Paul Krugman outlines the tricks that are used to obscure the distribution of the tax burden:
The Distributional Effect of Tax Cuts — A Brief Note, by Paul Krugman: With taxes on the wealthy on the political radar, we’re going to drowning in a vast wave of double-talk and smothered by vast amounts of fuzzy math. Still, one has to try. So, a couple of notes.
One is that you have to beware of the old trick of saying “taxes”, then slipping into “income taxes”. Most Americans pay more payroll than income taxes, but the reverse is true at high incomes. So focusing only on income taxes makes it seem as if the rich pay much more of the burden than they really do.
Another, more subtle trick involves comparing percentage changes in taxes as opposed to tax changes as a percentage of income.
The starting point is that federal taxes are indeed progressive on average (although there are billionaires who pay a lower rate than their secretaries). And this in turn means that you have to be careful about the question when evaluating a change in taxes.
Suppose that it’s 1979, and individual A is a member of the working poor, paying 12 percent of his income in taxes — basically payroll tax and not much else. Meanwhile, individual B is very wealthy, and pays 40 percent of his income in taxes — as the very wealthy did on average 30 years ago.
Now suppose that 30 years of conservative governance lead to a fall of a quarter in both individuals’ average tax rates; A’s rate falls from 12 to 9, B’s from 40 to 30. Would it make sense to say that they have gained equally from tax cuts?
Clearly not. A’s after-tax income has risen from 88 to 91 percent of pretax income, a gain of 3.4 percent. B’s after-tax income has risen from 60 to 70 percent of pretax income, a gain of 16.7 percent. The distribution of after-tax income has become substantially less equal. And that’s the calculation I was doing here.
Now, right-wingers come back and say that this is what has to happen when you cut taxes. No, it doesn’t. And anyway, cutting taxes is itself a choice — and they’re a choice that then leads to demands that we cut programs for the poor and middle class to close the deficit those tax cuts created.
The point is that yes, tax policy these past 30 years has been very much tilted toward benefiting the rich.
Another trick is to say that taxing the rich won't raise much revenue, certainly not enough to close the debt gap, with the implication of why bother at all if it won't fix the problem? But while it's true that raising taxes on the wealthy isn't enough by itself, it can still make a hefty contribution:
Is There Enough Income at the Top to Make a Difference to the Deficit?, by Kash Mansori: In response to Obama's proposal (pdf) to let the Bush tax cuts expire for high-income households as one of the ways to close the budget deficit in future years, I've heard and seen a number of commenters assert that there simply aren't enough people at high levels of income for that particular idea to make much difference to the federal budget deficit. Are they right?...
[I]f the US is only willing to raise taxes on the very top of the income distribution, the US's medium-term budget problems cannot be solved through additional revenue alone. However, tax increases that are limited to just the very top of the income distribution, while not sufficient by themselves, would actually probably get us about one-third of the way toward fixing the US's medium-term deficit problems. So while not a cure, it would make a significant dent in the problem.
And if the wealthy don't pay their share, guess who will, one way or the other?