Friday, October 19, 2012

After the ACA: Freeing the market for health care

This is an essay, based on a talk I gave at the conference, “The Future of Health Care Reform in the United States,” at the University of Chicago Law School. The pdf version on my webpage may be easier to read than this version, which is a bit long for a blog post. Also, I'll update the pdf over time as I collect comments, but not this blog post.

Update 2/6/2013 I revised the essay on my webpage which is now better than this one. 

Clearly, two important items on the policy agenda are, if we could get rid of the ACA and Dodd-Frank, what would we replace them with? This essay thinks about ACA, I'll be back on Dodd-Frank. Here goes:

After the ACA: Freeing the market for health care
John H. Cochrane1
October 18 2012

Most of the current policy debate, and the optimistically-named “Affordable Care Act,” focuses on health insurance. I think we need to move on to think about the economics of health care. If the ACA is repealed, we still have a mess on our hands, and just fixing insurance will not be enough to clean up that mess.


I’ve written a lot about how to fix health insurance, so I won’t repeat that all here.2 To summarize briefly, health insurance should and can be individual, portable, life-long, guaranteed-renewable, transferrable, competitive, and lightly regulated, mostly to ensure that companies keep their contractual promises. “Guaranteed renewable” means that your premiums do not increase and you can’t be dropped if you get sick. “Transferable” gives you the right to change insurance companies, increasing competition.

Insurance should be insurance, not a payment plan for routine expenses. It should protect overall wealth from large shocks, leaving as many marginal decisions unaltered as possible.

Preexisting conditions, lack of insurance by the young and healthy, and spiraling insurance costs– the main problems motivating the ACA -- are neatly addressed by this alternative. Why do we not have a system? Because law and regulation prevent it from emerging. Before ACA, the elephant in the room was the tax deduction and regulatory pressure for employer-based group plans. This distortion killed the long-term individual market and thus directly caused the pre-existing conditions mess. Anyone who might get a job in the future will not buy long-term insurance. Mandated coverage, tax deductibility of regular expenses if cloaked as “insurance,” prohibition of full rating, barriers to insurance across state lines – why buy long term insurance if you might move? – and a string of other regulations did the rest. Now, the ACA is the whale in the room: The kind of private health insurance I described is simply and explicitly illegal.

So, the alternative is clear. Getting there will be a long hard road. It’s not a simple matter of “deregulation,” given how deep and widespread the offending restrictions are, and the many legitimate purposes which they purport to serve, and sometimes do. We need to construct a different, but wiser, legal and regulatory regime. I know an interest group when I see one: Don’t worry, there will be lots of jobs for health economists, policy analysts, and lawyers.

Problem solved? Not really. Solving health insurance – who pays -- will not solve the evident inefficiencies and absurd cost of our health care markets.

Health care supply

We all agree what we’d like to see: Health care needs to become efficient, innovative, and provide high quality care at lowest possible cost.

Cost reduction and innovation: some examples

How will this happen? Well, we have before us many good examples. Walmart and Home Depot revolutionized retail. Airlines are dramatically cheaper than in the 1970s. Consumer electronics, telecommunications, computers, and even cars are much better and cheaper, for what you get, than ten or twenty years ago.

These revolutions are not just about technology. In most of these cases, we see process innovation, reorganizing activities to deliver complex services at lower cost and with better and more uniform quality. This process efficiency is most glaringly absent in health care .

Southwest Airlines turns a plane around in 20 minutes, and has finally figured out how to get people on it without the chaos at United and American. Walmart and Home Depot are as much about organizing and standardizing the motion of people and inventory as they are about adopting technology, outsourcing supply, or negotiating lower prices. Honda assembles a car with 30 hours of labor. As Atul Gawadne puzzled4 in the New Yorker, the Cheescake Factory delivers a complex service-oriented product with remarkable quality, efficiency and cost. Why can’t hospitals do the same?

Beyond stories: Amitabh Chandra and Jonathan Skinner5 summarized the academic literature, writing “there is increasing evidence of the potential for cost-saving technologies (with equivalent or better outcomes) in the management and organization of health care to yield substantial productivity gains. But these types of innovations are unlikely to diffuse widely through the health care system until there are much stronger incentives to do so”

But our hopes for health care go beyond the obvious need to streamline of process and delivery and adopt cost-saving technology. We don’t want 1950s care at cheaper prices. Technical innovation is, fundamentally, why we can be so much healthier than our grandparents. Health care markets need to bring that innovation as fast as possible-- and then diffuse it quickly down to the mass market.

My example industries are also great at technology innovation and diffusion. Health care is a paradox, that innovation is widely reviled as a cause of increased costs, where by any economic definition the opposite is true. The answer that you’re mistaking “cost” for “price,” and that a new $500,000 treatment represents a reduction in cost over a less effective but still available $50,000 older treatment is correct as a matter of economics. But it’s unsatisfying, because we all see the monstrous inefficiencies in health care.3

Why does Moore’s law not apply to medical devices? Why has the price of cell phones, GPS, and computers come down so fast relative to the prices of medical technology? Where is the home MRI? There is nothing deeply different about medical and other technology. The answer is that supply and demand – in the current highly regulated system – is not producing the Moore’s law incentives.

In my examples, innovation doesn’t always mean lower cost. I paid $1500 in 1982 for an IBM PC with 16 k and one floppy disk drive. I paid about the same (nominal) for my most recent laptop, with vastly more power. Nissan is going to sell6 $3,000 cars in China and India – with no airbags. We have chosen much better cars for slightly higher prices.

In each case, however, the industry has done a good job of pushing the cost/innovation/quality frontier out to its limits, and then discovering where people really want to be. If we “spend more” today, we know we’re getting a good deal, and simply choosing a different point on a far better frontier than we faced 20 years ago. What we need in health care is to push that quality-cost-innovation frontier back. If we then choose higher cost, it will indeed be cause for celebration, not hand-wringing.

These industries do not cut costs by selling shoddy products. Instead, they provide consistent quality on the dimensions people turn out to really care about, and save on those that people don’t really care about. Southwest gets you where you want to go at convenient times, with a good on-time record, and admirable safety. And seats 27 inches apart, feeding you peanuts. The iphone error rate is a lot lower than the medical error rate. Walmart shirts use inexpensive materials, and they are sold in environments far less sexy than Michigan Avenue boutiques, but it’s rare to find one torn, or missing buttons. The fear, so often expressed in medical contexts, that unregulated competitive suppliers will pawn off shoddy merchandise on consumers seems exactly false. Restaurants tremble at a poor yelp review. The corporatization and standardization, which we to some extent bemoan, is a good part of their ability to deliver consistent quality. If each airplane and pilot were a different practice, quality would vary a lot more!

How will this change come about? My examples share a common thread: Intense competition by new entrants, who put old companies out of business or force unwelcome and disruptive changes. Microsoft displaced IBM, and Google is displacing Microsoft. Walmart displaced Sears, and may displace Wal-Mart. Typewriter companies didn’t invent the world processor, nor did they adapt. The post office didn’t invent FedEx or email. Kodak is out of business. Toyota gave us cheaper and better cars, not Ford/GM/Chrysler competition. When the older businesses survive, it is only the pressure from new entrants that forces them to adapt.

My examples share another common thread. They remind us how painful the cost-control, efficiency, and innovation processes are. When airlines were regulated, artificially high prices didn’t primarily go to stockholders. They went to unionized pilots, flight attendants and mechanics. Protection for domestic car makers supported generous union contracts and inefficient work rules, more than outsize profits. A look at a modern hospital and its supply network reveals lots of similar structures. “Bending down cost curves” in these examples required cleaning out these rents, through offshoring, elimination of union contracts and work rules, mechanization, pressure on suppliers, and internal restructurings.

The fact that so much cost reduction comes from new entrants, not reform at the old companies, is testament to the painfulness of this process, and the ability of incumbents to protect the status quo. The big 3 still take 40 hours to build a car relative to Toyota’s 30. And two of them went bankrupt, while Toyota sits on a cash reserve. American and United are still struggling to match Southwest’s efficiencies, after 30 years. The parts of Kodak invested in film simply couldn’t let the company exploit its technical knowledge in optics and electronics. Chicago’s teacher unions are fighting charter schools tooth and nail. And a quick look at a modern hospital, and its suppliers, suggests just how wrenching the same transformations will be.

Supply and Competition

So, where are the Walmarts and Southwest Airlines of health care? They are missing, and for a rather obvious reason: regulation and legal impediments.

A small example: In Illinois as in 35 other states7, every new hospital, or even major purchase, requires a “certificate of need.” This certificate is issued by our “hospital equalization board,” appointed by the governor (insert joke here) and regularly in the newspapers for various scandals. The board has an explicit mandate to defend the profitability of existing hospitals. It holds hearings at which they can complain that a new entrant would hurt their bottom line.

Specialized practices that deliver single kinds of service or targeted groups of customers cheaply face additional hurdles, as they undermine the cross-subsidization provided by “full service” hospitals. For example, the Institute for Justice is bringing a major suit8 by a specialty colonoscopy practice in Virginia, which local “full service” hospitals managed to ban.

This is exactly the form of regulation put in place by the Civil Aeronautics Board until the late 1970s, which produced airline prices much higher than they are today. Airlines had to show “need” for a new route, and incumbents defended monopoly rents on the grounds that they cross-subsidized service to small airports. Its removal is pretty much centrally what brought us cheap airlines now.

Revealingly, CON laws were an earlier round of “cost containment,” and were federally mandated for a while. The idea was sensible enough, and you could imagine it echoing through conferences such as this one. On a fee-for-service system, there can be an incentive to buy too many MRI machines, and then prescribe “needless” scans, which insurance companies and the government would be forced to pay for. Well, said an earlier round of health-policy experts, we’ll patch that up by having a regulatory board review the “need” for major investments or hospital expansion to avoid “needless” overinvestment. Even if the theory is true, it’s an interesting story how an attempted regulatory patch to one broken system (poor incentives in fee-for-service reimbursement) turned in to a barrier to competition and wound up increasing costs.

How occupational licensing is captured to restrict supply and push up prices should be obvious by now – Milton Friedman wrote his PhD dissertation on it. If you’re a parent, you’ve been through it. It’s 2 am in a strange city. The kid has an ear infection. He needs amoxicillin, now. Getting it is going to be a 3 hour trip to an emergency room, hundreds of dollars, so a “real doctor” can peer in his ear, then off to the pharmacy to fill the prescription. A nurse practitioner at the Wal-clinic could handle this in 5 minutes for $15.

I’m not arguing that we have to get rid of licensing. But licensing for quality does not have to mean restriction of supply to keep wages up, including state-by-state licensing, restriction of residency slots, or restrictions that encourage overuse of doctors where they are not needed.

