Intellectual Garbage Pickup

Yes, it's time for our once-every-three months websurf over to Donald Luskin. Why? As a public service: somebody needs to lay down a marker that he simply does not know what he is talking about, and that anyone who believes anything he writes without very careful verification is asking for big trouble.

And it is unbelievable. You don't have to read a dozen paragraphs before you run across something so bats--- ignorant that it should cause every National Review editor and writer to resign in shame, move to Rwanda, and take up a life of anonymous service to others.

This time Luskin denies the existence of the entire discipline of statistics--the idea that a properly designed random sample can tell you important things aout a much, much larger population:

The Conspiracy to Keep You Poor and Stupid: The Times and the Journal cite many authoritative-sounding studies.... But to get an accurate picture... you'd have to track hundreds of millions of individuals.... [N]one of this is reliable... the Panel Study of Income Dynamics... tracks only 8,000 families out of a U.S. population of 295 million individuals.

And then goes on to reveal that he has never eyeballed the time series on growth and inequality:

Times of great prosperity have been associated with greater income inequality (for example, the 1920s), and conversely times of economic decline have been associated with greater equality (the 1930s). The lines of causality here are complex, and no doubt run in both directions: Prosperity is both the cause and the effect of inequality, and decline is both the cause and the effect of equality. So ideological advocates of income equality for its own sake ought to be careful what they wish for...

The relationship between growth and inequality in the U.S. in the twentieth century? None--neither positive nor negative: inequality is high in the fast-growing 1920s and low in the faster-growing late 1950s and 1960s; inequality is not low but high in the depressed 1930s.

http://www.j-bradford-delong.net/movable_type/2005-3_archives/001040.html

Stupidest man alive.

Comments

[pgl]

Brad: Luskin followed this piece with one entitled "Still Moving On Up" (or something to that effect) where he claims EVERY quintile including the lowest has enjoyed increases in real per capita income. Not sure what time period he refers to but this does not sound right for the more recent period (say from the mid-1970's). Your thoughts?

[no name]

Okay, I see a polar bear, pen in hand, gazing upward as if to decide what to write next. What does everyone else see?

[Maestro]

How would the graphs look if we use different definitions for inequality, or different lengths of time? Such as top 10% and 5 year growth? I don't of one, but surely there's a measure of income inequality that summarizes the whole spectrum, like Herfindahl-Hirschman for industry concentration?
In any case, that comment about tracking 8000 families surprised me. Wow. He works in finance at a high level, he should understand rudimentary statistics. Maybe he's an even bigger hack than I realized, and I always knew he was a big hack.
As a side note, don't you think it's a good idea to link to the stuff you write about? Linking to the site but not the post is kind of weak.

[Panochia]

It's all Lord Beaconsfield's fault.

[howard]

statistics are biased against reality.

[hrned]

Luskin also has Ann Coulter's disease - he'll link to "support" for one of his arguments - but when you follow the link, it actually undermines his argument.

[nk]

The graph "Growth and Income Inequality" has a triangle that reminds me of the tail of an airplane.

any thoughts on the airline industry?

[Brad DeLong]

I thought that was Andrew Sullivan/Glenn Reynolds disease?

[Napoleon Dolemite]

anyone email him this yet?

[Oskar Shapley]

http://www.trendmacro.com/a/about/luskincv.asp
"Education : 1973 to 1974 Yale University New Haven CT
Completed freshman year; dropped out to pursue career"

http://www.trendmacro.com/a/about/default.asp

"Traditional economic analysis is concerned primarily with backward-looking statistical aggregates that have little market relevance by the time they are published."

"Our tools: depth commentary and real-time analysis from principals Donald Luskin and David Gitlitz, and research advisor Bruce Bartlett"

"Our model: a free-markets supply-side view of politics, policy, economics, technology, and markets themselves"

[hrned]

Luskin also has Ann Coulter's disease - he'll link to "support" for one of his arguments - but when you follow the link, it actually undermines his argument.

[no name]

no name> Okay, I see a polar bear, pen in hand, gazing upward as if to decide what to write next. What does everyone else see?

nk> The graph "Growth and Income Inequality" has a triangle that reminds me of the tail of an airplane.

An airplane!? That's just weird.

[Ottnott]

I'm extremely impressed with Luskin.

Luskin says:
"But to get an accurate picture... you'd have to track hundreds of millions of individuals"

TrendMacro says:
"Our tools: depth commentary and real-time analysis from principals Donald Luskin and David Gitlitz, and research advisor Bruce Bartlett"

Are the rest of you not awed by a man who can conduct real-time analysis of hundreds of millions of individuals?

[Atanu Dey]

Ottnott wrote thusly "Are the rest of you not awed by a man who can conduct real-time analysis of hundreds of millions of individuals?"

I am as impressed as I am with Santa Claus's visiting millions of homes on one single night and putting neat stuff in everyone's stockings (if you are an American that is, as David Sedaris noted.)

[Atanu Dey]

I see a map of Santa Claus's visit to the US on Christmas Eve. He starts off at the top left point (north pole), then visits all those cities in the US using a approximate solution to the traveling salesman problem, and then goes back to his north pole home.

This fits in nicely with the polar bear theme and the aircraft tail them seen by others.

[Atanu Dey]

Brad concludes with "Stupidest man alive."

Well, that should come as a consolation to some others who may be otherwise in danger of winning that prize. So I guess Shrub lost out on that one.

