Tuesday, September 29, 2009

Greg Mankiw Misreads Larry Summers

Are mandated benefits, such as a compulsory health insurance, simply taxes? Greg Mankiw (“Is a mandate a tax?”, his blog, Wednesday, September 23) thinks so and quotes Larry Summers’ 1989 paper (“Some Simple Economics of Mandated Benefits”, The American Economic Review, May 1989) for sustaining this point of view.

I think that Mankiw is wrong and did not read Summers thoroughly. Summers wonders indeed:”Are there differences in the real effects of mandated benefits and tax-financed programs?” And he goes on “…I find that there are important differences in the efficiency and distributional consequences of standard public provision and mandated benefits programs. Essentially, mandated benefits are like public programs financed by benefit taxes. (My emphasis. And this last sentence is the only one quoted by Mankiw).”

But then Summers adds in section 2, titled “Mandated Benefits or Public Provision”:
“It is often asserted that mandated benefits are just hidden taxes with the same efficiency and incidence implications as taxes, so that the choice between public provision and mandated benefits should depend only on the relative efficiency with which employers and the government can provide a service. I challenge the equivalence of these two methods of provision below.”

A first reason for doing so is that privately supplied mandated benefits preserve a margin of choice regarding the quality of service. Public provision supplies the same service to all, while some of the insured, given a choice, would select a lower quality and lower cost service, while others would prefer higher quality and cost. There is thus a welfare loss inherent in a one-size-fits-all insurance service and price.

More importantly,though, Summers sees differences in welfare deadweight losses:

“Another argument in favor of mandated public benefits rather than public provision is that mandated provision avoids the deadweight loss of tax-financed provision”. And he quotes various evaluations of the importance of the deadweight loss in the literature (Stuart, Browning, and others).

This is precisely the argument I use in my proposal for a reform of the health insurance systems in France and Europe (“Comment gagner plus?” Commentaire, 2009, downloadable on my personal homepage) advocating a repeal of the current dual purpose tax-financed public provision, in which the tax serves the dual purpose of covering the health care expenses (an insurance scheme) on the one hand, and redistributing income from the high wages earners to low wage earners through the payroll tax, in order to supply to the latter the same public provided health insurance coverage but at a much lower cost than to higher wage earners, on the other hand.

What I suggest is to limit the use of a payroll tax to this second, redistributive, aspect (about one third of the current tax), and to replace the publicly supplied insurance by a privately purchased but compulsory health insurance. This would reduce the payroll tax by about two third, and the deadweight loss by the same proportion. It would thus allow for a major reduction in the level of taxes in Europe, given that the payroll tax financed health insurance amounts to about 8 to 12percent of GDP. A reduction of two third of these taxes would represent a tax cut of about 5 to 8 percent of GDP -- a major tax rebate, without any cut in the services provided to insured people.

Definitely, such a mandated benefit formula cum the same income redistribution characteristics as the present systems (a health care insurance voucher for lower income households) would constitute a Paretian improvement over the present European tax financed and publicly supplied health insurance systems. And it would boost growth too.

I would be glad to know what Greg Mankiw thinks of such a reform.

No comments: