In a post last week, my fellow co-blogger, Arnold Kling, writes about Obama OMB director Peter Orszag's thoughts on controlling health care costs. I think Orszag misses an important point and so does Arnold.
First, Orszag. He takes as given that the government should try to control health care costs. But why? The usual answer is that health care spending as a percent of GDP is rising. But lots of spending as a percent of GDP is rising. I haven't checked lately but a few years ago spending on airline travel as a percent of GDP was rising. I didn't hear anyone call for reducing or reversing that growth. So what's the relevant distinction between health care spending and spending on airline travel? There is one and it's that so much of health care spending is people spending other people's money. This is true in one of two senses. For government programs such as Medicare and Medicaid, where spending as a percent of GDP has risen much more than private health care spending as a percent of GDP, people are spending taxpayers' money. For private health insurance, even though people contributed either directly or implicitly via lower wages and salaries from their employers, at the point of purchase most people are spending mainly the insurance company's money.
But given that that's the important distinction, our solutions should focus on that. Which brings me to Arnold's point. I don't particularly disagree with what Arnold wrote, but he started his analysis in the middle rather than at the start. Arnold asks:
Next, imagine that the best hope is a treatment that costs $100,000 and offers a chance of success of 1 in 200. Would I want her [his daughter] to get that treatment? Absolutely.
Who wouldn't? But let's ask a prior question. What if Arnold or you or I could buy health insurance that covered treatments that give a 1 in 200 chance of survival and cost $100K? Or we could buy insurance from a company that refuses to cover such high-cost, low-probability treatments. Of course, ceteris paribus, the second kind of insurance would be lower-price. Which would you buy? If the former, then fine: you paid the expected cost. What's the problem? If the latter, then fine: you didn't pay the expected cost. And again, what's the problem?