Health policy: Excerpt from Economics in Two Lessons

Another excerpt from my book-in-progress, Economics in Two Lessons (partial draft here). As usual, praise is welcome, useful criticism even more so.

On most measures, health care is among the largest industries in a modern economy, and its share is likely to grow in future. Health costs are a major item of expenditure for Individuals, families, employers and governments, amounting to 16 per cent of national income in the US (costs are lower, but still substantial in other developed countries with more sensible policies). Yet only a tiny fraction of this expenditure takes the form of standard consumer market transactions; that is, purchases of medical services by patients at a price equal to the opportunity cost of their provision.

The majority of private health expenditure is funded, in whole or part, through insurance. Once the premium is paid, all or most of the cost of visiting the doctor or going to hospital is borne by the insurer. So, the opportunity cost to the household of using medical services is far below the opportunity cost of providing those services.

The same is true of publicly provided health care. Public health services are funded from tax revenue, sometimes specifically allocated to health care and sometimes derived from general revenue. Either way, the opportunity cost for patients of using these services is far below the opportunity cost to society of providing them.

More generally, the health sector of a modern economy involves a complex mixture of public provision, for-profit corporations, non-government organisations and individual professionals such as doctors. Systems for providing and financing health care differ radically, from comprehensive public systems like the British National Health Service to predominantly private systems like that in the US. But nowhere are such services provided primarily on the basis of prices that reflect the opportunity cost of their provision.

This was not always the case. Until the late 19th century, markets for health care were much the same as other service markets. Doctors and pharmacists provided services and medicines as patients (at least those who could afford them) demanded, and were paid fees, out of the patient’s pocket, in return. Those who could not afford to pay (the majority of the population) relied on charity or went without. This ‘fee-for-service’ system has not entirely disappeared, and remains dominant in some poor countries. But developed countries have abandoned it almost completely.

To understand why health care systems are the way they are, we must apply the Second Lesson. This will help us to understand why simple markets for health care do not work to tell us about social opportunity costs. The more difficult problem is to work out the best way of making prices and other market signals work where they can, and how to replace them where they do not work.

Rationing and opportunity cost.

Economists are frequently criticised for assuming (as they mostly do) that the wants of consumers are limitless. Why, it is reasonably asked, can we not learn to be satisfied with what we have?

One field in which the assumption of limitless wants seems plausible is that of health care. New and improved treatments are being developed all the time, offering us longer and healthier lives, and cures for previously entreatment ailments and disabilities. Whatever we might think about consumption in general, few of us are so stoical that we would willingly choose to accept death or disability if it can be prevented or cured by medical treatment.

The problem is that, while the range of potentially beneficial treatments is effectively limitless, the resources available to the health care system are not. In a simple market system, this problem resolves itself automatically; those willing and able to pay for health care services receive them, and those who are unwilling or unable do not.

As we have already seen, however, markets in health care do not work well. For this reason, they have been replaced by a mixture of public provision and insurance. Both for public health services and insurers the problem then arises: which health services should be funded?

The general idea may be seen by an example. Consider a hospital which finds that it has a small amount of additional money that can be allocated either to the orthopaedic ward, where it would fund additional knee reconstructions, or to the emergency ward, where it would increase the survival chances of patients involved in car crashes.

The opportunity cost in this example is clear. The opportunity cost of improving the mobility of orthopaedic patients is that fewer emergency patients will survive. The patients involved may or may not be the same people (since car crashes are a common cause of severe injuries requiring orthopaedic surgery). T

There is no simple way of making this choice. Obviously, saving a life is of greater significance than a knee replacement, but the opportunity cost trade-off is unlikely to be one for one.

There are various ways in which the trade-off may be assessed. Some methods rely on the expertise of doctors. These face the problem that, being human, most doctors regard the conditions treated by their own speciality as being more significant than those treated by others.

