Among all the magical tax cuts in supply siders' arsenal, one is believed to have almost magical powers—cutting capital gains taxes. Advocates take it as a matter of faith that cutting tax rates on profits from sales of assets spurs so much more selling that revenues must increase.

Today, the Wall Street Journal editorial page opined, in the latest of a long-running series of paeans to capital gains tax cut reductions:

Wow. The [capital gains] tax rate fell to 15% from 20%, yet revenue collections have climbed by 152% in four years.

The first President Bush was so convinced that capital gains tax cuts paid for themselves that he complained in his last state of the union address (delivered exactly 16 years ago) that opponents of capital gains tax cuts were just mean-spirited puritans who didn’t want rich people to have any fun:

And I'll tell you, those of you who say, 'Oh, no, someone who's comfortable may benefit from that,' you kind of remind me of the old definition of the Puritan who couldn't sleep at night, worrying that somehow, someone somewhere was out having a good time. [Laughter]

It's certainly plausible that capital gains tax revenues would be sensitive to tax rates. Investors can avoid the tax by simply holding onto the asset. If they hold onto assets long enough, they win the ultimate prize, complete forgiveness of all capital gains tax on assets held until death (what columnist Michael Kinsley called "the angel of death loophole").

Nonetheless, there is overwhelming evidence that capital gains realizations are not very sensitive to permanent changes in tax rates. Careful econometric research shows that the prime determinant of capital gains is the level of asset prices, while the capital gains tax rate has a surprisingly modest effect—much too small for capital gains tax cuts to pay for themselves.

Capital gains ballooned in the past few years because the stock market was booming. The same thing happened throughout the 1990s, even when capital gains tax rates were almost double their current levels. Yes, revenues are higher than forecasters expected, even before the capital gains tax rate cuts, but that is because the stock market has grown so much more than anticipated.

Besides for the compelling econometric evidence, there's this picture, which I think is worth a thousand words. Based on research by U. Conn professor George Plesko, the chart shows that individual and corporate capital gains follow a virtually identical pattern over time. Why is this revealing? Because individual and corporate capital gains tax rates rarely changed at the same time. The two tax rates are almost uncorrelated (correlation =0.22). While this doesn't rule out a role for taxes, it strongly suggests that other factors are far more important.

Or just look at the Wall Street Journal's "money and investing" section. Their investment strategies often involve a lot of selling and rarely mention taxes.