• An international case for all-payer rate setting

    The following is a guest post by John Nyman, Professor of Health Policy & Management, University of Minnesota School of Public Health.

    Below is a table that I use in my health econ classes. It is now a little outdated, but shows that our G7 peers spend half as much as we do on healthcare but go to the doctor twice as often and spend twice as many days in the hospital as we do. This suggests to me that the big difference between the US and the other G7 countries is the prices. There are some caveats: we have more outpatient care, our use of technology might be greater, we might have a different (younger) population, and so on. While accounting for these differences may reduce the magnitude of the difference in costs, the central story, I think, remains true.

    Juxtapose these data to the use of all-payer rate setting by government agencies in France, the UK, Japan, Canada, and possibly Italy, and you have a strong argument for a causal relationship. In Germany, sickness funds negotiate prices with the providers (and global budgets are imposed on spending), so there is less monopsony power, and Germans pay relatively more for healthcare as a result. Insurers in all these countries, including Germany, are non-profit.

    Because of the ascendant health insurance theory in the US (purporting to show that the additional health care consumed when insured is not worth the costs), the results of the RAND Health Insurance Experiment (purporting to show that additional care has no significant effect on health), and the political interests of the providers, we have spent much of the last 40 years trying to reduce healthcare costs by reducing the quantity: cost sharing, managed care, consumer-driven health care. According to the OECD data, we have succeeded, but costs are still double on a purchasing power parity basis. It seems to me that it is time to switch our focus to prices, rather than quantity, as has been suggested by a number of others, including Murray, Volsky, Mahar and Austin. There are a number of policies, including all-payer rate setting and single payer and possibly the public option, that would likely accomplish this.

    • Isn’t it our “culture” to spend more and get less? Why would we want to live in a socialist health care system which gives us twice as much medical care at half the cost?
      Of course, you are right, it is the price of medical care. We are overpaying for everything due to our slavish devotion to the “free market”. The problem is that the regulated (doctors, hospitals, pharma, insurance, etc.) have captured the regulatory process and our political system is deeply corrupt so don’t expect any regulatory relief.

    • Here’s an interesting experiment for you: take a stroll through the doctors’ parking lot at a US hospital, and then do the same for one of the other developed countries. The difference in number, size and luxury levels will be obvious.

    • I don’t begrudge MDs their German cars. They (MDs, not the cars) are a valuable resource in our economy, and I’d like to think that we reward them for that. I DO think that the fee-for-service model is a big part of the problem. Why can’t we agree on a system that allows the free market to compete over patients where they are incentivized to keep customers healthy?

      Also, I have a problem with some of these metrics. What would be a “good” number for average hospital LoS? Hospital discharges per 100 pop?

    • The fact that we ever took our eyes off of “prices” in lieu of anything else is astonishing! Health care has so many factors which make it an irregular market to judge. Consumers are typically removed from the point of purchase or equilibrium in several ways so it can be very understandable that there are substantial distortions.

      The first barrier to consumers and suppliers of health care is usually insurance policies, which often consumers don’t even decide upon. Insurance markets are often thought to be asymetrical information markets, but I’ve often wondered which side had more info. Insurance also functions as a way of bundling goods, which gives the consumer less freedom. Jonathan Levin wrote a working paper about how insurance premiums don’t reflect individual differences in costs because out of pocket expenses are generally not risk rated. (Bundorf, Levin, Mahoney, Pricing and Welfare in Health Plan Choice). It is likely that this issue of bundling erodes large parts of the consumer surplus in demand, resulting in higher prices.

      There are also so many other reasons that our health care markets are abnormal. Our doctors have been given license to operate as a monopoly of sorts (Arrow, Uncertainty and the Welfare Economics of Medical Care, 957). Also, almost half of our health care is paid for by the government, which will functionally increase demand and correspondingly raise prices. Health care is probably the most regulated industry in the United States which, using the Richard Posner idea that regulations function as a tax, create exogenous price increases on the equilibrium. Many of these factors that raise prices would likely decrease demand (and to a certain degree have) but largely have not because of bundling issues.

      I’m a economics student that hasn’t really studied health care, so if I’m way off… I apologize. This is just my two cents on why U.S. health care markets are abnormal, which I think we can all agree that they are.

    • I haven’t had time to peruse your entire oeuvre here, but I can’t help but wonder if you are aware of the problems with using life expectancy and child mortality as meaningful tools to assess variations in the quality of care between developed countries.

      In the case of infant mortality, there are significant differences in what counts as a live birth from one country to the next. The US counts babies as live births that would be considered stillborn elsewhere, and this introduces significant registration artifacts into the data. Even after correcting for those defects, you’d still have to normalize by maternal age, etc to get a valid metric.

      In the case of life expectancy the data are compromised by variations in mortality that doctors and hospitals can do little or nothing about – accidents, murders, suicides, lifestyle choices, etc, etc.

      In short – these are terrible metrics to use to make determinations about variations in health system performance between first world countries. They tell you little or nothing about clinical efficacy, and completely ignore morbidity. This matters quite a bit – since there are an infinite number of conditions that can make your life much more difficult and painful, but will not make you die prematurely. I’d wager that a significant fraction of all spending on medical care addresses conditions that are painful but not fatal.