Wednesday, March 31, 2010

The French Model

Greg Mankiw compares per capita taxation:

The most common metric for answering this question is taxes as a percentage of GDP.  However, high tax rates tend to depress GDP.  Looking at taxes as a percentage of GDP may mislead us into thinking we can increase tax revenue more than we actually can.  For some purposes, a better statistic may be taxes per person, which we can compute using this piece of advanced mathematics:

Taxes/GDP x GDP/Person = Taxes/Person

Here are the results for some of the largest developed nations:

France
.461 x 33,744 = 15,556

Germany
.406 x 34,219 = 13,893

UK
.390 x 35,165 = 13,714

US
.282 x 46,443 = 13,097

Matt Yglesias says that Mankiw probably doesn't believe that this analysis makes any sense, but he likes the conclusion.

3 comments:

MYOS said...

Does that include State + Federal + local taxes?

It seems to me to be about equal: what you pay in taxes here you'd pay anyway in the US, if not in "taxes" -- glaring example : health insurance (the premium for the kind of coverage the French take for granted must be pretty astounding... and no pre-existing condition clause, here. That alone saves you money.)

Although, clearly, it seems more favorably tipped to the higher brackets if you compare with the lowest ones - I assume one pays more taxes in the US if one is making little, than in France, but only marginally less for the top bracket. (I'm not speaking of the wealthiest families, those in the upper 0,1% ..)

Also, what about tax credits? Those exist in both countries but they seem very generous in France. I count "assurance familiale" stipend, aide au logement, "part" for children, etc, etc.

Tex_Exile said...

Fun stuff, but these numbers ignore the invisible side of government expenditure, which is particularly large in the US.

Tongue firmly in cheek, the late David Bradford used to propose a great way to reduce taxes without cutting government services. Instead of spending billions out of the federal budget on weapons procurement, Congress could introduce a “Weapons Supply Tax Credit” that would give arms manufacturers a tax credit for delivering weapons that met specific criteria (determined, presumably by the Pentagon) to the government. The value of the credit would equal what the government might otherwise have had to spend to buy those weapons.

I recall this, because, whatever their other problems and merits, Mankiw’s numbers omit tax expenditures. If the estimates included the value of tax breaks given to particular groups in order to achieve specific policy objectives, the numbers in all cases would be different. One cannot know how different the numbers would be, because relatively few countries make any serious attempt to define, let alone account systematically for, tax expenditures. To its credit, the US GAO has tried, coming up with a figure of about 7.5% of GDP for the US federal budget in 2005 (the Australian Treasury did likewise and came up with a figure of 4%).

That said, the likelihood is that the upward revision of the numbers for the US would be more radical than most, because the US seems to rely more heavily on tax expenditures than most countries. As Edward Kleinbard recently observed, ‘Tax expenditures have grown in importance to the point where they are now the dominant instruments for implementing new discretionary spending policies.’ He cites such recent examples (which are not far off Bradford’s proposal) as new tax credits administered by the IRS and the Department of the Environment for participation in gasification projects and ‘advanced coal projects’.

The notion that the US federal government is ‘small’ by international standards thus rests, at least in part, on a failure to appreciate the extent to which the federal budget process blurs the distinction between revenue and expenditure.

There are, to be sure, times when intervention via the tax system is likely to be preferable to direct expenditure, but the scale on which such interventions occur in many countries is absurd. Moreover, because tax expenditures are harder to quantify and observe, they are less transparent than simple expenditure. Lobbyists therefore love them.

Unknown said...

Excellent point, Tex. In fact, consider this one item alone (from Gail Collins' column in today's Times):

" It turned out that the $1 billion goes back to the famous 2003 Medicare prescription drug entitlement passed by a Republican-controlled Congress and paid for through their innovative pretend-it’s-not-there financing system.

In order to keep businesses from ending their drug coverage and dumping their retirees on the federal system, Congress provided a 28 percent reimbursement for the benefits. And, the companies got to deduct the entire cost of the drug plans from their taxes. Including the government subsidy."