EDITOR’S NOTE: This story, first posted on February 24, has seen some interesting developments, all documented in four addenda at the end of the piece. The fourth addendum is the clincher: a correction from the Times’s Paul Krugman. Read on …
Try to imagine this scene taking place in the ivied halls of Princeton University. Economics professor Paul Krugman — who happens also to be America’s looniest liberal pundit — has decided it’s time to ask for a raise. So he marches into the office of Princeton president Shirley M. Tilghman and makes his demand.
”I was hired in September, 2000,” Krugman says. “I’ve been making $250,000 a year for six years. It’s time for a raise.”
Tilghman says, “Okay, let’s make it $300,000 a year. And a reserved parking place for your very old Volvo, too.”
“What?” Krugman screams. “I asked for a raise and you gave me an 80 percent pay cut! And by the way, I ride my bike to school whenever I can.”
“But Dr. Krugman,” Tilghman ventures timidly. “Isn’t $300,000 a 20 percent raise from $250,000?”
“I’m an economics professor,” Krugman shoots back. “So let me straighten you out. $250,000 for six years is $1.5 million. And the $300,000 you are offering me is 80 percent less than that! You call that a raise?”
“Surely, professor,” Tilghman says, scarcely believing what she’s hearing, “you can’t compare all six years of your prior earnings to just the one year that I’m talking about.”
But Krugman can. In fact he’s done it before — and more than once. No, not in a salary negotiation, but most recently in his New York Times column of January 30 (subscription link via TimesSelect; free link via Truthout), when he tried to show that the Jack Abramoff scandal is a purely Republican affair. Here is Krugman, talking about how political contributions to Democrats from Indian tribes who employed Abramoff don’t really count:
A study commissioned by The American Prospect shows that the tribes’ donations to Democrats fell by 9 percent after they hired Mr. Abramoff, while their contributions to Republicans more than doubled. So in any normal sense of the word “directed,” Mr. Abramoff directed funds away from Democrats, not toward them.
But that study doesn’t “show” that at all. An American Prospect article indeed claimed it — but the data from the study commissioned by the Prospect shows that this claim is based on the same fuzzy and self-serving arithmetic used in Krugman’s imagined salary negotiation.
The data show that contributions from Abramoff’s seven tribal clients to Democratic politicians totaled $868,890 before they retained Abramoff as a lobbyist. After Abramoff was retained, the total contributions fell to $794,483 — the drop of 9 percent that Krugman is talking about. But, what Krugman doesn’t say is that the average period before retaining Abramoff was 9.8 years, while the average period after retaining Abramoff was only 3.5 years. So, unless Abramoff had directed the tribes to almost triple their contributions over those 3.5 years, they could not have possibly even equaled the contributions racked up over the 9.8 years before Abramoff was retained.
If we look at contributions per year — which is the fair, apples-versus-apples way to look at this, just as it would be the fair way to look at Krugman’s salary — we see that the average contribution of Abramoff’s seven clients to Democrats rose from $11,908 per year to $25,691 per year — an increase of 115 percent.
To be fair, which is something Krugman never seems to be, the seven tribes’ contributions to Republicans rose even more when measured this way. But that’s okay, because conservatives have never told the lie that the Abramoff scandal doesn’t touch Republicans. It’s liberals who are telling the lie that the scandal doesn’t touch Democrats. But clearly it does — big time. Unless you try to hide it with a lame-brained math error like the one used by the Prospect’s reporter Greg Sargent.
Perhaps we can forgive a mere reporter like Sargent for confusing an increase of 115 percent with a drop of 9 percent. But we’d like to think that Krugman, an economics professor, would have checked the numbers before reproducing Sargent’s fallacious claim in the pages of America’s “newspaper of record.” But no. It took a blogger — Pat Curley, of Brainster’s Blog — to break this story. Krugman was too obsessed with proving, as he wrote in the same column, that “There’s nothing bipartisan about this tale, which is all about the use and abuse of Republican connections.”
And the same thing goes for all the other pundits on the Angry Left who have seized on the Prospect’s story as “proof” that the Abramoff scandal never touched any Democrats. As but one example, Brainster’s Blog’s Curley points out that another economist-pundit, Brad DeLong, wrote on his blog that “The American Prospect performs a public service — one that the Washington Post would have long ago performed, were it a real newspaper.”
By the way, does the pattern of this lie about Abramoff and the Democrats sound faintly familiar to long-time readers of the Krugman Truth Squad column? It should.
Remember back in 2003, when Krugman was pulling out all the stops to try to prevent President Bush’s tax cuts on dividends and capital gains from becoming law? He claimed in his April 22 Times column that the tax cuts would cost $726 billion and create 1.4 million jobs. That’s $500,000 per job, when the average salary in America is only $40,000. What it took this column to point out was that the $726 billion cost was spread out over ten years, and that the average salary is only for a single year. Couldn’t an economics professor divide by ten?
