Don't Believe Everything You Read In the Blogs

I recently finished reading Seth Mnookin's Hard
News: The Scandals at the New York Times and Their Meaning for the Media blah blah. It's a good read, and its assessments of Howell Raines, Gerald Boyd and Arthur Sulzberger, Jr., seem just about right, even if without larger "Meaning." So I read the saga of the Times's "public editor" Daniel Okrent's grim parting jab at Paul Krugman as a kind of coda to that story, since Okrent's appointment was the most visible attempt to admit to and remedy the Times' problems.

In his last column, you'll recall, with the title "13 Things I Meant to Write About, But Didn't," Okrent said, "Op-Ed columnist Paul Krugman has the disturbing habit of shaping, slicing and selectively citing numbers in a fashion that pleases his acolytes but leaves him open to substantive assaults."

Obviously a serious charge, without evidence. Krugman and Okrent have now had a back-and-forth exchange, which you can read here, or with annotations by economist Brad deLong, here.

This is mostly pretty arcane stuff, with Okrent accusing Krugman of manipulating numbers of discouraged unemployed workers because he failed to adjust them based on a 1995 research paper, etc. Trust me, Krugman wins every point.

As to why he didn't raise any of these issues with Krugman before, Okrent says that he dealt with him on another issue, and basically Krugman was such a pain in the ass that "it wasn't worth it." What a bizarre admission: The whole job of the ombudsman or public editor or whatever the Times wants to call it is to be independent and invulnerable. He's got a one-year contract that can't be renewed; no editor or columnist can touch him. If he thinks a columnist is misusing data, his job is to say so.

Imagine, by analogy, a CBS executive saying that he wanted to say something about the National Guard documents, but Dan Rather had yelled at him over something else, so he decided not to.

Part of this story, I suspect, is in the still-awkward relationship between traditional media and the free-for-all world of blogs. Okrent didn't take on Krugman out of his own ideological bias, but got played, in this case apparently by National Review Online blogger Donald Luskin, unforgettably dubbed "The Stupidest Man Alive" by DeLong, who claims an e-mail exchange with Okrent totalling 40,000 words. Okrent claims no economic expertise, so his knowing citations to "the Polivka-Miller study" presumably come not from immersing himself in the methodological literature regarding BLS data, but from someone else, e.g. TSMA.

What Okrent doesn't seem to understand is that he's hearing only one side of a debate with two sides. He seems to assume that where there's smoke -- e.g. Luskin's obsession with discrediting Krugman, which goes back to "proving" Krugman's incompetence on the grounds that he predicted continued inflation, in 1982 -- there must be fire.

But the amazing thing is that Krugman is the most accountable columnist alive, with David Brooks perhaps second. Everything he writes is ripped apart within seconds, and litigated to death, and almost every time he comes out ahead. (Take, for example, Krugman's citation of Dean Baker's challenge on Social Security, which was to find a likely economic scenario under which the Trust Fund does as poorly as Bush predicts, while simultaneously stocks do as well as predicted for private accounts.) The problem is that you have to read in several places to follow the debate about any given Krugman column. But if he makes a serious mistake, he'll be called on it within minutes.

Krugman and Brooks are subject to this kind of public accountability because they mostly work with publicly available information. William Safire deserves sharp criticism -- which he didn't get from Okrent -- for repeating the thoroughly disproven assertion that Mohammed Atta met with an Iraqi agent in Prague, but no blogger is in a position to challenge a columnist's assertion about what sources told him. Another Times' columnist who seems to escape scrutiny altogether is Bob Herbert, who's at least as liberal as Krugman. I don't know why that is.