Health Care Policy Isn't so Hard

Last July, as the last Republican Obamacare bill was imploding, Greg Mankiw wrote "Why Health Care Policy is So Hard" in the New York Times. For once, I think Greg got it wrong. Health care policy isn't hard at all, at least as a matter of economics. (Politics, and ideological politics, is another question, but not Greg's question nor mine.)

There are some important underlying themes uniting how Greg's piece goes wrong (in my opinion)

• A little bit of economic education can be a dangerous thing

While most opinionated people and most "policymakers" are blissfully unaware of any economics, a little bit of economics education can sometimes mislead. Economics is full of pretty fairy tales, passed on through the decades or even centuries. The day after one sees the beautiful tale of the natural monopoly, or the externality, or the public good, then like a two-year-old with a hammer to whom everything looks like a nail, one starts to see natural monopolies, externalities and public goods all over the place. Wait a moment. Just because it's in the textbook -- even Greg's textbook -- doesn't mean every single industry and case fits.

The other rhetorical error is of the type, "well, we can't have homeless people who get heart attacks dying in the streets." No, of course not, but, is every single line of the ACA and tens of thousands of subsidiary regulations absolutely necessary to provide for homeless people who suffer heart attacks? Why must your and my health insurance be so totally screwed up -- and so totally micromanaged by the Federal government -- just to solve the problem of homeless people heart attacks? I'm struggling to find just the right category for this sort of argument

• Gross disregard of the size of effects. 
• Straw man -- a theoretical problem with a completely free market justifies any regulation. 
• Disregard of the choice at hand -- it's not benevolent perfection vs. free market. 
• Using problems as talking points. If the same "problems" exist elsewhere and you don't want to or need to fix them, then you're not serious about that "problem" for health. 

Maybe we can come up with a better one sentence characterization later. (There must be a Greek word for these rhetorical tricks!)

Let's review Greg's "why health care policy is so hard" problems.

"...free market sometimes fails us when it comes to health care. There are several reasons.

Externalities abound.  Take vaccines, for instance. If a person vaccinates herself against a disease, she is less likely to catch it, become a carrier and infect others. Because people may ignore the positive spillovers when weighing the costs and benefits, too few people will get vaccinated, unless the government somehow promotes vaccination. 

Another positive spillover concerns medical research. When a physician figures out a new treatment, that information enters society’s pool of medical knowledge. Without government intervention, such as research subsidies or an effective patent system, too few resources will be devoted to research." 

Well, ok. We require vaccinations to enroll children in schools. And basic research might be under funded. But basic chemistry research might be underfunded too. Does the Federal government need to buy half of all chemicals in the country and intensely regulate the other half just to keep basic chemistry research going? There are externalities everywhere. A neighbor mowing his lawn on a Saturday morning might wake you up. Does this justify the entirety of America's exclusionary zoning codes, or make "housing policy hard?" We do have research subsidies and a patent system, by the way. People like Greg and I are paid pretty handsomely to do research!

These "problems" exist in many markets -- and the ACA, or even pre-ACA regulation, is hardly a minimalist solution to the problem of vaccination and basic research!

The logical connection from "free markets sometimes fail us" to "and therefore the Federal Government needs to take a heavy hand as it does for health care" deserves its own place in the pantheon of fallacies. We have a choice between imperfect alternatives.

"Consumers often don’t know what they need. In most markets, consumers can judge whether they are happy with the products they buy. But when people get sick, they often do not know what they need and sometimes are not in a position to make good decisions. They rely on a physician’s advice, which even with hindsight is hard to evaluate."

"The inability of health care consumers to monitor product quality leads to regulation, such as the licensing of physicians, dentists and nurses. For much the same reason, the Food and Drug Administration oversees the safety and effectiveness of pharmaceuticals."

I am surprised that Greg, usually a good free marketer, would stoop to the noblesse oblige, the cute little peasants are too dumb to know what's good for them argument. This argument applies equally to car repair, tax advice, contracting, home repair, computer setup and repair, economics teaching... and just about everything else in our economy. We purchase complex personal services from people who know more than we do. It seems to work out ok.

