Friday, April 22, 2011

Progressivity, Political Strategy, and Consumption Taxes

-- Posted by Neil H. Buchanan

In response to my post last Friday, a loyal Dorf on Law reader/commenter raised some interesting points about consumption taxes. He suggested that, even though consumption taxes are not (contrary to much political hype in the U.S.) inherently less complicated than income taxes, there could nonetheless be good reasons (even for those who believe in progressive redistribution) to support a move to a consumption tax. He posed two questions to explore this possibility, one of which I will address here. (I hope to address the second question in a later post.)

Can a consumption tax system can be designed to be more progressive than an income tax system? Answering this question requires some important background work.

Because non-rich people tend to save very little of their incomes, all (or nearly all) of their incomes would be subject to taxation under a consumption tax, whereas the high-saving rich folks would see large amounts of their incomes exempted from taxes in a consumption tax regime. If the new system will be used to raise revenues at approximately the same levels as the old system, this would represent a regressive change in the tax system, with tax liability shifting downward from higher- to lower-income people.

For tax types, this is old news. Most real-world consumption tax systems (such as state sales taxes) include exemptions for "necessities," such as food and medical care, which requires higher rates to raise the same amount of revenue. Similarly, most academic proposals to adopt a consumption tax in the U.S. include some kind of "demo-grant," which is a cash benefit from the government that is sent to all (or, under means-testing, some) taxpayers. If, for example, everyone received a $10,000 annual grant from the government, this could effectively offset the regressivity of the consumption tax (and then some, depending on the size of the grant). The tax rate would, again, have to be set at a higher level to pay for the grants. The idea, however, is that you would then have a tax system that is not an income tax, but that shifts the tax burden (net of tax benefits) onto higher income earners.

This is, admittedly, an appealing idea. Indeed, the liberal democracies of Northern and Western Europe are able to combine consumption taxes with benefit programs that significantly narrow the gap between rich and poor in those countries. It is, therefore, realistically possible to have a fiscal system that is progressive overall, even if it includes a consumption tax.

Even so, I think that the realities of U.S. fiscal politics make it a bad idea to agree to a grand bargain by which we adopt a consumption tax coupled with progressive redistribution. My concern is that, even if we were able to negotiate an agreement that would be more progressive than the current (at best weakly progressive overall) system -- a highly unlikely possibility, I should say -- we would be much more likely to see movement in regressive directions almost immediately, for reasons that are unique to the design of a consumption tax.

The U.S. political debate is, after all, deeply preoccupied with "giveaways," the idea that the government should not give people something for nothing. Progressive consumption tax regimes tend to rely on mechanisms that can all too easily be mischaracterized as giveaways. The most obvious target, of course, would be the demo-grant. Even though the grants could be designed specifically to offset the consumption tax liabilities of poorer citizens, it still amounts to the government cutting checks to poorer people.

The current debate over the refundability of the Earned Income Tax Credit suggests just how easy it is to deride as giveaways programs that are designed to offset tax liabilities. An effective demo-grant system would have to be much broader, involving much larger checks, to offset the effects of switching to a consumption tax regime.

Admittedly, it would be more difficult to attack progressive exemptions and progressive caps as giveaways to the poor -- or, at least, not inherently easier than attacking similar provisions in an income tax system. The problem is that those provisions are more necessary in a consumption tax system to guarantee overall progressivity than they are in an income tax system.

It seems a safe prediction, therefore, that the first thing we would witness after adopting a progressive consumption tax system would be a concerted attack on the parts of it that are most essential to making it progressive. In addition, the politics of setting tax rates under a consumption tax regime would tilt the playing field against progressivity. Under an income tax system, there are simple ways to raise money from only the wealthiest taxpayers, most obviously by raising high-income rates. Under a consumption tax system with a single rate (the preferred system for most consumption tax advocates), you can only enact a progressive change if you raise rates while increasing the demo-grant (or while widening the other exemptions to benefit the non-rich).

We could, on the other hand, adopt a consumption tax system with a graduated set of rates (which is only administrable, by the way, if we continue to have people submit annual tax returns, rather than setting the system up as a sales-tax equivalent). Doing so, however, does not get us where we need to be. Because higher-income people's incomes would be largely exempt from a consumption tax, it requires (compared to an income tax system) larger increases in rates to collect the same amount of additional revenue from richer taxpayers. The usual arguments against high rates would thus be intensified under a consumption tax.

Progressivity in both spending and taxation are under fierce attack. Nothing that I have written here should be interpreted to suggest that I think that our income tax is sufficiently progressive, nor that I think that it is easy to achieve progressivity under an income tax regime. On balance, however, I think that the income tax system provides -- even under our current political constraints -- a better framework for defending, and perhaps increasing, progressivity in the U.S. fiscal system. Any system could be designed to be more progressive. The income tax requires less work to make it so.

1 comment:

egarber said...

One thing I've read about regarding consumption taxes is that they tend to make revenues more directly tied to business cycles.

Of course, any tax framework carries such a correlation, but I'm wondering what the latest downturn would have done to tax revenues under a consumption tax. With all the saving, it seems to me the downward swing in revenues would have been worse than under the current income tax code. The minimum spending on basic needs would be largely exempted, so cutbacks on higher value items would seemingly have a severe effect.

Even under our actual framework, the hit has been devastating. Why experiment with something that might be much worse?

Has this been studied in any detail?