By Thomas P. Stossel

There’s a mounting crusade to impose price controls on prescription pharmaceuticals. Democratic presidential hopefuls Hillary Clinton and Bernie Sanders have called for deep price discounts for drugs sold through Medicare, importation of drugs from countries like Canada that have price controls, and limiting drug companies’ marketing exclusivity rights. “Cancer treatment shouldn’t bankrupt patients” is the gist of their emotional argument.

That politicians with no medical training or business experience would have a myopic understanding of drug pricing is unsurprising. But such misconceptions also afflict leading cancer specialists.

An Aug. 27 article by several oncologists from Emory University and public-health experts concludes that a “reasonable price” for treating lung cancer patients for three weeks with necitumumab, a new drug under development, should be no more than $1,309. That’s what the authors judge to be the value of the two-month life extension the drug seems on average to confer. Typically, new cancer drugs cost several thousand dollars a month. Physician advocates for price controls have seized on this difference. But medical schools don’t teach students about the difficulty and grim economics of drug development.

Policy makers and doctors pushing for price controls believe they can reduce the return for a successful new drug while still preserving the incentive to innovate. This is a dangerous form of magical thinking.

Leaving aside the arrogance of monetizing survival—how do you “price” the value of two additional months that might allow a dying cancer patient to attend a daughter’s wedding or a son’s college graduation?—the difficulties and economics of drug innovation demand premium prices. Here’s why.

One reason is that with rare exceptions, progress in treating cancer with drugs has been slow and incremental—advancing survival by weeks at a time. Indeed, the drug discussed in the oncology article is the first treatment in decades to extend life for patients with a particular form of advanced lung cancer.

Another reason, the most important, is that in contrast to other high-tech enterprises such as building skyscrapers or jumbo jets based on predictable engineering principles, drug development suffers from the immense uncertainty of biology. This unpredictability has enabled us to avoid extermination by highly mutable microorganisms that attack us, but it also causes seemingly promising drug candidates to fail.

When pricing new medications, pharmaceutical firms must take into account the research and development costs of that medicine—but also of the many drugs that didn’t pan out. According to the Tufts Center for the Study of Drug Development, only about 12% of all medicines entering clinical trials make it to market, and the failures are big money-losers. Cancer drugs have even higher failure rates. Of the 78 potential treatments for brain cancer that underwent clinical trials from 1998-2014, three proved safe and effective enough to win Food and Drug Administration approval.

Firms have to find some way to compensate. Since “cancers” are really many different diseases that respond idiosyncratically to treatment, patient populations eligible for specific therapies can be small. Hence, paying for failure means charging higher prices. Yet even though drug prices rose more than other health-care costs in 2014, they constituted less than 14% of overall health-care expenditures.

Like it or not, financial investment, not good intentions, fuels such high-risk innovation, and investors want returns. Price controls on drugs are a surefire formula for sending investors elsewhere. According to a 1994 analysis in Health Affairs, when President Clinton threatened price controls in 1993, the stock prices of biotechnology companies dropped, some by as much as 40%, before recovering when the threat subsided.

It would be a huge mistake to skew the risk calculus for investors so they conclude that the billions of dollars in research funding required for a potential miracle medicine would be better invested in a new dog food or the latest Silicon Valley startup. Large pharmaceutical companies may survive, but small, barely financed startup companies at the cusp of innovation will not.

Policy makers need to be more attentive to the economic realities that are enabling drug producers to extend lives and improve quality of life. Price controls will lead to shorter and harder lives for future patients beyond the reach of current treatments. Rather than demonize drug prices, policy makers—and physicians—should encourage the development of lifesaving medicines and aid patients in finding assistance programs to help pay for expensive drugs.