For example, a Johnson & Johnson report argues that the cost of a drug should not only reflect research costs related to that drug, but also should include expenses from “drug candidates that fail in development.” The company contends that “pharmaceutical companies and the rest of the scientific community can learn from these failures to improve the research process,” suggesting that consumers should pay for these failures. The report’s approach epitomizes how the pharmaceutical industry justifies high prices for medication.

But while science may learn from failures, that has nothing to do with patents or what they’re intended to do. In fact, the entire notion of compensating for failed research puts the modern application of patent law on a collision course with the history and theory of patents, reaching back to this nation’s inception.

The patent system is designed to reward success. One does not receive a patent for an invention one tried and failed to create. Similarly, a patent’s reward reflects the successful invention, rather than compensation for other attempts gone bad.

And that has been the case since the nation’s founding. In 1790, Congress enacted the first patent statute and George Washington signed the first U.S. patent, which went to Samuel Hopkins for an invention related to making potash. Since that time — and through various iterations of the patent statute — the law has consistently required patent holders to disclose their invention so that those “skilled in the art” can make and use it after the patent has expired. Both sides of the patent deal — society’s and the inventor’s — are strictly and carefully limited to the precise invention that the patentholder can prove to have created.

In fact, there was not a single patent law or court case between 1790 to 1865 stating or even indicating that a patent grant was intended to compensate the patentee for the costs of developing a failed (that is, never-patented) invention. Indeed, with limited exceptions, early patent law does not even suggest that a patent should compensate for the costs of developing the successful invention being patented, let alone costs beyond the patented invention. It was about granting intellectual property for that invention.

During the 1930s and 1940s, the courts developed a doctrine of law prohibiting patent misuse, which is broadly defined as an impermissible attempt to expand the time or scope of a patent. The goal was to ensure that the nation’s patent laws did not provide a pass for violating antitrust law. And so, when a patent holder tries to extend its patent power outside the bounds of the patent, it acts improperly.

All of this shows how closely the patent system has clung to the specific boundaries of the invention disclosed in the patent, while not concerning itself with costs of other research.

This all changed in 2006, when the Medicare Modernization Act went into effect, providing full Medicare coverage for prescription drugs. With a vastly expanded and secure marketplace, pharmaceutical companies began raising prices on existing and new medications. To justify soaring price points, companies needed a way to legitimize their high profit margins. Compensating for the failed drugs — the ones that didn’t make it to market — fit the bill.

Ensuring that rewards cover the costs of failure might appear logical at first glance. After all, if one wants inventors to invest in new research, shouldn’t the payoff be large enough to compensate for the pain of slogging through the long, cold winters of failure?

As appealing as the concept may sound, however, allowing the patent system to compensate for the cost of failures has a perverse effect, particularly when industry uses it to lobby for expansion of patent protection. The argument has become: We to need to hold off competition more, so that we can charge more to compensate for past failures. But that distorts the purpose of a patent: encouraging companies to succeed — and in the most efficient manner possible. If patents reward failure, then the more a company fails, the longer and broader of a monopoly they’ll need when they do succeed. But why would we want a patent system that rewards companies more when they are less successful?

In a perfect world, one might expect purchasers to create a natural brake on the system. In theory, one cannot charge a price unless buyers are willing to pay. Health care is a strange market, however. Patients may consume the medication, but doctors are in charge of choosing it, and insurance pays for much of it. As a result, consumers do not have full information and do not bear the full burden of the costs. In addition, modern strategic behaviors in the drug approval and reimbursement system can diminish the competition that might otherwise prevent drug companies from raising prices unchallenged. Thus, the normal pressures that would limit prices are dampened in the market for prescription medicine.

Incentivizing failure is particularly problematic in light of a historic shift in the pharmaceutical industry over the past decade. Faced with stagnating research results, the industry has shifted to outsourcing innovation. Specifically, the majority of innovation in the pharmaceutical industry comes from academia or from small life-science companies. Large pharmaceutical companies then shepherd the drugs through the FDA approval process and into production.

While there should be little room for excess returns at the top, that’s not how the system works in practice. Consider Gilead’s hepatitis C cure, Sovaldi. The company more than recouped what it paid for the drug in the first year of sales alone. And after five years, the company reaped in excess of $58 billion dollars from sales of the drug, more than five times what it paid to acquire the drug from the start-up that took the initial risk and engaged in the innovation. Or consider Merck’s immunotherapy drug, Keytruda. In 2020 alone, the drug’s sales topped $14 billion, with no signs of flagging. Forbes estimates that the value of Keytruda is $200 billion — a far cry from the $300 million the company paid to acquire it.

And Keytruda’s boatload of patents insulates the company’s pricing scheme from competition. Yet the true innovator — that is, the company that took the research risk — isn’t the one reaping the lion’s share of the reward. Rather, the bulk of the patent reward is going to the company that walked the last mile.

Examples such as these show how modern pharmaceutical markets allow large companies to be over-rewarded while innovators are under-rewarded. If society actually wanted a patent system that compensates for failure, the dollars would go to those who make the investment in the research and take the risk of failure — the inventors. None of that is happening here. Instead, by passing so little of the profit to those who perform the successful research, the system dilutes the incentive for research. We are not just incentivizing failure, we are sending those dollars to the least inventive part of the innovation chain.

In short, incentivizing failure is as counterproductive as the phrase sounds. Why would we want a patent system in which the less efficient person — the one who fails more along the way — gets a larger reward? Unless we recognize that problem, the nation may find itself sliding quietly into an approach that undermines the contours of the patent system from time immemorial, hampering our ability to innovate in crucial areas.