« The Double Switch | Main | Pirke Avot Chapter Five, verse twenty-five »

More fun with Politifactuality

So. Your Humble Blogger has been cranky of late about—well, actually, I’ve just been cranky. But what I’m writing about started with Mitt Romney starting to use in his stump speeches and on the web site the line that “Women account for 92.3 percent of the jobs lost under Obama.” The utterly useless Politifact (about which YHB has ranted before) called it mostly false. Greg Sargent over at the Washington Post outright called it a falsehood in a note titled Keep an eye on the Big Lie. The NBC political blog First Read asks Does anyone believe, with a straight face, that nine in 10 job losses over the past three years have been women workers?

Y’all know where I’m going with this, right? Yes. Mitt Romney’s claim happens to be true. You should believe it, with a face as straight as you like, because nine in ten job losses over the past three years have been women workers.

It’s not a lie.

Now, it’s being used in a misleading manner, sure. Mr. Romney seems to be implying that this disparity in job losses stems from some action of the Administration, either through deliberately destroying jobs for women or through a particular kind of neglect that happened to affect women more than men. In fact, many of those job losses were in the first couple of months of Our Only President’s administration, before his policies had come into effect. Mitt Romney has not identified what particular policies he thinks led to this disparity, nor is he saying how he (or anyone) might have done anything differently that would have led to fewer job losses among women in those first three months of 2009. So it winds up being a kind of free-floating fact-without-context, much like blaming Our Only President for the batting collapse of England’s cricket team in the Test Series against Pakistan last year. The argument is crazy, but the fact is, well, a fact. Not mostly false, not a falsehood, not a big lie.

And then, when you look into the fact further, you see some things I find interesting. For one thing, it had never occurred to me that the curve of job loss and job recovery would find women trailing men in both parts of the curve. This seems to be one of those things that economists know (or “know”) but aren’t common knowledge. So there’s that. And, when you think about it, it makes some sense that layoffs hit factory workers and construction workers first, and then, if things are still bad, the support staff goes (perhaps as the whole business goes under). Given the gender breakdown as things are at present, then, the curve would be pushed to one side. The inauguration of Our Only President coincided (mostly coincidentally) with the beginning of the turnaround in jobs, so it happens that women’s job losses were still going up at that moment, while men’s job losses had crested. That coincidence of timing is not, perhaps, all that Interesting, but it’s still a fact.

It also seems to be the case that the bulk of the disparity comes from a disparity in public sector job losses. Women make up a disproportionate number of public sector workers, and the layoffs in 2009 were largely in the public sector, as states reacted to revenue losses by cutting their costs. This means that the disparity in the first three months of 2009 was in fact a partisan issue: it was largely due to the Other Party’s policies prevailing, and to my Party’s policy’s being stymied. Had more of what we called stimulus money been made available to the states, or had just more money been made available to states, or even had more Democratic legislators in more states found more ways to pander to the public sector unions, then fewer women would have lost their jobs during that time. Mitt Romney doesn’t talk about that.

The thing that really drives me crazy, though, about this whole thing—other than seeing people whose other stuff I want to trust call a true statement a lie—is that the campaign, in cherry-picking a statistic to make a (to my mind) fraudulent case against Our Only President, managed to find one that is technically true, but which nobody would believe. As the NBC people said, it doesn’t pass the straight-face test—not because it’s false, because it’s not, but because it’s implausible. Literally incredible. This is heightened by the ethos problem that Mitt Romney is known to say things that are not, shall we say, operative at a time other than the speaking of them. As a result, the reaction people have to Mitt Romney (or his surrogates) making an implausible statement is, I would think, simple disbelief. This isn’t very persuasive. I would imagine that the campaign wants people to hear “Women account for 92.3 percent of the jobs lost under Obama” and think This is a failed presidency or even I may be willing to support an anti-abortion, anti-contraception, anti-equality candidate because job losses disproportionately hit vulnerable groups (or just because job losses hit their own demographic). I think that instead people hear “Women account for 92.3 percent of the jobs lost under Obama” and think No way, he’s making shit up again.

