Administrative costs, a simple point or two

Andrew Gelman serves up some links.  Mankiw's addendum serves up some more.

Public sector programs usually have higher administrative costs — all relevant costs considered — than corresponding private sector programs.  The public sector program is funded by taxation.  That means the public sector doesn't have to worry so much about marketing or meeting payroll on commercial revenue alone.  That will bring significant cost savings on administrative matters.  But you can't stop counting there.

The deadweight loss from taxation is perhaps twenty percent or more.  (It depends on which tax you consider as "the marginal tax" and there is not a simple factual answer to that question.)  That should be factored into any comparison, even if you define that cost as "not an administrative cost." 

The public sector also engages in less monitoring of who receives its services.  If you're 67 and have worked a lifetime in this country, usually you can receive Medicare benefits.  The "indiscriminate" nature of the program may be either a net social cost or a net social benefit but certainly it should not be counted as zero or ignored.

If you favor "indiscriminate" programs over targeted programs, OK.  But the accompanying lesson is not one about the relative efficiency of the public sector at a comparable task.  The lesson is that sometimes the public sector can be more effective when you don't wish to discriminate in supplying a particular kind of service.



If marketing is the price for voluntarism that's a deal. There also seems to be no recognition from the left that a product that is for everyone had better be darn near perfect.

I'd rather the government provide a technical service that is easily compared to competitive innovations rather than things everyone likes because they take what they can get for free and thus provide no motivation for improvement. Everyone needs water, so maybe the government can supply that. And if someone figures out how to supply cleaner, cheaper drinking water it will be easy to replace the government program.

Maybe the government should just provide free checkups, but not major procedures. A diagnosis of "you are not average" kicks you out of the public good domain. No offense, but curing your particular rare disease is not a public good.

The lesson is that sometimes the public sector can be more effective when you don't wish to discriminate in supplying a particular kind of service.
In other words, if you don't spend money trying to deny healthcare, you can provide healthcare at a lower price.

Tell me again why it's not significant that France can provide better healthcare than the US at half of the US price.

Don't forget the fact that they can "push down" their administrative costs to their counterparties via edict. The government does have the ability to make laws, ya know.

What do economists mean when they say that there's no such thing as a free lunch? Is air a free lunch? If you're asked to give a lecture, and you would have happily paid for your own lunch, and your hosts pay, does that count as a free lunch? Does anyone deny that there's such a thing as a cheaper lunch? Isn't a cheaper lunch, properly framed (on your way to the restaurant, you find $22.37 on the sidewalk) a free lunch? Couldn't France's healthcare system count as a cheaper lunch?

Economists can't deny that there are exchanges in which everyone is better off, since that's a market. Does that count as a free lunch? Or is it just a lunch at a good price? Nor can they deny that some government systems provide more utility at less cost, since that's a consequence of the critique of socialism. Is not having a the East German government of 1980 a free lunch? Is avoiding the anarchy of Somalia a free lunch?

Suppose you lived in the developed state that had a system of governmental regulation and governmental service that did the worst job of providing health care (for the money) of any state in the world. Would moving to a different, better system be a free lunch? I say: TISATAAFL.

Don't forget continued rent-seeking!

Brian: Deadweight loss is most certainly a problem during a recession. It could even be a cause of one. If taxes on (say) labor are too high, the price employers pay for labor will be much higher than the price workers receive. The higher the tax, the bigger the gap. The bigger the gap, the more instances in which the maximum wage cost employers results in a net wage too low for workers to except -- that is, the more unemployment. And therefore, the worse the recession is.

Imagine the extreme case -- a 100% income tax. No matter how much you make, you actually receive nothing. Therefore, no one works. Result: 100% unemployment and GDP=0.

Obviously, smaller taxes will have less extreme effects. But increasing a tax can surely make a recession worse.

The private sector provides deadweight losses as well: employer provided health insurance creates some deadweight loss just as surely as taxation would.

In the current discussion the only alternatives for the US seem to be either the fragmented and completely intransparent private insurance market of today or a presumably inefficient single payer.

The Swiss system looks like the perfect compromise: Private insurers compete with each other while the government creates market transparency by defining a package of minimum coverage that each insurance has to offer. The insurers compete on price and service quality for the minimum coverage and are free to offer additional packages on top. Insurance is mandatory to spread the risk and there are subsidies for the poor. Works great!

The big remaining problem IMHO is how to create a functioning market between insurers and service providers, i.e., doctors, pharmacies, hospitals etc. I don't know of any system that gets anywhere close to providing a solution there.

"The lesson is that sometimes the public sector can be more effective when you don't wish to discriminate in supplying a particular kind of service"

Exactly, and that's a pretty good argument for the public option. Insurance companies (and HMOs, etc) seem to spend a great deal of effort avoiding the sick or potential sick while attracting the relatively healthy. Their ability to do that well may be the number one factor affecting their profitability. Of course, this is a negative-sum game. It's far cheaper to simply cover EVERYONE.

I don't get the TANSTAAFL slogan. There often IS a free lunch (or pareto improvement) when we can identify a more effective way to organize economic activities (like healthcare). In fact, isn't identifying (and showing how to mitigate or even eliminate) inefficiencies like adverse selection and rent dissipation one of the most valuable things economists can do?

Leftists seem to strangely unable to divorce the concepts of "costs" and "benefits". We spend more money than any other country in the world - yes; because we receive larger benefits from that spending, and people rationally want to spend a greater portion of larger incomes on health-care, as was shown in some research paper I won't look up, because of course you won't listen anyway and will continue to ply the same dishonest nonsense elsewhere. The costs of biomedical research may be well 3% of the GDP, but surely the BENEFITS (social, private) are greater than that. That's why the INVESTMENT is undertaken in the first place. Because the BENEFITS are greater than the COSTS.

@Oreg: France doesn't include babies that die soon after birth into their infant mortality or life expectancy stats. Be careful of using any medical stats from the french gov, they are very suspect.

@Oreg: To attribute a higher life expectancy to the quality of the health care system seems to require a leap of faith that is not warranted by the facts - at least the few facts that are usually reported. I think we would all benefit from seeing data from a credible source that controls for such factors as suicide/homicide rates, high-risk pregnancies rsulting in live births, lifestyle choices (i.e. rates of chronic disease resulting from obesity, lack of exercise, drug use, stress, poor nutrition, alcohol consumption, etc.), demographic variations, etc. I would also be interested in seeing exactly how spending per capita is calculated, and what is included/excluded from the calculation. The figure on its own tells us nothing.

Comments for this post are closed