Rhetorically, it's a good example of an argument that isn't serious because it isn't uniform. Why haul this out just for health care?

Again, is the ACA a minimal solution? All policy is a choice among alternatives. Do you really think government run insurance systems are better for figuring out what you "need?" Does Greg think he and his family are too dumb to make medical choices, so wishes for a government bureaucracy to determine his and his family's care?

Is inability to monitor quality a central economic problem? How much of the ACA is devoted to that? How much of the ACA and surrounding regulation is instead devoted to stopping the free flow of information,  to stop competition over quality, to maintain the illusion that all doctors are equal?

Licensing.. In this age when the Obama administration started to sound like the Cato institute on the subject of occupational licensing, 70 years after Milton Friedman showed how the AMA uses licensing to restrict supply and keep their earnings up, and as London Transport brazenly bans Uber, Greg gives us this vision of the wise benevolent government licensing for our protection? Those unlicensed dog-walkers sure are a national disgrace. And let's not start on the FDA's wise overseeing of the safety and effectiveness of pharmaceuticals, like, say, the epi-pen.

"Health care spending can be unexpected and expensive. Spending on most things people buy — housing, food, transportation — is easy to predict and budget for. But health care expenses can come randomly and take a big toll on a person’s finances." 

"Health insurance solves this problem by pooling risks among the population. But it also means that consumers no longer pay for most of their health care out of pocket. The large role of third-party payers reduces financial uncertainty but creates another problem." 

Greg surely knows better than this. Spending on houses and cars is not easy to predict and budget for -- when the house burns down or the car crashes. That's why we have insurance, regulated and perhaps over-regulated, but nothing like health insurance.

"Consumers no longer pay for most of their health care out of pocket" is not a necessary consequence of insurance. Insurance, in a free market would not cover routine predictable expenses, just as car insurance does not cover oil changes. This is entirely an artifact of regulation.

Let me skip to the last, most common and most important argument, most illustrative of how a little economics education can be a dangerous thing.

"Insurance markets suffer from adverse selection....If customers differ in relevant ways (such as when they have a chronic disease) and those differences are known to them but not to insurers, the mix of people who buy insurance may be especially expensive. "

"Adverse selection can lead to a phenomenon called the death spiral. ...Suppose that insurance companies must charge everyone the same price.... the healthiest people may decide that insurance is not worth the cost and drop out of the insured pool. With sicker customers, the company has higher costs and must raise the price of insurance. ...As this process continues, more people drop their coverage, the insured pool is less healthy and the price keeps rising. In the end, the insurance market may disappear."

We have all been to that beautiful econ 1 class, where we hear Ken Arrow's asymmetric information insurance spiral, or George Akerlof's justly famous proof that the used car market does not exist.

But are these fables true of our world, or is this a case of two year old with hammer? In the fable, you know things about your health that a pure free-market health insurer, armed with your entire history, every scan and test they can dream up, cannot know. In reality, the information advantage is exactly the opposite! They know a whole lot more about you than you do. That's not the asymmetric information of this fable.

In fact, a few paragraphs ago, Greg make exactly that opposite argument! Health care must be run by the government because the poor peasants don't know how sick they are and what to do about it, but now health insurance must be run by the government because the crafty little buggers know exactly what they need and private health insurers can't tell them apart.

We do have asymmetric information and a death spiral -- because the government forbids insurers to use information they have! The government forces insurers to take everyone at the same price, so only the sick sign up.  Maybe that's good or bad, but it's not the  fundamental asymmetric information problem of the fable. And somehow life insurers, car insurers, home insurers, and carmax exist.

Greg is a careful writer. "the mix ... may be expensive... the insurance market may disappear." Yes, every fable is a possibility. But we have to think whether in fact this is a real problem, whether it is a central problem, whether we advocate the same policies uniformly when we see this problem or whether it's just a talking point for policies advocated for other reasons, and whether the ACA or other regulation is a minimally crafted solution to this problem.

"One thing, however, is certain: The existence of a federal law mandating that people buy something shows how unusual the market for health care is."

