In his latest letter from America, Angus Deaton highlights graphically the problems of asymmetric information in the US market for healthcare.

Although I read and write about health, I have been fortunate to have had little direct experience of the American healthcare system. So when I was advised last June that I needed a hip replacement, a routine procedure, but one that is classified as medium risk, I was apprehensive on both financial and medical grounds. The mortality rate is about half of one percent, half of which is associated with unpredictable post-surgi;cal deep vein thrombosis, recently familiar to long-haul airline passengers as ‘economy-class syndrome’. But there was also a modest opportunity to do empirical research on a topic that, once again, has become central in American policy discus;sions. The Bush administration has embraced the idea of consumer-directed healthcare, and put in place schemes that give people incentives to seek out good value in healthcare. Saving a few dollars was hardly my main consideration, but I was happy enough to try to be a good consumer, subject to an appropriate quality constraint. And hip replacements are big business in the US; more than 150,000 hips are replaced each year, at an average cost (hospital charges only) of around $50,000 each, which is $7.5 billion in total. But to make good choices, I needed good information, and information on either quality or price was remarkably hard to come by.

Checking on quality

Although US News and World Report ranks hospitals along various dimensions (as it does universities and university departments), there are no such reports on orthopaedic surgeons. Although individual surgeons advertise and promote themselves, and although there is lots of gossip (‘he’s the guy who did the Pope, but he’s past it’) the medical profession has successfully resisted the publication of any official guide. It might be reasonable to suppose that, just as it is possible to find out from colleagues who are the top people in, say, game theory, or industrial organization, one could talk to one orthopaedic surgeon about the others. But this is simply not the case, and while I eventually found a senior night nurse who, when sufficiently bored at 2.30 a.m., was prepared to tell me which of the surgeons in my hospital knew what they were doing, that conversation was during my recovery, and so of limited use for decision making (unless I need another replacement). Of the several surgeons I talked to in advance, I could tell whether I liked or felt comfortable with each, but that seemed of limited relevance, and neither they, nor my primary care physician, nor friends and acquaintances, could tell me more. Indeed, the only useful information that I had before the surgery (and the validity of which was strongly confirmed by my own experience) is the well-known rule to go to a hospital and a surgeon who do the procedure frequently. So I lined up a surgeon who had done 10,000 previous hip replacements, and who works in a hospital that is highly ranked by US News and World Report, and where nearly a hundred other hips were being replaced on the same morning as my own.

…and on price

Information on price, surely, would be simpler to find. Not so. The surgeons were forthcoming with their fees, between $7,000 and $8,000, although, from the first, it was clear that these were somehow negotiable, if not with me, then at least with my insurers. Much less straightforward was the price list for the other associated services, anaesthetists, physical therapists, pain management specialists, and what turned out to be the largest item of all, ‘board and lodging’ in a semi-private room (shared with one other person) which, incredibly, cost more than $10,000 per day, although it should be admitted that my room was a large one, with private bath, and a splendid view of one of New York’s rivers, with its constantly moving shipping that provided fine entertainment, though perhaps enhanced by the morphine pump. Telephone and television were extra. But none of these prices are what they seem. Each insurance company negotiates its own prices with each of the hospitals and physicians with which it deals, and these prices are closely-guarded secrets. Of course, the insurance company tells its customers what it will and will not pay for, but that contract, like the hospital’s price list, is much less useful than it appears. If the ‘provider’ is ‘in-network’, the company will reimburse a large fraction of the charges. For ‘out of network’ providers, the reimbursement is a smaller fraction, although still usually 80 percent or more. So it would appear that my personal liability was fairly limited. Not so, because my insurance company pays 90 percent of the ‘secret’ price, not of the full price. So if, for example, the anaesthetist (who seems like the wrong person to antagonize, and who, in my case, asked me to sign an ‘informed consent’ form for an experimental procedure immediately before rendering me unconscious) bills me for $6,000, and the insurance company believes that the appropriate price is $4,500, the reimbursement is 90 percent of the latter, not the former, so that my exposure is not $600 (10 percent of $6,000), but more than three times as much, $1,950 (the uncovered $1,500 plus 10 percent of the covered amount.) If there is a way of knowing these amounts in advance, I could not discover it. Nor, at any stage before, during, or after my hospital stay did anyone ask me whether or not I wanted any of the many procedures and services that I ‘purchased’.

So much for informed choice based on price. This situation has been compared by my Princeton colleague Uwe Reinhardt to shopping blindfold in a department store, and then months later being presented with a bill on which some items are charged at full price, and some at some fraction of full price, but with no advance knowledge of either what one has bought or what it will cost. And this is for those who are fortunate enough to have insurance. The more than sixteen percent of the American population that has no insurance is charged the list price, known as the ‘chargemaster’ price, and Reinhardt gives the example of someone who spends many years paying off a debt of $30,000 for a procedure that would have cost Medicare (the government insurance scheme for the elderly) $6,000. Hospital debt recovery procedures involve relentless persecution by collection agencies, something that is threatened on almost every one of the blizzard of bills that come to (even insured) patients for many months once the surgery is over.

Controlling the health bill

The favoured instrument of the Bush administration for controlling the rapid increase in health spending is the health saving account. Consumers (and their employers) are permitted to make annual contributions of untaxed dollars into these accounts provided they purchase a health insurance policy with high deductibles. Health expenses can be paid from the accounts and any unspent balance is rolled forward, ultimately providing a retirement nest egg from any unspent balances. Because people are spending their own money, they have an incentive to find out about prices, and to pick providers who offer the best value. The hope among the proponents of the scheme is that the very existence of the accounts will put pressure on providers to give the sort of information that will permit comparison shopping. Certainly, that information is not available now, as my own experience showed. Opponents of the scheme point out that a large fraction of health expenditures are incurred by a small fraction of people whose expenditures are so large that they could not conceivably be covered by such a scheme. So even if people were to shop around more effectively, total savings are likely to be small. The accounts also reward good health and penalize the sick. Glenn Hubbard, a distinguished economist, and George W. Bush’s first chairman of the Council of Economic Advisors, argues that health saving accounts ‘are probably the best thing to happen to healthcare in a generation’, and that they would ‘give people a way to profit financially from their own good health’. None of the proponents has explained how this is consistent with one of the administration’s other avowed aims, which is to reduce the gap in health status between the rich and the poor. Indeed, health saving accounts would seem to be uniquely well-designed to generate a correlation between health and wealth, even if none previously existed.

And how is my hip? Just fine. And, like almost everyone else who has had the procedure, I would be happy to recommend my surgeon and my hospital, for all that that is worth. How is my wallet? Although it will take many months before the final accounts are settled, something like $7,000 lighter, some of which probably reflects a quality premium for the providers I selected. There was no wait, and I could schedule the surgery at my convenience; for comparison, the OECD says that in 2000, the average wait in England was 250 days. Because my employer offers a modified version of a health savings account (but with no rollover of unspent funds,) most will come out of pre-tax earnings, though that worked only because I knew well in advance when I was going to have the surgery. If I were poor, unemployed, or uninsured, I would be bankrupt or, more likely, still hobbling around on an arthritic and increasingly painful hip