if we had made a deal with the rich to raise taxes on the bottom while lowering them at the top in exchange for a promise to pay for Social Security later ... well, the figures say that we reneged on that deal decades ago.

From Megan McArdle. Excerpt:

"There's a kind of legend about the Ronald Reagan tax cuts in which the rich were supposed to have paid their "fair share" of taxes from the 1930s to 1980, then got Reagan to give them an unfair handout that persisted until Bill Clinton undid some of the worst of the damage, while still leaving a perversely unequal distribution of income and benefits. The story is a little more complicated than that. I'm not saying Reagan didn't cut taxes for higher incomes in the late 1980s, because he did. But that doesn't fully explain the Reagan budget deficits, and it certainly doesn't account for the Social Security surplus. The complications come from three factors: the broad nature of the tax cuts, bracket creep, and the difference between marginal and effective tax rates.

A little history: Reagan pushed through two major tax bills, one in 1981 that was an across-the-board tax cut of about 23 percent and one in 1986 that lowered marginal rates while hacking away at the number of deductions. Because both bills lowered the top marginal tax rate -- it went from 70 percent in 1980 to 28 percent in 1987 -- it sounds like these were enormous giveaways to the wealthy. If you look at effective tax rates -- the percentage of their income that people paid, rather than the top marginal rate they paid on income above a certain level -- reality is a bit more complicated. For starters, as I said, the 1981 reform lowered rates across the board, not just at the top. Moreover, you can argue that it was actually a correction for a big problem that had developed in the 1970s known among budget wonks as "bracket creep": Basically, effective tax rates were increasing faster than real incomes, especially at the top.

Why did this happen? Inflation. We're used to the IRS publishing new tax tables every year, which, if you didn't know, is done to account for inflation. The idea is that tax rates should apply to increases in real income, not increases in the inflation rate. But this was a Reagan-era innovation. Before then, tax rates were set by law and fixed on an ad-hoc basis. The result was that receipts as a percentage of gross domestic product tended to creep upward in high-inflation periods until Congress intervened. By 1980, they had reached a postwar high of 19.1 percent of GDP, a level they did not reach again until 1998, when the dot-com boom delivered a windfall of highly inflated capital gains taxes to the Treasury. Essentially, inflation was crowding upper-middle-class people into tax brackets that had been designed for the ultra-wealthy years ago, middle-class people into rates designed for the affluent, and so on. But because the rates designed for the ultra-wealthy were really, really high, the effect was most pronounced at the top.

The changes Reagan made to the tax code lowered marginal rates and, starting in 1985, indexed them to inflation. This brought the tax take down, but by 1983, the year of peak deficits, the individual income tax was still taking in 8.2 percent of GDP, which was above the average for the 1970s. That figure does not include payroll taxes, which rose from 5.4 percent of GDP in 1979 to 6.5 percent a decade later.

During Peak Deficit, the corporate income tax yielded only 1 percent of GDP, but that was an anomaly due to the severe recession the economy was suffering. In 1978 -- the year before the oil crisis -- the corporate income tax was delivering 2.6 percent of GDP in revenue, while personal income taxes yielded 7.9 percent; in every year that followed, except for 1983, the corresponding figure for corporate income tax receipts was at least 1.4 percent of GDP. This decline in corporate income tax revenues, combined with a slight increase in personal income tax yields, cannot explain why the budget deficit had gone from 2.6 percent of GDP in 1978 to nearly 6 percent in 1983.
What does explain it? For starters, inflation fell faster than the Reagan administration had expected. Because indexing wasn't phased in until 1985, that meant the administration ended up with much lower revenues than it had expected, as Bruce Bartlett wrote in his book on Reagonomics, because there was less bracket creep. Paul Volcker's radical interest rate hikes also caused a severe recession, which lowered GDP, especially on the corporate income tax side, since corporate profits tend to be more sensitive to downturns than individual incomes. (In 2009, individual income tax receipts fell to 6.3 percent of GDP, down from 7.8 percent the year before. Corporate income tax revenues, meanwhile, declined by more than half to, yes, 1 percent of GDP. In 2007, the last healthy year for the economy, they had been 2.6 percent, about the same level as they were in 1978.) Also, Reagan was spending a lot of money. That helped.

OK, but that's deficits. What about the distribution of taxes? Wasn't there a massive shift from the wealthy to lower-income groups?

The 1981 tax reform definitely lowered income tax rates on the top, though it also lowered them on the bottom. And because the bottom pays more of its income to payroll taxes than to progressive income taxes, those people did see an effective total tax hike after 1983. By 1989, the bottom quintile had an effective total federal tax rate of 7.6 percent, compared with 7.4 percent in 1980. The quintile above them saw their total average tax rate fall to 13.5 percent from 14.1 percent. The top quintile had gone to 25 percent from 27 percent, the top 1 percent to 28 percent from 33 percent. Since the top 1 percent of the country had about 15 percent of the income, this amounts to something less than 1 percent of GDP.

Moreover, that was the tail end of a low period. In 1986, in the face of a persistent budget deficit of roughly 5 percent of GDP, the Reagan administration undertook a massive tax reform that lowered marginal rates but also got rid of most deductions, which actually ended up raising effective taxes on the highest-income groups; the total average tax rate for the top 1 percent jumped from 24.6 percent in 1986 to 30.3 percent a year later. That's why you could lower the top marginal rate to 28 percent from 70 percent and only see effective tax rates decline by five percentage points over that period. But even that didn't last. The George H.W. Bush administration did a big budget deal that raised taxes. The Bill Clinton administration raised them again, and the effective tax rate for the top 1 percent peaked at 35.3 percent in 1995, slightly higher than it had been at the previous peak in 1979. Even after the Bush tax cuts, the effective tax rates of this top group ran somewhere slightly north of 30 percent, or about where they'd been in 1981, before Reagan's tax reform took effect. They only dipped back into the 20s under Barack Obama, because of the lasting effects of the recession.

And that doesn't tell the whole story: Meanwhile, tax rates on the bottom also fell. The average household in the bottom quintile was paying an effective 7.5 percent of their income in taxes in 1979; by 2006, that was 5.7 percent (and it dropped to basically nothing during the recession). The next quintile went to 9.9 percent from 14.5 percent by 2006; the middle to 13.9 percent from 18.9 percent; the upper-middle to 17.7 percent from 21.5 percent; and the top quintile to 25.4 percent from 27.1 percent, while the top 1 percent went to 30 percent from 35 percent. The earned income tax credit, the middle-class tax cuts that Obama made permanent in 2010, and various deduction expansions had amounted to an across-the-board tax cut for everyone, not just the wealthy.

In other words, while there was a short period during which tax rates dropped a lot on the very top and rose slightly on the bottom, that period ended in 1986. Then taxes started increasing, mostly on the highest incomes, until they came close again to their bracket-creep-driven 1979 peak. There was no "deal" to permanently cut taxes for the wealthy while raising them on everyone else. All in, when you look at the numbers from 1979 to now, you see the middle and the very top getting roughly equally sized cuts (though the middle's cuts were larger in proportion to what they paid in taxes), while the bottom, who already had low tax rates, and the upper middle class, who have high incomes but little ability to structure their incomes for tax advantages, got somewhat smaller cuts.

In other words, if we had made a deal with the rich to raise taxes on the bottom while lowering them at the top in exchange for a promise to pay for Social Security later ... well, the figures say that we reneged on that deal decades ago."