Certificate-of-Need Laws Make a New Clinic Ask Its Rivals for Permission

Adam Smith, who was not a conservative but whose book is the closest thing to our scripture on markets, made an observation in 1776 that every legislature should have to read aloud before voting on a licensing bill. In The Wealth of Nations he wrote:

People of the same trade seldom meet together, even for merriment and diversion, but the conversation ends in a conspiracy against the public, or in some contrivance to raise prices.

Now imagine a state law that does not wait for the meeting. It writes the meeting into the statute. That is a certificate-of-need law, and it is the subject of this post.

The claim, and the definitions

My thesis is simple. Certificate-of-need laws should be repealed. Let me define the terms before I argue it.

A certificate of need, or CON, is a permission slip from a state agency. In states that have the law, a provider that wants to build a new hospital, add beds, buy certain expensive equipment such as an MRI scanner, or open certain kinds of facility, such as an ambulatory surgery center, must first apply to a state board and demonstrate that the community needs the new capacity. The board decides. The details vary widely by state, in which services are covered, which thresholds apply and which agency runs the process.

The part that matters most is who gets to speak at the hearing. In many states, existing providers that would face a new competitor are entitled to participate in the review. They can submit comments, appear, object and in some states appeal a decision they do not like. The applicant must persuade a panel that the market needs it, in a room that includes the people whose market is at stake.

I will argue five numbered premises.

1. CON replaces the market's question, "will customers choose this?", with a bureaucrat's question, "do we need this?"
2. The process hands incumbents a veto, or at least a very large lever.
3. The history shows that the laws were adopted for a reason that was abandoned long ago.
4. The evidence from the states that dropped them does not match the fears.
5. The best defense of CON, that it protects rural hospitals, fails on its own terms.

Premise one: "need" is a word that does the work

Start with the central phrase. A free market does not ask whether something is needed. It asks whether somebody will pay for it. A grocery store does not apply for a certificate proving that the neighborhood needs a second grocery store. It opens, and the neighborhood decides.

The applicant, in effect, is asked to prove a negative about his own competitors: that nobody already in business minds. It is the only exam in America where the other students help write the grading rubric.

The reply is that health care is different, because patients do not shop the way they do for groceries, and because someone else usually pays. I will concede part of that, since third-party payment really does blunt price signals. But notice what the concession proves. If the problem is that prices do not work well in health care, the remedy is to make prices visible and to let patients and payers choose, not to forbid new suppliers. A rule that blocks new entrants preserves the distortion it claims to fix. It treats a symptom of weak competition with a cure that is guaranteed to weaken it further.

"Need" is also a word with no natural stopping point. Who decides how many MRI machines a county needs? The answer in practice is a state formula, a planning document or a board's judgment. Each one is a prediction about the future, written by people who will not bear the cost of being wrong and who have no way of knowing what a new provider's innovation might be. The whole case against central planning, from Hayek on, is that the information needed does not exist in any central office. It lives in millions of decisions by people who use the service and people who provide it.

Premise two: the incumbent's veto

Here is where the structure becomes a problem, and where I want to be careful and fair. A CON board is not a smoke-filled room of villains. It is staffed, generally, by public employees and appointed members who are trying to apply a statute. The incentives are the problem, not the character of the people.

Consider the incentives. A new competitor's benefit is spread thinly over thousands of future patients, none of whom know to show up at a hearing, because nobody can attend a hearing about a clinic that does not exist yet. The cost of the new competitor is concentrated on a handful of existing providers, who know exactly what is at stake and who employ lawyers and lobbyists. Put those two groups in front of a board and you do not need a conspiracy for the result to be predictable. You only need the ordinary logic of concentrated interests against diffuse ones, which James Madison analyzed in Federalist No. 10 when he wrote about the danger of faction.

That is the structure Smith warned about, converted into administrative law. It is hard to design a system more efficiently tilted toward the people already inside the market. If a casino built its house edge this plainly, the gaming commission would call it a scandal. A health planning board calls it Tuesday.

A fair question is whether the evidence of such objections exists. It does, in the form of the process itself: the law in many states lets competing providers intervene in an application and challenge the result. You do not need to take my word for it. Read your own state's statute, which will say who may be an "affected party" and what rights that status carries. I will return to this in the closing section.

Premise three: the history

The origin story matters because it shows that CON was a product of a particular moment, one that has passed.

New York adopted the first CON law in 1964, in what was known as the Metcalf-McCloskey Act. The idea spread slowly through the 1960s and early 1970s, built on the assumption that hospitals were overbuilding and that third-party payment, in particular Medicare, which began in 1965, was funding capacity that no one had to justify. If the government was going to pay the bills, the thinking went, it ought to have a say in how many beds there were.

Then, in 1974, Congress turned a trickle into a flood. The National Health Planning and Resources Development Act of 1974 pushed every state toward adopting a CON program by tying federal health planning funds to it. Within a few years nearly every state had a program. That is a plain example of what I have called, in other posts, the pattern of federal money buying state uniformity: a policy that the states would not have adopted on their own was adopted by all of them because a check was attached.

And then Congress changed its mind. In 1986, Congress repealed the federal mandate. Once the federal incentive was gone, a number of states dropped their programs. Others kept them, and the survivors are what we have today.

I want to underline the point, because it is the one that cuts through the argument. The federal government created the nationwide CON system and then repealed it. The states that kept it are defending a program that its own author gave up on four decades ago.

