In the latest debate: Paul Krugman attacks Greg Mankiw for linking to a study by Robert Book arguing that administrative costs under Medicare are not as low as many people think.  Book defends against Krugman's attack here.  I find the debate peculiar for a number of reasons:

1)  Picking out one measure of health care "costs" to compare systems is sadly reminiscent of the arguments for socialism.  Do you remember those arguments?  Under socialism:

• "Think of how much money we will save on advertising!"

• "Socialism will lower costs by maximizing economies of scale!" 

• "Money will be used for production not profits!"

Exactly these arguments are regularly trotted out in the debate over administrative costs in health care so color me unimpressed.  To be clear, the point is not that these statements are false – the point is that these premises to the argument are all in some sense true it's just the conclusion, socialism is more efficient than capitalism, which turned out to be false.  We tried that and it didn't work. In other words, you have to compare systems not arbitrarily pick out for comparison one type of costs.

2)  Closely related to this point is the bizarre habit of taking about costs without mentioning benefits.  The implicit argument appears to be that administrative costs are simply waste – this is the ancient cutting out the middleman fallacy.  Administrative benefits, for example, reduce fraud and are a necessary consequence of making it easy for patients to get second and third opinions from different doctors.

3)  Even if we could switch from a private to a public system and save administrative costs, the deadweight costs of taxation will far exceed any reasonable savings.

4)  Any savings on administrative costs is a one-time level effect but the real issue with health care costs is growth as a share of GDP.  (By the way, this same point explains why the debate over whether the public plan will discipline private monopolies is not especially important, monopoly–even if it is a  problem–could at best explain a level effect not a growth effect which is where the action is.)

5)  I'm not surprised that administrative costs under Medicare and under Canada's system suggest some potential cost reductions from moving to a single-payer system–again, Lada did save on marketing expenses–but it's a complete blunder to use Medicare administrative costs as an argument in favor of a "public option."  The whole point of the public option, so we are told, is to compete on a level footing with private plans which means marketing expenses and all the rest.     

Addendum: n.b. this post is about administrative costs not other reasons for preferring one system to another.  See also Tyler on administrative costs further below.

Adverse selection is an easy story to tell but a hard story to verify.  In fact, empirical studies indicate that adverse selection is not an important (equilibrium) effect in the market for used cars, or used trucks, or of auto, life insurance or health insurance.  See my earlier post, Adverse Selection is NOT the Problem, for reasons why markets handle asymmetric information better than most economists think.

In two excellent posts (here and here), Bryan Caplan further points out that the adverse selection model does not explain current regulations.  Adverse selection, for example, implies that it's the low-risk consumers who drop out of the market so it's the low-risk consumers who need a mandate to buy insurance.  But…

When you actually look at these regs, you'll notice some peculiarities:

1. Mandatory insurance is most prominent in the auto insurance industry. But these regulations don't target low-risk drivers. Their main purpose, contrary to the adverse selection model, is to make sure high-risk drivers get insurance.

2. Even more shocking: The regulations usually go on to somehow subsidize the rates that high-risk drivers pay. This is necessary because, contrary to the adverse selection model, insurance companies are able to detect high-risk drivers, and do not want to cover them at a loss.

3. Economists usually mention adverse selection in the context of health insurance. But in the market for individual health insurance – precisely where you'd expect adverse selection problems to be most severe – governments very rarely mandate insurance coverage. Instead, they focus on mandatory employer-provided health insurance, where the adverse selection problem is likely to be milder.

4. When governments do mandate health insurance, they almost always subsidize the rates that high-risk buyers pay. This is once again necessary because, contrary to the adverse selection model, insurance companies are able to detect high-risk customers, and do not want to cover them at a loss.

Bottom line: Real-world insurance regulation has little or nothing to do with economists' "moral hazard and adverse selection" mantra. The "intellectual" bases of real-world regulation of insurance are rather populism and paternalism: Big bad insurers won't cover people unless it's profitable, and simple-minded consumers don't care enough about their own health to pay for it themselves.

See also Tyler's post on this issue which makes many similar points.

The US spends considerably more per-capita on medical care than other countries, without an obvious increase in life expectancy.  Yet what we make of this depends a great deal on the value of human life.

The value of a quality adjusted life year (QALY) is often set at $50,000 although more recent research puts it at $100,000 to $300,000 or even higher. Kidney dialysis, for example, costs $70,000-$100,000 per year and the quality of a life-year on dialysis is estimated at about half the value of a fully-healthy life-year which suggests that Americans are willing to spend $140,000-$200,000 for an extra quality-adjusted life year.  Let's go with $100,000, you may adjust as you see fit.  

Let's imagine that all of the extra spending in the US adds one QALY to US citizens.  How much is that worth?  Well $100,000*300 million is $30 trillion but we don't all get the QALY at the same time.  We could do some fancy discounting by age but let's instead imagine that the QALY goes annually to the people who are dying – that is, we will assume that the people who died this year lived one QALY more than they otherwise would (since everyone dies this involves no double counting). 2.5 million people die annually in the United States so the total QALY increase per year is worth $250 billion ($100,000*2.5 million). 

US health care spending is around 15% while in many other advanced countries it's 10% so call the extra spending 5% of GDP or $670 billion.  Thus, on this calculation we spend 2.6 times as much as is justified by a one year increase in QALY; alternatively, one QALY must be worth at least $260,000 for our spending to be justified.  The latter number is high but not outside the ballpark.  Of course, if medical spending results in less than one QALY to US citizens the value of QALY must be higher to justify such spending.

More generally, when people say we should cut "wasteful" health spending they should specify what they think a QALY is worth.  Politicians who say that they can balance the budget by elminating "health care waste" are selling the same line as politicians who say that they can balance the budget by elminating "government waste."  In particular, it's naive to think that we can save a lot of money by eliminating spending with 0 QALY.  More reasonably, we can eliminate spending with high costs per QALY.  For example, dialysis for the sickest patients (top 10%) costs more than $240,000 per QALY and some heart pumps costs more than $500,000 per QALY.  

Cutting waste means cutting medical care which costs more per QALY than a QALY is worth.  So what is the value of a QALY?  And who does the cutting?

Hat tip to Robin Hanson for discussion.