Because most philosophies that frown on reproduction don't survive.

Tuesday, April 19, 2011

Responding to Stiglitz on Inequality

There's a Vanity Fair piece on income inequality by Nobel Price-winning economist Joseph Stiglitz, "Of the 1%, by the 1%, for the 1%", which has been cited again and again in the commentariat lately, and it's a frustrating piece because of the extent to which is makes logical leaps or simply distorts reality. Scott Winship of The Empiricist Strikes Back does a good job of going through the piece and addressing it point by point, including taking on a few of the talking points which are increasingly becoming things "everybody knows" in the wonk community but which don't actually mean what they seem to.

One of the problems with our modern society's fixation on "data" is that people, even very educated people who should know better, often fixate on a given metric (for example, the claim that "While the top 1 percent have seen their incomes rise 18 percent over the past decade, those in the middle have actually seen their incomes fall. For men with only high-school degrees, the decline has been precipitous—12 percent in the last quarter-century alone.") without taking the time to dig into what we can discover of the realities that underlie that measure. Sometimes those realities do not fit with the ideological picture which makes the original metric so appealing. (Winship's responses to the just quoted claim, both in the main article linked above and in this older one, are fascinating.)

Definitely worth a read.

4 comments:

Anonymous said...

Stiglitz's article is more accurate than Winship's. It's amusing to me that Winship thought he had dissected Stiglitz's article until he got smacked in the face by a commenter who pointed out that he was missing important data. I'll give Winship credit for being honest enough to include the new data in new bracketed paragraphs, though he still fails to acknowledge that it destroys what he had thought was the strongest point of his argument.

Also, Winship's grasp of logic does not appear to be any better than his grasp of data. He writes, regarding unequal income growth, "when you start from an unequal distribution, if everyone experiences the same rate of income growth, a disproportionate share of gains will go to the top." This is true: If a group takes in 25% of aggregate income, and everyone experiences equal income growth, then that group ought to take in 25% of the growth. But according to data that he just cited, that is not what is happening. The group taking in 25% of income is also getting 22-33% of growth. Is 33% the same as 25%? Winship seems to think it is.

I could go on, but this isn't worth my time. I could just as well read Ray Comfort "fact-checking" a paper on evolution.

Joel

Darwin said...

I'll give Winship credit for being honest enough to include the new data in new bracketed paragraphs, though he still fails to acknowledge that it destroys what he had thought was the strongest point of his argument.

Um, no. Stiglitz used a different income table out of the same paper that Winship cited -- one which Winship argues would provide lumpier data, and arguably a less accurate read.

The group taking in 25% of income is also getting 22-33% of growth. Is 33% the same as 25%? Winship seems to think it is.

Is 22% more than 33%?

Depending on your you calculate the share of income, and the increase, you get different spreads. But in no case is "most" of the increase in income going to the top 1% as Stiglitz claimed. there is an amount slightly more than their share of income (as in, their income is growing faster) but not by that much.

I will concede this much, though: Winship gets too mired down in quibbling with Stiglitz specific statistics and doesn't spend enough time on Stiglitz incorrect or overly broad generalizations. Generally speaking, Stiglitz's data isn't bad, though he's selecting the figures which most fit his argument. But his generalizations and conclusions go beyond his data and are excessively hyperbolic.

Anonymous said...

"one which Winship argues would provide lumpier data" - yes, and Winship is a moron for arguing that, which his commenter was correct to note. The difference in the models is capital gains income, which is indeed lumpy from year to year, but just leaving it out is ludicrous, since capital gains is a huge chunk of income for many of the wealthy. Both the McCains and the Kerrys get more than half of their income from capital gains. They are hardly unusual in their class.

Joel

Darwin said...

No, he's not a moron, he's just trying to have a more apples to apples comparison. The table that Stiglitz used included only taxably realized capital gains. This means that it would have showed the capital gains of people like the Kerry's and the McCain's who have large fortunes invested in brokerage accounts, but it wouldn't show the capital gains of people like you and me who have our investments in 401ks.

Thus, if I have $50,000 in a retirement account and earn 10% during a boom year, while Kerry has $5,000,000 in an investment account and earns 8% -- the table which Stiglitz is using would show Kerry's income based on capital gains to have been $400,000 and mine to have been $0.

Either way, it's unquestionable that Kerry would have made more than me, but the fact that the table shows only realized capital gains means that it would make it look like he is capturing all the income, while I get none, thus exaggerating the appearance of income inequality.

That's why Winship wants to use either the table excluding capital gains (after all, it's the same market that both wealthy and poor invest in, and if one is going to be seriously concerned about growing inequality it's increased salary and other compensation that one wants to look at, not investment returns) or else use the CBO data which includes both realized and unrealized capital gains.

You may differ from Winship on whether this is the best way to look at things if you like, but his stance is hardly moronic.