Really? Does the existence of every federal law show how unusual the underlying market is? Agricultural subsidies prove how unusual the food market is? Solar panel subsidies show how unusual the market for energy is? Tariffs and quotas show how unusual steel is?

"policy wonks of all stripes can agree that health policy is, and will always be, complicated."

As a matter of economics, this wonk disagrees. 95% (made up number) of health expenses are relatively predictable complex personal services, bought by savvy shoppers who buy houses cars and cell phones. I will agree that it always will be complicated only because our government will always be screwing it up. But not that it must be complicated.

OK, health care policy is hard. But it's hard because so few in our political and commentary class have any trust that markets actually can work, and that by and large thoughtfully getting the heck out of the way can lead to a better system for health, as it has for just about everything else where it has been tried. Allowed to do so, competitors will come in and provide better service at lower prices. People and the businesses that want to serve them will find a way to overcome econ 101 problems. CarMax does exist, despite the lemons theorem. Companies really care about their reputations.  What a lot of economics education can do -- including a bit of economic history -- is to patiently remind people of these fact, rather than to give them excuses for endless mindless dirigisme.

Greg is careful, and this is a good review of the potential theoretical problems of health care and insurance markets, as presented in a standard (his!) econ 101 textbook. Greg does not say that the ACA, or even 5% of the ACA, is a necessary solution to these problems. But Greg does not say the opposite either. That these are small, manageable problems, which a government bureaucracy will likely mismanage for health as it does everywhere else, is absent in Greg's column. The average New York Times reader will come out thinking Greg's on board with the basic architecture of the vast complex mess coming out of Washington. If Greg thinks, as he may well do, that a regulatory system about 5% of the size of the ACA could handle all of these economic problems with your and my health insurance, that the rest of the ACA is a vast mess mostly designed to cross-subsidize health care from one group to another,  maintain rents for incumbents, and hide the cost of it all, you wouldn't know it from this article. Greg is a great writer, and knows his audience and the context in which he is writing, so it is a puzzling sin of omission.

I suspect I know what happened. It sounded like a good column idea, "I'll just run down the econ 101 list of potential problems with health care and insurance and do my job as an economic educator." If so, Greg failed his job of public intellectual, to help us digest just which economic fables are actually relevant.

(The last section of After the ACA goes through all these arguments and more, and is better written. I hope blog regulars will forgive the self-promotion, but if Greg hasn't read it, perhaps some of you haven't read it either.)

Update: Greg Responds. Thoughtfully, politely, and unlike me, concisely, as one expects. Yes, there is a great question as to what the role of an economics educator should be! Do we run through the standard list of theoretical possibilities for market failure? Or do we go to the second step of questioning just how much they apply, how central they are, how much they actually drive the regulatory outcome, how effective regulations are at addressing them; making sure they aren not just turned into talking points for political outcomes and rent seeking? All in 900 words or less!

Update 2: Sometimes I'm really slow. It occurs to me only this morning that both Greg and I missed the elephant in the room. The number one lesson that econ 1 has for health policy is: The demand for health care, and health care quality, is highly elastic. And Lesson 2, the income elasticity is pretty high too.

The standard vision in the policy world, the public, and too many health “economists” is that we “need” health care and it is a homogenous good. Translated to economics, they suppose a vertical demand curve.  The hard fact is exactly the opposite. Perhaps less obviously, quality is highly price elastic too. Your back hurts. Do you “need” surgery? (and if so what kind, performed where?) steroid injections? Ibuprofen? Physical therapy (an incredibly varied and price elastic service)? Many people looking at the cost go to chiropractors.

For the world of policy, this fact is what upends all health care schemes. If the cost is low, people will expand their demand for health care services enormously. If the demand curve were vertical, the supply curve could be flat. Sadly, if the demand curve is very flat, the supply curve must rise, and if not through price, through rationing. Someone else will decide what you "need."

Income elasticity is huge. What else is there to spend your money on, if you can? Plus, like business class, people are willing to pay a lot as income rises for the ancillary parts of health care services.

Update: Noah Smith thinks my blog posts and essays aren't long enough. Perhaps a book-length asymmetric-information literature review is a good idea. Someday.