Einer Elhauge 9 examines “fragmentation” of medical care in detail, the fact that even in hospital settings care is bought essentially from different doctors and specialists rather than in an integrated manner, as, say airline travel is, where you do not separately purchase pilot, flight attendant, fuel and baggage services. My examples suggest a consolidation, integration, and corporatization of overall health service provision, as restaurant chains displace individual stores. What stops this defragmentation? He surveys research concluding that nothing in the nature of health care seems to require this structure, as hospitals in other countries have salaried doctors, and concludes instead (p. 11):

The dominant cause of fragmentation instead appears to be the law, which dictates many of the fragmented features described above and thus precludes alterative organizational structures.

He lists a long string of legal impediments, including Medicare reimbursement rules, laws against corporate medicine and tort doctrines. Referring to private insurance (p.12):

…State laws generally make it illegal for physicians to split their fees with anyone other than physicians with which a physician is in a partnership. More important, alternative payment systems, such as paying a hospital (or other firm) to produce some health outcome or set of treatments, would make sense only if it has some control over the physicians and other contributors to that outcome and treatments. And other laws preclude such control, as detailed in the chapters by Professors Blumstein, Greaney, Hyman, Madison, Cebul, Rebitzer, Taylor, and Votruba. The corporate practice of medicine doctrine provides that firms—whether hospitals or HMOs—cannot direct how physicians practice medicine because the firms do not have medical licenses, only the physicians do. Although some states allow hospitals to hire physicians as employees, that change in formal status does not help much if the employer cannot tell the employee what to do. Even if the law did not prohibit such interference, tort law generally penalizes firm decisions to interfere with the medical judgments of individual physicians, making it unprofitable to try, as Professor Blumstein observes. Further, hospital bylaws usually require leaving the medical staff in charge of medical decisions, and those bylaws are in turn required by hospital accreditation standards and often by licensing laws. By dictating autonomy for the various providers involved in jointly producing health outcomes, these rules largely dictate separate payments to each autonomous provider.

Private insurer efforts to directly manage care have likewise been curbed by the ban on corporate practices of medicine and the threat of tort liability. In addition, states have adopted laws requiring insurers to pay for any care (within covered categories) that a physician deemed medically necessary, banning insurers from selectively contracting with particular providers, and restricting the financial incentives that insurers can offer providers.
My cost-cutting examples are all for-profit companies. About 70% of hospitals and 85% of health-care employment is in non-profits,10 whose legal and regulatory treatment protects much inefficiency from competition.

If United didn’t have to pay taxes, Southwest’s job would have been that much harder.

Maybe for-profit companies pay too much attention to stock prices. But non-profits can go on inefficiently forever, with no stockholders to complain. The whole point of a non-profit is to pursue goals other than economic efficiency.

More importantly, if a for-profit company is inefficiently run, another company or a private-equity firm can buy up the stock cheaply, replace management, and force reorganization. Non-profits (and their management especially) are protected from this “market for corporate control11.”

Many non-profit hospitals are too small or, by definition unable to issue equity, undercapitalized.

Recognizing some of these pathologies, there is a wave of mergers, and transfers between for-profit and not-for-profit status. But there is lots of gum in the works. When a nonprofit is sold or converts to for-profit, the state attorney general and courts can weigh in on the sale; legally to ensure that the proceeds benefit a charitable cause related to the non-profit’s original mission. This is a great opportunity for competitors to block the change.1

The FTC is ramping up antitrust action against hospital mergers.13 Hospitals need economies of scale for expensive, specialized modern medicine and to comply with the avalanche of regulatory and insurance regulation. The FTC worries about local monopolies able to raise prices, especially given the inelastic demand by insurers and government reimbursement. So here we have the government forcing small size in order to boost competition with one hand, stopping entry to protect hospitals from competition with another, trying to force larger “networks” through “Affordable Care Organizations” to obtain the needed economies of scale with the third, but laws preserving doctor independence with the fourth.

On reflection, it’s amazing that computerizing medical records was part of the ACA and stimulus bills. Why in the world do we need a subsidy for this? My bank computerized records 20 years ago. Why, in fact, do doctors not answer emails, and do they still send you letters by post office, probably the last business to do so, or maybe grudgingly by fax? Why, when you go to the doctor, do you answer the same 20 questions over and over again, and what the heck are they doing trusting your memory to know what your medical history and list of medications are? Well, this is a room full of health policy wonks so you know the answers. They’re afraid of being sued. Confidentiality regulations, apparently more stringent than those for your money in the bank. They can’t bill email time. Legal and regulatory roadblocks.

So, medical records offer a good parable: rather than look at an obvious pathology, and ask “what about current law and regulation is causing hospitals to avoid the computer revolution that swept banks and airlines 20 years ago,” and remove those roadblocks, the government adds a new layer of subsidies and contradictory legal pressure.

The impediments to supply-side competition go far beyond formal legal restrictions. Our regulatory system has now evolved past laws, past simple, explicit, and legally challengeable regulations, to simply hand vast discretionary power to officials and their administrative bureaucracy, either directly (“the secretary shall determine..” is the chorus of the ACA) or through regulations vague enough to let them do what they want. Witness the wave of discretionary waivers to ACA handed out to friendly companies. Those administrators can easily be persuaded to take actions that block a disruptive new entrant, and with little recourse for the potential entrant. (Lobbying government to adopt rules or take actions to block entrants is legal, even if those actions taken directly would violate anti-trust laws, under the Noor-Penington doctrine.)

Forget about Wal-clinics; Chicago and New York kept the food and clothes part of Wal-Mart out for years, at the behest of unions and competitors, by denying Wal-Mart all the necessary permits and approvals. So many citizens, especially our poor and vulnerable, continue to live in employment and retail deserts.

The increasing spread of medical tourism to cash-only offshore hospitals is a revealing trend. Why does this have to occur offshore? What’s different about the hospital location? Answer: the regulatory regime.

So, what’s the biggest thing we could do to “bend the cost curve,” as well as finally tackle the ridiculous inefficiency and consequent low quality of health-care delivery? Look for every limit on supply of health care services, especially entry by new companies, and get rid of it.

The reregulation path

Now, this is of course not the way of current policy. The ACA and the health-policy industry are betting that new regulation, price controls, effectiveness panels, “accountable care” organizations, and so on will force efficiency from the top down. And the plan is to do this while maintaining the current regulatory structure and its protection for incumbent businesses and employees.

Well, let’s look at the historical record of this approach, the great examples in which industries, especially ones combining mass-market personal service and technology, have been led to dramatic cost reductions, painful reorganizations towards efficiency, improvements in quality, and quick dissemination of technical innovation, by regulatory pressure. I.e., let’s have a moment of silence.

No, we did not get cheap and amazing cell phones by government ramping up the pressure on the 1960s AT&T. Southwest Airlines did not come about from effectiveness panels or an advisory board telling United and American (or TWA and Pan AM) how to reorganize operations. The mass of auto regulation did nothing to lower costs or induce efficient production by the big three.

When has this ever worked? The post office? Amtrak? The department of motor vehicles? Road construction? Military procurement? The TSA? Regulated utilities? European state-run industries? The last 20 or so medical “cost control” ideas? The best example and worst performer of all,..wait for it... public schools?

It simply has not happened. Government-imposed efficiency is, to put it charitably, a hope without historical precedent. And for good reasons.

Regulators are notoriously captured by industries, especially when those industries feature large and politically powerful businesses, with large and politically powerful constituencies, as in health insurance or as in most cities’ hospitals. In turn, regulated industries quickly become dominated by large and politically powerful businesses. See banks, comma, too big to fail. (Several insurance companies were bailed out in the financial crisis, so too-big-to-fail protection is not a distant worry.) This is not to say that regulators are not well-meaning and do not put great pressure on many industries. But the deal, “you do what we want, we’ll protect you from competition” is too good for both sides to resist.

Needless to say, price controls have been an unmitigated disaster in every case they have been tried. Long lines for gas in the 1970s are only the most salient reminder. Their predictable result is, vanishing supply. Try finding a doctor who will take new Medicare or Medicaid patients.

The current regulatory approach is not really well described as simple price controls, e.g. $3 per gallon of gas, but rather as fiddling with a payment system of mind-numbing complexity and endlessly discovered unintended consequences. The past record of “cost control” and “incentive” efforts should warn us of how likely adding more complex rules is to work.It seems instead to be a challenge to the next generation of planners.14

But that’s only the beginning. Real cost reduction is a hard process, as my examples remind us.

Can a regulator in a democracy really become a union-buster, force painful concessions on workers, suppliers, and other “stakeholder” beneficiaries of rents? Can a regulator realistically demand that jobs be outsourced or replaced by software? Can a regulator really preside over a wave of turnover in which new businesses send old ones to the dustbin, firing their management?

Consider a small example now in the news. Hospitals are starting to outsource the reading of x rays, even to India. This is still heavily regulated – the radiologists are still US trained and certified. But already it’s a cause celebre for the potential to cost jobs. When the obvious happens – Hmm, we have some good Indian doctors who can read the x rays just as well – you can imagine the scandal. And doesn’t every American deserve the best – a US radiologist on staff and present 24 hours a day, ready to consult with the doctor? Personal-injury law firms are already lining up to sue based on the “inferior quality” of outsourced readings, with requisite horror stories.15 How could a regulator demand outsourcing radiology and using Indian doctors?

My examples also do a remarkable job of getting rich people to voluntarily pay through the nose, covering fixed costs for medium-income consumers. Two words: Business Class. A politician who proposed taxing people this way to provide air travel would be hanged as a socialist. And a regulator who consigned middle-income patients to seat 25d while wealthier patrons got business class would be hanged as a fascist.

Our current system tries to accomplish such cross subsidies, but at massive inefficiency: to cross-subsidize Medicare, Medicaid and emergency rooms, we overcharge cash customers and private insurance, and protect inefficient hospitals from competition.


Now by being concrete, and therefore realistic, I invite obvious complaints. What, I like airlines and Walmart? Have I been to an airport lately or shopped at Walmart? (Yes to both, incidentally.) But I think the examples are good to remind us what efficiency looks like, how it is achieved, and to keep us from fantasies about what health-care can look like and what outcomes regulators are likely to be able to achieve.

We love to complain about airlines. But aside from the TSA’s security theater and air traffic control – both run by the government – what we really want is 1970s service at 2010 prices. Sorry, we can’t afford private-jet medicine for everyone. Southwest medicine has to be the goal.

Shop at Walmart? Walmart is putting all those cute mom and pop stores out of business. It’s putting pressure on union jobs, the main reason Chicago kept it out all these years. It pushes suppliers relentelessly. It buys from China. Aren’t I being heartless?

No. I’m being realistic. The lesson from all our experience with other industries is that “cost control” and innovation are a hard and brutal process.

