Wednesday, May 21, 2014

Unique economics of healthcare

I was prompted to write this follow-up on health economics after seeing a recent post by blogger Noah Smith, who weighs in with some reasonable views after some intense criticism of the ‘freakonomic’ Chicago-boy Steven Levitt. In a meeting with UK PM David Cameron, Levitt and his co-author apparently made some rather absurd remarks about health care.
They told him that the U.K.’s National Health Service -- free, unlimited, lifetime heath care -- was laudable but didn’t make practical sense.  
"We tried to make our point with a thought experiment," they write. "We suggested to Mr. Cameron that he consider a similar policy in a different arena. What if, for instance...everyone were allowed to go down to the car dealership whenever they wanted and pick out any new model, free of charge, and drive it home?"  
Rather than seeing the humor and realizing that health care is just like any other part of the economy, Cameron abruptly ended the meeting... 
This nonsense reminds me that what constitutes economic debate in the US is often laughable at best.

Health care is obviously not like most other parts of the economy. As I said last week medical services are credence goods - goods which we don’t know whether we need, and even once we’ve consumed them, still don’t know if they were good value. In economic terms, these goods suffer from the worst possible information failures, particularly with respect to the asymmetry of information between the seller (in this case the doctor) and the consumer.

For these goods the demand curve may slope any which way, and people are often left to use price as the only signal of quality (or quantity for that matter). This means that a socially optimal level of medical service provision cannot be determined using basic marginal economic analysis.

Not only that, there are substantial positive externalities to most health care services. Vaccinations are the obvious example, but the same principle applies more broadly.

Once you’ve accepted that health care and medical services don’t fall neatly into preexisting economic models, you need a better way to think about the potential efficiency of any health care system. Rarely is this step taken in public debate.

One way to assess any health system is in terms of the two main sets of incentives - those of the patients, and those of the medical service providers (doctors and suppliers of medicines, equipment and accessories).

We often hear about the patients, with the archetypal case of the lonely hypochondriac making a few extra trips to the GP or emergency department when the service is provided free of charge. Sure they exist, but as I’ve said before, pricing these visits deters both the time-wasters and those with genuine medical needs. Making prices for patients reflect production costs in health care systems has the benefit of reduced health expenditure, but comes at a cost of poorer health outcomes.

But in general no one wants major medical services even at a zero price. Here’s a comment from Noah’s blog making that point
Most medical treatments are painful, unpleasant, and time consuming, and are only desired when non treatment is worse. While making treatment costly will deprive some of access, it will do little to make treatment more undesirable than it already is.
On the other side of the ledger we have the incentives of doctors and other businesses involved in the supply of medical services - drug companies, suppliers of medical equipment and so on.

Here there are usually financial incentives to over-treat or over-prescribe. To take just one example, new evidence from Australia’s two-tiered system shows that in private hospitals there are more medical interventions for low-risk births than in public hospitals.

Indeed there is clear evidence of these interacting incentives following the announcement of the $7 GP co-payment in the Australian federal budget. Some medical centres have been text-messaging patients to remind them that they are not charging fees, possibly due to lower patient numbers. On display is the doctor’s incentive to earn a living providing medical services by marketing to customers, interacting with the uninformed patient reacting to a price that doesn’t even exist yet in a way they can't possibly know is beneficial for them or not.

As a final point, we rarely hear about the monopsony buying power of having a single medical provider, which can be significant. A single national (or State level) healthcare entity is in quite a position of power in negotiating supply contracts when they are the only game in town. In a world were drug companies are heavy political hitters, having incentives within the government to reduce drug costs to economise on health spending seems an important consideration.

If we are going to have an intelligent debate about efficient health care we need to remember three key points
  1. We must consider benefits as well as costs (including externalities) 
  2. There are serious moral judgments necessary about the scope and priorities of health care 
  3. Medical services are credence goods, hence there are unique incentives at play


  1. This recent study in JAMA pediatrics (linked) supports Murray's take here. The study strongly suggests that low-income patients with chronic illnesses suffer adverse consequences through the use of cost sharing (co-pays). That even such minor bows to market pricing, co-pays, tends to produce poor outcomes shows the folly of shoehorning the medical system into existing economic theory. And it also points to one of the principal reasons economics continues to be rather dismal: theory often trumps empirical evidence. For example, theory has it that QE must, perforce, lead to inflation and that raising the minimum wage must increase unemployment. That neither is true empirically does not matter to many economists who cling like automatons to the theoretical worlds in which they dwell.

