TurboTax notwithstanding (think about this: is it right that calculating your taxes should require a computer, or a professional adviser with a computer?) the docility of the American taxpayer surprises me. The compliance costs of the tax code have been estimated at 20 cents per dollar collected, and to me that feels like an understatement. Simplification surely ought to be a priority if the opportunity for meaningful tax reform should arise. Allowing the AMT to take over by degrees, rather than all at once, scores very low on that scale. Come on. Isn't one tax code enough?
An appealing step toward an outright AMT takeover—remembering that the need to collect more revenue is motivating this discussion—would be to abolish virtually all exemptions and deductions in the existing code, except for one standard deduction. The AMT would no longer be needed. Together with some spending restraint, this would raise more revenue than is required, in fact, to balance the budget. The surplus could be shared among lower rates, a bigger standard deduction, an expanded Earned Income Tax Credit (a necessary exception), and a means-tested health insurance subsidy.
A reform like this would so simplify the system that it would spare tens of millions of households the bother of filing a tax return in the first place. Too radical? Another milder possibility, a midpoint between a zero-exemption regime and where we are now, would be to convert all deductions into fixed-rate credits (so that tax breaks would be worth the same to middle-income earners as to people with very high earnings).
All kinds of blue-sky tax-reform blueprints are perpetually in circulation. Such has been the accretion of distortions and anomalies since the last big tax reform (in 1986) that almost any of these would constitute a huge improvement over what we have. But radical reform is a mighty political challenge—and even when it works (as in 1986) it can veer unpredictably out of control. Where you end up can be a matter of luck.
Radical systemic reform would be great if we could get it. The appetite for change seems to be growing, but a root-and-branch redesign may be beyond the country's political capacity. Short of that, one must hope for piecemeal improvements that push in the direction of fairness and efficiency. Abolishing or limiting some deductions is one such avenue. Another, again remembering that more revenue is needed, is a carbon tax. As I have previously argued, this would attack two significant forms of pollution—greenhouse-gas emissions and traffic congestion. It would make sense to levy such a tax even if additional revenue weren't needed (you could instead use the money to cut other taxes).
If Washington does gird itself to take on tax reform, one avenue is especially worth avoiding. This is the line of thinking, popular with some Democrats, that says, let businesses pay. Raise corporate taxes; shift public spending onto companies, with mandates of one kind or another; tax capitalists more so that hard-pressed workers can pay less. The first two ideas are based on a straightforward misconception—that corporations, as opposed to people, can be made to carry part of the fiscal burden. They cannot. The burden always gets divided between owners, workers, and customers. The third idea is based on the unrealistic hope that you can tax owners of capital more heavily than workers without discouraging investment. Discouraging investment really does, in the end, make everybody worse off.
Fairness and efficiency require that capital and labor be taxed at the same rates. What does that principle say about the Bush tax cuts on dividends and capital gains? It says they should be tweaked, but not abandoned altogether. If, but only if, profits have already been taxed at the corporate level, the distribution of those profits in the form of dividends and capital gains should not be taxed in full as ordinary income. The basic idea is simple: Tax people with high incomes at higher rates than you do people with low incomes. That is good policy. But it is bad policy to discriminate—in either direction—between income from capital and income from labor.