Many of you are probably still squirming in your seats. You want some other way. You want to keep unionized jobs, “living wages,” “worker protections,” or “keep our community hospitals going.” Perhaps you mourn the bank tellers replaced by ATM machines, and jobs sent to China.

More deeply, you are probably squirming in your seats at my observation that quality varies enormously in efficient industries: some fly economy middle seat, and some fly in private jets. Some get shirts from Walmart and some get shirts from Macys. Surely, doesn’t every American deserve the best when it comes to health care?

If so, you’re not serious about reducing costs, i.e. finding the efficient point on the quality-cost curve. This is simply a fact: you’re adding other goals to the mix, so you’re accepting rising costs to fund those other goals. Or you’re fantasizing that you can have it both ways.

And if you’re having trouble putting those other considerations aside and accepting a Walmart / Southwest airlines model for health care, imagine how unlikely it is that the department of health and human services will force that model to emerge through its regulatory power.

Health-care demand

The demand side of the health care market is also severely distorted.

Most basically, with either government provision or private insurance, health care is bought in “payment plan.” You pay a tax or a premium, then your expenses are “covered.”

We all understand that when somebody else is paying, people don’t economize on expensive services shop for better deals, or accept less convenient but cheaper alternatives. More importantly, I think, demand affects supply: it’s a lot harder for new entrants to attract business in the current payment system.

Is there something about the nature of health care, as an economic good, that necessitates payment-plan provision? Thinking about it, I think the opposite is true: Health care, as an economic good, is a particularly poor candidate for payment-plan provision.

I think people have in mind a simple wound, or a broken arm. Even if it’s free, nobody is going to overuse that—nobody will have a good arm put in a cast or have stitches just for fun. Pretty much any qualified doctor can handle it; you don’t need to find one that’s “really good at setting bones” but charges a higher price. So, the “good” is well defined, it’s a pretty generic commodity, the demand curve is very steep, and what you “need” is clearly observable.

But this is a very misleading anecdote. The actual demand curve for health care is incredibly elastic. When provided at low cost, people consume prodigious amounts of health-care services. Every cost estimate for government provision or subsidy, from the UK NHS, to medicare, medicaid, and beyond has missed the mark by orders of magnitude. And, though it’s common to disparage “overuse,” in health policy circles, the elastic demand curve is real. These are real people, with painful illnesses, and the “extra” test or visit to the specialist might just be the one to finally help them. Conversely, when asked to pay more, consumers economize rapidly, refusing “too much” care in the judgment of the medical community.

So, we have attempted payment plans with limits – insurance rules, managed care, effectiveness panels, “affordable care organizations” and so on – to cut off the flat demand curve. Ezekiel Emanuel, Neera Tanden, and Donald Berwick, writing in the Wall Street Journal 16 explained the idea: “Instead of paying a fee for each service, providers should receive a fixed amount for a bundle of services or for all the care a patient needs.”

Hmm. “From each according to his ability, to each according to his need.” It has a nice ring to it. Why do I feel a certain foreboding?

Would this work for clothes? Your employer gives you “access” to clothes by including it in your benefits. Then your primary style consultant will determine how many shirts you “need,” which you can pick from the preferred provider network (K mart). Home repair? The home-repair effectiveness board will conduct peer-reviewed research on material for kitchen counters. Sorry, granite is off the approved list, you don’t “need” it.

Health care? For many patients, just getting through the diagnosis to decide what treatment they might try is an expensive and inconclusive nightmare. How much diagnosis do you really “need” in these circumstances?

Many diseases are chronic, requiring widely-varying and individual-specific treatment plans. Nothing really works, and we’re trading off different options with different bad side effects, and needing different levels of commitment from the patient.

End-of-life care, care for elderly, infirm, and handicapped, are very expensive, and all lie on a long string of quality vs. quantity choices. Does grandma really “need” a 5 star nursing home, a helper (a highly personal service! – could insurance or government “provide” housecleaning services successfully?) or just support from family? Does “need” without considering cost, i.e. willingness to pay, really even begin to describe the economics of this decision? Should a family that decides to provide care, saving the nation hundreds of thousands, receive no benefit?

I had a back pain episode recently. (Somehow health policy always ends up with here’s-where-it-hurts personal anecdotes!) Did I “need” an MRI to really see the structural problem? Cortisone shots? Surgery? Physical therapy, or just a Xerox of recommended exercise? Therapy at the hospitals here, or at the specialty sports-rehab clinic that patches up the Bears? Or just a handful of ibuprofen and let it heal? Did my planned trip to Europe matter in this medical “need?”

And now the dirty little secrets. For most patients, “stop smoking, exercise and lose some weight” is the best advice they could take. Patient’s awful compliance is an open secret. How much drugs and treatment do patients “need” who won’t stop smoking, lose weight, exercise, do the physical therapy, or comply with drug regimes?

Another dirty little secret: Quality, both actual and perceived, varies enormously. Rates of medical errors, infection rates, rates of success in difficult procedures, just getting basic diagnoses right, or even washing hands often enough, vary widely. The quality of service provided, including everything from waiting times to convenience of making an appointment and whether the doctor answers emails varies as well. Concierge medicine is emerging really targeted to people tired of the whole runaround. And medicine is not perfect. For a range of conditions, we have imperfect treatments, and scientific knowledge of what works or doesn’t is changing fast.

If only it were so simple to determine “need.” If only people like me went away quickly when told we don’t “need” an MRI to find out why our backs hurt. Or if people with hard to diagnose illnesses like food allergies quietly went away rather than hold out hope that the next specialist will figure out the problem.

So what does “need” really mean for services like this? The only sensible economic definition I can think of is that “need” is the bundle of services you would choose if you were paying with your own money at the margin. You “need” that MRI to make sure your back pain won’t just heal after 6 weeks of ibuprofen if you’d be willing to shell out $1,000 of your own money to get it. (I am!) And you “need” it delivered at a convenient hour, tomorrow, rather than next week across town if you’re willing to pay that extra cost.

“At the margin” is important because intuitive thinking soon mixes up “what you’d rather spend money on with “what you can ‘afford.’” Suppose we offered each patient the choice, “well, your doctor prescribed this MRI. You can have the MRI or you can have $1000 in cash.” You “need” the MRI if you forego the cash and go through with the MRI.

This is an important and unsettling conceptual experiment. If I offer the cash and the patient decides if he wants to take the treatment or find a cheaper alternative, you can’t argue the patient “can’t afford” treatment. It’s unsettling, because we may suspect lots and lots of people would take the cash. So there is a lot of paternalism in health care, which we might be more upfront about.

In any case, once defined, it’s pretty clear that this “need” is essentially impossible to measure externally for a personal service with so much variety, and imperfection, as health care.

Moreover, many more people would “need” MRIs if competition and innovation drove the price down to $50, by any definition of “need.”

So, we’re just arguing about who makes the cost/benefit decision. What you “want” is where you make the cost/benefit decision. What you “need” is what I – or some panel of bureaucrats -- think you should get.

I think the word “need” also has a moral tone, “what society owes you.” This seems even harder to define or measure. How much back treatment did society owe me?

(A little digression for economists: You could quibble with my definition of “need” as “what you are willing to pay for,” because I left out income effects. Perhaps “need” can mean “what you would be willing to pay if you earned $500,000 a year?” Alas, we, as a society don’t have the resources to pay for that definition of “need.” We simply cannot all fly on private jets.

So, while private jet stories are fun, given the budget constraint, the relevant question is whether someone earning $50,000 a year would give a much different answer than someone earning $80,000 per year. And remember that the question is on the margin, with an insurance payment, voucher other lump-sum subsidy to offset income effects. Care for the very poor and indigent is a separate question, which I discuss below.

Now it’s not so obvious that income is the greatest source of variation in “willingness to pay,” in this relevant range. Variation across people within income categories is far greater for every other good, and complex service, so is likely to be greater for health care as well. So, while a relevant quibble, in the end I think an argument based on income effects in the definition of “need” is distraction.)

Bottom line

In sum, health care is a complex, highly varied personal service, not a simple well-defined commodity. The demand curve is as elastic as any in economics. When, where, how, how much, by who are vital components of that service. Objective and subjective quality, and corresponding cost, varies tremendously. The distinction between “need” and “want” is at best unmeasurable and at worst simply meaningless. The broken arm is a horrendously misleading anecdote.

But health care is an economic good. Health care is not that different from the services provided by lawyers, auto mechanics, home remodelers, tax accountants, financial planners, restaurants, airlines or college professors.

Payment-plan provision, with rationing by some external determination of “need,” is based on the opposite and false assumptions and thus pretty hopeless for health care. No planner can mimic the market outcome in which what you need is what you’re willing to pay for.

To some extent, private unregulated insurers can offer high quality vs. generic plans to sort patients ex-ante by quality vs. wiliness to pay. But regulation makes that sorting much harder: Once we force guaranteed issue at the same price, it’s next to impossible for insurers to maintain bare bones vs. fancy plans. The minute a bare-bones customer gets sick, he will demand to be issued a fancy plan at the same cost as everyone else. And health insurers will respond by tailoring plans to attract healthy consumers – free health club benefits – and discourage sick ones. The whole guaranteed issue + mandate arrangement assumes that health insurance is a generic good, not one with good-better-best quality and price points. Or it will soon be forced to be generic! And regulatory rationing cannot say that anyone should shop at Walmart.

The whole stated point of regulation is to ensure quality, of course, but it does a poor job on the dimensions we care about. Regulators can impose minimum standards, requiring degrees, certification, inspections, etc. and keep out really dangerous quacks. But beyond that they are terrible at pushing for higher quality, especially when quality is so much in the experience of a customer in a service-oriented business. Restaurant regulation keeps restaurants reasonably safe, but there’s no pressure for Joe’s Tacos to use better cuts of beef, let alone to adopt molecular gastronomy. Yelp ratings do that in a way no regulator can hope to do.

I conclude that at the margin, the consumer needs to be paying a lot closer to full marginal cost of health care, or, equivalently, receiving the full financial benefits of any economies which he is willing to accept.

The health-care market – supply and demand

The obvious problem here is that the cash market is dead. Making people pay, and shop, is unrealistic.

If you walk in to the University of Chicago Hospitals and say, “I don’t have insurance. I have a bank account. I’ll be paying cash,” their eyes will light up. “We’ll pay for 100 Medicare patients with this guy.” That’s like walking up to United Airlines and saying “I want to go to Paris, first class. Sell me a ticket.” Actually, it’s worse – at least United will quote you a price up front and on its website, and let you compare with American. So, insurance companies now function as purchasing agents, negotiating complex deals on our behalf.

