The Health Care Laws You Should Be Paying Attention to (and Probably Aren't)

Providers in states without CON laws can focus on what caregivers do best: providing care. They don’t need to ask permission to provide care. They don’t need to wait years for that permission to be granted. And they don’t need to spend thousands of dollars countering the arguments of their would-be competitors. Patients would be better served if policy makers paid closer attention to patient outcomes in CON and non-CON states.

“Really? This is a thing?” That’s the reaction I typically get when I tell people about certificate-of-need (CON) laws that limit the provision of health care services. These little-noticed state regulations ought to get at least a sliver of the attention afforded to the Affordable Care Act. And unlike that hopelessly politicized legislation, repeal of CON may appeal to those on both the left and the right.

Found in 35 states and the District of Columbia, CON laws require medical providers hoping to purchase equipment, open a new practice, or expand an existing practice to first obtain permission from a state regulator.

Are you interested in opening a burn care center in Alabama? You’ll need to get permission from the CON board. Want to buy a CT scanner in Iowa so your patients can have a less-invasive alternative to a colonoscopy? You’ll need a CON for that. Want to open a neonatal intensive care unit in rural Virginia—and hopefully avoid more tragic deaths of babies born too far from such a unit? Sorry, you’ll need to ask for a CON (and don’t count on it being granted).

As a general rule, CON regulators do not assess quality or competence. Those things are already regulated through other means such as licensing and certification boards. Instead, as the name suggests, the CON process aims to determine whether a community “needs” a service in question. In almost every other industry in the country, entrepreneurs themselves assess the viability of a service before they risk their own money. But in states with CON laws, health entrepreneurs can spend years and tens or even hundreds of thousands of dollars proving to regulators that their community “needs” the service they hope to offer. In the process, other providers often go before these boards and try to persuade them that, all things considered, they’d really rather not have any competition.

If you think this sounds like a blatantly protectionist measure designed to protect incumbent providers from competition, you are in good company. Antitrust officials at the federal Department of Justice and at the Federal Trade Commission have long taken this view, arguing before state legislatures that CON laws undermine competition to the detriment of patient care.

So why do CON laws exist? Over the years, those who benefit from these regulations have offered a number of rationales. They’ve claimed that CON laws ensure an adequate supply of health care. They don’t. They’ve argued that these rules promote the use of cheaper alternatives to care such as ambulatory surgery centers. They don’t. They’ve said that these rules encourage rural health care (they don’t) and charitable care (they don’t).

An alarming link between CON and hospital quality

Two recent studies puncture two of the most common claims of CON proponents. The first is coauthored by Professor Thomas Stratmann of George Mason University, who has done a great deal of CON research in recent years. Working with GMU PhD student David Wille, Stratmann assesses the effect of CON on hospital quality.

Though—as I have noted—the CON process typically does not assess a provider’s quality or safety record, advocates have recently claimed that CON can nevertheless promote better care. By restricting the supply of providers and channeling more procedures through fewer facilities, the argument goes, CON laws may make these facilities more proficient at certain procedures. As far as theory goes, the argument is so-so. After all, other well-supported economic theories suggest that quality tends to go down, not up, whenever competition is restricted. But what do the data say?

Stratmann and Wille assess this question using Hospital Compare, a database maintained by the Centers for Medicare and Medicaid Services (CMS). Previous researchers have tried to test the effect of CON laws on quality by looking at a single dimension—say mortality following coronary artery bypass graft. But the advantage of Hospital Compare is that it uses multiple dimensions—everything from patient satisfaction surveys to heart attack readmission rates to pneumonia mortality rates.

In order to get a clean estimate of the effect of CON, they assess the regulation in health care markets that happen to span across CON and non-CON states. This allows them to control for other, unobservable, factors that might affect health outcomes such as culture and socio-economic conditions. They also control for several observable demographic and hospital characteristics that might confound the estimate.

Stratmann and Wille’s findings suggest that not only does CON fail to promote quality care, but it appears to be doing some real harm. Controlling for other factors, they found that hospitals in CON states have statistically significantly higher mortality rates for pneumonia, heart failure, and heart attack. They also found higher mortality rates among surgical inpatients with serious treatable complications. When they narrowed their focus to only those states that regulate four or more services through CON, they found that hospitals there have higher readmission rates for heart failure and heart attack, and that patients in these states are less likely to rate their hospital experience highly.