Another approach is to rely on the expressed preferences of potential patients. In the example above, we might ask people who are not currently in need of a specific treatment to make judgements of the general form “I would (or would not) prefer a treatment yielding 20 years of additional life with full mobility to one yielding 25 years of additional life with a requirement to use crutches to move about”. [A formal version of this approach, using the concept of Quality Adjusted Life Years or QALYs is sometimes used. This approach and its strengths and limitations will be discussed in an Optional Extra section]

Judgements about the relative benefits of alternative outcomes may be translated into medical decisions in various ways. The simplest is to specify a list of treatments approved for funding. An alternative, referred to as case mix or activity-based funding relies on providing a specified payment for each kind of treatment, based both on the cost of provision and the requirement that benefits should exceed costs.

These price-based approaches can provide useful signals when hospitals or other health care providers are making treatment decisions in combination with their own judgments about what policies are most consistent with a mission of providing care. The price paid for an activity provides an indication of the social benefit perceived by the government. However, given a sufficient surplus, providers can choose to provide services that they judge to be desirable, even though they are funded at less than the cost of provision.

As with other price-based policies however, activity-based funding models are vulnerable to exploitation in a system dominated by profit motives. Profit motives are obviously dominant when services are provided by for-profit corporations. The may also dominate when resource constraints are so tight that a non-profit provider has no alternative but to maximise returns by providing only those services with a price sufficient to cover the opportunity cost of provision.

In these cases, various forms of cost-shifting may emerge. For example, doctors funded on an activity-based model may attempt to divert ‘unprofitable’ patients to the emergency wards of public hospitals rather than spending the time necessary to treat them. Similarly, it may be cheaper, from the provider’s perspective, to give a patient a prescription (the cost of which is borne elsewhere in the system) rather than to spend the time necessary to treat the underlying condition.

Ultimately, given scarce resources and effectively unbounded wants, some form of rationing is inevitable. The opportunity cost approach makes explicit the fact that, for any given level of resources, providing one kind of medical treatment comes at the expense of the alternative treatments that are not provided.

No approach is perfect, but reliance on markets and incentives is likely to lead to highly unsatisfactory outcomes. The best options involve a combination of explicit policy judgements on the general value of particular treatments and reliance on the expertise and commitment of medical professionals to provide the best possible treatment within the constraints of those policies.

14 thoughts on “Health policy: Excerpt from Economics in Two Lessons

  1. The emergency ward versus orthopaedic ward example largely doesn’t arise in practice in Australia. If you need a knee replacement and are prepared to pay the market price minus insurance reimbursement, you go to a private hospital. Very few private hospitals have emergency wards, so it’s just not the case that a knee recon is using resources that might otherwise be used in emergency surgery for a car crash victim. The knee reco does use resources that might otherwise be used in a hip replacement, but that is the type of situation where a market-based allocation is called for.

    With public hospitals, emergency surgery comes first. If you’re in for a knee reco and a car crash victim comes in, your surgery gets postponed until the operating theatre is free. There is no price mechanism at work.

  2. The example above happens regularly in the Australian public health system. As a nurse in the public system I have seen many instances where a new emergency department is built or an old one enlarged to cope with the increase in presentations, without a proportionate increase in inpatient beds, required by the increased number of patients requiring admission. Thus patients wait longer in the ED for an inpatient bed. Don’t forget much of the population does not have the resources to go private for their knee replacement. Many with Health Insurance only have basic cover which explicitly does not cover joint replacements, cardiac interventions or ICU admissions (required following many major surgery interventions)

  3. What you haven’t fully factored in John in your analysis is the enormous waste in the health system at present. We could reduce health expenditure by a third by eliminating those services which are harmful or add very little benefit, and by substituting high cost services and products with lower cost services and products which have the same benefit. (In this latter area there are lots of easy gains in the pharmaceutical field in particular). There is duplication everywhere whether it is in tests or record keeping. If we cut the waste we could afford any health service or product that any reasonable person would want. The only rationing we would need to do is to exclude from government subsidy any services or products which cost more than $200,000 per QALY.

