Asymmetrical Information

From the desk of Jane Galt:

My facts are better than your facts

I slapped back at Henry Farrell's condemnation of an article in The Economist the other day; turnabout being fair play, he has rejoindered:

. . . This is an argument from authority, a kind of argument with which Ms. McArdle has a rather unhappy history. More to the point--it's a bogus argument from authority. McArdle's claim is that newspaper reporters are more authoritative than op-ed writers, because they don't leave out "all the inconvenient facts," and because they're "at least nominally interested in the truth, not the conclusion." Now this is a claim that I'm prepared to buy, up to a point, with newspapers that maintain a clear separation between editorial content and reportage. The Wall Street Journal, for example, does some first rate economic reporting, even if its editorial pages are a cesspit. But as a defence of The Economist, it isn't even laughable; it's pitiable. The Economist has never sought to disguise the fact that it's a magazine with a strong pro-free market agenda, which pervades not only its editorial content, but its reporting. It doesn't try to present both sides of the question and never has; its reportage is shot through with opinionated assertions and undefended value judgments about the need for "reform" of lamentably social-democratic West European countries, to marketize the education system &c&c&c. Nor does it tend to report developments which might call its preferred policy stances into question with any great degree of enthusiasm. Now there's absolutely nothing wrong with that in principle--I'm obviously in favour of strongly opinionated political writing, or I wouldn't do it myself. But it certainly doesn't put Economist reporters in a very good place to criticize op-ed writers and political magazine journalists, or more generally to assume a lofty position from which they may criticize the pell-mell of ideologically driven debate beneath. The activities that op-ed writers and Economist reporters are engaged in aren't nearly as far removed from each other as Ms. McArdle might wish to suggest.

Which brings us to the more particular matter under discussion. Herbert's piece rested on a set of factual claims--if she wants to take issue with the article, she should, one would think, concentrate on whether these claims are in fact correct, rather than appealing to general arguments about the inferiority of op-ed writers. My original post suggested precisely that "inconvenient facts have been left out so they won't annoy the reader." As I claimed, if you want to take an undocumented immigrant worker's experience in Smithfield Foods' meatprocessing plant as a proxy for the Mexican-American dream, it's hardly irrelevant that Smithfield Foods has an established track record of abusing aforementioned undocumented immigrant workers' rights, and threatening to report them to immigration authorities if they should dare to organize themselves. If this isn't an "inconvenient fact" for the Economist's preferred narrative, I'm not sure what would be.

For starters, Mr Farrell seems to have confused an argument from authority with an argument from methodology. If I argue with a creationist about the age of the earth, he could say that I--not having done the scientific research myself--am just making an "argument from authority", and that there's no reason to privilege the authority of the scientists over the authority of The Divine Architect. But while I think far too many people do believe in evolution for pretty much the same reason that fundamentalists believe in creationism--because their neighbours and parents and other authority figures told them it's so--that is not why I am accepting evolution over creationism. I am placing my bets on evolution because the scientific method seems to me to be a better way to find out the truth than searching for answers in a book written by some enterprising middle-eastern shepherds ca. 800 BCE.

I've written opinion columns and I have, as Mr Farrell points out, written reported pieces for The Economist. I would seem, therefore, rather more facts at my disposal than Mr Farrell, who has, AFAIK, done neither. No matter how much it pleases Mr Farrell to think so, the procedures for the latter are very different from, and to my mind, superior to, the procedures for the former, even at such an august institution as the New York Times. (Perhaps especially at such an august institution; it took Paul Krugman's accusing the Secretary of the Army of being a major player in the Enron scandal to end the NYT's previous policy of leaving corrections up to the columnist.)

At least in the minds of all the editors I've ever dealt with at The Economist, the purpose of an Economist article is not to sell the gospel of free markets; it is to find out the truth, and print it. That means that if you turn in an article full of right wing talking points (not that I've ever tried such a trick), you can expect to find an irritated editor grilling you about the other side of the story, and demanding that you revise it to be more balanced. If you get facts wrong, you are in for an embarassing session with the fact checker. When you write an op-ed or a column, by contrast, everyone knows that you are putting in only the facts which put your case in the best light, and I've never heard of an opinion piece anywhere being fact-checked.

That doesn't mean that Mr Herbert is wrong. Just as it is possible that Yahweh created the world in six days and rested on the seventh, Mr Herbert may have the right of it. But if I have to choose between believing an opinion column, and believing a reported piece from any paper, including The Economist . . . including even most of the political papers that are donor supported . . . I'll go with the reported piece every time.

