Defining Ideas

In Defense of an "Expensive" Drug

Thursday, April 30, 2015
Image credit: 
Steven Depolo

In "The Drug that is Bankrupting America” (Huffington Post, February 16, 2015), Columbia University economist Jeffrey Sachs argues against the high price that Gilead Sciences charges for Sovaldi, a drug used to treat Hepatitis-C. Sachs does not fault Sovaldi for being ineffective. Instead, his argument is simply that the $84,000 price for a full treatment is far above its production cost.

Sachs says that he wants a “rational drug pricing system.” So do I. And I’ll point out some changes that could be made to get to such a system.

But, first, let’s consider the relevant question about Sovaldi: Is its value higher than its price? For many people, it is. Indeed, when we understand just how valuable the drug is—not to Gilead, but to patients—it’s clear that Sovaldi is an incredible bargain. Consider how doctors treated Hepatitis-C shortly before Sovaldi came along. According to WebMD, “the typical treatment for hepatitis C was a combination of interferon and ribavirin with two antiviral drugs,” telaprevir or boceprevir, both of which were introduced in 2011. This combination led to cure rates between 30 and 80 percent. But the side effects were awful: flu-like symptoms, fatigue, anxiety and depression, and anemia. Sovaldi has only mild side effects and a 90-percent cure rate. Also, the previous treatments took 28 to 48 weeks to cure Hepatitis-C, whereas Sovaldi takes only 12 weeks to work.

Moreover, according to Jonathan M. Fenkel, a doctor who directs Thomas Jefferson University’s Hepatitis C Center, the price of a Sovaldi treatment is “on a par with the costs of telaprevir or boceprevir.” So the price of Sovaldi is about the same as the combined prices of telaprevir or boceprevir; the cure rate is higher; the side effects are much milder; and the therapy takes less time. Sovaldi sounds like a bargain.

Professor Sachs admits all of the drug’s virtues, calling it “a remarkable, life-saving medicine at the cutting edge of science.” He points out that Sovaldi could save “millions of Americans” and perhaps “hundreds of millions of people around the world.” He also calls it a “godsend.”

Whatever role God may have played, God didn’t invent Sovaldi. People did. These people worked under Professor Raymond Schinazi, a biochemistry professor at Emory University. People also brought the drug to market. After the drug was invented, Gilead Sciences bought the rights to the drug and now produces it.

People need incentives to discover, not just to produce life-saving drugs. High drug prices give people such an incentive. Sachs doesn’t disagree. He writes, “With a rational U.S. drug pricing system, private investors would expect to earn a reasonable multiple of their R&D for a highly successful drug, perhaps even 5 to 10 times the R&D outlays, in order to reflect the long time horizons and high uncertainties surrounding drug development.” He points out that the multiple for Sovaldi is “40 times or more.” But how does Sachs know what the right multiple is? He doesn’t. Nor do I. What we know, as I’ve noted, is that Sovaldi, compared to what existed before, is a bargain.

Why can Gilead charge so much for the drug? One main reason is that the people who use it do not typically pay for it. Sachs points out that federal and state governments—he presumably has in mind Medicare and Medicaid—as well as private insurers will often pay for all of the cost. If our health insurance systems were different, where a large percent of patients paid even a small percent of the cost, drug companies would almost certainly price drugs lower. Without the favorable tax treatment that the government gives employees’ private insurance, there would be more cost sharing. And more cost sharing by patients should certainly be injected into Medicare and Medicaid. Is Professor Sachs advocating more cost-sharing by patients so that prices will fall? He doesn’t say.

There is another reason that Gilead Sciences has been able to charge such a high price for Sovaldi. That factor is a government monopoly in the form of a patent. No one besides Gilead Sciences is legally allowed to produce and sell Sovaldi.

Is Sachs' problem with Gilead the fact that it has a monopoly? I don't think so. The only way you could get rid of Gilead's monopoly would be to end its patent. And he doesn't propose doing so. Is his complaint that Gilead is using the full extent of its monopoly power to charge high prices? That appears to be it. But then what is the "right" price? As noted above, Sachs seems to know that it's one that gives the company 5 to 10 times the R&D outlays, but he never tells us how he knows that. And which outlays: the early high-risk outlays or the later low-risk ones? And does this account for failures in the clinic, where 198 out of 200 products fail?

It may seem strange to justify Gilead's monopoly. But a brief tour of the economics of intellectual property and FDA regulation shows that if we want new drugs and if we keep FDA regulation, we will need patents. 

The classic argument that economists have made for patents is that they imperfectly solve a market failure. If someone invents something and tries to charge a high price to recoup not only production costs but also his high cost of invention, other producers can copy the item and make money by selling it for a price above their production costs but below the price charged by the inventor. A potential inventor, looking ahead and seeing this, will have less of an incentive to invent. Why is this an imperfect solution? Because the downside is that the inventor has a legal monopoly. Unless he can perfectly price discriminate, charging lower prices to people willing to pay less, he will price some people out of the market who would have been willing to pay more than his production cost.

Much innovation would occur without patents. It would occur because curious people still will want to do research. Also, people often invent things by happenstance. So we have a tradeoff. On the one hand, patents will give monopolies to people for inventions that would have been invented even without patents. On the other hand, patents will cause many things to be invented that otherwise might never have been invented or, at least, might not have been invented so soon.

That’s a tough tradeoff, and, as British economist Arnold Plant pointed out in the 1930s, the choice between having and not having patents is not obvious. But there is one area where one can pretty clearly make the case that patents are, on net, good. That area is pharmaceuticals.

The main reason for that is regulation by the Food and Drug Administration. The FDA's rules for being able to produce and sell a drug put companies through multiple layers of testing. According to a November 2014 study by Joseph DiMasi of the Tufts Center for the Study of Drug Development, Henry G. Grabowski of Duke University’s Department of Economics, and Ronald W. Hansen of the Simon Business School at the University of Rochester, the cost of bringing a drug successfully to market in the United States is $2.558 billion. That includes, of course, the cost of "dry holes"— the drugs that companies abandon because they are not safe or effective or because to figure out whether they are safe and effective is too costly. What drug company would be foolish enough to spend that amount of money without some kind of protection from competition?

The FDA’s regulations matter in another way also. The FDA’s rules require disclosure of the drugs’ contents. That, combined with the long time taken for approval, makes it easier for competitors to produce copycat versions.

With such extreme FDA regulations and with no patent protection, we would probably have no new drugs. Given that the FDA is unlikely, any time soon, to lose its regulatory powers, the patent system seems like a reasonable solution to a difficult problem.

There is a further solution. Ironically, one of the federal government’s regulations makes it difficult for drug companies to charge low prices to low-income people who want their drugs. The government mandates that when a drug company sells to Medicaid, it must charge Medicaid a price at least as low as it charges anyone else. That law is what prevents many drug companies from charging low prices to the uninsured who have little wealth. Drug companies would love to engage in the “price discrimination” that I mentioned earlier. If the production cost of Sovaldi is, say, $500 for the whole regimen, and the market price is $84,000, there are probably many uninsured people willing to pay well below $84,000, but well above $500, to save their lives. Gilead Sciences could charge them, say, $2,000. But as long as Gilead is stuck having to charge that price to Medicaid, it won’t charge that price to anyone.

One of the big breakthroughs in economics was the “Marginal Revolution” of the 1870s, when economists figured out that the value of something is not the same as the cost. The value of Sovaldi to many people is far above the production cost. Allowing companies to charge high prices that reflect that value will give them an incentive to keep on researching, discovering, and producing high-value—and sometimes life-saving—drugs.