Nobody in this room really needs health insurance for anything less than catastrophes. We pay for transmission repairs, leaking roofs, and vet bills out of pocket. Most people in this room send our kids to private schools, throwing away our right to expensive public education. We could easily “afford” most of our routine medical expenses, and even pretty big unplanned expenses, especially if we were paying commensurately lower health-insurance premiums.

But we all have health insurance, and deal with the paperwork nightmare.

Why? You don’t need an “insurance” company to negotiate your cellphone contract, home repair and rehab, mortgage, airline fare, legal bills, or clothes, and pay as we do for health. Why do you and I need a professional negotiator masquerading as an insurance company? Moreover, Dr. Jones is in Humana’s network, Dr. Smith is in Blue Cross’. What economic principle means I shouldn’t see Jones, just because some arcane negotiation took place behind the scenes? And what about the new low-cost specialty clinic that Dr. Thomas is setting up, which can’t get into either network?

The answer: we’re missing robust supply-side competition. Hospitals would never get away with obscure pricing, hidden rebates, or massive cross-subsidies if they were facing serious competition from new entrants who could peel you away – and peel you away from your expensive “price negotiator” as well.

The cash market is also dead because of the demand-side distortion: too many people have payment plans. Competing for cash customers just does not make enough money to keep a hospital going, and the pool of such customers is a lot sicker.

And a hospital must choose, basically to be all insurance or all cash. If it offers clear transparent prices to consumers, it can’t also play the game with insurance companies. (The spread of “concierge medicine,” the equivalent of private schools for people so fed up they just throw away health insurance, is an interesting phenomenon. But it’s still too small to affect the overall market. There aren’t any concierge, cash-only hospitals. That business seems to have to be off-shore. )

In a vicious circle, the absence of a functional cash market lies at the heart of many insurance and government “cost control” problems. Insurance functions best when it is a small part of a market, in which prices are set by “marginal consumers” paying cash, and competitive businesses supplying them.

With little price discovery left in health care, health insurers have to do all the price negotiation in a vacuum.

Airlines, restaurants, and car repair work reasonably well even though in each case a large fraction of consumers are not paying with their own money. Each has competitive supply, and a remaining fraction of consumers who feel marginal decisions, enough to allow price discovery and competitive pressure for efficiency.

The cash market is also dead, because of the vast system of cross-subsidies and implicit taxes in our health-care markets. Medicare and Medicaid pay less than cost. Protected insurance companies go along with partially cross-subsidizing them. The poor solvent cash customer cross-subsidizes everyone else.

Part of the reason for phony pricing is that hospitals know most “cash” customers won’t end up paying, so they will end up negotiating charity care. Nicholas Kristof’s story17 in the New York Times, of the travails of an uninsured friend who got cancer, unwittingly illustrates my point beautifully. The article cites completely ridiculous prices, then explains how his friend applied for charity care and had a $5500,000 bill knocked down to $1,339. But, just to reiterate how ridiculous the cash pricing is, wanted to charge $1,400 for an ambulance ride.

In sum, freeing up either supply or demand without freeing up the other will do little good. Increasing copays can help to ration expensive or overpriced services, but does not stimulate supply or efficiency as long as new entrants can’t come in and compete for business. And allowing new entrants in doesn’t do any good as long as few consumers are able to vote with their money.

We need to free up supply, demand, and health insurance!

Health Insurance

If cash markets were functional, health insurance could become what it should be: a way of protecting lifetime wealth from catastrophic shocks, like life insurance. Such insurance would, of course, be a lot cheaper. It would not have to be a negotiator and payment plan for routine expenses.

“Access” should mean a checkbook and a willing supplier, not a Federally Regulated payment plan. Insurance means your large-scale standard of living isn’t enormously impacted by rare events.

If there were functional cash markets, health savings accounts could substitute for much of the necessarily cumbersome functions of insurance. Health borrowing accounts, i.e. HSAs with a preapproved line of credit, which you can tap for unexpected expenses but are not insurance in the sense of transferring overall wealth, would help even more. Without functional (competitive) cash markets, HSAs are not that helpful.

Generic objections

The idea that health care and insurance can and should be provided by deregulated markets, and that existing regulations are the main source of our problems, is, perceived to be fairly radical within the current policy debate. Let me deal with a few of the standard objections.

The poor

“What about the homeless guy with a heart attack?”

Let’s not confuse the issue with charity. The goal here is to fix health insurance for the vast majority of Americans –people who buy houses, cars, and cell phones; people who buy insurance for their houses and life insurance so their families.

Yes, we will also need charity care for those who fall through the cracks, the victims of awful disasters, the very poor, and the mentally ill. This will be provided by government and by private charity. It has to be good enough to fulfill the responsibilities of a compassionate society, and just bad enough that few will choose it if they are capable of making choices. I wish it could be better, but that’s the best that is possible. For people who are simply poor, but competent, vouchers to buy health insurance or to refill health savings accounts make plenty of sense.

But supplying decent charity care does not require a vast “middle-class” entitlement, and regulation of health insurance and health care for everyone in the country, any more than providing decent homeless shelters (which we are pretty scandalously bad at) or housing subsidies for the poor (section 8) requires that we apply ACA style payment and regulation to your and my house, Holiday Inn or the Four Seasons. To take care of homeless people with heart attacks, where does it follow that your and my health insurance must cover first-dollar payment for wellness visits and acupuncture? The ACA is hardly a regulation minimally crafted to solve the problems of homeless people with heart attacks!

The straw man

There is a more general point here, which will appear time and again as I answer each criticism. The critics adduce a hypothetical anecdote in which one person is ill served, by a straw-man completely unregulated market, which nobody is advocating, with no charity or other care (which we’ve had for over 800 years18 , long before any government involvement at all). They conclude that the anecdote justifies the thousands of pages of the ACA, tens of thousands of pages of subsidiary regulation, and the mass of additional Federal, State, and Local regulation applying to every single person in the country.

How is it that we accept this deeply illogical argument, or that anyone in making it expects it to be taken seriously? If you can find one person who falls through the cracks, the government gets to regulate the whole market, not that we craft a minimal solution to fix that person’s problem.

But wait, will not one person fall through the cracks or be ill-served by the highly regulated system? If I find one Canadian grandma denied a hip replacement, or someone who can’t get a doctor to take her as a medicare patient, why do I not get to conclude that everyone must be left to the market?

Adverse selection

We all took that economics course, in which asymmetric information makes insurance markets impossible due to adverse selection. Sick people sign up in greater numbers, so premiums rise and the healthy go without. George Akerlof’s justly famous “Market for lemons” proved that used cars can’t be sold because sellers know more than buyers. Interestingly, Car Max is still in business.

Does a patient, with knowledge of aches and pains, really know so much more about likely cost than an insurance company, armed with a full set of computerized health records and whatever tests it wants to run? Life, property, and auto insurance markets at least exist, and function reasonably well despite the similar theoretical possibility of asymmetric information. Life insurance is also “guaranteed renewable,” meaning you are not dropped if you get sick.

Now, the “adverse selection” phenomenon, that sick people are more likely to buy insurance, and healthy people forego it, is a big problem. But the insurance company charges the same rate, not because it can’t tell who is sick – a fundamental, technological, and intractable information asymmetry. The insurance company charges the same rate because law and regulation force it not to use all the information it has. If anything, we have the opposite information problem: insurers know too much.

This source of adverse selection is a legal and regulatory problem, not an information problem, and easily solved. If insurance were freely rated, nobody would be denied. Sick people would pay more, but “Health status” insurance shows how to solve that. See footnote 2 for references.

Shopping paternalism.

Defenders of regulation reiterate the view that markets can't possibly work for health decisions,19

“A guy on his way to the hospital with a heart attack is in no position to negotiate the bill.” “One point I cannot agree with is that competition can work in health care, at least as it does in other markets. I cannot fathom how people faced with serious illness will ever make cost-based decisions”

“What about those who currently don't have the background and/or the economic circumstances to consume health care, (e.g. take anti-hypertensive medicine instead of [buying] an iphone)?”

Ezra Klein,20 trying to understand why health-care prices are so high and so obscure,

Health care is an unusual product in that it is difficult, and sometimes impossible, for the customer to say “no.” In certain cases, the customer is passed out, or otherwise incapable of making decisions about her care, and the decisions are made by providers whose mandate is, correctly, to save lives rather than money.

In other cases, there is more time for loved ones to consider costs, but little emotional space to do so — no one wants to think there was something more they could have done to save their parent or child. It is not like buying a television, where you can easily comparison shop and walk out of the store, and even forgo the purchase if it’s too expensive. And imagine what you would pay for a television if the salesmen at Best Buy knew that you couldn’t leave without making a purchase.

New York Times columnist Bill Keller put it clearly, in “Five Obamacare Myths:21

[Myth:] The unfettered marketplace is a better solution. To the extent there is a profound difference of principle anywhere in this debate, it lies here. Conservatives contend that if you give consumers a voucher or a tax credit and set them loose in the marketplace they will do a better job than government at finding the services — schools, retirement portfolios, or in this case health insurance policies — that fit their needs.

I’m a pretty devout capitalist, and I see that in some cases individual responsibility helps contain wasteful spending on health care. If you have to share the cost of that extra M.R.I. or elective surgery, you’ll think hard about whether you really need it. But I’m deeply suspicious of the claim that a health care system dominated by powerful vested interests and mystifying in its complexity can be tamed by consumers who are strapped for time, often poor, sometimes uneducated, confused and afraid.

“Ten percent of the population accounts for 60 percent of the health outlays,” said Davis. [Karen Davis, president of the Commonwealth Fund] “They are the very sick, and they are not really in a position to make cost-conscious choices.”

Now, “dominated by powerful vested interests and mystifying in its complexity” is a good point, which I also just made. But why is it so? Answer: because law and regulation have created that complexity and protected powerful interests from competition. And is the ACA really creating a simple clear system that will not be “dominated by powerful vested interests?” Or is it creating an absurdly complex system that will be, completely and intentionally, dominated by powerful vested interests?

But the core issue is these consumers who are “passed out, or otherwise incapable of making decisions about her care,” ”strapped for time, often poor, sometimes uneducated, confused and afraid,” and “not really in a position to make cost-conscious choices.”

Yes, a guy in the ambulance on his way to the hospital with a heart attack is not in a good position to negotiate. But what fraction of health care and its expense is caused by people with sudden, unexpected, debilitating conditions requiring immediate treatment? How many patients are literally passed out? Answer: next to nothing.

What does this story mean about treatment for, say, an obese person with diabetes and multiple complications, needing decades of treatment? For a cancer patient, facing years of choices over multiple experimental treatments? For a family, choosing long-term care options for a grandmother?

Most of the expense and problem in our health care system involves treatment of long-term, chronic conditions or (what turns out to be) end-of-life care, and involve many difficult decisions involving course of treatment, extent of treatment, method of delivery, and so on. These people can shop! We actually do a pretty decent job with heart attacks.

