What’s a good health-care system?

There’s little doubt that the US health care system is not good. The US health care costs are around 20 percent of GDP, or $10,000 per capita per year. That’s unreasonably high compared to other developed nations. There are many reasons for this but the primary reason, as I see it, is the tacit collusion between the health insurance business, the hospital business, the pharmaceutical business, medical professional bodies, and governmental regulatory agencies.

Insurance has an important role to play in any large, modern society. Random events can be insured against in a population. Insurance distributes losses arising from random events across the insured population. Suppose in a population of 100 people, it is statistically certain that within a year one person at random would incur a loss of $200, then to cover that loss, if everyone paid an insurance premium of $2, then the loss can be spread over the entire population instead of just that unlucky person bearing the entire loss. In effect, everyone in the insured population bears a small definite loss so that no one bears a huge loss. When we buy insurance, we trade a guaranteed small loss against an uncertain big loss.

We note in passing, without going into the details, that insurance is  plagued by the usual adverse selection and moral hazard problems.

It is clear that for such events that happen quite rarely but randomly, which are statistically predictable, and which impose unbearably large losses when they do happen, agents would rationally insure against such events. For insurance to work, every event has to have only a very small chance of leading to a very large loss. If every event was highly likely to lead to a small loss, the insurance mechanism would not work. Similarly if every event was highly likely to lead to a huge loss, then too insurance would not make sense.

The probability of a commercial jetliner being involved in a catastrophic accident leading to fatalities and hull loss is very, very tiny but when it does happen, the cost is measured in hundreds of millions of dollars. That’s a low probability but extremely high cost event. Therefore insurance makes immense sense. In effect, in their ticket price every of the hundreds of millions of passengers who do not suffer any loss pays a fraction of a dollar to cover the losses of those rare commercial aviation accidents.

Now let’s move on to health. Common colds, mild fevers and stomach upsets, little aches and pains, a sprained ankle or a wrist, a minor cut in the kitchen, etc, are a part of living. They are high probability in any extended period but low loss events. They do not require heroic medical interventions. They should not necessitate insurance. Insuring against them incurs costs that outweigh any benefits.

But what about catastrophic health events such as cancer, myocardial infarctions (commonly called heart attacks) and other events like those? Should there be insurance for them? The simple answer is, “Yes there could be but like all other forms of insurance, if the people are willing to pay the premiums for it, then there are no reasons not to have it. But if the affected relevant group is not willing to cover the costs of insurance, they cannot be insured against them.”

The main point here is that routine, certain to happen events that affect everyone in the relevant population cannot be insured against. News flash: everyone is going to die. Because death is 100 percent certain, death cannot be insured against. Of course insurance markets exist for untimely death but that is only probabilistic and not certain.

Health insurance that attempts to cover the costs of all kinds of maladies–from little aches and pain to cancer and beyond–does not make sense. It inflates the costs without commensurate benefits.  What are the costs? The major one is that it creates perverse incentives for all parties concerned.

When you know that it costs you very little to go to the doctor because you have medical insurance, you get yourself to the doctor because of a mild stomach ache. The doctor, in turn, knows that you have insurance, and therefore recommends that you get extensive costly tests done for your stomach ache. He does not want to run the risk of not making sure that you don’t have appendicitis. His malpractice insurance premiums — which already run into tens of thousands of dollars per month — would go through the roof.

So off you go to get various tests. They all cost a bomb but your co-payment is still not astronomical. The insurance companies pay but their premiums are high enough that many cannot afford to buy insurance. So the government gets into the act and makes it mandatory for everyone to buy insurance. The insurance companies essentially pay the politicians to pass laws that coerce people to buy insurance.

Doctors charge high prices because insurance companies pay high prices. People don’t economize on using medical services because those who use the services don’t pay the full price. And so the circus goes on.

So what’s a good health care system? One that entirely market based. That means, health care only partly based on insurance. Which part should be insurance based? The part that people are willing to voluntarily buy insurance for. Maybe there is a market for insurance against little aches and pains. If so, those who want that coverage would be able to get it.

(What does “a market for insurance for x exist” mean? It means that the premiums from people willing to buy the insurance cover the losses of the insured.)

And the part of health care costs that are not covered by insurance has to be covered by direct payment for services. That is a good health care system.