Premise four: what the repeal states and the federal agencies found

Some states never adopted CON, and others dropped their programs after the federal mandate ended. More recently, Florida repealed major parts of its program in 2019, and in 2023 South Carolina's governor signed a repeal while North Carolina's reform narrowed its program. The map keeps changing, so check your state's current status on your legislature's website.

At the national level, the two federal agencies charged with protecting competition have been consistent. In July 2004, the Federal Trade Commission and the Department of Justice issued a joint report, Improving Health Care: A Dose of Competition, and its message on CON was blunt. The agencies said that CON laws can act as a barrier to entry and can protect existing providers from competition, and they recommended that states reconsider them. Since then, FTC staff have repeatedly sent comment letters to state legislatures making the same point. That is not a partisan source. These agencies served presidents of both parties.

Researchers have studied the question, and a 2011 study by Eichmann and Santerre in the Journal of Health Care Finance, for example, found that states with CON programs have fewer hospitals per person than states without them. I will leave the specific numbers to the studies, since they depend on the services and the method, and encourage you to read the primary papers instead of the summaries. What the research does not produce is a clear finding that CON lowers costs. That was the promise made for it, and I have not seen the research that shows it kept.

The steelman: the rural hospital argument

Now the objection I take most seriously. Defenders of CON say that it protects rural hospitals and the communities they serve. The argument runs like this. A small rural hospital depends on a mix of patients. Some are profitable, such as those needing elective procedures with good reimbursement. Others are not, such as emergency patients who cannot pay. If a new surgery center opens nearby and skims off the profitable procedures, the hospital loses the revenue that subsidizes its emergency room, and eventually the community loses its hospital. CON, on this view, prevents that cream-skimming.

It is a coherent argument, made by sincere people who care about their towns. Let me answer it in four points.

1. It proves too much. If it is legitimate to protect a hospital's profitable services from competition so that it can fund unprofitable ones, then the same logic would justify protecting any business from any rival. Taken seriously, it is a rationale for a permanent monopoly, with the community paying the cross-subsidy through higher prices on everything.
2. It misdescribes the remedy. If the community values the emergency room, the honest policy is to fund the emergency room directly, through a subsidy, a payment formula or a designation for rural facilities, and let competition work elsewhere. A hidden tax on every patient, collected through the absence of competition, is the least transparent way to pay for anything.
3. It does not match the geography. CON statutes usually apply statewide. A rule justified by a vulnerable hospital in a remote county is applied to a new clinic in a growing suburb a hundred miles away. The law does not distinguish, because a statewide law cannot.
4. The evidence does not support it. Researchers who have compared rural access in CON and non-CON states have not produced a convincing finding that CON states have healthier rural hospital networks. Some of the work has pointed the other way. I would not stake the argument on one study. I would say that the people who claim CON saves rural hospitals bear the burden of showing it, and the burden has not been met.

If you want the full range of options for covering the cost of rural and unprofitable care without a barrier to entry, that is a topic for a separate post. The short version is that a subsidy you can see is better than a barrier you cannot.

Premise five, restated: a verdict on the defense

I said the best defense fails on its own terms. Here is what I mean. The rural hospital argument assumes that the CON board will use its power to protect the hospital in a rural county. But the board's power is general. It does not carry a label that says "use only for rural hospitals." It will be used by every incumbent, in every county, for every service, including the urban hospital system that wants to keep a competitor out of its own wealthy suburb. A power defended as a shield for the vulnerable will be used as a sword by the strong, because the strong have lawyers and the vulnerable do not.

That is the argument of the Federalist about every grant of power: you must assume that it will be used by the people who are best positioned to use it, not by the people for whom you intended it. Madison's phrase in Federalist No. 51, that "ambition must be made to counteract ambition," is the right test for a design. CON puts ambition on only one side of the table.

What a repeal would look like

I am not proposing the abolition of safety rules. A new hospital must meet building codes, staffing standards and licensing requirements, and no one is arguing otherwise. Licensing asks whether the provider is competent and the facility is safe. CON asks whether the market needs it. The first question is the government's business. The second is the public's.

A sensible repeal would do four things.

• End the "need" test for hospital beds, equipment and surgery centers, while keeping licensing and quality standards in place.
• Replace any genuine rural-access goal with direct, visible support, rather than a barrier.
• Require public reporting of prices and quality so patients can shop among the providers who open.
• Sunset any remaining requirements on a schedule, so that no program continues out of habit.

For related reading, arizona's universal license recognition and the case for copying it covers how a state can open its professions to qualified workers from elsewhere, and Hospitals Went to Court to Keep Their Prices Secret, and They Lost shows the same instinct at work: an industry that would rather the public not look too closely. Both pair well with the argument here. If you want a laugh at the logic, see A Certificate of Need for Pizza Shops, a Thought Experiment in Planning, which applies the rules to a pizzeria.

The case in brief

1. A certificate of need makes a provider prove that a community "needs" a new facility or device to a state board, and in many states existing competitors may take part in the review.
2. That structure gives concentrated incumbents influence over diffuse future patients, which is the problem Madison described in Federalist No. 10.
3. The system began in New York in 1964, was pushed on every state by the National Health Planning and Resources Development Act of 1974, and was repealed by Congress in 1986.
4. The FTC and the Justice Department said in their joint 2004 report that CON laws can protect incumbents, and states that have repealed have not seen the predicted collapse.
5. The rural hospital defense would justify a monopoly for anyone, and a visible subsidy would serve rural communities better than an invisible barrier.

What you can do: find your state's certificate-of-need statute on your legislature's website, read who counts as an "affected party" in a review, and ask your state representative whether the law should be repealed or narrowed.