And even then... have they no families? If I’m on the way to the hospital, I call my wife. She’s a heck of a negotiator.

Moreover, health care is not a spot market, which people think about once, at 55, when they get a heart attack. It is a long-term relationship. When your car breaks down at the side of the road, you’re in a poor position to negotiate with the tow truck driver. That’s why you join AAA. If you, by virtue of being human, might someday need treatment for a heart attack, might you not purchase health insurance, or at least shop ahead of time for a long-term relationship to your hospital?

And what choices really need to be made here? Why are we even talking about “negotiation?” Look at any functional, competitive business. As a matter of fact, roadside car repair and interstate gas stations are remarkably honest. In a competitive, transparent market, a hospital that routinely overcharged cash customers with heart attacks would be creamed by Yelp reviews. Competition leads to clear posted prices, and businesses anxious to give a reputation for honest and efficient service.

It’s not even a realistic anecdote.

OK, some conditions really are unexpected, and incapacitating. Not everyone has a family. There will be people who are so obtuse they wouldn’t get around to thinking about these things, even if we were a society that let people die in the gutter, which we’re not, and maybe some hospital somewhere would pad someone’s bill a bit. But now we’re back to the straw man fallacy. The idea that ACA is a thoughtful, minimally designed intervention to solve the remaining problem is starting to look more and more ludicrous.

Take a closer look at Keller and Davis’ statement, ”strapped for time, often poor, sometimes uneducated, confused and afraid,” and “not really in a position to make cost-conscious choices.”

We’re talking about average Joe and Jane here, sorting through the forms on the insurance offerings to see which one offers better treatment for their diabetes-related complications. If Joe and Jane can’t be trusted to sort through this, how in the world can they be trusted to figure out whether they want a fixed or variable mortgage? Which cell phone or cable plan to buy? To deal with auto mechanics, contractors, lawyers, and financial planners? How can they be trusted to sign marriage or divorce documents, drive, or ... vote?

We have a name for this state of mind: legal incompetence. Keller, Davis, and company are saying that the majority of Americans, together with their families, are legally incompetent to manage the purchase of health insurance or health care. And, by implication, much of anything else.

Yes, there are some people who are legally incompetent. But straw man again, Keller and Davis are not advocating social services for the incompetent. They are defending the ACA, which applies to all of us. So, they must think the vast majority of us are incompetent.

This is a breathtaking aristocratic paternalism. Noblesse oblige. The poor little peasants cannot possibly be trusted to take care of themselves. We, the bien-pensants who administer the state, must make these decisions for them.

Let me ask any of you who still agree, does this mean YOU? When you are faced with cancer, do you really want to place your trust in the government health panel? Or is this just for the benighted lower classes, and you, of course, know how to find a good doctor and work the system?

And choice is always between alternatives. Sure, some people make awful decisions. The question is, can the ACA bureaucracy and insurance companies really do better? Yet you would not trust them to buy your shirts?

Really? Does this entire bureaucratic garganuta follow, not on the proposition that there is some fundamental economic market failure, but because…Americans are no good at shopping?

If anything, the opposite seems to be true. Where is it easier to shop, Southwest Airlines, or your average hospital? In the name of the consumer, who finds it hard to shop, we have created an arcane system where it is, in fact, nearly impossible to shop.

No. It’s not “health is too important to be left to the market.” It’s “health is so important --and so varied, so personal, and so subjective – that it must be left to the market.” If you don’t trust the vast majority of people to make the most important decisions of their lives, you’re a devout patrician, not a “devout capitalist.”

Theory and experience

I’m often told, “Well, fine, but this is just theory. Free market health care hasn’t been tried in a modern economy. All countries regulate health care or governments provide it.”

That’s the point of my extensive examples of other industries. As an economic good, there really isn’t much difference between health care and other complex personal services such as auto repair, legal services, home repair and remodeling, or college education. Yet these markets no not require “insurance” for access, nor must bureaucracies decide what every American “needs” even though the providers have considerably more expertise than the customer. Once upon a time all governments had monarchies. That observation didn’t prove monarchy was a better system.

Moreover, the pockets of health care that are out of the insurance system and allowed relatively competitive free entry operate reasonably well. Plastic surgery and dentistry are not disasters. Radial keratotomy (corrective eye surgery) is a good example, as specialization and competition has led both to lower costs and increased quality. I am not the first dog owner to notice how easy and relatively inexpensive cash-and-carry veterinary medicine is compared to the same treatment for humans. Concierge medicine is taking off.

If anyone is guilty of theorizing, it would seem to be the faith that the next round of brilliant ideas for layering on ACA-style regulation will lead finally to successful “cost control” that is not simply rationing, or will induce the radical quality improvement and innovation that we need, where the past ones have all failed.

Realistic freedom, help and vouchers

I do not require that you follow me to some unrealistic libertarian nirvana. “The unfettered free market,” where the improvident die in the gutter is another ridiculous straw man. Southwest’s pilots have FAA licenses. Walmart’s products pass the consumer product safety commission. We can argue about this stuff, but we don’t have to. A little freedom will go a long way. The market can survive a lot of regulation, even silly regulation.

In addition to the need for genuine charity care, there can stlll be lots of help in various places.

But a central principle of economics is, “don’t transfer income by distorting prices, or providing services.” The vast majority of any help and transition-smoothing can and should be given in the form of on-budget, lump-sum subsidies or vouchers, leaving marginal incentives intact.

When we transition to freely-rated lifelong individual insurance, individuals who are already sick face high premiums. That problem is easily solved with a voucher, or a lump-sum payment to their health savings accounts

The same principle applies to genetic diseases. Economics has long recognized the principle that insurance can’t insure events that have already happened, so lump-sum transfers are appropriate. But one-time, lump-sum transfers based on clearly defined events over which no one has control, such as a DNA marker, are much less distorting, or subject to abuse, than perpetual regulation and intervention in a market.

If we want to subsidize health care or insurance for old or poor people, give them a voucher. There is no reason the government should try to run an insurance company, and less reason to do it pass an implicit tax, by mandating that businesses “provide” insurance.

If we want to subsidize emergency rooms, let’s just do it. That will be much more efficient than forcing a big cross-subsidy scheme and blocking competition to keep them afloat. (Letting Walmart set up clinics would be a lot cheaper too!)

If you think people don’t get enough checkups when paying with their own money, give them a voucher. That’s much easier than passing a mandate that every company must provide first-dollar health payments with a long range of mandated benefits.

More generally, there is an income paternalism at work in health care policy, somewhat more reasonable than the “they can’t shop” paternalism I decried above, worth making explicit. Most people, when spending their own money at the margin, are likely to choose less health care than we, the self-appointed advisers to “policy-makers” would like. Already, they evidence tradeoffs that imply less health than we would like – they drink sugared sodas, eat fast foods, and don’t exercise enough. In my example that patients were offered an MRI or $1,000 in cash, I think we suspect that a lot of patients would choose the cash.

A good libertarian would say, well, let people choose more iphones and less health if that’s what they want. But we don’t have to have this argument. If you think people will spend too little on health overall, give them vouchers in a health-savings account. This maintains the efficiency of patient-driven choice, distorts the overall health vs. non-health price, without distorting relative prices or writing ten thousand pages of regulations and supply-side restrictions that gum up the entire system.

Now, you might object, that all these subsidies and vouchers will raise “costs” on the budget. But this happens simply because of phony accounting. If the government mandates that cardiac patients cross-subsidize emergency rooms this is exactly the same as a tax on cardiac services and an expenditure on emergency rooms. Actually, it’s a lot worse because the distortion of the current system is much greater. So any economically relevant accounting would recognize that we save money overall. Fixing the accounting is a lot better and cheaper project than keeping our ridiculously inefficient health care system.

"Politically feasible"

Well, my typical critic concludes, maybe you’re right about all this as a matter of economics, but it’s not politically feasible.

No, not now. But the alternative is not economically feasible, a sterner taskmaster. And what was not feasible today, can quickly become feasible tomorrow if it is correct, and once people understand it, and understand there is no option. Our job as economists is to figure out what works and explain it, not to bend reality to some notion of what today’s politicians are willing to say in public.

Our political conversation is truly lunatic. It is taken for granted in policy discussion that no American can be asked to “pay for” (directly, rather than through taxes) one cent of health cost risk. While they routinely pay for broken and crashed cars, destroyed houses, suffer huge risks in the job market, and shoulder housing, transport and other expenses much greater than the cost of health care. Yet while pretending nobody should pay for things, unfortunates who fall through the cracks can be handed ridiculous $550,000 bills for cancer treatment.

We can start by saying, out loud, health care is a good like any other. It is ok to ask Americans to pay for it, and to allow American companies to competitively supply it, just like any other. It is ok for insurance to retreat to its proper role, of protecting people from large shocks to wealth, rather than being a hugely inefficient payment plan. As car insurance does not pay your oil changes – after you fax in the forms in quintuplicate, obtain permission from your mechanic, go to the in-network mechanic, and wait 6 weeks, and answer a 20 page questionnaire about your repair history and driving habits.

Bottom line

Health care is a complex personal service, with wide variation in quality, both along measures of health outcomes and along more subjective dimensions of satisfaction. Its demand curve is very elastic – people will consume a lot at subsidized prices. The distinction between “want” and “need” is conceptually fuzzy, and nearly impossible to measure.

The big improvements in health care come from better technology. But big improvements in its delivery and average quality are also attainable. They come from much better human organization, as has happened recently in many other industries that have witnessed revolutionary supply competition. Yet achieving those improvements will displace lots of entrenched interests.

From these observations, simple conclusions follow.

Health care markets need a big supply-side revolution, in which the likes of Southwest Airlines, Walmart and Apple enter, improving business practices, increasing quality and transparency, and spurring innovation. And disrupting the many entrenched interests and cross-subsidies of the current system.

I outlined a long string of restrictions on competition that can be repealed, or modified to allow competition. At a minimum, every new regulation should be evaluated by its effect on competition by new entrants, or protection of incumbents, a consideration not even spoken in policy discussion today.

Health care is singularly ill-suited to payment-plan provision, either by government directly or by heavily regulated insurance by a few large well-protected businesses. A functional cash market must exist in which patients can realistically feel the marginal dollar cost of their treatment, or (equivalently) enjoy the full financial benefits of any economies of treatment they are willing to accept, and are not patsies for huge cross-subsidization and rent-seeking by an obscure system negotiated behind the scenes between big insurance companies, hospitals, and government.

Both supply and demand must be freed. Without supply competition, asking consumers to pay more will do little to spur efficiency. Without demand competition, new suppliers will not be able to succeed.

