Miles Kimball summarizes the economics of American health care; in particular, he reviews how, even before the Affordable Care Act, “There [were] at least five big departures of our health care system from a classical free market”:
1) Health care consumers almost cannot, by definition, be well-informed. Or, at any rate, there are informational asymmetries between health care providers and patients.
2) People have empathy. Strangers getting sick and not receiving health care makes the typical person feel poorly.
3) Regulatory practices artificially restrict the supply of medical providers in the country–there are a very limited number of medical school spots, and, given the acceptance rates and number of qualified applicants, there are almost certainly qualified individuals that can’t get into med school. Furthermore, scope of practice regulations limit the number of procedures that (even qualified) PA’s and nurses can perform (which is especially troubling in rural areas with fewer doctors).
4) Third-party payment systems remove consumers from the cost of medical care. Combine this with the traditional fee-for-service model that doctors employ, and there are few incentives to hold down costs at the point-of-delivery.
5) Patents drive up the cost of new medical technology by allowing pharma companies and the creators of new medical technology to charge monopoly prices.
While this is a fairly comprehensive list of the market failures in American health care (and teh list makes it pretty obvious why reforming health care and bending the cost-curve isn’t completely straightforward); I thought there were a couple points that would be useful to add:
1) One of the biggest market failures that insurance markets, in general, suffer from, is adverse selection. Basically, insurance pools need a large number of healthy people paying insurance premiums to cover the costs of the sick. However, healthy people tend, at the margin, to under-invest in health care in the US. This could be because they are young and think they don’t need it, or it could be because if they show up in an emergency room, they will still receive emergency care.
This tends to drive up costs in a couple of ways–first, it makes insurance premiums more expensive. With fewer people in the insurance pool, the % of unhealthy people is higher, meaning that costs need to be higher to cover the same number of procedures. Additionally, emergency care is basically the most expensive medical care you can receive in the US. It would be far cheaper for people to go to the doctor more-regularly and cut down on ER trips. [It’s probably obvious, but the whole point of the individual mandate in the ACA is to address this specific issue].
2) Second, I wish Kimball had paid more attention to the issue of both patents and the third-party/fee for service payment system. Via The Incidental Economist: the data makes it very clear that technology is the overwhelming driver of health care cost-growth over the last 50 years.
Now, obviously technology is an extremely vague term. But basically our entire medical payments infrastructure offers perverse incentives for adverse behavior. First off, new medical technology tends to be over priced relative to the marginal cost of production because of patent protection. Second, our payments system incentivizes more procedures rather than fewer procedures–patients bear little to none of the cost, and some doctors are paid on a fee-for-service basis; not to mention there might be liability issues involved in not recommending a new treatment for a patient if it could potentially be effective. Last, developing effective metrics to measure the impact of new medical technology is tough. How do you measure the cost-effectiveness of a new drug treatment? If, a new drug will allow 5% of people to live a year longer than they otherwise would, but 95% of people that receive the treatment don’t get any benefit, and the drug is very expensive and leads to a lower quality of life, on average, is that drug worth the cost? Questions like that are incredibly difficult to quantify and answer.
However, they are the questions that are going to become increasingly important as we hope to bend the health care cost-curve. Look at this chart, via Wonkblog:
This chart plots the cumulative % of health care spending against the cumulative % of the US population. What you see is a classic example of a power-law distribution, or the 80/20 rule–20% of individuals in the US account for 80% of all health care spending. 5% of people account for 50% of healthcare expenditures. 1% of the population–just over 3 million people–accounts for over 20% of all US healthcare expenditures. To put that into context, health care is a roughly $3.2 trillion industry in the US–20% of that is $640b. 3 million Americans account for $640 billion in health care spending. That’s over $200,000 per person, and the evidence suggests that most of this spending is concentrated among individuals with chronic illnesses. Figuring out a way to treat these patients more cost-effectively will be the key to bending the health-care cost curve moving forward.