URLs du Jour

2018-09-02

[Amazon Link]

  • At EconLog David Henderson discusses a recent Paul Krugman NYT column and associated tweet. As you might guess from the title, New Hampshire is involved: Live Free and Die?

    Here's an excerpt from the Krugman column:

    The other day I had some fun with the Cato Institute index of economic freedom across states, which finds Florida the freest and New York the least free. (Is it OK for me to write this, comrade commissar?) As I pointed out, freedom Cato-style seems to be associated with, among other things, high infant mortality. Live free and die! (New Hampshire is just behind Florida.)

    Well, first, an obvious point: our state motto is "Live Free Or Die". Who knows if this was a stupid mistake, a Freudian slip, or some effort at being "clever"?

    But it gets worse. The "had some fun" link goes to a Krugman tweet:

    "Clear positive association" implies that the higher a state ranks in freedom, the more likely it is that babies are dying. And the clear implication of Krugman's "live free" snark is that NH is one of the exemplars of this rule.

    But, as recommended in David's post, click over to the CDC's latest Infant Mortality Rates by State page. There, NH has the second lowest infant mortality rate among the states. (You can kind of see this from Krugman's own chart.) Making Krugman's "live free" invocation kind of pointlessly stupid.

    And, perhaps even worse, you might expect from PK's snipe that Florida would have an exceptionally high infant mortality rate. But it doesn't. Among the states, its rank is nearly exactly in the middle. And it's actual rate (6.1) is just a smidgen higher than the US average (5.9).

    And—wait, it gets even worse—David's post points to an even more detailed takedown of Krugman from Robert P. Murphy of the Mises Institute. Who arguably gives Krugman a more serious analysis than he deserves:

    Stepping back, even the scatter plot as a whole, doesn’t really accomplish what Krugman wants. Visually, he is seeing what he wants to see: Krugman thinks it’s clear (especially if you read his tweet about it) that the best-fit line would be upward sloping in the chart. Yet that is largely because of New York and California. If you exclude them from the analysis, the remaining cloud of states looks like it might exhibit a downward slope.

    We don’t need to speculate. I asked Jason Sorens, one of the co-authors on the Cato study, to crunch the numbers. He did it a few different ways. First, he included all of the states, and found that the correlation (specifically, the Pearson coefficient) between infant mortality and “freedom” was 0.20. (Keep in mind that a correlation of 1.00 occurs when two variables are perfectly positively linearly correlated, while 0.00 means they are not at all linearly correlated.) Even here, the correlation wasn’t statistically significant; there was too much variability / not enough data points to be confident in the observed relationship.

    Similar points are made in the replies to PK's tweet.

    Bottom line: when it comes to issues involving freedom, don't trust Paul Krugman to analyze data fairly or accurately.


  • Jonah Goldberg reports in the G-File that he is On the Road Again. He starts from the observation that motivating passions of Donald Trump and John McCain are essentially the same. ("I am honestly not sure what word best describes it: Vanity? Ego? Pride?")

    But that plays out obviously differently:

    McCain’s egoistic passion led him to surrender himself to the faith of his fathers, or a cause larger than himself, as he might put it. Trump’s egoistic passion is dedicated to making himself as large a cause as possible. The irony is that the former’s approach made McCain seem the larger man, while the latter gets smaller by the day.

    I'm not a fan of either man, but I think Jonah's insightful here.


  • At Power Line, John Hinderaker takes a look at the Neil Armstrong movie First Man, which avoids the planting of the American flag on the lunar regolith: Revising History, Moon Landing Edition.

    We all know the biases that underlie Hollywood’s editorial decisions. There is no need to belabor that point here. What bothers me most about this incident is the rewriting of history. When most people watch “First Man,” they will assume that depictions of actual, historic events in the movie are accurate. If there is no flag in the film, it will not occur to them to wonder whether there was a flag in real life. If Armstrong is depicted in the movie as a citizen of the world, it will not occur to them to wonder whether in real life, he was an American patriot.

    We see this transmutation of history in films all the time, often in more brazen forms. Oliver Stone made “JFK,” which depicts the crazed and despicable Jim Garrison as a hero and peddles absurd conspiracy theories about the Kennedy assassination. The movie “Truth” enacts a wholly false account of the Rathergate controversy, and portrays Mary Mapes, who tried to swing a presidential election by publishing smears against President Bush that she had good reason to know were false, as a heroine.

    Hollywood lies, as John asserts, "always lean in the same direction."