The alternative, doubling down regulations on an already highly regulated system, full of protected and politically connected incumbents and rent-seekers, has little chance of achieving these goals. Whether in the post-office model (government provision), or the 1950s-style regulated airline, utility or bank model (the ACA) this effort will just produce less efficiency, more costs, and another generation of bright ideas dashed. Oh ye reformers, remember that the last 20 bright ideas did not fail simply because the people in charge weren’t as smart as you are, or as well-meaning!


1 John Cochrane is a Professor of Finance at the University of Chicago Booth School of Business, Adjunct Scholar of the Cato Institute, Senior Fellow of the Hoover institution and Research Associate of the NBER. Address: 5807 S. Woodlawn Chicago IL 60637,, This is an expanded version of remarks given at the conference, “The Future of Health Care Reform in the United States,” at the University of Chicago Law School, October 12 2012. I am grateful to Anup Malani for extensive and very helpful comments.
2 “Health-Status Insurance.” Cato Institute Policy Analysis No 633.(2009); “Time-Consistent Health Insurance” Journal of Political Economy, 103 (1995), 445-473; "What to do about pre-existing conditions Wall Street Journal August 14 2009; “Forget about the mandate” Bloobmerg Business Class July 12 2012; What to do on the Day after Obamacare Wall Street Journal April 2 2012; “The Real Trouble With the Birth-Control Mandate” Wall Street Journal February 9 2012; all available on my website, see footnote 1.
3 If personal experience is not enough to remind you how inefficient the current system is, I recommend Jonathan Rauch’s YouTube video, “If air travel worked like health care” Hat tip to Einer Elhauge who showed it at the conference.
4 Gawadne, Atul, 2012, “Big Med: Restaurant chains have managed to combine quality control, cost control, and innovation. Can health care?” New Yorker (August 13)
5 Chandra, Amitabh, and Jonathan Skinner, 2012, “Technology growth and expenditure growth in health care,” Journal of Economic Literature 50, (September 2012) 643-680.
9 Elhauge, Einer, ed. (2010) The Fragmentation of U.S. Health Care -- Causes and Solutions Oxford: Oxford University Press, quotes from the introductory chapter available at
10 Lakdawalla, D., and T. Philipson (2006), “Non-Profit Production and Industry Performance”, Journal of Public Economics, v 90 (9), 1681-98.
11 Fama, Eugene F. and Michael Jensen, 1983, “Agency Problems and Residual Claims” Journal of Law and Economics, 26, 327-349. Fama and Jensen note that the presence of donors on boards of directors is an imperfect substitute for knowledgeable insiders and market discipline.
12 For a description of the process, with however a view that it needs more not less regulation, see, Horwitz, Jill R. 2012, “State Oversight of Hospital Conversions: Preserving Trust or Protecting Health?” The Hauser Center for Nonprofit Organizations, The Kennedy School of Government,
13 For an example of recent news coverage see “Regulators Seek to Cool Hospital-Deal Fever” Wall Street Journal, March 18, 2012.
14 At the conference, Meridith Rosenthal gave a wonderful presentation highlighting a wide range of complex payment schemes, and how they didn’t work out, a wider range of bright new ideas, and how little we know about how they work. Her conclusion was that lots more research will lead to something workable to patch up each leak. Mine was that jiggering health payment systems is the best modern example of the hopelessness of central planning. You can get some idea from Rosenthal, Meredith B., 2009, “What Works in Market-Oriented Health Policy?” New England Journal of Medicine 360, 2157-2160 (May 21) And especially the table in Rosenthal, Meredith B., 2008, “Beyond Pay for Performance — Emerging Models of Provider-Payment Reform,” New England Journal of Medicine 359, 1197-1200 (September 18, 2008)
15 For example,, complete with a link, “Read the tragic story of a now brain-damaged young woman who had 2 sets of x-rays - yet no one diagnosed her brain abscess,” and “Outsourcing radiology abdicates 3 of 4 of the core responsibilities of radiologists.
17 Kristof, Nicholas D. “A Possibly Fatal Mistake,” New York Times October 12 2012, Sunday Review,
18 One reference: Founding of the Misericordia charitable hospital in Florence, 1244.
19 These quotes are from commenters on my blog, not a very authoritative source, but they put the view so clearly I couldn’t resist.
21, July 15 2012


  1. "if we could get rid of the ACA and Dodd-Frank, what would we replace them with?"

    A free market?

    1. Actually this would result in the single most dramatic decrease in health care costs ever. Insurance insulates people from the true cost of care. People complain about a $20 co-pay for a drug or a doctor's visit? Imagine their reaction if they paid the whole bill. Same for tests, hospital services, etc.

      And most people are insulated from the cost of the insurance since their employer pays for it. The consumer is twice-removed from the cost of care. What happens when people are spending someone else's money? The U.S. Postal Service, or just about any other government program anywhere in the world.

      I wonder if perhaps a system could be implemented whereby insurance paid a set fee for a given medical service and providers of services could charge whatever they wanted above that. If someone was really good and in demand they could command a premium over the floor price and there would be an incentive to be in demand, i.e., provide premium service. Right now the worst doctor usually gets the same fee as the best one.

      I've told many of my colleagues who want a free market in medical care that they really don't want to find out their true value.

  2. There is something truly remarkable and puzzling about this long, sincere, disjointed essay on US health care reform: there's no mention of the fact that every other advanced country is able to provide healthcare for its people at least comparable and in some cases superior to what we can manage, at half our per capita cost or less. These countries also provide universal care for their citizens. We don't for ours. Half the cost! Universal care! Have they already adopted most or many of the good professor's policy prescriptions? Have they minimized government involvement and regulation in their health care systems and optimized for a free insurance market with the least oversight? Have they banned unions for doctors and other health care workers? Not at all!

    Shouldn't we at least look take a look at those systems that provide expanded coverage of comparable quality and succeed in cost management compared to us for ways to reform our system? Why is this not the place to start?

    1. Actually that is there, but it's so long and disjointed you didn't get to it. See "theory and experience." Also, believe it or not, I cut out a long section on "I went to France and they took care of my toothache quickly cheaply and politely, why can't we have that." It would have made it even more long and disjointed. I'll have to save that for another day. Short version, talk to any european/canadian with something seriously wrong.

    2. Also, even if it does work, don't forget the cost. You can have European benefits. If you pay European taxes. 40% payroll taxes, 50% income taxes, 20-23% VAT (sales) tax on everything you buy, $6/gallon and up gas... and it's still quite enough as they're having a debt crisis.

    3. Talk to anyone "with something seriously wrong" and you're unlikely to get an upbeat evaluation. And I think many studies show a universal dissatisfaction, though the particular problem areas differ from one system to another. But if we concede the point that other advanced countries offer a similar quality of care and similar outcomes to what we provide, only it costs us a lot more, why not look to them for efficiencies we could adopt here and not be restricted by the shackles of libertarian (or other) ideology?

      (Sure, Europeans pay more taxes. But at least in health care they're getting a lot more value for the cost than we are, no?).

    4. Europe's taxes are not the cause of its problems, which is well know to those of us skilled in location economics.

      Miles Kimball has this information on his blog

      Differences in metropolitan populations may help explain gaps in productivity and incomes. Western Europe’s per-person GDP is 72% of America’s, on a purchasing-power-parity basis. A recent study by the McKinsey Global Institute, the consultancy’s research arm, reckons that some three-quarters of this gap can be chalked up to Europe’s relatively diminutive cities. More Americans than Europeans live in big cities: there is a particular divergence in the size of each region’s “middleweight” cities, those that teem just a little less than the likes of New York and Paris (see chart). And the premium earned by Americans in large cities relative to those in the countryside is larger than that earned by urban Europeans.

    5. I don't think Sweden, Austria, Finland, Netherlands, Switzerland, Denmark, Canada, UK are having any sort of debt crisis

      Yet they are able to provide healthcare for roughly half the cost. So, perhaps we should really try to copy what it is that they are doing well, without necessarily replicating bad parts of their systems.

    6. Europe and Canada (and others) do not research, invent, innovate and produce by a longshot the amount of new technologies, new medicines and new procedures that are invented and developed in the United States. All those countries with nationalized health care systems they just take without producing, they take advantage of the externalities, without paying for it.
      And try to sue a doctor for malpractice in Europe or Canada. See how far you can go. My wild guess? You will not go even 10% as far as in the United States. And also, see how much malpractice insurance a UK doctor or Holland doctor carries, and how he has to pay in premia.
      Thus, before we all engage in Canada and Holland and UK healthcare worshipping, let us all compare apples to apples, as good economists do.

    7. "Europe and Canada (and others) do not research, invent, innovate and produce by a longshot the amount of new technologies, new medicines and new procedures that are invented and developed in the United States."

      This is just a myth. Countries like Germany, France and in particular Switzerland do much more research and innovation than the US on a per capita basis.

    8. John wrote:

      "There is something truly remarkable and puzzling about this long, sincere, disjointed essay on US health care reform: there's no mention of the fact that every other advanced country is able to provide healthcare for its people at least comparable and in some cases superior to what we can manage, at half our per capita cost or less."

      Cochrane replied:

      "Also, even if it does work, don't forget the cost. You can have European benefits. If you pay European taxes. 40% payroll taxes, 50% income taxes, 20-23% VAT (sales) tax on everything you buy, $6/gallon and up gas... and it's still quite enough as they're having a debt crisis."

      With respect to the cost factor, John did posit that European countries provide health care at "half the cost", so he clearly did not "forget about the cost". I suspect the "half the cost" is exaggerated; however, I would also suspect that the average cost of healthcare is less per capita in most European countries than it is in the US. Whether health care is financed through taxes or through private payments (most often via insurance intermediaries) one will pay one way or the other. The other claim John makes that those countries have equal or better healthcare outcomes is likely also hyperbole, particularly when one excludes the negative effects of the higher US crime and obesity rates, etc. When these are excluded, studies show the effectiveness of care is probably better (costs aside).

      Data in this area is notoriously inaccurate; but, isn't the real issue which system is more efficient and therefore cheaper without sacrificing quality? If John is right that the per capita cost in European countries is cheaper than in the US, they must be doing something right, or at least not doing things as badly.

      The problem in comparing these various systems is largely that nearly all of them are hybrids. There is a common misperception that "universal care" equates with government-supplied health care. In fact, most European countries (with the exception of the UK) deliver more healthcare directly through the private sector than does the US. The US government, via Medicare, Medicaid, the VA, TriCare, CHIPS, etc. has a much heavier hand in the market than does, say, the Netherlands which has no equivalent of Medicare or Medicaid.

    9. ""Europe and Canada (and others) do not research, invent, innovate and produce by a longshot the amount of new technologies, new medicines and new procedures that are invented and developed in the United States."