  4. “Until the late 19th century, markets for health care were much the same as other service markets.” Until the 1930s, medicine was to a first approximation useless. The rare interventions that worked were mostly forms of surgery: extremely dangerous, and, before anaesthetics, agonizingly painful as well, imposing non-financial costs on patients quite sufficient to discourage casual over-use. The health cost problem of today is the product of the arrival of medicine that worked, starting more or less with the sulfonamides.

    “One field in which the assumption of limitless wants seems plausible is that of health care.” I’m disappointed that JQ does not challenge this piece of CW. There is a lot going for it of course, especially the rise in health costs in OECD countries with efficient socialized health care. The case against relies on Lewis Thomas provocative 1965 scheme of three levels of medical technology: primitive, cheap and ineffective (the general case till the 20th century); intermediate, with elaborate and expensive palliative measures (iron lungs, insulin injections, and I would now add cancer and HIV drugs); and advanced, with cheap real cures like vaccinations, antibiotics for syphilis, Barry Marshall’s discovery on ulcers). Blog post on the search for real cures. Wishful thinking? Possibly. But possibly not.

    There is a curious test population that (absent a proper study) at first sight supports the case for satiability of health care: European royalty. Another blog post. They are a pretty random sample genetically, especially since they started marrying commoners. They enjoy the best health care available in their countries. They have healthy lifestyles, with interesting and varied work, and not too much of it: a future qualitatively available to most of us. And they are very healthy, working on until their 80s and even 90s, with few chronic maladies and surgical interventions. I see no reason to think their lifetime medical costs are high.

  5. Actually, nobody knows what the health costs for royalty are. The concept of limitless wants may well apply to them and others because there are all sorts of screening tests , scans etc that can be used ad infinitem.

  6. O dear James. There is so much wrong with your post it is hard to know where to start.
    First, just because in previous centuries, the health interventions were not very effective does not mean there was not a lot of effort put into trying to fix things and to care for people who were sick. What were the ship surgeons doing for example and what was Florence Nightingale doing in the Crimean?
    It is true that in previous centuries most health care was carried out by the household sector rather than in the market sector, but the market sector was surprisingly strong and getting stronger even in the eighteenth century as is illustrated by the extract below from the preface to the Rev John Wesley’s book ‘Primitive Physick’. Primitive Physick was a home doctor book used by many in the US and UK in the eighteenth and nineteenth century which contained mostly herbal remedies but also treatments using new fangled technologies like electricity (from batteries). (this link is to an edition published in 1846, but the first edition was in 1747).

    .. ‘Medical books were immensely multiplied ; till at length physic became an abstruse science, quite out of the reach of ordinary men. Physicians now began to be had in admiration, as persons who were something more than human. And profit attended their employ, as well as honour ; so that they had now two weighty reasons for keeping the bulk of mankind at a distance, that they might not pry into the mysteries of the profession. To this end, they increased those difficulties by design, which began in a manner by accident. They filled their writings with abundance of technical terms, utterly unintelligible to plain men. …….
    Those who understood only how to restore the sick to health, they branded with the name of Empirics. They introduced into practice abundance of compound medicines, consisting of so many ingredients, that it was scarce possible for common people to know which it was that wrought the cure ; abundance of eloties, neither the nature nor names of which their own countrymen understood ; of chymicals, such as they neither had skill, nor fortune, nor time to prepare ; yea, and of dangerous ones, such as they could not use, without hazarding life, but by the advice of a physician. And thus both their honour and gain were secured, a vast majority of mankind being utterly cut off from helping either themselves or their neighbours, or once daring to attempt it. ….
    In all that have yet fallen into my hand, I find many dear and many far-fetched medicines besides many of so dangerous a kind, as a prudent man would never meddle with. And against the greater part of those medicines there is a further objection ; they consist of too many ingredients. The common method of compounding and de-compounding medicines can never be reconciled to common sense. Experience shows that one thing will cure most disorders, at least as well as twenty put together. Then why do you add the other nineteen? Only to swell the apothecary’s bill ; nay, possibly, on purpose to prolong the distemper, that the doctor and he may divide the spoil. ‘

    The profit motive has been distorting health care for centuries.

