Tax my consumption, please

Continuing with my musings about the President’s Advisory Panel on Federal Tax Reform (and here’s the link, since I forgot to provide it in the last message). When the Panel issues its report in July, we can expect that the main message will be a call to scrap the current tax code and replace it with a consumption tax. We can expect Democrats to oppose this on the grounds that a consumption tax is fundamentally regressive: since the rich save a larger portion of their income than the rest of us, more of their income is shielded from taxes. I (pre-emptively) oppose this argument for three reasons: (1) it’s wrong, (2) it puts the Democrats in a weak position in the political debate as defenders of the current tax system, (3) it ignores the real danger of the Panel’s (i.e. Bush Administration’s) proposals.



The primary benefit of a consumption tax, according to its proponents, is that by exempting savings from taxation it promotes savings, investment, and growth. If you plug numbers into the standard growth model, you find that a pure consumption tax would raise per capita GDP by maybe 6% relative to a simplified version of the current system. That’s not a bad chunk of change (6% of GDP is about $700 billion, enough to fight the wars in Iraq, Afghanistan, and five or six other countries), but I don’t believe the effect would be that big. Nevertheless, in terms of economic growth I don’t see any downside to a consumption tax and there’s the possibility of an upside, so on that basis I don’t have a problem with it. The only serious downside to a consumption tax would be its regressivity. However, the claim that a consumption tax is necessarily regressive is just plain wrong.



A progressive tax is commonly defined as one in which those who earn higher incomes pay a higher average tax rate. A consumption tax is said to be regressive because while most people consume all of their income, high income earners save some, so pay taxes on a lower base. Suppose, for example, that there is a uniform tax rate of 25% on consumption. A person making $40,000 a year has no saving, and so pays $10,000 in taxes, or 25% of her income. A person making $400,000 a year saves half of his income, so gets taxed on only $200,000. He ends up paying $50,000 in taxes, or 12.5% of his income.



The first problem with this reasoning is that annual income is not the best measure of a person’s well-being, and therefore not the best metric by which to measure progressivity or regressivity. The lifecycle model of consumption says that though a person’s income tends to bounce up and down over the course of his life, he tries to maintain a relatively constant level of consumption by borrowing when income is below average and saving when income is above average. Since, as we all know, “you can’t take it with you,” the smart person will die leaving no net assets or debt. In other words, the theory predicts that every individual consumes, over the course of her lifetime, an amount exactly equal to her lifetime income. So over the course of a person’s lifetime, a flat-rate consumption tax is neither progressive nor regressive, but proportional, exactly as a flat-rate income tax would be. What about the (relatively few) people who leave bequests to their children? True, they don’t get taxed during their lifetimes (but then they never enjoyed the fruits of their labor during their lifetimes). But to the extent that their children, thanks to their inheritance, are able to consume more than their income over the course of their lives, they pay a higher fraction of their income in taxes than the poor. I’m happy with a tax code that rewards Conrad Hilton for creating immense amounts of wealth and penalizes Paris Hilton for squandering it.



The second problem with the argument that a consumption tax is necessarily regressive rests on the false assumption that a consumption tax must also be a flat-rate tax. It is certainly true that all of the consumption tax proposals that have been offered are also flat-rate taxes. And it is difficult to imagine how a sales tax, which is one way to institute a consumption tax, could be made anything but flat-rate (you could exempt food and rent from the sales tax, but it’s not clear to me that the rich spend a smaller fraction of their income on food and rent than do the poor). But there’s more than one way to skin a cat. I could demonstrate mathematically that one way to convert the current system to a consumption-based tax system would be to do two simple things: (i) retain the current income tax system, but exempt income from interest and dividends, and (ii) replace the current tax on corporate profits with a tax on cash flow (revenues minus wage payments), allowing for complete expensing of new investment. Alternatively, one could eliminate the corporate tax entirely, keep the tax on interest and dividends, but allow households to deduct savings. Both of these systems retain the individual income tax, and so both can be made as progressive as you want. There’s nothing to stop Democrats from proposing a consumption-based tax system with, say, an exemption for the first $40,000 of income and a top rate of 90 percent.



Democrats will also box themselves into a difficult political position by opposing a consumption tax in principle. In taking this stance they will be accepting the position of defenders of the current tax code, just as in the debate on Social Security they have accepted the position of defenders of the current Social Security system. The difference is, Social Security is wildly popular, and the tax code is just the opposite. Democrats ought to be out their proposing their own alternative tax reform; otherwise they’ll be sitting ducks for the Republican spin machine.



Finally, if Democrats focus on the inequities of the consumption tax, they will take attention away from the real problems with the Republican proposal. First, it is quite likely that the Bush administration will propose something very much like a flat-rate system (perhaps with exemptions for very low income people). They will try to piggy-back a flat-rate tax on their consumption tax proposal, with the result that the obscenely rich will end up shouldering no more of the tax burden than the middle class. But as I described above, the two issues are easily separable: if the Democrats would accept the idea of a consumption tax, they could take on the issue of the flat rate directly. Second, the Panel's statement makes clear that any proposal they come up with will be “revenue neutral,” that is, the proposal will not be used as cover for a new tax cut or an increase in taxes. But the Council says that it will be using the “baseline in the President’s budget” as the basis for revenue neutrality. The baseline in the President’s budget assumes that the 2001 tax cuts, which are scheduled to expire in six years, will instead be made permanent. So a “revenue-neutral” tax reform proposal will turn out to be a very sneaky way of achieving the priority that the president laid out in his 2004 State of the Union address – making his earlier tax cuts permanent.



So if I were Nancy Pelosi or Harry Reid, I would within the next month put together a vision for tax reform that is easily distinguishable from the Republican’s proposal. The plan would be consumption-based, so Democrats could not be attacked as either defenders of the current system or as anti-growth. I would insist that the plan be made at least as progressive as the tax code was before the 2001 tax cuts. I would insist that any reform generate revenues sufficient to cover the federal government’s spending over the long run. Bush’s “revenue neutral” plan would raise tax revenues in the amount of about 17 percent of GDP (that’s what the government is currently taking in). But for the last forty years, through Democratic and Republican administrations alike, through Democratic and Republican Congresses alike, the federal government has been spending about 20 percent of GDP. So any new tax code ought to raise about 20 percent of GDP. The Republicans would package this as a tax increase; Democrats would have to package it as a budget-balancing plan. Finally, I would sweeten the tax proposal with provisions that would make my liberal base (not to mention my liberal self) very, very happy. First, as long as we’re taxing consumption, we might as well tax the most harmful type of consumption especially heavily: the consumption tax should fall especially heavily on fossil fuels, gasoline above all. The reform now is not only a bold step forward on tax policy, but an essential component of a market-based attack on global warming. Several European countries have already instituted similar “pro-growth green tax reforms”, but we don’t have to bring that up. In addition, we would recognize that for middle income Americans, the two most important types of saving are not the purchase of stocks and bonds, but spending on housing and education. My reform would treat the purchase of a new home as saving (thus replacing the mortgage interest deduction, which would have to be eliminated in the move to a consumption tax, hence dodging a major political bullet) as well as college tuition and fees. Now we’re not only pro-growth and pro-environment, but pro-home ownership and pro-education as well. Not a bad place to be.