Republicans were very certain about one thing in 1993. “Three hundred billion in new taxes,” Newt Gingrich declared at the time, “is going to shrink the economy, put people out of work, lower tax revenues.” Op-ed after fearful op-ed echoed this party line: higher tax rates would bring in lower revenues.
Of course, just the opposite happened—the economy grew fatter, millions more went to work and revenues soared—and supply-siders haven't had an explanation. But on June 5 The Wall Street Journal printed the first effort by a voodoo economist—in this case, Stephen Moore, director of fiscal policy studies at the Cato Institute and former adviser to Dick Armey—to reckon with the inconvenient lack of a recession following the 1993 tax hike. It's just one measly op-ed, but it's worth dwelling on because it shows so clearly how an economic theory that continues to hold sway over the political system is completely impervious to data. Also, on a more ignoble level, it's rare that history repudiates a theory so utterly, and it's just fun to watch an intellectual apologist squirm.
THE DAUNTING FEAT of historical revisionism Moore undertakes requires especially devious methods to torture the data. Moore, however, is up to the task. The first thing he does is change the question. Moore obviously can't argue that tax revenues dropped after 1993. Instead he notes that tax revenues rose, after inflation, just 21.6 percent in the 1990s, while they shot up 24.1 percent during the 1980s. “Total federal revenues grew at a slightly faster pace,” he notes. While Moore tries to pass this off as vindication, it's actually a telling concession. After all, in 1993 Moore flatly predicted that Clinton's plan would “torpedo” the economy. Now he is reduced to merely arguing that it produced slightly less miraculous results than Reagan managed.
But, even for this, Moore has to turn a few somersaults. Instead of comparing the rate of revenue growth under Reagan against the rate of growth under Clinton, he selects his years carefully. While the Reagan tax cuts began in 1981, Moore starts with 1982, a recession year. Since economic growth and tax revenues always shoot up quickly after a recession, this makes it look as if the natural results of the business cycle were actually the magic of Reagan's tax cuts. (Moore did not invent this trick; if you read nothing but the Journal editorial page, you would think that Reagan took office in 1982.) To further stack the deck, Moore starts the Clinton era in 1990. There's a certain fairness here, because Moore is trying to argue that tax increases lead to lower tax revenues, and there was a tax increase in 1990. The problem is, there was also a recession that year. So the “Reagan” years begin with the economy climbing out of a recession, and the “Clinton” years begin with the economy sinking into one.
Pretty sneaky, huh? Well, that's not the half of it. Remember, Moore is trying to prove that lower income taxes bring in higher revenues. But, instead of comparing income tax revenues, he compares total tax revenues. That's not just income taxes. It also includes revenue from the payroll tax, which Reagan hiked. In other words, Moore uses the extra revenues brought in by Reagan's payroll tax hikes to hide the revenue lost by Reagan's income tax cuts. Rattling around in Moore's brain, somewhere, is the germ of a realization that something about his calculation is a tad unfair. So later in the essay he makes amends, sort of. “Even if we examine the path of only individual income tax collections over the past 15 years,” Moore writes magnanimously, as if bending over backwards to be fair rather than correcting his own fundamental error, “the story is not much different.” Here is where Moore is going to show us how income tax revenues, like total tax revenues, rose faster during the 1980s. This is the really beautiful part of his essay. It's worth quoting in full:
From 1982 to 1989 income tax receipts climbed from $298 billion to
$446 billion—a 50 percent increase. From 1990 to 1997 the income
taxes rose from $467 billion to an estimated $710 billion—a 52
If you didn't just slap yourself on the forehead, read that passage again. And possibly again. Because Moore surrounds his numbers with ringing declarations of victory, it took me until my third perusal to realize that the Clinton-Bush number is higher than the Reagan number. Moore has repudiated his own point. Maybe he has failed to grasp the basic mathematical concept here: the number 52 is greater than, not less than, the number 50.
Actually, it might be generous to call Moore ignorant, given the way he arrived at the 50 and 52 figures in the first place. Keep in mind that, when Moore compared total tax revenues, he rightly used the percentage increase after inflation. But when comparing just income tax revenues—which is the relevant statistic—he doesn't factor out inflation. Why not? Because inflation was higher under Reagan. Ignoring inflation makes the comparison look closer than it is in reality. If you take out the part of the increase that was due to inflation, income tax revenues from 1982 to 1989 rose just 16. 5 percent, while from 1990 to 1997 they rose 24 percent—half again faster. Hmm. I wonder why Moore forgot to control for inflation this time?
The funny (and frustrating) thing about this is that Moore and his ilk have actually won the debate. In 1995, Clinton publicly conceded that he raised taxes too high. And, in the latest budget pact, he has agreed to institute a supply-side tax cut. Sometimes historical vindication is small consolation.
Jonathan Chait is a senior editor of The New Republic.