  1. This post is brought to you by a question asked in the last AMA.
  2. The picture at the head of this post is an image of Dhanwantari, the physician of the gods. He is the authority on Ayurveda, and is the bearer of amrita — that which confers immortality.
  3. Recently a friend of mine who lives in Oakland CA sprained his neck in his sleep. He visited a hospital instead of just taking it easy for a couple of days. At the hospital they took a couple of x-rays and recommended that he take it easy. The hospital billed him only $4,600 for it.  Instead of going to the hospital, if he’d asked me I would have told him to take it easy for a couple of days. I’m not a medical doctor but in such cases, even a PhD in econ works. Our bodies are self-healing. We just have to pay attention to our aches and pains. They signal the need for rest and recuperation.

6 thoughts on “What’s a good health-care system?

  1. Thanks a lot Atanu for this post. Some thoughts:

    Your explanation of insurance being suitable only for low-probability but high-cost maladies is nice. Cancer fits that bill. However, if tomorrow cancer becomes a household malady, insurance companies should be allowed to discontinue cancer-insurance-schemes. Otherwise, the entire insurance will become unsustainable due to runaway cost. True, it will create hardship for customers caught during the transition, but it may be beneficial for greater-good.
    In spite of all imperfections, US is the freest society. Why is market unable to auto correct itself in the US? I wonder.
    Accidentally (and without any conscious intention), big cities in India (like Bangalore) have the best medical care at most affordable prices for middle/upper-middle-class. Due to the huge number of abysmally poor/lower-middle-class population, who do not have any insurance cover, the medical system is forced to come up with affordable and good solutions (plain market economics at work). With that system in place, middle/upper-middle-class are having it really easy. For example, many middle/upper-middle-class people in India can easily afford a child-birth procedure with 2 months of salary, in a high-brow private hospital.


    1. Dear Atanu,

      Been a long while since I wrote something here, but, as they say, I will bite the proverbial bullet, and post a reply here…

      I have for some time wanted ask you what you, as a PhD in the “dismal” science, think of, on the topic of insurance for agriculture (including farmers, not just their produce) in India: the idea, the practice, the reasons for the practice’s (obvious) failures (else we would not have demands for loan-waivers etc. very periodically and then, depending on mere weather, also aperiodically), and the possible and practical remedies there could possibly be. …

      … No, you don’t have to do another PhD. But the issue, I think, is relevant to India, and you could keep it somewhere at the back of your “economizing” mind. Post about it, or its aspects, purely at your own convenience.

      Best wishes and regards,



    2. In spite of all imperfections, US is the freest society. Why is market unable to auto correct itself in the US? I wonder.

      Atanu has already covered that point in first paragraph. Healthcare in USA is one of the most heavily regulated sector perhaps more than in India. For example I don’t think any Indian doctor or hospital worries too much about malpractice law suites, in USA every doctor has to buy insurance against it and recover the cost from patients. More regulations invariably hurt both producer and consumer.

      But then look at the unregulated/less-regulated aesthetic medical procedures. Fixing your node costs you around $18k in USA, increase boobs size costs $12k, LASIK on one eye costs around $3k. Adjusting for PPP I think this is way cheaper than India or most other countries. Not to mention the technology and techniques used in these surgeries has only gone better with time.

      That is what market competition generally does. I recommend you watch this :


  2. In the AMA edition you refer to, you also commented:
    “Thank you for raising the important points you raise. I have to write a rather detailed post on Gurumurthy’s talk with Malhotra. In it I will address many of the points you raise.”
    Waiting keenly for that post to come.:)


  3. I want your answer on a very important question which is quite unrelated to this post.
    With scams tumbling out of Indian public sector banks the demand to privatise them is rising, even among the general population. But such demands evoke a horrified NO! from many, who say “look private banks caused the 2008 meltdown, you want that?”…So what would you say to them??


    1. TheFreeIndian,

      You write, “But such demands [to privatize public sector banks] evoke a horrified NO! from many, who say “look private banks caused the 2008 meltdown, you want that?”…So what would you say to them??”

      The answer in brief is that the 2008 financial crisis was the result of government intervention in the banking sector. The government, through its regulations, forced lending agencies — private, semi-private and public — to make loans that had little chance of repayment, the so-called “sub-prime” loans. That was the cause of the crisis, and it had little to do with whether the banks were private or public.

      I recommend an awesomely funny video that explains the subprime loan crisis. Go check out this post from Jan 20018 – over 10 years ago: The Financial Crisis According to the Long Johns.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s