What’s the incentive, then, for Mitt Romney to stop telling outright lies? When Politifact or Steve Benen treat outright lies the same as artfully-misleading truths, where’s the incentive to stop with the outright lies?

Tolerabimus quod tolerare debemus,


Hmm. It sounds to me like the core of your argument here is that instead of calling it “mostly false,” they should have called it something like “technically true, but wildly misleading” or “true in its numbers, but false in its implications.”

If that's what you're saying, then I see your point—but I'm not entirely convinced that the label makes all that much difference. Especially because it seems to me that despite the “mostly false” rating, their article does explain, in a fair bit of detail, that the aspect they're calling false is the implication that it was Obama's fault, rather than the specific numbers.

(I should say that the only one of your links I followed was the Politifact one; if your objections were more aimed at the others, then my comment may be irrelevant.)

In my opinion, a fact-checking site that is claiming to provide a truth value for candidate's statements should simply have marked this one true and presented the BLS data. An analysis site could add to that the information about the date range and a comparison to other recessions. An advocacy site could take on the statement and refute its implications with the relevant policy stuff. The Politifact people don't know what they are, but it seems to be something between a fact-checking and analysis. But by refuting their interpretation of the implications of the statement, instead of checking the statement itself, they went a long way beyond that. Which is their business, I suppose, but it destroys their credibility with people who support Romney, and then with the other side when they do it the other way (as they do).

And then--the other sites, the ones I linked to and others, take the top-line info from Politifact (and similar places) and use it in their advocacy and even analysis. So there are references to Polifact-debunked claims, which is completely understandable and predictable. The label makes a big difference. In fact, the label makes far more of a difference than anything else. They are all about the labels. And if you look at the Politifact Mitt Romney Page what stands out are all those labels that point the wrong way, don't you think?

Thank you,

The number is only true from a warped perspective that privileges an ever-updated balance sheet over people's actual experiences.

If someone loses their job in 2010, and finds a new job in 2012, that person will think "I lost my job in 2010." Romney's statement utterly denies that, by ignoring every job loss that was subsequently followed by a same-gender hiring.

If 200 men lose their jobs and later find new ones, and 16 women lose their jobs and do not find new ones, does that mean that 100% of the job losses were women? If 16 of those 200 men do not find new jobs, does that mean that women account for 50% of the jobs lost? Only if you ignore what actually happened and elide several caveats.

It's an easy claim to campaign against, because people remember when they lost their jobs.

"Do you remember when the plant closed down, and many of you lost your jobs? Romney doesn't. Romney thinks that's ok as long as you eventually find another job, even if it doesn't pay as well. He thinks you didn't really lose your job."

Employment numbers are essentially a cash flow problem, in that it's not just where you start and finish, but how it goes along the way.

"The only business that doesn't care about cash flow is the business with so much money that the cash flow doesn't matter. Romney has so much money he doesn't care about cash flow. But it's worse than that. Romney has so much money that he doesn't understand why anyone cares about cash flow. And he's never had to worry about getting or keeping a job, so he doesn't understand why you care that you lost yours."

"'I like firing people.' Romney doesn't understand that the people he fired lost their jobs. He doesn't understand that when he closed down companies, people lost their jobs. This is why he claims that women account for almost all the jobs lost. This is why his party keeps trying to take away unemployment benefits. This is why his party would rather make our whole economy worse or shut down our government if it might help them get elected. There's only one person who deserves to be unemployed right now, and that's Mitt Romney. 'I'm also unemployed.' Let's keep it that way."

Women did not experience 92.3% of the job losses under Obama. Romney's statement isn't even reasonably ambiguous where one of the meanings is true and one is utterly false. The only interpretation of Romney's statement that is true is a narrow minority definition of jobs lost that doesn't include most of the jobs that were lost. "Women account for 92.3 percent of the jobs lost under Obama", except for all the people who lost their jobs. That's a rather large caveat to elide, don't you think?

You could come up with a breakdown of jobs lost from adding up a running total of new unemployment claims each month. That number would be far more accurate, and far different.

Comments are closed for this entry. Usually if I close comments for an entry it's because that entry gets a disproportionate amount of spam. If you want to contact me about this entry, feel free to send me email.