The authors run a series of robustness checks using different sample sizes and alternative measures. No matter how they looked at it, hospitals in CON states were no better—and in many cases seemed quite a bit worse—than those in non-CON states.

CON and spending: No, less supply does not mean lower cost

The second recent paper to deflate a common claim of CON proponents concerns the effect of CON laws on health care spending. I authored this paper, and compared with Stratmann and Wille, my task was relatively simple. Since researchers have assessed this question many times and in many ways, I didn’t need to reinvent the wheel. All I needed to do was read and summarize four decades of research on CON and spending.

Before I get to the findings, however, I should clarify what we really mean when we talk about spending. Most of the time, when people talk about the price or the cost of something, they speak in per unit terms. The price of gasoline is only meaningful if we talk about it as the price per gallon or per barrel. The cost of a pizza is only informative if we talk about it as the cost per slice or per pie. These per unit measures are helpful because they relate financial sacrifice to some tangible measure of the good or service enjoyed as a result of that sacrifice.

When we refer to health care costs in per unit terms, we mean the cost of a medical procedure such as a colonoscopy, or of a service such as a checkup, or of a device such as a stent. And basic economic theory is clear about what should happen to these per unit costs if supply is limited through something like a CON: All else being equal, a supply restriction will tend to raise per unit costs so long as demand is not perfectly elastic (i.e., flat). And even if demand is perfectly elastic, there is no reason to expect a supply restriction to reduce costs.

That is why it is no surprise that, in my survey of the literature, I find no evidence that CON laws decrease per unit costs or prices. In fact, the balance of evidence suggests that these laws are associated with higher per unit costs.

The more interesting question, perhaps, is why anyone would think that a supply restriction would lower per unit costs. Near as I can tell in talking with state policy makers over the past several years, few of them actually think that a supply restriction could do this. Instead, they prefer to think about health care spending in a way that anyone who spends any time budgeting would find unhelpful. Instead of focusing on per unit costs, they focus on how much is spent per patient or per person in a given time period.

This is the medical equivalent of a family’s annual pizza or gasoline budget. It isn’t a particularly useful measure since it tells us nothing about the value obtained as a result of that sacrifice. State A may have lower per capita medical expenditures than State B. But if this is simply because A’s residents have access to fewer convenient, health-improving, life-saving services, it may not be a good thing.

In any case, unlike the effect of a supply restriction on per unit measures, the effect of a supply restriction on total expenditures is ambiguous. If there were a Certificate of Pizza Convenience limiting the supply of pizza shops, families might end up spending more or less on pizza depending on whether the price-increasing effect or the quantity-reducing effect of the supply restriction dominates.

So we have to turn to the data. I looked at four decades of research, comprising about 20 peer-reviewed, academic studies. None of the published studies I reviewed found any evidence that CON was directly associated with lower patient expenditures, while a majority found CON to be associated with higher expenditures.

Since CON proponents also claim that it leads to lower investment expenditures (which, again, may or may not be a good thing, depending on whether those investments are worthwhile), I also looked at studies that examine this question. Here again, I found no evidence that CON reduced aggregate investment.

Finally, I looked at studies that assess hospital efficiency—essentially, how well they transform inputs into outputs. Here the evidence is mixed, with some studies suggesting CON hospitals are more efficient than non-CON hospitals and others finding them less so. But even if CON hospitals do transform inputs into outputs better than non-CON hospitals, this may not be a good thing if it means needy patients in places where CONs are denied are stuck without proper care.

The bottom line: Neither economic theory nor empirical evidence suggest that CON laws reduce cost in the way that most people think of it. In theory, they might reduce per capita or per patient health expenditures. But the evidence suggests that they don’t do this either.

Another way

These studies and others on the effect of CON rely on the fact that policy makers in 15 states have chosen not to restrict the supply of health care services in this way. Instead, caregivers and health entrepreneurs in these places may decide for themselves whether their communities “need” a particular service. In making these calls, they are guided by the signals of prices, profit and loss which guide all investments in a market economy.

They don’t need to ask permission to provide care. They don’t need to wait years for that permission to be granted. And they don’t need to spend thousands of dollars countering the arguments of their would-be competitors. Instead, they can focus on what caregivers do best: providing care. Patients would be better served if policy makers paid closer attention to patient outcomes in CON and non-CON states.