And Alterman just keeps putting his foot in his mouth whenever he talks about anti-Semitism. What's going on here? In the latest issue of The Nation, he raises the issue of anti-Semitism as he defends George Soros' right to contribute his millions to anti-Bush initiatives, such as the Center for American Progress, at which, Alterman discloses-cum-brags, "I am a senior fellow" (I would shudder to see the junior fellows). The Center for American progress might object to its senior fellow's characterization of it as the recipient of Soros' generosity, he being "so moved by the behavior of Bush & Co. that he decided to invest...to defeat them." Without Alterman's help, it characterizes itself only as "a nonpartisan research and educational institute."

But I digress... Alterman characterizes as "worrisome" that a "writer on the conservative website GOPUSA.com termed Soros -- get this -- a 'descendant of Shylock.'" Did you "get that"? Good. Well, get this. Alterman just can't seem to remember, no matter how often I remind him, that his idol Paul Krugman has himself said some even more "worrisome" things about Soros. In a November 8, 1998 article in the New York Times Magazine, Krugman said"
 

"When the occasional accusation of financial conspiracy is heard — when, for example, Malaysia's Prime Minster blames his country's problems on the machinations of Jewish speculators — the reaction of most observers is skepticism, even ridicule. But even the paranoid have people out to get them. Little by little, over the past few years, the figure of the evil speculator has reemerged."

And who's the example of the "evil speculator" given in Krugman's very next sentence? George Soros -- a Jew.

And get this: Alterman accuses the Wall Street Journal of anti-Semitism on the grounds that it once noted in an editorial that Soros "reportedly chipped in $20 million to the Center for American Progress, a new liberal think-tank that is financing the likes of Bush-hating pundit Eric Alterman." Though the Journal hedged its characterization of the amount as being only "reportedly" $20 million, it subsequently published a correction, noting it was only $3 million. Now Alterman is finding anti-Semitism in the fact that the error exaggerated the contribution "by a Satanic 666 percent." Did you get that? Satanic? That's Alterman's Bible code for an anti-Semitic slur. How can The Nation or Slate actually publish Alterman's crap? We know how the Center for American Progress can...

And at least now we know whom Alterman was talking about when he wrote:

"One problem with anti-Semitism - the genuine problem - is anti-Semitism is the easy, anti-intellectual smear."

So we find in Krugman's new article, "The Death of Horatio Alger" in the January 5, 2004 edition of The Nation, this statement in support of his thesis that there is dramatically increasing "income inequality" in America:

"According to estimates by the economists Thomas Piketty and Emmanuel Saez--confirmed by data from the Congressional Budget Office--between 1973 and 2000 the average real income of the bottom 90 percent of American taxpayers actually fell by 7 percent. Meanwhile, the income of the top 1 percent rose by 148 percent, the income of the top 0.1 percent rose by 343 percent and the income of the top 0.01 percent rose 599 percent. (Those numbers exclude capital gains, so they're not an artifact of the stock-market bubble.)"

Our friend Steve Antler published on his Econopundit blog some correspondence from Jim Glass, in which Glass notes that he can't find any such estimates in Piketty and Saez's paper, "Income Inequality in the United States, 1913-1998" (NBER Working Paper, No. 8467, September 2001). Actually they are there, at least through 1998 (in Table A4). And Emmanuel Saez pointed me to a comparable table available on his website that has been updated through 2000.

Table A4 shows average real income of the bottom 90% of American taxpayers declining by 6.74% from 1973 to 2000. OK, it's only a minor sin that Krugman rounded up to 7% to make it look that much more dramatic (but I've never once seen a case in which Krugman's rounding didn't sex up whatever case he was trying to make). But there are more significant distortions in the way Krugman presents this evidence.

First, why did Krugman pick 1973 of all years as a starting point? Because -- well, what do you know! -- that year just happens to be the high water mark for real income in the entire history presented in the table, from 1917 to 2000. If he'd chosen, say, 1955 (a year in which real income was about at the average value for the entire series), we'd see not a 6.74% drop but a 43.13% increase. Or if he'd chosen 1993 (a recent low point), we'd see an 11.4% increase (and in just seven years, too). We've seen this classic Krugman fraud before -- remember in his New York Times Magazine article "The Tax-Cut Con," in which he invented his own private business cycle peak and trough in order to make GDP growth under the Reagan tax cuts look bad?

