Consumer complaint: Left wing econo-blogging in bad shape.


Krugman is not the only left wing economist who is loosing his analytical edge from an overdose of partisanship. Brad Delong of may be level headed compared to Krugman, at least in the economic material. But his political writing is very uneven, and when it comes to Bush or Republicans it is becoming more and more unhinged.

It�s hard to take someone seriously who writes that he is a Democrat because the Republicans were turned into �The Party for People Who Don't Like Black People�

On what basis might one ask? Delong doesn�t offering any evidence or reasoning to support his case, the mark of the true scholar. What was the last racist law Bush or the Republican Congress enacted? He doesn�t say. Historically we all know that the Southern Democrats were the party that officially supported segregation. Wallace, Byrd, Gore Sr and W. Fulbright were Democrats. Twice as many Democrat�s as Republicans voted against the civil right act, etc.

Name-calling 210b

OK, maybe we can ignore his political writings and only look at economics. Even there he is slipping. Delong, under the tasteful heading �Intellectual Garbage Pickup� yesterday called Donald Luskin �Stupidest man alive�. Why? Because Luskin wrote that �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).�

If choice to you call an opponent writing garbage and call them stupid you should have a pretty strong case. As it happens Luskins quote above it perfectly true.

In 1920 the richest 1%s share of national income (including capital gains) was 14.9% By 1929 it had risen to 22.5%, the highest in the 1913-1998 dataset. By 1939 it had again gone back to 16.2%. This exaggerates the change, as the figures fluctuate. However if we look at the average under the 1920-29 period the figure is 17.9%, compared to 16.6% in the 1930-1939 period. Excluding capital gains the pattern is very similar. Just as Luskin wrote, the 20s had more income inequality and the 30s less.

I also know there is a debate between Harvard�s Gregory Mankiw and Delong about this subject , (Mankiw argues that the rich are disproportionably effected both in boom and busts). I don�t care about the debte itself. Maybe there is a statistical relation, maybe there isn�t. I would only note that income for all categories of Americans has gone up.

But what matter is that Delong, a professor at one of the world�s most prestigious Economics Departments is calling someone who is factually correct �Stupidest man alive�.

Delong also claims that �Luskin denies the existence of the entire discipline of statistics�. The reason is that Luskin considers the sample size for the US income study to be too small, 8000 individuals of 295 million. Now should know that sample size does matter if you want to use income data in policy. While 8000 is a large sample overall, for many detailed questions it may well be too small.

As there are so many different categories of people in society Individual sub samples may become too small statistically (the sample would only contain 13-14 people from Wyoming!). That is one reason Sweden�s LINDA database includes fully 3% of our population, compared to 0.003% for PSID.

We on the right need opponents we can respect

Luskin is a grownup and has his own history of names-calling, so that�s not really the issue. But Delongs tone and more importantly nonexistent basis for calling Luskin an idiot still worries me. As Arnold Kling points out these people are the role models for less accomplished leftwingers, and they are setting the bar pretty low.

And we on the right need strong leftwing economists to challenge our ideas, not just the same bitter �Bush Lied� and �Republicans are Idiots� drivel reposted five times a day.

DeLong and Krugman could learn something on manners from their free-market peers. Look at or Both are simply class acts, always keeping a civil tone and focus on discussing ideas, not demonizing their opponents. Can you imagine Tyler Cowen calling writing �Stupidest man alive�?

And the comment pages reflect this. On Becker-Posner-blog you can often read intellectually stimulating discussions by the readers. Delongs comments section is just an echo-chamber since he seems to delete dissenting opinions, the entries looking cut and pasted from Democratic Underground. It would seem even in econo-blogging you reap what you sow.


Ever heard of the Civil Rights movement? The Dixicrats left the Democratic Party and joined the Republicans after that as any analysis of voting patterns over the last fifty years shows. So DeLong is correct.

To argue that Luskin is "factually correct" because you find some data points in a long time series consistent with his claim is simply ridiculous. DeLong's point is exactly that there is essentially no correlation between inequality and growth in the US, which is the only relevant measure of whether Luskin is correct. He is not, and neither are you.

Finally, Luskin explicitly compares 8,000 to 295m, which is an utterly irrelevant comparison for anyone with basic knowledge of statistics. The only relevant metric is the sample size, population size does not matter.

