the Orphan Drug Act work the same way. Both stop other manufacturers from continuing to sell.) This is a coercive and messy process. It spawns litigation over the validity of patents and their scope. It encourages the use of strategies like evergreening and pay-for-delay settlements. It encourages consumers to buy cheaper drugs abroad while also requiring the U.S. government to prevent them from doing so by policing imports. It imposes deadweight losses on society and only weakly links inventors’ returns to the risks and costs that were incurred. Finally, it enables manufacturers to take advantage of the payment system’s inability to impose meaningful limits on the amounts they can charge. A prize system would have none of these shortcomings.
Of course, a prize system would cost taxpayers money, because the dollars needed to fund the prizes would come from them. But, in light of the many problems existing arrangements generate, it seems better to use the tax system to fund prizes than to impose costs as the patent system does. Today, those costs fall on consumers, insured populations, and taxpayers, whose dollars buy drugs at inflated prices for millions of people, support research, and also fund drug-related tax preferences. This scattershot approach has many downsides. It creates problems for people with diseases that require expensive medications. It creates problems for insurers, who must figure out how to deal with these people while pricing insurance in ways that maintain stable risk pools. It creates problems for the public, which shares the cost of developing new drugs only to find itself exploited by drug companies when discoveries pan out. Finally, it creates problems for orphan populations that are too small to motivate drug companies to look for needed treatments. By comparison, taxing people to fund prizes seems simple, straightforward, and fair.
Of course, the big advantage of a prize system would be that, post-invention, prices for drugs would be set competitively. Once the FDA approved a new drug or determined that an untested old drug was effective, all companies would be free to make it. Competition should then force prices downward toward manufacturers’ production costs. This is the efficient, pro-consumer price that emerges naturally when markets are competitive.
The fundamental economic problem in the pharma sector is that it costs an enormous amount to create a safe, effective new drug but only pennies to manufacture the actual pills. (Pharmaceuticals are not unique in this regard; computer software and blockbuster movies from Hollywood share these characteristics.) If drug companies sold pills at their marginal cost, they’d never recoup the billions they spend on R&D. That’s not a viable business model, and it would leave patients with a suboptimal level of pharmaceutical innovation.
The existing patent monopoly model enables inventors to recoup their sunk R&D costs by pricing medicines well above the marginal cost of production. As we have seen, this approach has enormous downsides. Of particular interest is that it ties drug companies’ returns to their sales, not to the risks and costs they incur. It is the latter attributes that should matter in setting a market price. The combination of R&D costs and success odds define the gambles that pharma companies take when exploring new drugs. When the gambles are good ones, because the expected costs and the odds of succeeding are high, small prizes may provide all the inducement that is needed. By contrast, when the costs are high and the odds are long, the returns on success must be large. Under the existing patent regime, inventors will take these gambles only when they can charge exceedingly high prices or sell large volumes—or, better still, both.
A prize-based approach would provide compensation for sunk R&D costs directly and would remove sales revenues from the picture. An innovator wouldn’t necessarily have to sell a drug at all. It would only have to obtain FDA approval.
To create good incentives, the checks cut by the government will have to be large. According to the Pharmaceutical Research and Manufacturers of America (PhRMA), the trade association for the drug industry, its members spent $51.2 billion on R&D in 2014.6 Because, as other sources report, “95% of the experimental medicines that are studied in humans fail to be both effective and safe,” many of the molecules on which that money was spent won’t pan out.7 To enable drug companies to stay in business, a prize regime would have to reward them lavishly when they produce winners. It would have to treat them like trial lawyers who work on contingency and give them bonuses for success.
Exactly how much the government would have to spend cannot be determined with precision. One reason is that PhRMA’s numbers may exaggerate the cost of research. As Dr. David Belk has explained, drug companies sometimes treat costs associated with advertising and marketing as R&D.8 A recent study, which pulled numbers from drug companies’ public filings with the U.S. Securities and Exchange Commission, found that, for 10 cancer drugs that were approved by the FDA, the median cost was $648 million and the median revenue was $1.7 billion. The reported median cost was substantially lower than the $2.7 billion that is often said to be the outlay required to bring a new drug to market.9 An auditing process is needed to ensure the accuracy of the cost figures on which prizes would be based.
The process of estimating the cost of a prize system is also complicated by the existing pharma business model, which causes enormous variation in R&D costs per approved drug. According to Matthew Herper, who writes about medical economics at Forbes, some companies spend billions of dollars per new FDA-approved drug while others spend far less.10 To some degree, this variation reflects the connection between invention and expected sales. If sales were removed from the picture and all compensation for R&D costs came from prizes instead, the pharma sector would reorganize itself. Presumably, companies that perform R&D inefficiently would get out of the business or hire someone else to do it for them.