I suspect that Mr Farrell would be happy to endorse my methods if I were using them (as I do, when writing), to decide that a paper on health care delivery from a tenured academic has more credibility than one from the Heritage Foundation. I interrogate the facts when I can, which is how that I decided that the AEI's first estimates of the cost of John Kerry's health care plan were likely to be more reliable than either their second, higher estimate, or the ludicrously low second estimate from teh Kerry campaign's pet health care economist (whose first estimate was within shouting distance of the AEI's first estimate). But if I don't have time--and journalists on deadline unfortunately often don't--I tend to assume that a tenured academic's work is more likely to be correct than a paper from a think tank where the donors determine within what limits a paper's answer may fall. Perhaps this is just another in my long and disreputable history of arguments from authority¹?

That is not to say that The Economist is not a classical liberal paper with a bias towards free markets and free trade; obviously it is. But just as at papers with a left-wing bias (almost all of them), that bias takes the form of which arguments the editors and reporters find more convincing, NOT a conscious decision to selectively report the truth. I might have a harder time selling an Economist editor a story that said that free trade was bad than I would selling such a story to, say, the Boston Globe. But then I might have an easier time selling them a story on how the minimum wage transfers money from poor people to middle-class teenagers, a view endorsed by that well known source of right-wing propaganda, the Clinton-era HHS.

Neither is an indictment of the news orgnaisation; it's the natural human tendency to find things more believable when they agree with what you already believe. Mr Farrell, I'd suggest, simply doesn't notice it in other papers because, well, they agree with him more, and hence he finds them more believable. The Economist is no less methodologically rigorous than any other paper anyone I know has written for; indeed, it is rather more rigorous than most about things like fact-checking. The difference is that The Economist states its opinions, rather than maintaining a facade of neutrality while slanting the article so that the readers come to the same conclusion that the reporter did. This, of course, is more irritating if you happen to disagree with the analysis, but it is not measurably more "objective". Indeed, Mr Farrell's criteria for "objective" seems to be "left wing", that being the ideological tenor of all the papers he cites, including those which, like the Wall Street Journal, are quite opinionated².

I am not particularly interested in interrogating Mr Herbert's facts, and I certainly don't have the time--any more than Mr Farrell could be bothered to call up the subject of The Economist's story and ask him whether he liked his job, rather than taking Bob Herbert's word that he couldn't possibly. I have no doubt that most or all of the things that Mr Herbert printed are factually true . . . which is irrelevant, as Mr Farrell's very point is that one can tell a materially false story even when everything in it is factually true, simply by leaving out inconvenient facts. For example, the NLRB may have thrown out the previous two union elections at Smithfield, and the majority of workers at Smithfield may not want to unionise--except that for a large swathe of the left-wing commentariat, that is a logical impossibility, like pulling yourself up by your own bootstraps. Such unstated assumptions are how we know that The Economist is lying and Bob Herbert is telling the truth.

Now, as to the question of what facts are selected, I think Mr Farrell is simply wrong here: Mr Herbert's allegations about Smithfield's union practices were, at best, tangential to the story. I have no doubt that Mr Farrell sincerely believes that they belong there, just as I would sincerely like to see every story about the retail industry include a section on the manifold benefits of interstate trucking deregulation. But now that I've spent some time as a journalist, I recognize that in a world of limited resources (not the least of which is reader attention), most of the time other journalists are going to have to use their precious space to write about whatever they were writing about, rather than flogging my pet political projects. This recognition is not shared by most bloggers, left and write, engaging in freelance media criticism.

This is the opening of the story in question:

A HUNDRED years ago, a sensational novel attacking the meatpacking industry prompted Congress to draft the first federal food-safety laws. The author of “The Jungle”, Upton Sinclair, was disappointed. He had hoped to persuade Americans to embrace socialism. For him, the important point was not that the slaughterhouses of Chicago were unsanitary, but that they were “the spirit of capitalism made flesh”—a system in which “a hundred human lives did not balance a penny of profit.” The book's central character, a Lithuanian named Jurgis Rudkus, had come to America believing that through hard work he could grasp the American Dream. But he found that “the whole country...was nothing but one gigantic lie.”

Rarely has a great novelist been so wrong about so much. No one now worries about the poverty of Lithuanian-Americans. But many still worry about the health of the American Dream. Can immigrants still work their way up from the bottom? Can they become American?