[chris]

Between this post and the helicopter money boy Bernanke home alone post, why aren't more of you scared Santa isn't going to make it to your home? PBS is compiling lists of who's conservative (nice), and who's liberal (Taliban and/or Al Qaeda if not terrorist summer camp sponsoring Iranian). This year I'm trying to be nice and wishing for a Hummer.

[kharris]

This is one of those stunning cases in which just a tiny familiarity with the situation in question could have led to better results. Could it be that Luskin is unaware of the level of unemployment during the Depression? That is a huge hint that bad times were worse for those who lived by their labor than for those higher up the scale. It isn't proof, of course, but it's the sort of thing that should make any thinking individual question whether the high end did worse than the low end. Kick in enough zeros at the low end, and it is very hard for declines in wealth and income of any magnitude at the high end to make up the difference.

Dumbest man, indeed.

[philipw2]

Is the income share of the top 1% the best measure of income dispersion? I know it is famously George Bush's base ("The haves and the have mores")but aren't there gini coefficients (or others I forget from graduate school) that track income distribution over the entire population?

[Fundo the Clown]

It's called Faith Based Economics. They refuse to teach at our universities because they are controlled by liberals. Why do you liberals support gay marriage for flag burning illegal aliens?

[anne]

KHarris

There is a determined effort to revise the history of the Depression and New Deal in terms generously described as "imaginative." There is however something to be said for imagining history, I guess, after all as Tolstoy understood that was much the strength of Napoleon :) By the way, do read Isaiah Berlin, "The Fox and the Hedgehog," on tolstoy and history. A small gem.

[chris]

In the early nineties Forbes magazine in some article tried to accentuate the positive about the Great Depression with a line similar to "but three quarters of the population was happily working." After they came out with the lame-o Forbes Global around that time I let my subscription lapse.

[Auros]

PhilipW: You're certainly right that there are more sophisticated ways to determine income distribution, but if you look at histograms for a bunch of different years, you'll see that, for the most part, the overall shape stays the same, while the dispersion -- and hence the length of the right tail -- changes. So measuring one feature, like the area of the top percentile, does provide a reasonable representation of the whole. And it's a lot less work than the better ones. *g*

PGL: Has Luskin repeated the error referred to here?
http://www.j-bradford-delong.net/movable_type/2004_archives/001388.html

[kharris]

Anne,

Thanks for the recommendation. I've always enjoyed Irving Berlin...

[anne]

KHarris

Well, you do have a wonderful sense of humor as well. But, beyond name dropping, I have been reading Berlin these several days and realizing yet again how important a political theorist he was, how contemporary in effect. A wonderful writer.

[Redleg]

Luskin must be practicing "Faith-Based Economics" (or he hopes his readers do). Thank you for keeping tabs on this sorry excuse for a human.

[save_the_rustbelt]

Luskin wrote a piece in which he explained how the current scandals in Ohio government are to be blamed on the Democrats.

Wildly inaccurate (I'm an Ohio Republican and this is a GOP scandal) to say the least.

I sent him an e-mail with the facts and his response was "Fair enough. I just want to make sure that the Democrats get their fair share of the blame."

Problem is, the Dems only deserve about .0005% of the blame.

What a meatball.

[RT]

Nah, Luskin isn't even the stupidest member of the commentariat. I *think* that dubious honor goes to Cal Thomas, although there may be someone out there who is even lower on the food chain, but is nonetheless paid for his/her opinions.

And I've met people who make Cal Thomas look like a freakin' genius. That's the scary part.

[ry blingston]

Luskin may be the stupidist man alive, but he isnt a trained economist. What is your excuse, Brad, for using a cloudy crystal ball to insist that the dollar will almost certainly fall over 30% in the next decade?

[The fact that the dollar will almost certainly fall by more than 30% over the next decade. Countries with large current-account deficits see their currencies lose value. The U.S. current-account deficit means that the supply of dollar-denominated assets to be held by foreigners goes up. For a while, demand can keep pace. But not forever.

It's just supply-and-demand.]

[Brad DeLong]

The top 1% income share moves with all other sensible inequality measures. Income and wealth inequality, however, do move somewhat differently...

[Maestro]

"The top 1% income share moves with all other sensible inequality measures"
Show me, please.

[A. Zarkov]

“The whole purpose of the science of statistics is to tell us that this is simply not true.”

Finite population sampling is not the *whole* purpose of statistics. It’s a very important activity within statistics, but not the only one. It also has its controversies. For example do we use model-based inferences in population sampling? There are whole books dealing with this question.

“ ... law of large numbers tells us that the sample average you compute will converge to the true population average at a frighteningly rapid speed.”

This statement is generally true, but borders on dogmatism. If the finite population itself is a sample from a stable distribution, then the statement is false as far as inference goes about the parent distribution.

For a simple thing like gross income, a longitudinal study consisting of a relatively small sample does provide some information income trends for the US population as a whole. So I completely agree with De Long that Luskin statement is ignorant. I’m not sure how PSID handles the effects immigration. Perhaps someone can expand on this.

It’s important to realize that finite population sampling becomes gets difficult when dealing with high dimensional data. For example the National Halothane Study, which dealt with anesthesia and postoperative hepatic necrosis ran into this problem. Moses et al did some very clever things to get the correct inference about the safety of Halothane. Even though the study had a large sample size (850,000) the analysis wasn’t easy.

[derek]

no name writes: "Okay, I see a polar bear, pen in hand, gazing upward as if to decide what to write next. What does everyone else see?"

The Sorting Hat, from the Harry Potter films.