    1. Thanks for the comment and the link.

    2. "reasons economics continues to be rather dismal: theory often trumps empirical evidence. For example, theory has it that QE must, perforce, lead to inflation and that raising the minimum wage must increase unemployment. That neither is true empirically does not matter to many economists who cling like automatons to the theoretical worlds in which they dwell."

      Thank you, well said. I've been trying to make this point for the last decade.

  2. Cameron K Murray,

    I really like making #2 explicit. I think often times what happens in conversations like this is that the proponents of universal health care want to use arguments that their interlocutors will be forced to accept because logic or commitment to neutral values like cost (be it in arguments with centrists, libertarians, conservatives, neoliberals, etc).

    Obviously everyone knows that a Randian anarcho-capitalist and a socialist have so little in common in terms of their moral visions of society - there's almost no point in those two talking policy. But I think often we run into much finer versions of that, and if that's not noticed, then we'll end up talking past each other. Talking through this can be pretty tough and the conversation will be at least as philosophical as economic.

    About health care being like any other part of the economy, what people on the left often bring up are very serious and expensive medical procedures, which would tend to be covered by Milton Freidman/Steven Levtitt type universal catastrophic plans. In other words, we agree! Or, we seem to. Of course, Levitt said the opposite, but what I'm saying is that his seems to be the latest in the line of many kinds of plans that acknowledge, through their design, that at a certain point health care is a unique good.

    It's hard for me to balance many things:

    Whether being a bit more stingy in preventative care would result in higher spending later (a plausible claim) - whether to think that costs for routine care would go down if a Freidman type plan prevailed - whether to be moved by CBO's claims that preventative care doesn't actually save money - whether to think that lengthening people's life spans will make total costs go up - whether to think that medical innovation would slow in a socialized health care system and if so if government spending would have to make up the gap, etc, etc, etc.

    It may be that the distinctions between something like Singapore - -
    on the one hand or Canada and the UK on the other are so fine that even good liberals (in the broadest sense of the term) are hard pressed to say which one is better (as good liberals are supposed to care a bit about cost to the taxpayer, and display a modicum of compassion) without tipping the scale with moral commitments.

    1. Thanks for the comment. If you want a fee other systems to look at, try Australia, which has State-level public health plus incentives for the wealthy to privately insure. It's a kid of middle road that I believe finds a good balance.

    2. Exactly. What we need is a basic plan that covers all citizens, paid for with a dedicated tax so it's deficit neutral, then maintain a private for profit coverage system to supplement the basic plan so the wealthy can still maintain better coverage and service (something the wealthy take very seriously)...

  3. "Obviously everyone knows that a Randian anarcho-capitalist and a socialist have so little in common in terms of their moral visions of society - there's almost no point in those two talking policy."

    Which did not stop Rand herself from ending her life as a Medicare client. The fact that according to her the program is immoral and should not didn't count when the chips were down.

  4. See this counterargument to Krugmans post on same topic

  5. Thank you for this post! It is very educational. Hopefully some talking head economists read it and amend their usual free-market rantings.

    It also makes me wonder how many other markets are like the market for health care: special cases. If supply and demand curves don't work for health care, which makes up a big part of modern economies, and a number of other markets, how can they work for the economy as a whole? Or, rather, how can they be the basis for a general theory? Economics might do well by conducting empirical studies of local markets and finding out how many "special cases" there really are.

    Thanks for the great blog. Keep it up!

  6. The influence of IT departments within the health trade is outstanding. With the evolution of the web and technology, the dependency of the medical considerations thereon solutions have mature within the past few decades, for more information click here Private Duty Home Care West Palm Beach.