Streetwise Professor

March 3, 2011

Omitted Variable Bias: Krugman v. Texas

Filed under: Economics,Politics — The Professor @ 4:14 pm

Paul Krugman is obsessed with Texas. He repeatedly attacks the state, primarily because it has chosen to cut spending to address its budget shortfalls, rather than raising taxes.  This is an anathema to Krugman, so he lambastes the state repeatedly.

As an I-got-here-as-soon-as-I-could Texan, Krugman’s antipathy is actually a badge of honor.  And he adds to the pride with his most recent column.  Perhaps in anticipation of Texas Independence Day (yesterday, 2 March), on Monday Krugman launched an all-out assault.  His main target was Texas schools:

But here’s the thing: While low spending may sound good in the abstract, what it amounts to in practice is low spending on children, who account directly or indirectly for a large part of government outlays at the state and local level.

And in low-tax, low-spending Texas, the kids are not all right. The high school graduation rate, at just 61.3 percent, puts Texas 43rd out of 50 in state rankings. Nationally, the state ranks fifth in child poverty; it leads in the percentage of children without health insurance. And only 78 percent of Texas children are in excellent or very good health, significantly below the national average.

But wait — how can graduation rates be so low when Texas had that education miracle back when former President Bush was governor? Well, a couple of years into his presidency the truth about that miracle came out: Texas school administrators achieved low reported dropout rates the old-fashioned way — they, ahem, got the numbers wrong.

It’s not a pretty picture; compassion aside, you have to wonder — and many business people in Texas do — how the state can prosper in the long run with a future work force blighted by childhood poverty, poor health and lack of education.

Well, there are a lot of factors in addition to spending that go into determining educational outcomes (and health outcomes for that matter).  But Krugman fails to mention those altogether.  It is well known by first year econometrics students, let alone Nobel Prize winning economists, that omitting variables from an analysis usually leads to biased estimates.  So by not controlling for these other factors, Krugman’s denigration of Texas performance is quite likely subject to bias.  (Krugman? Bias? In the same sentence? Heaven forfend!)

Provoked by Krugman’s attack, non-Texan and hilarious satirist Iowahawk decided to play it straight and see how Krugman’s criticism holds up when you include a key omitted variable.  Specifically, Iowahawk compared Texas’s educational performance to that of Wisconsin, that bastion of progressivism and empowered teacher unions. More specifically, he compared scores on the standardized NEAP math, science, and reading tests for 4th and 8th graders within three ethnic categories.  That is, Iowahawk controlled for a major source of variability across states–the racial/ethnic makeup of the student body.  With all due disclaimers, which I second:

[T]his has nothing to do with innate ability or aptitude. Quite to the contrary, I believe the test gap between minority students and white students can be attributed to differences in socioeconomic status. And poverty. And yes, racism. And yes, family structure. Whatever combination of reasons, the gap exists, and it’s mathematical sophistry to compare the combined average test scores in a state like Wisconsin (4% black, 4% Hispanic) with a state like Texas (12% black, 30% Hispanic).

The results were not pretty–for Wisconsin.  Texas, that Dickensian hellhole of cruelty to children in Krugman’s telling, outperformed enlightened Wisconsin in 17 out of 18 categories (3 subjects x 2 grades x 3 ethnic categories).  If Texas and Wisconsin were equal performers, the odds of that happening are microscopic: .00007, to be exact

Now, Iowahawk doesn’t control for everything, but he certainly controls for something important that Krugman does not. And terrible Texas comes out on top.  If Krugman or his numerous acolytes believe that other controls would reverse the results then: (a) he/they should admit that the original column is fatally flawed because it controls for nothing, and (b) support his/their claim with a rigorous analysis based on a defensible set of controls.

Iowahawk begins his post by making plain that it is not one of his usual–and invariably funny–satirical efforts:

Please pardon this brief departure from my normal folderol, but every so often a member of the chattering class issues a nugget of stupidity so egregious that no amount of mockery will suffice. Particularly when the issuer of said stupidity holds a Nobel Prize.

Iowahawk doesn’t have to satirize, because Krugman has become self-satirizing, especially when it comes to his Ahab-like obsession with Texas.  (New motto: “Don’t Obsess With Texas.”)  Indeed, Krugman’s  opinion writing generally has gone completely pear shaped.  But as this photo suggests, that may be part of a pattern.

To quote a reader: “Poor cat.”

Print Friendly, PDF & Email

2 Comments »

  1. @SWP “Well, there are a lot of factors in addition to spending that go into determining educational outcomes (and health outcomes for that matter). But Krugman fails to mention those altogether. It is well known by first year econometrics students, let alone Nobel Prize winning economists, that omitting variables from an analysis usually leads to biased estimates.

    Well, I am happy to see that you realize this. Tell me, when you do analysis, do you yourself follow this wisdom? For example, when talking about various political and Russia-related issues, do you always present all sides of the story, and describe all significant variables?

    Comment by Ostap Bender — March 4, 2011 @ 5:44 pm

  2. “Tell me, when you do analysis, do you yourself follow this wisdom? For example, when talking about various political and Russia-related issues, do you always present all sides of the story, and describe all significant variables?” Heck no, and Jacksonianism is for Amurricans ONLY, not furreners.

    Comment by The Other Ivan — March 7, 2011 @ 6:01 pm

RSS feed for comments on this post. TrackBack URI

Leave a comment

Powered by WordPress