If the exact cost of a prize system cannot be estimated with precision, we can nonetheless be sure that the government will have to spend billions. But the cost of innovation should still be much lower than it is now. One reason is that existing patent monopolies let pharma companies capture 100 percent of the returns on new drugs even when they bear far less than 100 percent of the R&D–related risk. Taxpayers already bear a large fraction of the cost of drug-related R&D through research grants and other means. But taxpayers currently get little financial return on these outlays because the dollars generated by sales go to drug companies. A prize system would compensate drug companies for only the R&D costs they actually bear. Consumers would keep the rest of the gains.
Another reason is that a prize regime would enable the U.S. government to share the cost of R&D with governments elsewhere. This could be a boon to American consumers. Currently, the United States accounts for over 40 percent of pharma companies’ total revenues, even though it contains only 4–5 percent of the world’s population.11 American consumers subsidize drug innovation for the rest of the world, while the rest of the world does little or nothing in return. But, if sales monopolies were eliminated, prices here would fall and subsidization by that means would end. The U.S. government could then reduce the cost of prizes further by negotiating deals with other countries, which could help their own citizens by eliminating their sales monopolies and kick in a few bucks to help with the cost of prizes. Even if those other countries insisted on continuing to free ride, we would still be better off with a prize system than the current approach of combining patents with open-ended reimbursement at whatever price point the drug manufacturer decides to set.
Finally, a prize regime has already been tried in the United States with some success.12 In 2007, the federal government sought to incentivize drug makers to develop treatments for certain neglected diseases. It offered a prize in the form of a priority review voucher (PRV) that a successful drug maker could use to shorten the FDA approval time for any drug it had in the pipeline, including one that might be vastly more profitable. Winners could even sell their vouchers to other drug companies that wanted expedited review for their discoveries. United Therapeutics reportedly received $350 million from another drug maker for the PRV it won for developing a cancer treatment for young children.13 The bounty supplemented other incentives drug makers had to produce the desired medications.
Having touted prizes as substitutes for patents, we emphasize the limits of our approach. We do not recommend, and would not support, a policy of replacing patents with prizes across the board. This is partly because, in most sectors of the economy, prices are subject to a natural ceiling: consumers’ willingness to pay. The lack of a similar limit in the health care payment system makes the prospect of adopting a prize regime for new drugs unusually attractive. Competition also appears to be less robust in the pharmaceutical sector than in other areas. As shown in Chapters 1 and 2, drug makers seem to have figured out how to maintain high prices even when branded drugs are close substitutes. Elsewhere, competition by rivals who invent around patents is more likely to minimize the deadweight losses that patents entail. Finally, we should start small and sort out which of the many competing prize regime designs work best.
Every time we think we know which drug costs the most, we learn of a new one that costs even more. In 2016, we first learned about Soliris, a treatment for paroxysmal nocturnal hemoglobinuria (PNH) that received FDA approval in 2007. PNH is a rare blood disease that afflicts about 10,000 people in North America and Europe combined. Until Soliris came along, the only treatment for it was a bone marrow transplant, a procedure that wasn’t very effective. Soliris doesn’t cure PNH but does improve sufferers’ quality of life, so doctors prescribe it often.
Soliris is prohibitively expensive. It costs $440,000 a year, and PNH sufferers have to take it their entire lives. One patient, Janet “Bunny” Williams of Albuquerque, New Mexico, has already received more than 200 Soliris infusions at an estimated total cost of more than $13 million.14
Why does Soliris cost so much? Partly because it was expensive to develop and partly because manufacturers can charge whatever they want, knowing that insurers and public payers will pay up. Alexion Pharmaceuticals, the company that makes Soliris, reportedly spent nearly $1 billion over 15 years to develop the drug. That’s a lot of money, but the returns are far larger. Alexion “reported revenue of $2.6 billion for the drug in 2015 alone.” The effect of Soliris sales on Alexion’s market capitalization has been astounding. “The company’s stock has soared more than 300 percent in the past five years, boasting a faster rate of growth than Apple Inc. over the same period.” Alexion’s stock price rose 130 percent in 2010 alone.15
Whether Alexion bore the amount of risk that it contends is seriously disputed. According to Sachdev Sidhu, a University of Toronto researcher, “the public science is well over 80 or 90 percent of the work. . . . Public resources went into understanding the molecular basis of the disease, public resources went into the technology to make antibodies and finally, Alexion, to their credit, kind of picked up the pieces.”16 Sidhu added that Soliris is cheap to make too. “It probably costs less than 1 percent of the price of the drug to make the drug.”17
Soliris provides an example of how a cost-based prize system might work. During the 15-year period in which research on Soliris was underway, a condition of eligibility for a prize would have been that Alexion file periodic cost reports with the FDA, along with a final report when FDA approval was sought. The reports would have been audited by an established means. Upon approval being granted, Alexion would have received a prize based on its documented costs and a reward formula that was known in advance. Generic manufacturers would then have been free to make and sell Soliris. Competition would then have driven the price down toward the actual cost of production, and neither members of the orphan population of PNH sufferers nor their insurers would have been gouged. America’s taxpayers would have borne the full cost of the R&D, but they had already paid for much of that by supporting university research, and the remaining cost would be spread across all governments that chose to participate in the United States’ prize regime. The only losers would have been Alexion’s shareholders, who would have received normal stock market returns instead of spectacular ones.
