A machine works on purifying a potential hepatitis C drug at Gilead Sciences in Foster City, Calif. (David Paul Morris/Bloomberg)

Zain Rizvi is a student at Yale Law School. Amy Kapczynski is a professor at the school and co-director of Yale’s Global Health Justice Partnership. Aaron Kesselheim is an associate professor of medicine at Harvard Medical School and a faculty member at Brigham and Women’s Hospital.

When the government launches an ambitious public-works project — say, building a railroad or airport — one problem it can face is a “strategic holdout.” People who know that the government needs their land to complete a project may demand a price for their property far beyond its actual value.

Today, we face a similar problem with certain high-cost drugs.

Some important drugs have no reasonable clinical substitute. When a pharmaceutical company has a patent on such a drug, it holds the same privileged position as the last landowners along a proposed train route. Because only the manufacturer can sell the drug, the only available “route” to treatment runs through its intellectual property.

Predictably, drug manufacturers sometimes use this leverage to demand excessively high prices. Take Gilead, for example. According to a Senate Finance Committee investigation last year, before the company decided on its $84,000 price tag for a 12-week course of its new hepatitis C drug, Sovaldi, executives spent months debating how high they could set its price. The price of the drug reflects not what the drug costs to manufacture (a couple hundred dollars), or what it cost to develop (in the hundreds of millions of dollars, the investigation indicated), but the ability that the company has to hold out from potential buyers, some of whom are required by law to provide the drug. Facing ballooning costs, payers have restricted access to the medication, putting treatments out of the reach of hundreds of thousands of people with the virus and raising the risk that some will transmit the infection and experience potentially fatal liver damage. Even with recent discounts, Gilead has earned staggering sums on Sovaldi and a related drug: nearly $36 billion in less than three years.

So what can the government do against such holdouts?

In expanding a railroad or airport, the conventional answer is obvious: The government can use its power of eminent domain, allowing it to acquire the land for a reasonable price. When expanding access to important new medicines, the response, as two of us argue in an article in this month’s edition of the journal Health Affairs, should be similar. The government should employ an analogous power — government patent use — to negotiate lower prices, or buy low-cost, generic versions of drugs for use in government programs. This is possible because existing law gives the federal government limited immunity to challenges from patent holders: Patent holders cannot stop the government from making or buying products that infringe on their patents, and can sue only for reasonable compensation.

Federal agencies have previously relied on this mechanism to purchase items produced by companies other than the patent holders, ranging from lead-free bullets to electronic passport readers. But using the power effectively sometimes means just discussing it publicly. After the anthrax attacks in 2001, the government suggested that it would consider importing generic versions of an antibiotic treatment, ciprofloxacin (Cipro), under this provision. Bayer, the manufacturer, agreed to a significantly reduced price shortly thereafter.

The pharmaceutical industry’s central objection to the government-use approach would probably be to point to its potential impact on innovation: Research and development are expensive, drug development is risky, and manufacturers must be rewarded, handsomely, lest they stop investing in innovation.

This innovation narrative may or may not be correct. (Drug companies spend far more on marketing than research and development, for example.) In any case, compensation could be set to ensure an adequate return and protect companies’ incentives to invest in innovation. When courts set a “reasonable” payment for government patent use, they should account for the amount invested in the drug, adjusted for the risk of failure, and add a premium to provide adequate profits, particularly where a drug has high clinical value.

By invoking this power, the government could transform our approach to important but drastically overpriced new medicines. Consider the implications for just this one class of medicines: In 2014, the government treated only about 2 percent of all Medicaid recipients who have hepatitis C with Sovaldi, for nearly $1 billion. If it uses our approach, it could treat all those left untreated for less than $150 million, plus a reasonable royalty for the company. Since Gilead has surely more than recouped its risk-adjusted investment and then some, a small royalty would be appropriate. The precise details could be worked out in negotiations or in court, but even under the worst-case scenario for the government, it could almost certainly treat far more people in federal programs at no more than what it plans to pay.

The debate about excessive drug prices — and the de facto rationing that results — often assumes that the government has little power over prices in the short term. But one solution is hiding in plain sight, and it requires no additional congressional action. It might take as little as a phone call — backed by a credible commitment to follow through, if needed — to get companies to return to the bargaining table and to bring some sense to the drug-pricing environment.

Will anyone in the Obama administration pick up the phone?