  7. @Vegetarian
    Do you have any evidence for you speculation on the health costs of royalty? I don’t have much for mine, but what there is points in the same direction: few surgeries, and active longevity. They may well pay more for top private GPs, who will spend more time with them than the norm – and will not advise useless tests. The big costs are from hospitalisation, which the celebrity mags cover, and dementia, which is obviously rare. A proper scientific study would be well worth doing.

  8. @John Goss
    Read Lewis Thomas. His typology is from a memo he sent LBJ in 1967, not 1965. He personally witnessed the arrival of the sulfonamides in the 1030s: before the, the treatment for pneumonia say was to watch. Scientific medicine was all about prognosis, not cures. “Guérir parfois, soulager souvent, consoler toujours”, as François 1er’s surgeon Ambroise Paré put it. That was what Florence Nightingale was doing.

  9. @James Wimberley
    I can hardly have evidence for something I said nobody knows about.
    And of course no doctor would order tests s/he knew to be useless, but it’s often a case of checking to be on the safe side. However there is no easy way to know whether say a PET scan ($1000) for a persistent cough will turn up cancer, or if it was just a lingering cold.
    Good health is often the result of a genetic lottery. A sample size of one family proves nothing.

  10. @James Wimberley

    Do royals have healthy life styles? Every British monarch since Victoria has died of a smoking related disease. It’s unlikely the Queen smokes, but you can be sure she did as a young woman, because just about every woman of her generation did. Her sister certainly smoked .

  11. @Smith

    Netflix have produced a decent soap called Crown about E II’s life. It presents her as a nonsmoker who makes Philip give up smoking when she gets engaged. (He is still going today at 95, so he stopped in time.) Looks plausible to me. Her father died of lung cancer. We should take account of the fact that royals get top-notch doctors, and the antismoking advice could well have reached her early. Richard Doll published his first epidemiological paper fingering smoking in 1950 in the widely read BMJ.

  12. @Vegetarian
    It’s not just one family. Read my blog post. The longevity includes the reigning monarchs of Denmark, Sweden, Norway and the Netherlands. The families are much less interrelated now than they were in 1900, when everybody was related to Victoria one way or another. I left out Spain. Juan Carlos abdicated two years ago, age 77, for reputational not health reasons. His father never reigned but lived to 79 too. It’s not a big sample, but it is a population, not an anecdote. The interest is that unlike other groups of very rich people, they got there entirely – not just significantly – by genetic lottery,

    Smith: Are royal lifestyles healthy? Of course. They have to stay presentable, keep a schedule, read fairly challenging state papers, and interact socially with lots of different people. Very strong reputational pressures inhibit abuse of alcohol and drugs – the scandals are generally about sex, which is not unhealthy. They mostly have outdoor activities, horses, sailing, shooting. skiing, walking. They have few money worries, and servants do the housework. A pretty nice life.

  13. James: I’m not really following the argument about royalty, but as others have said, it seems to go the wrong way for you. We can assume that royals are hardly ever denied health care from which they can benefit and, as you say, they seem to live a long time. That would seem to suggest that if everyone got unlimited access to health care, there would be an increase in longevity. Even allowing for the removal of inefficiencies, that would imply a large increase in total expenditure.

    Am I missing something here?

  14. @John Quiggin
    Healthy longevity – they keep working long after normal retirement age. If you were going to study this seriously, the metric could be lifetime medical costs or medical costs per year of life.

    IIRC a remarkable proportion of US medical costs are incurred in the last year of life, in a system with strong incentives for heroic measures and a weak system if palliative care. These costs should not be affected by longevity. A priori, they could even fall. When you start on the last slide over 90, every body component is worn down and it can go faster. The obviously low incidence of royal dementia (expensive and often long-lasting) is a point in favour of my speculation.

    I don’t however see how we can take this further absent a proper study. It would be a good project for a graduate student – it’s a small population, massively and systematically covered by the press.

Leave a Reply

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

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

Google photo

You are commenting using your Google 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 )

Connecting to %s