      This is just a myth. Countries like Germany, France and in particular Switzerland do much more research and innovation than the US on a per capita basis."

      Well, actually, the myth is mostly true. A quick look suggests that two researchers, Burke and Monot, have done quite a bit of work on estimating the total health R&D spent by various countries. Here's one source that suggests, comparing R&D to GDP, the US ranks only below Iceland, Switzerland, Sweden and Denmark. If per capita spending is the measure, the US would likely place above Sweden and Denmark. I don't know about Iceland; however, spending in tiny Switzerland is explained by the presence of some very large pharma companies resident there (perhaps, logically, for tax reasons).

      You may also consult this source (focus on Figure 6) which shows the US spends more than half of total health R&D spending. Canada, is way, way down the list.

      Of course, this is the wrong question. The amount spent on R&D does not necessarily factor in to how high per capita care is in a given country. The question you should be asking is why drugs and other relevant hardware, particularly those manufactured by US companies, cost so much more in the US than many other places in the world to which those goods are exported.

    10. Vivian: I agree with your reply to Anonymous 1:40 PM.
      I would only pin it down further. Where are most of the drugs developed, is it in the US, or is it in Europe or Japan? Where is most of the medical hardware, medical technology developed, is it in the US or Europe or Japan? Where is most of medical innovation taking place?
      I am not completely sure that R&D medical or health expenditure really measures this; I think one has to be more specific that just such expenditure (like per capita patents in medical technology and pharma?).
      As for Switzerland: I cannot prove it, but I have this gut feeling that, even though the Swiss do have some well known pharma companies, where are most (or at least a big chunk) of their research labs sitting? Are they really in Switzerland?

      As to your final question: why are prices of medical hardware and drugs so much higher in the US than in the rest of the world? Because, I think, the US consumer absorbs most (if not all) of the R&D innovation cost and expenditure, and the rest of the world (much of it with nationalized health care systems) free rides on such innovation and inventions. The US consumer basically pays for the rest of world to enjoy that positive externality.

      I think it was a few years ago that Michael McClellan (Medicare Administrator I think under GW Bush) once asked/demanded/requested that other countries (like the ones in Europe) pay higher prices for US medical innovation in technology and drugs. In other words, he was explicitly addressing that free-ridership issue that other countries enjoy and US health care consumers have to pay for.
      I wish I had a link or citation to show this.

    11. It may be the case that Swiss companies DO the R&D in Switzerland, but then make their money on it in the US market. This sounds like a reasonable possibility for pharmaceuticals. What matters is who pays for the innovation, not where the innovation is located. (Well, what matters for this discussion, obviously it's not a great thing for the US if all the high tech innovation is moving elsehwere.)

    12. "It may be the case that Swiss companies DO the R&D in Switzerland, but then make their money on it in the US market."

      That may be the case. On the other hand, it may very well be the case that they don't make the bulk of it in Switzerland *or* the US.

      One of the interesting facets of intellectual property is that it is highly mobile. Under US tax rules, for example, a patent is considered owned and royalties taxed in the place where that patent has been financed (that is to say, the location where the risks and rewards are borne). If a US pharmaceutical company wishes to put profits in, say, Ireland, one of it's Irish subsidiaries might finance R&D to be conducted in the US (or somewhere else). That Irish company owns the patent under this contractual arrangement and is therefore able to charge royalties on the manufacture and/or sale of the drug and to be taxed at the relatively low tax rates in Ireland. The Tax Code tries to backstop this through "Subpart F", but meeting the "active licensing" exception to those rules is quite possible.

      The same is probably true of Switzerland and other countries that do not have as sophisticated anti-avoidance rules as does the US (e.g., Subpart F). Even though the rate of tax in Switzerland is comparatively modest, it can, and is, often improved through this type of planning. It is certainly not to the advantage of most companies (particularly Swiss ones) to make most of their money in the US due to the relatively high corporate income tax rates here. We often here of the disadvantage US business has due to the relatively high corporate income taxes. What is often missed in these analyses is the amount of income (and tax revenue) that is missed because foreign corporations plan around those rates, too. If the US rates were more competitive, there would be less incentive to do so.

  3. Thank you for a thoughtful well done piece.

    When I make similar arguments as yours, the main objection I receive (mostly from doctors, naturally) is a free market doesn't work in health care. When I respond it works in Lasik and plastic surgery they go silent. Are there other examples you can give of success of the free market in providing health care?

  4. Prof. Cochrane:

    John, wrote, above, "There is something truly remarkable and puzzling about this long, sincere, disjointed essay on US health care reform: X"

    There is something else truly remarkable and puzzling about this long, sincere, disjointed essay on US health care reform: the idea market you describe is perfectly legal now and has been for the last 100 years, save the certificate of need laws. That fact means you should have asked, why is the system as you describe it?

    The answer is very simple. The system is designed and functions to make up for the unwillingness of people to pay the taxes directly necessary to pay for healthcare for the elderly and poor. The system is designed and intended to create a cross subsidy: those with health insurance (whose care is subsidized by the employer deduction and exclusion from income provisions of the tax code) make up for taxes they would otherwise have to pay for the elderly and poor. The cash buyer, who has no political clout, is the one who isn't hidden behind a tree, hiding (so to speak) from the tax man.

    I will start listening to what you have to say when you become serious and explain how you would raise taxes to pay for the cross subsidy.

    I don't for a moment believe that you would pay higher taxes so as to eliminate the cross subsidy. You want to cut taxes and send around carts with drivers yelling, as in India, throw out your dead.

  5. Yes, the PDF version is better: never seen a youtube video academically footnoted like this!

  6. From physician's perspective this is the first article about health care that even came close to understanding the problem. It may seem disjointed but the problem is disjointed.

    Comparing health care delivery to things like Wal-Mart and Southwest Airlines is not valid. It reminds me of the time that Bill Gates criticized the airline industry and said it should be more like the computer industry, to which an airline executive replied that he didn't think the public would like airplanes that crashed several times per day.

    The biggest difference is that people are biological systems. It is not like moving objects from point A to point B, nor is it like repairing a car. Patients are unique and seen serially. You have to work on people while the motor is running and you can't just disassemble them. There are complaints for which there are no tests. There is no gadget that you can run someone through that tells you where their pain comes from or why somebody is hallucinating or depressed.

    People want a personal relationship with their doctor and the SWA model certainly doesn't provide that. You get the same peanuts and the same width seat regardless of personal preference or body habitus. Unlike SWA, hospitals now provide extra-wide wheelchairs to accommodate the increasing number of morbidly obese patients, and they don't charge for an extra seat. They also offer diets tailored for the patient. You really don't want cookie-cutter care.

    I am so sick and tired of hearing about how great the other countries provide health care. I travel to Europe frequently and try to get to know the locals wherever I am. When they learn that I am a doctor they start talking about their health problems. We often keep in touch by email or snail mail. I am appalled at how their systems work and the level of care provided.

    Americans don't think they have had adequate care if they haven't undergone tests. I have literally lost patients because I recommended a treatment based just on the history and physical exam; they expected x-rays and MRIs and so forth.

    If you think Canadian health care is so wonderful I suggest you go to the web sites that publish wait times for various procedures and tests. No American would stand for that delay. Sweden literally would not allow a man to use his own money to pay for denied chemotherapy. If the British NHS was so great there wouldn't be a two-tiered system. I recall seeing a documentary where they went to Cuba and showed a man getting his broken leg set for $2. My first reaction was, "Are they going to come back and film how they deal with the subsequent mal-union?".

    I am singularly unimpressed by the whining about filling out forms. I'm a patient too. I go to the doctor, I fill out the forms. I don't bitch about it. It isn't that great a burden. Put on your big boy pants.

    Consider this: Do you really want your personal health information stored in a central data bank? They're so secure, and we don't have to worry about the government snooping on us. /sarcasm off

    Insurance plans actually encourage overspending by disconnecting the consumer from the cost. If I want to start a patient on a new drug they ask for a 30-day supply because they have a fixed co-pay for a month's supply. They get 30 pills @ $2/pill, take one and vomit it it up, and toss $58 worth of pills in the toilet. Suppose they had to co-pay 20%? They would ask for a 3-day trial @ $1.20.

    Somehow we need a scheme that has patients considering price without setting the price so high that they defer needed care.

    Consider yourselves lucky that there is a limit to post size. This is about 10% of my potential rant. ;-)

    1. "Put on your big boy pants". Doctors should probably say this to their patients more often.

      More generally in relation to this (refreshingly long) blog post, it's good to know all that excellent privatisation literature still has useful shelf life.

      Even if the supply side revolution in health care were only partial, the entry of e.g. Walmart and Apple would shake things up positively and make people cross about cross-subsidies.

  7. Nothing but praise – for this thoughtful essay. The change is necessary but the change will become socially possible only after we drastically improve our rhetoric.

    But there is something profoundly different about health care if you compare it with Wal-goods (shirts) and Wal-services (restaurants) - it's that all medical outcomes are uncertain. Our ability to manage this variability is gradually improving but will always remain limited. What matters is that we are dealing with complex, resilient, adaptable systems - the human body and mind. When they fail - they fail in a multitude of complex and tangled ways. The system where inefficiency is punished by personal choices and market forces will likely improve the quality of care and perhaps even reduce the variability of outcomes - but only to a degree.

  8. Hmmm ...

    I like cash markets, they work. The absence of a cash market it a symptom of a Ponzi-like 'conduit scheme' which is what USA Health-care Inc. has devolved to. The patients are conduits between credit sources and beneficiaries, the patients receive abstractions while being on the hook for the debts taken on.

    Conduit Schemes:

    "Sticking with the current system is a bit like sticking with the Titanic because it has a nice piano bar. Best thing to do is walk away from the system en mass and its false promises and start thinking and acting independently of the establishment. If enough people must step aside the ... rackets will unravel. Schemes and bubbles require a constant flow of new funds. The education racket insists its product has value, how much value can it have when police start breaking heads in its defense?

    "Will the establishment throw people in prison in order to ‘insure’ that the same people have adequate (whatever that means) health care? There is something absurd when education and medical care can only be delivered at gunpoint."

  9. Thanks for a very thorough and thought-provoking essay. The world needs more examples, such as provided here, of why healthcare as we know it doesn't work and costs too much, and how introducing free-market incentives would make things much better. If I had to choose just one thing to change to get the ball rolling, it would be to get rid of the current tax preference that allows only employers to deduct the cost of healthcare insurance.

  10. Excellent piece. Really enjoyed it. Thanks for sharing

  11. """ every other advanced country is able to provide healthcare for its people at least comparable and in some cases superior to what we can manage, at half our per capita cost or less. These countries also provide universal care for their citizens."""

    Those countries provide universal *insurance*, and universal take-a-number, not universal medical care.