That makes it 3:0 for DeLong over you - it seems the "left-wing" econ bloggers are doing fine, but we need a better class of right-wing hacks.

It should also be noted that Luskin actually puts into practice what he preaches as a money manager. There was article recently in the LATimes which points out that many economist don't.

Well...Wait a minute. Luskin's claim isn't just about the 20s and 30s. It's a quite general claim (that growth and inequality are positively correlated), with the 20s and 30s as an example. Being correct about the 20s and 30s isn't enough to make his general claim "factually correct". But, looking over the graph on Delong's website, this proposed relation doesn't really jump out at one. Is there something wrong with the graph? Do you want to defend Luskin's general claim?

Couple of points:

"On what basis might one ask? Delong doesn�t offering any evidence or reasoning to support his case, the mark of the true scholar."

This one pretty much stands unrefuted. Just noting that the Republicans have captured the south since the sixties hardly counts in supporting the claim in question.

As for Luskin, income inequality, etc. - the point was *explicitly* DeLong behaving immaturely. As stated above, there is a debate on about the subject in the academy. (I believe a link was even provided)

The hack comment was sweet though. (Another piece of empirical evidence!)

I know the history of the Civil Rights movement and the southern Stretegy pretty well, that�s partially why Delongs crazy comment annoys me.

The 1964 Civil Rights Act had 34 Republicans voting against it, and 96 Democrats!!! Even taking into account that Republicans were the smaller party the Democrats were twice as likely to oppose the bill. This is no small or trivial matter. The left has infiltrated history education, and this is now forgotten.

In the Senate it was even worse, 6 Republicans and 21 Democrats voted against the civil rights act when it finally passed.

The Southern Democrats were the party of racism. This can never be forgotten. Republicans (slowly) took those regions only when segregation stopped being a political issue the Democrats could win on. Republicans first increased in the outer south (Texas, Florida, Virginia), not the deep south (Alabama, Mississippi, Arkansas etc) where Democrats and Segregation were more popular. This proves the point that they won on OTHER ISSUES.

And as pointed out most the most prominent racists never became Republicans, like Wallace or Gore Sr or Byrd. They staid in Delongs party.

Today the Republicans hardly hold the south based on racism, which is dead in politics, but on low taxes, pro-military, social conservatism, harsh on crime etc.

Also if you think Republicans are the party for people who do not like blacks you should have a really easy time finding examples of recent anti-black laws or legislations, speeches etc. Delong didn't provide any, because he doesn't ahve any.

You have NON, zip zero. African-Americans vote Democrat, but that does not prove Republicans hate blacks, same was the military voting republicans 70-80% does not prove Democrats hate the military. Which party appoints blacks to higher position? Which party supported Clerence Thomas or Janice Rogers Brown?

PS. You are also wrong that population size does not matter. It does matter if you want detailed statistics on sub-groups and if the population (like the US) is heterogeneous. In this case both are true, as I said once you come down to sub-categories you will have very few people in each. How many rich African Americans from the pacific northwest will be in a random sample of 8000 Americans?

Tino, Luskin's title was, I regret to say, earned for a long series of blunders that demonstrated incapacities to grasp the most basic analytical concepts. Brad is indeed prone to yell at opponents on both his right and left (e.g. the Guenther Grass imbroglio), but Luskin is a mere idiot, lacking redeeming qualities.

You fail to grasp that the sample-size question had to do with the adequacy of a given sample and methodology FOR A PARTICULAR QUESTION. Obviously you can always dream up finer-grained questions for which a given sample is inappropriate.

Both Krugman and DeLong have serious records of publication in peer-reviewed journals, as of course do Becker and Posner, and if you know the literature you can see the connections between that and their policy writing; DeLong regularly brings up scholarly work in his blog.

I'm much less shrill than, say, David Frum... or Andrew Sullivan... or Powell's Chief-of-Staff Larry Wilkerson... or George H.W. Bush's friend Brent Scowcroft.

But let me agree that Alex Tabarrok and Tyler Cowen are class acts.


Powered by Movable Type 5.02

About this Entry

This page contains a single entry by published on June 29, 2005 2:27 AM.

Labor Market Reforms was the previous entry in this blog.

Comments on a excellent inteview is the next entry in this blog.

Find recent content on the main index or look in the archives to find all content.