The Danger of Managing Drug Prices
Ezekiel Emmanuel says drug prices are too high. Very expensive drugs, for rare conditions or cancer, can run well into six figures. What we need, he argues, is a national board that sets prices, or at least sets some caps, rather than letting the free market (and excessively generous Medicare reimbursements) drive costs into the stratosphere.
Well, maybe. But I'd like a bit more evidence before we leap in with red pencils blazing and start slashing away.
I understand Emmanuel's frustration. The market for pharmaceuticals is characterized by a set of conditions that makes pricing difficult, and inevitably controversial:
1) People who are dying are price insensitive; they would pay any amount to get a few more months of life. I don't think it's any accident that you see high prices clustered in diseases such as cystic fibrosis, cancer, and Hepatitis C -- which historically had few good treatments.
2) The tab for pharmaceuticals often is picked up by third-party payers. I may be willing to pay any amount to cure my cancer, but in practice, the amount I can pay is limited by my bank account, or the equity in my house. Third-party payers don't have the same hard limits. And though they do have incentives to hold the price down, law or public opinion often limit what they can refuse to pay for. The worse the condition, the harder it is to refuse treatment -- companies probably can safely turn down a $300,000 a year treatment for athlete's foot, but cystic fibrosis is another story.
3) Much of the cost of producing a drug is the up-front outlay required to develop a new pill, and get it approved by the Food and Drug Administration. Once that's done, the marginal cost of producing another dose is comparatively small. This process of drug discovery is insanely risky and expensive.
4) Pharmaceuticals are protected by patents; a competitor can bring out an alternative cancer drug, but until the patent runs out, they can't produce yours.
All this adds up to prices that often seem arbitrary, excessively high, and ripe for trimming. And, in fact, it's true: if the U.S. mandated that prices for drugs had to be much lower, we would pay much less for drugs, and those pills would keep rolling off the assembly line.
But would we get new drugs? Or rather, how many new drugs would we get? There's the rub: Those high profits provide strong incentive for pharmaceutical innovation.
Products with a high fixed cost and a low marginal cost always look as though they ought to cost a lot less. After all, it only costs a few cents a pill to roll it out! How dare they charge me so much? But over the long run, someone has to pay for the fixed cost -- the research. If you don't pay for it, you don't get any more.
Emmanuel acknowledges this, but suggests that pharma profits are so high that we could easily slash them by a huge fraction, and still get plenty of new drugs: "Regardless of the risks, many drug companies are making huge profits. Gilead, maker of Sovaldi, has profits of around 50 percent. Biogen, Amgen and other biotech firms have profits of around 30 percent. Merck and Pfizer are seeing profits of 18 percent or more. Even if profits were cut by a third or a half, there would be sufficient incentive to assume the risks of drug development."
These numbers are accurate, as far as they go. Unfortunately, that isn't very far. For starters, Merck and Pfizer are among the more profitable pharmaceutical firms. Eli Lilly's profit was 12 percent last year. Could we cut that in half, too? How about the profits of all the biotech firms that aren't making money just now?
More fundamentally, he seems to be thinking about this as a budget problem, rather than an investment problem. When your income is fixed, you start with the amount of money you have, and divvy it up to various uses. Could we cut the "profit" line and still make drugs? Sure. Would Eli Lilly still be a desirable stock? At some price, absolutely.
This is very sensible, if you're just thinking about how to divvy up your household income. But that's not how investment decisions are made. Instead, you consider two things:
1) What is the risk-adjusted expected value of this investment?
2) If I compare this investment to other uses for the money, do I want to invest in making new drugs, or should I put the money into something lower-return, but safer?
Some biotechs have giant profit margins. But there's tremendous survivor bias among these companies. A large number of them got started, and burned through phenomenal amounts of money looking for drugs. Many of them didn't find any. All that money vanished, along with the company. The only ones left to compare before making your decision are the ones that found a lucrative drug. So you look at all these companies raking in big profits and think, "This can't be necessary to compensate investors for risk -- look how much money they're making!" You forget that when the money was initially invested, no one knew which companies were going to make it, and which were going to crash and burn. Investors needed to be well compensated for taking that risk, because biotech is close to an all-or-nothing proposition: Either you get a drug approved, or you vaporize millions of dollars with absolutely nothing to show for it except some secondhand lab equipment.
In fact, you should expect the successful companies in such a risky business to make big profits. No one would invest in an industry with enormous risks if it has a 4 percent return. You'd just stick the stuff in an index fund, or a government bond.
The same is broadly true of internal decisions at companies seeking new drugs: If the profit margins on innovations are low, then it probably doesn't make sense to invest in finding them. It makes more sense to find some safe, steady business that will deliver the same returns at lower risks.
Right now, the U.S. accounts for a disproportionate share of the profits that make it attractive to keep looking for new drugs, precisely because we do not have a pricing board that attempts to hold down reimbursements to levels closer to marginal cost. That means we're providing a disproportionate share of the incentive for new research. Every so often, there is a clamor about lowering our prices and forcing other countries to pay their "fair share" of research costs, but there is no practical way to do it. So the only question is, are we willing to subsidize new research?
I am not arguing that the current prices of these drugs reflects some platonic ideal of value, or that the free market has found the best possible combination of affordability and incentive for innovation. As I mentioned above, the pharmaceutical market has all sorts of strange wrinkles that make it hard to assess the value they are delivering for the prices they charge.
However, people who are advocating for a government price-setting board cannot simply say, "Well, there are some issues with the way that pharmaceuticals are priced." The onus is on them to show that the government would do a better job of determining the tradeoff between innovation and current prices than the market already does -- that it would not, for example, set prices artificially low in order to reap temporary political benefit, at the expense of future generations who would then have to go without beneficial treatments. And perhaps they will. But saying "Merck could still make drugs with a 9 percent profit margin" is hardly adequate.
Unfortunately, making that case is hard, because the systems in other countries are smaller, and benefit from that implicit subsidy from the U.S. There's no way to see what the world would look like if the U.S. decided to hold prices closer to marginal cost, except by moving to a pricing board and accepting the risk that we might be shooting the goose that's laying the golden eggs.
This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.
To contact the author of this story:
Megan McArdle at email@example.com
To contact the editor responsible for this story:
Max Berley at firstname.lastname@example.org