    The US system has cost problems and too many fall through the cracks. How could it not given the hurdles illustrated in the health-air video?

    But the quality of US medicine is, for most ailments (not all), second to none, something I is discovered when one actually gets sick and, after wading through the irrelevant chaff on 'lifespan' cast out by the planners, starts with the grim business of looking up of survival rates in literature for various countries, as I had to do.

  12. As a dog owner, I concur with your observation about veterinary care. I do think that dental care in the US is far less regulated than health care, and wonder if that market offers a sketch of what a more market driven healthcare system would look like (probably more like plastic surgery).

  13. Professor,

    Thank you for the excellent analysis. It may be disjointed, as some commenters have pointed out, but you had a disjointed

    set of problems to address.

    I note that you claim: "Health care is not that different from the services provided by lawyers, auto mechanics, home remodelers, tax accountants, financial planners, restaurants, airlines or college professors." I can think of several ways in which health care is different, and would appreciate your analysis of how these differences might impact the policy solutions you discuss.

    1. Health care contains a public good component that the other industries do not. If I contract a contagious disease, it should not be up to me what quality of care I should seek. Minimum standards must be maintained to protect others who might come in contact with me. Much of medical care is infection prevention/treatment. Licensing requirements are justified to protect us all from contagious infections. As you point out, when the government regulates an industry, the established firms and providers have the incentive to capture the regulatory agencies and use them to suppress competition.

    2. You talk a lot about the demand for health care vis-a-vis other industries. Correct me if I am wrong here, but my understanding is that demand derives from economic agents solving underlying constrained utility maximization problems. But nobody buys health care because it increases utility. They buy health care because they have suffered a shock to their health. It is poor health that lowers utility -- not lower consumption of health care (in other words, utility is a function of health and consumption of non-health products). Secondly, when a shock to a person's health occurs it lowers the person's productivity and this forces a tighter budget constraint as income decreases. These concepts are no doubt familiar to you from Kenneth Arrow's 1963 paper on the subject here: and/or in this model developed by Koijen, Philipson, and Uhlig:

    3. Hubbard, Cogan, and Kessler mention discretionary vs nodiscretionary health care here: Demand for discretionary healthcare would be far more elastic. Emergency medicine would be highly inelastic. In fact, I wonder if elasticity estimates for non-discretionary health care demand would not be an artifact of the data that really have no proper economic interpretation. Your example of the broken arm touches on this. Nobody goes to the emergency room to get the "two for one special". You go to get exactly the treatment you need. In these cases, the individual's demand is very nearly vertical if not completely inelastic. There is a maximum amount that you are willing/able to pay to have the bone set, and while you will happily pay less, no amount of price decrease will induce you to purchase more or less health care. You either get the treatment or you don't. The only variation can be in the quality of the care (or the illusion thereof). Since the aggregate demand is a summation of the individual demands, wouldn't demand be more properly modeled as a discontinuous step function?

    I am not sure these differences I mention imply any policy changes from what you suggest. But I have never seen you mention them in any blog post. Your analysis of health care/insurance always seems great, but I wonder if these points I raise from the health economics field change anything. I read an interesting article by Victor Fuchs at His research showed that health economists disagree with economic theorists not just on normative statements but on positive claims about the effects of health care policies. I wonder if you have noticed the same thing and might have an explanation.

    Best regards,

    Ben Wheeler

  14. "My views are evolving." But if I feel right that you think most of these problems belong to IO and political economy, not finance (or PF), I see your point. Thanks for this.

    That said, I think it's borderline-facetious to pretend that political economy problems can be solved by simple deregulation. I guess it did not go well for the financial industry. Yes, it worked better for retail, airlines or telecommunications, but we need to be more careful about what made the good examples work.

    Finally, I am not sure the insurance problems are that separable from the health care problems (IO and political econ). Let me try to argue why. I think our body politic, our nation, our community cannot bear the thought of people really getting no care — even if you could argue that they "chose" not to when they did not allocate their budgets accordingly. And I think most conservatives share this feeling. Maybe some philosophical training and some public campaigns could help matters, but I wonder how much. Sure, we only want to pay for "necessary" treatments and not elective MRIs, but to make those choices, regulated efficiency creeps back in. Actually, creeps back big time. But the alternative of trying to draw a line in the sand is not credible. Medicare Part D was an overt, irresponsible and potentially consciously dishonest mistake of conservatives. But well-meaning (Ryancare) vouchers might be more grievous a mistake, even if unconscious and honest. When the elderly start going bankrupt because of bad financial planning, Republicans would bail them out just as much (unless they cynically wait only as long as a Democratic administration picks up the bill before them). The kindergarten-worthy solution of limiting spending by simply promising not to spend more does not stand economic scrutiny. (Also, if actual health care spending, even private, is partly controversial and worrying, e.g. because of IO problems, then it is not much better to let it happen because of private demand without further public financing.)

    But if denying treatment is unacceptable, we are back to the insurance problems, public insurance that is, with all the IO and political econ, let it be "simply" regulation or public financing, horrible dicta, public provision.

    This is also a bit relevant while the public sector has a hard time letting the education sector go (or figure out how to let it go exactly). Arguably, education is somewhere between retail and health care on this spectrum.

    By the way, I wonder if you really meant Akerlof-unravelling and not Rothschild-Stiglitz unravelling. Cf.:

  15. Great essay.

    I'm a US expat living in Germany, and have physicians, working both in private practice and for a government-owned and run hospital, as friends, there are two major differences in US and Europe.

    Physician income and tort law.

    The physician friend who works at the hospital (in a stroke unit) works a 35-hour week, variable day length (i.e. she can be at work for 2 full days and then work just one day the rest of the week, depending on scheduling) and earns around €50k/year. She has a very nice benefits package and is pleased with her life-long employment promise and relaxed life style.

    The other, who is my physician, has around 700 private patients and works a 70-hour week, an internist specializing in diabetes (which I do not have). He takes in around €300k/year, which after praxis costs nets him around €160k/year.

    In both cases, their liability insurance is only around €2k/year: they have never been sued.

    Now, my kid brother is a diagnostic specialist in a doctor-owned clinic: his liability insurance is slightly larger than his annual income, but since the clinic pays for it, he doesn't notice it. But that price is reflected in fees. He also has never been sued.

    As long as you have tort lawyers able to have jury trials for medical errors and acts of both commission and omission that lead to multiple millions of dollars in payouts, you will not be able to reduce costs meaningfully in the health system. Period.

    Oh, and yes: it does mean that doctors who make errors don't pay much for these errors. German courts are loath to grant more than a few thousand in pain and suffering cases, and they're capped at around €50k in any case.

    It's a fundamental trade-off: lower costs in exchange for losing the ability to sue and make millions off a court case. Gonna have to fight the lawyers on that one!

  16. Well, the election is over, and the American People have decided they want to adopt the Greek system of political economy.

    I guess this will turn out like the public schools. Anybody with money will hire a concierge physician and go to a private hospital in the Caribbean, if he needs hospitalization. The proles will be in P.S 187, getting medical care that is just as good as the education the kids in public school get, which is to say almost worthless. Medical innovation? forget about it. Some will occur in the private system and will leak into the public system if it really saves money.

    The only virtue I see in this system is that it will accelerate the day that the Peoples Democrat Republic of what used to be the United States of America will collapse in utter bankruptcy.

  17. I would add that more than healthcare needs reformed. Our food supply, and for some lack of access, is part of the problem when it comes to handling our health. Our cities are designed unfriendly to pedestrians and bikes. In Europe, they get plenty of exercise as part of their every days life by walking and biking to work. They don't have to carve out extra time to go to the gym. But we are built with suburb housing in one location, work and shops in another.

    In order to healthcare prices to come down, we as a population need to have access to a culture that is more supportive of a healthy lifestyle.

  18. I would appreciate comment on the possible contradiction:

    On one hand regulation is the enemy because it "(...) does a poor job on the dimensions we care about. Regulators can impose minimum standards, requiring degrees, certification, inspections, etc. and keep out really dangerous quacks. But beyond that they are terrible at pushing for higher quality, especially when quality is so much in the experience of a customer in a service-oriented business." Thus, we should avoid resorting to the institutions which provide regulation on health care such as government (openly stated regulation provided by laws) or charity (internal regulations provided by charity policy). The criticism is that they arbitrarily determine who gets care, what quality of care, and how much.

    But on the other hand "we will also need charity care for those who fall through the cracks, the victims of awful disasters, the very poor, and the mentally ill. This will be provided by government and by private charity." Thus, either government or private charity is to determine who gets help, what quality, and how much. In other words, some groups should actually be exposed to regulation by government and charity.

    So my question is, why institutions which are the enemy of the system are at the same time the saviors for the most vulnerable? Why poor are to be treated differently than vast majority of Americans and cannot experience the benefits of the free market in health care?

  19. This will be offered by govt and by personal charitable organisation." Thus, either govt or personal charitable organisation is to figure out who gets help, what excellent, and how much. In other terms, some categories should actually come in contact with control by govt and charitable organisation.

    1. There's an alternative: Subsidize hospitals that provide charity care.

      Already, Medicaid is essentially a subsidy for charity care (it compensates for only 80-90% of overall treatment costs). But Medicaid is complex and expensive to administer, and it uses fee-for-service price controls which distort physicians' treatment incentives (since they're rewarded for providing treatments that are reimbursed at higher rates).

      Better than Medicaid, I think, would be for the government or a charitable organization to write the hospital a check for X% of its charity care or uncompensated treatment expenses. That way, the hospital could keep control of treatment decisions even as the third party helps to pay the cost of that treatment.

  20. Excellent piece. Really enjoyed it. Thanks for sharing

  21. Excellent article.


    I addition to most doctors being asshole the are the biggest thieves and killers in America and the world.

  23. I suppose it must be a separate article, but I see no possible way to implement a system implied by these ideas, practically or politically. And that's too bad. If such a proposal could be devised, we could at least start talking about it. I hate, too, that hatred and demonization in the political sphere have poisoned the very air in which any real discussion could take place.

    A few of my words of praise for the ACA have been that at least motion has been restored to the ocean in health care and I looked for future improvements. The author has depressed me by showing why entrenched interests are even further dug in. Unfortunately, I feel we will have to clean up our Congress before real reform can occur in health care.

  24. Your Post is very useful, I am truly to this blog which is specially design about the exceptiona healthcare It helped me with ocean of awareness so I really appreciate your blog.
    exceptiona healthcare Brownsville ER


Comments are welcome. Keep it short, polite, and on topic.

Thanks to a few abusers I am now moderating comments. I welcome thoughtful disagreement. I will block comments with insulting or abusive language. I'm also blocking totally inane comments. Try to make some sense. I am much more likely to allow critical comments if you have the honesty and courage to use your real name.