A prize system cannot help reduce the prices of generic drugs, which are already off-patent. What can be done to make them less expensive?
Regrettably, the most effective options with which to address generic drug prices require governmental action. For example, many of the problems we identified in Chapter 1 are attributable to insufficient competition in the generic drug market. Substantial price hikes occur and stick because only one company makes a drug or because manufacturers adopt tit-for-tat strategies or play other pricing games. Illegal, anti-competitive conduct may underlie some of these problems. Only aggressive antitrust enforcement can discourage this type of behavior.
The FDA also bears part of the blame. There is a sizable backlog of pending applications from generic drug manufacturers who want to enter the market. Congress should give the FDA the resources it needs to process these applications more quickly. Until that happens, the FDA should reprioritize applications in a manner calculated to bring prices down. Currently, it follows a FIFO (first in, first out) plan. That approach doubtless seems fair to the pharmaceutical companies, but it results in significant consumer harm. A more consumer-friendly approach would move applications for drugs that have experienced significant price spikes to the head of the queue. Approval of new entrants should reduce prices by expanding the number of suppliers. And, once incumbents realize that price hikes will result in new market entry, they may be deterred from jacking up prices in the first instance. In short, the FDA (which after all is the Food and Drug Administration) should start paying attention to drug prices—and use that information to inform judgments about the impact of agency action and inaction.18
More broadly, it is important to recognize that, because the FDA controls market entry, its actions and decisions can greatly influence the prices at which generic drugs change hands. Policy makers should consider liberalizing access to the U.S. market by relaxing the agency’s grip on entry. Why not let a company that qualifies to sell a generic drug in Canada, England, France, Israel, or some other developed country automatically qualify to sell the same drug in the United States—at least so long as the branded drug has already been approved by the FDA, and the marketing exclusivity provided by the Hatch-Waxman Act has expired? These countries have the expertise needed to protect their citizens from excessive risks and the desire to do so. Instead of requiring companies that meet their standards to waste time and money satisfying the FDA, we could simply waive them in to the U.S. market. The resulting increase in competition would surely benefit consumers.
In the meantime, consumers can expand their choices by engaging in another form of medical tourism: purchasing drugs abroad. They can do this legally by visiting foreign countries and bringing medications back with them. The U.S. Customs Department allows travelers to bring in a 90-day supply of a prescription medication purchased abroad, as long as the traveler has a prescription for the drug and the pharmacy fills the prescription as written. One hitch is that the FDA and the Customs Department don’t want travelers to use this option to bring unapproved generic drugs into the country, so these drugs are at risk of being seized. It is not clear how vigorously Customs actually enforces this requirement, particularly against ordinary citizens. Indeed, there may well be a speakeasy norm, where only the loud and obnoxious get their drugs confiscated and everyone else is waved through.
Another option is ordering prescription medications from online pharmacies in other countries, which ship drugs into the United States. Although it is illegal to do so, the practice is flourishing. According to a 2016 report, “[a]s drug prices have spiraled upward, tens of millions of generally law-abiding Americans have committed an illegal act in response: They have bought prescription medicines outside the U.S. and imported them.”19 As WebMD observes:
Current law says that if Granny decides she can get her heart medications more cheaply in Alberta than in Alabama, she could be busted for either bringing it over the border or having it delivered to her. Does that mean that dear Granny is likely to do a stretch in solitary? Hardly, experts say, because nobody wants to be seen putting the cuffs on elderly pensioners. Also, they’d have to arrest the governments of the states of Wisconsin, Minnesota, Illinois, Vermont, as well as many city governments and private employers who have turned north for lower-cost prescription drugs.20
The list of states that buy drugs for their own purpose in Canada or help others do so is now up to nine, and the cities, counties, and even school districts that do so number in the dozens.21 Given that they are flouting the law and that millions of Americans are doing so too, the prohibition on buying drugs abroad seems to fall into the same category as the speed limit.
As law professors, we are reluctant to encourage people to break the law. Instead, we repeat the point we made when discussing retail medicine. When consumers lead from below, politicians feel pressure to follow. Given the number of people and the number of government entities that are buying drugs abroad, we think it unlikely that Congress will decide to spend the dollars that would be needed to prevent this from continuing. To the contrary, as the number of people who are responsible for their own drug costs continues to grow, foreign purchases will become increasingly common and pressure will mount for Congress to call off the dogs. Recent efforts to beef up enforcement, including a crackdown on stores in Florida that help tens of thousands of people purchase Canadian drugs, may push the issue to the front of the policy agenda.22
Amarillo Slim was a famous and successful poker player. He is widely credited with the observation, “If you’re at a poker table and you don’t see a sucker, it’s you.” The decision to use health insurance to pay whatever price the pharmaceutical manufacturers have decided to ask for makes us all into suckers. Prizes, antitrust enforcement, revisions to the FDA approval process, and self-pay are the best ways we know of to start playing to win.