It's not a story about slaughterhouses or labour relations; it's a story about whether or not immigrants are getting a toehold on the ladder of prosperity. It isn't that you couldn't include Smithfield's unionization woes in a story about the economic fortunes of immigrants; it's just that it's not obvious to me that, as Mr Farrell asserts, you have to. I am only guessing, but I'd imagine that Mr Farrell believes that unionisation is a--even the--major factor in creating worker prosperity; to him, therefore, it would be natural to talk about Smithfield's labour practices. But not everyone believes this; many people think of unions as a way to redistribute benefits not to workers, but between them, and as with minimum wage laws, argue that unions are net destroyers of social value. To those people, more interesting issues would include labour mobility, economic networks, and human capital acquisition--the very topics that The Economist's article covers.

But here is the truly weird thing, which I forgot to mention the first time around: The Economist does mention the labour problems: it's there in the first sentence, even before talking to a worker:

But is a slaughterhouse a nice place to work? Smithfield does not let journalists in, for reasons of “biosecurity”. Human Rights Watch, a watchdog from New York, issued a report in 2004 entitled “Blood, Sweat and Fear”, which accused American meat and poultry firms of “systematic human-rights violations”. Slaughterhouses are harsh and dangerous places to work, said the report, and illegal immigrants, who form a large chunk of the workforce, find it hard to defy abusive employers.

That report, as you'll find if you follow through the link, is probably the source of many of the allegations that Mr Herbert made. That The Economist did not choose to make it the focus of the piece, nor to elaborate on the anti-union activities, rather than general allegations of abuse, is a sign of right-wing bias only if you first accept the unstated assumption that unions are a means to transfer value to workers, rather than between them; that unions are net creators of value for poor workers; that unions unambigiously make their workers better off--all of which are, to say the least, hotly contested.

Instead of focusing on unions, The Economist, having established that a widely respected human rights group thinks Smithfield is abusive, decides in a fit of right-wing madness, to ask a worker who had made it out of the slaughterhouse what he thought about the experience. Mr Farrell appears to think that asking people about their experiences, and then writing down what they say, is bad reportage. This must be some special, super-scientific academic method that those of us who do not have PhDs in Political Science are too dumb to understand.

Mr Queiroz takes a more benign view. Yes, the work is hard. The line goes fast and you have to keep cutting till your hands are exhausted. And yes, it is sometimes dangerous. He says he once saw a co-worker lose a leg when he ducked under the disassembly line instead of walking round it. But many occupations are risky. Taxi-drivers are 34 times more likely to die on the job than meatpackers.

Mr Queiroz does not think Smithfield was a bad employer. Wages of more than $10 an hour enabled him to buy a house back in Mexico. Cutting up pigs was easier than picking blueberries, he says, because he did not have to toil under the sun all day. And when he had had enough, he quit and set up a taco stand with his brother. That was five years ago. Now he owns a Mexican restaurant. America, he says, is “the land of opportunity”.

Mr Queiroz is already somewhat tangential to the story, an example, not a data point; Smithfield is doubly so. I agree that if you were writing a story about how Smithfield is making immigrants' lives better, you would have to mention its union troubles more extensively (though I am virtually 100% sure that if you did, good reporting would find the story much less black-and-white than the one Mr Herbert tells). But demanding that a journalist go off on a tangent to a tangent of a tangent because you are very interested in unionisation in this country is rather presumptuous--particularly if you've never had to write under a strict word limit.

This is emblematic of much of what passes for media criticism in the blogosphere. My economic paper is solid fact; yours is a partisan hack job. My issue is vital; yours is irrelevant. My facts are important; your facts are silly. Reporters have to wade through all of this and find some way to capture the essence of a story in 1,000 words or less. Academics who get to turn in papers of 30 pages plus-or-minus 3,000 words, and bloggers who have no word limits at all other than those on the muscles of their typing fingers, should try compressing their thoughts into 600 words a week before slinging around accusations of hackery and bias.

¹ Weirdly, the example that Mr Farrell links has absolutely nothing to do with argument from authority; its an argument that economists relying on revealed preference are more likely to produce correct answers than social scientists forced to rely on survey data. For which I was roundly blasted by all the sociologists, political scientists and so forth in the blogosphere. Was I correct? That's another blog post. But this reinforces my conclusion that Mr Farrell has a rather idiosyncratic--not to mention somewhat silly--definition of "argument from authority".

²The Wall Street Journal? Left wing? I hear you cry. The news section leans almost as far left as the editorial board leans right. Unlike many other bloggers, I see nothing wrong with this; they run an excellent news team.

Posted by Jane Galt at June 19, 2006 2:25 PM | TrackBack | Technorati inbound links