Krugman never officially corrected that gaffe, although our critique inspired no less than ten hapless responses on his personal website. But this time Krugman is going to have to publish a correction — just as this column forced the Times to correct Krugman’s lie that “Two different news media consortiums reviewed Florida’s ballots; both found that a full manual recount would have given the [2000 presidential] election to Mr. Gore.”
Now, as to that matter of Krugman’s salary. This week the faculty at Harvard proved what a liberal lynch mob can do when it forced president Lawrence Summers to resign. Maybe Ms. Tilghman ought to think carefully about that 80 percent cut in Krugman’s pay.
Since this column was published early Friday morning, I have spoken to Dwight Morris, president of Dwight L. Morris and Associates, the research firm that provided the analysis behind The American Prospect’s story. Concerning the amount of tribal contributions to Democrats once Abramoff was retained, Morris told me, “To say it dropped 9 percent is silly because you can’t compare those two timeframes. We did not prepare that number for them. In fact, it was not even in the reporter’s original draft.”
Byron Calame, the “public editor” of the New York Times, has told me that he will not pursue a correction of Krugman’s quotation of the Prospect’s erroneous figures. He told me,
Given that Mr. Krugman cited only one factor in classifying tribal donations — whether they occurred before or after the tribe hired Mr. Abramoff — I don’t think his statement constitutes a factual error. Is it unfair? Yes. But the fairness of columnists is beyond the mandate of the public editor.
But now I have new information on this case, based on another conversation with Dwight Morris, the consultant who produced the study for the Prospect. It turns out that Morris’s data — from which the Prospect derived the 9 percent number — does not, in fact, differentiate between contributions based on “whether they occurred before or after the tribe hired Mr. Abramoff.” Instead, the data differentiate based only on whether the contribution was made while Abramoff was retained, or while he was not. Contributions made at times when Abramoff was not retained, therefore, include periods both before he was retained and after he was fired.
So the Prospect’s claim that “the donations of Abramoff’s tribal clients to Democrats dropped by nine percent after they hired him” and Krugman’s version that “the tribes’ donations to Democrats fell by 9 percent after they hired Mr. Abramoff” — my emphasis in both cases — are more than “unfair.” They are outright not factual. The data simply do not support a before-and-after comparison. (Thanks to Pat Curley of Brainster’s Blog, who has doggedly insisted that I pursue this additional angle to the story.)
The consequence of this new understanding of the data is that my original estimate — that Abramoff’s tribes’ contributions per year to Democrats approximately doubled after he was retained — surely errs on the low side. If you remove from my calculations the contributions per year made after Abramoff was fired, the percentage-increase after he was hired can only be greater.
This new information has been provided to Byron Calame. It seems inconceivable that he would not pursue a correction now.
The American Prospect has appended a correction to its original story, confirming what we have said all along: “In the interest of accuracy, the Prospect asked Dwight Morris, the professional analyst who did the original research for our article, to take another look at the data. His conclusion is that the 9 percent figure — an overall average which was based on our reading of his numbers — can’t be validated statistically; indeed, he thinks it’s statistically invalid to do any before-and-after comparisons in this fashion.”
Can a correction of Paul Krugman’s citation of the same bogus 9 percent number in the New York Times be far behind?
Today, March 3, 2006, one week after this column called Paul Krugman on his citation of The American Prospect’s erroneous statistics, the New York Times has run a correction. Appended to the bottom of Krugman’s Friday column is this:
On Jan. 30 I cited an article in The American Prospect that reported that Indian tribes who hired Jack Abramoff had reduced their contributions to Democrats by 9 percent. Dwight Morris, who prepared the study on which the article was based, says on The American Prospect’s blog that “there is no statistically valid way to calculate this number given the way the data were compiled.” The American Prospect was sloppy, and so was I for not checking its methodology.
However, Mr. Morris goes on to say this is a minor point because other calculations show “an undeniably Republican shift in giving.”
Pre-Abramoff, the tribes gave slightly more money to Democrats than to Republicans; post-Abramoff, they gave 70 percent to Republicans, versus only 30 percent to Democrats. In other words, there’s nothing bipartisan about the Abramoff scandal.
The first paragraph is the correction. The next two paragraphs are Krugman’s attempt to claim that he was right anyway — or, as the Times once said of Dan Rather’s bogus documents about President Bush’s war record, “fake but accurate.” But deprived of his ability to lie by quoting the Prospect’s lies, there’s nothing Krugman can say anymore to substantiate his continuing claim that “there’s nothing bipartisan about the Abramoff scandal.” The statistics he cites now merely show that there was a shift toward Republicans under Abramoff, but that the Democratic giving went on without decline from prior levels. And that’s nothing but a big so what.
– Donald Luskin is chief investment officer of Trend Macrolytics LLC, an independent economics and investment-research firm. He welcomes your visit to his blog and your comments at email@example.com.