Second, Krugman notes that these "numbers exclude capital gains," and he's got his rationale for that all ready: "so they're not an artifact of the stock-market bubble." But if you include capital gains -- again, well, what do you know! -- it turns out that the decline in real income for the bottom 90% is only about half of the number Krugman cites: 3.49%. This is classic Krugman -- excluding (or including) inconvenient details that get in the way of the conclusion he'd already reached in his mind before he even looked at the data. In fact Antler chides Krugman for brushing aside studies that show that upward income mobility is alive and well by including "as the economist Kevin Murphy put it... 'the guy who works in the college bookstore and has a real job by his early 30s.'" Krugman prefers "Serious studies that exclude this sort of pseudo-mobility," and thus come up with the answers determined by Krugman to be politically correct.

Third, Krugman chooses not to point out a critical deficiency in the Piketty and Saez data. Emmanuel Saez pointed out to me that his numbers are based only on the kinds of income that gets reported on tax returns "and hence exclude[s] all transfers... such as Social security, unemployment benefits, welfare, etc... Transfers have grown pretty fast since the 1970s." Well, what do you know! It's gruesomely hypocritical for Krugman not to have noted this, considering that he has accused the Bush administration's Treasury Department of being politically corrupt for not using the holistic definition of income that includes transfers (more on that subject here).

Fourth, it appears from all the evidence I have at this time that Krugman is lying when he says that Piketty and Saez's estimates are "confirmed by data from the Congressional Budget Office." A highly placed Congressional Budget Office official told me that the only recent CBO data of this type was presented in the agency's August 2003 report, "Effective Federal Tax Rates, 1997 to 2000."  That report only looks at household incomes from 1979 to 2000 (not from 1973, Krugman's hand-picked starting year) and it defines income as "comprehensive household income," which includes all the elements that Piketty and Saez exclude (transfer payments, capital gains, and others). Also, the CBO report does not directly give the income of the bottom 90%. However, it reports the income and income share of the top 10%, so the income of the bottom 90% can be unambiguously computed.

Given all that, what does the CBO say that would "confirm" Piketty and Saez? According to Piketty and Saez, using their definition of income (but from the table that includes capital gains, to make the data as comparable with the CBO as possible), real income for the bottom 90% grew 1.81% from 1979 to 2000 (remember, Krugman said it fell 7% from 1973 to 2000). The CBO report, on the other hand, says that real income for the bottom 90% grew 21.61%. So, yes, the CBO "confirms" Piketty and Saez in the sense that their numbers for comparable periods at least point in the same direction (and the CBO certainly confirms Saez's comment to me that "Transfers have grown pretty fast since the 1970s"). But -- well, what do you know! -- the CBO certainly does not confirm Krugman's version. If there is some other CBO study that tells a different story, no doubt Krugman will let us know in a self-righteous posting on his website (in which he will no doubt ignore all the other criticisms being made here).

Fifth, Krugman fails to point out that Piketty and Saez's estimates are all based on pre-tax incomes. Considering that in his article in The Nation Krugman goes on at length about how Republican tax cuts for "the rich" are impairing economic mobility, you'd think he'd have mentioned this -- and perhaps even have asserted that "shifting the burden to the payroll tax and other revenue sources that bear most heavily on people with lower incomes" has made the plight of the bottom 90% even worse than Piketty and Saez suggest. But -- well, what do you know! -- the CBO data that supposedly "confirms" Piketty and Saez contradicts this supposition. The CBO data includes an after-tax series that shows real income for the bottom 90% rising even faster after tax -- 22.45% since 1979 -- than before tax.

None of this stuff should be any surprise. Krugman's not a serious force in economics anymore. He's just another policy entrepreneur, the very thing he was writing about a decade ago as the lowest form of life. But there are compensations. Well, what do you know! -- that's Krugman's book The Great Unraveling on Howard Dean campaign manager Joe Trippi's bookshelf.

Update... Bruce Bartlett reminds us that Krugman has lied about income statistics before -- all the way back to 1992, when he fed bad data to the New York Times. Here's Bartlett's coverage.

Posted by Donald L. Luskin at 4:42 PM | link