Nothing today. Go read CalPundit. He's right about everything.

My imaginary conservative correspondent asks, "Aren't you going to mention Bush's new money to fight AIDS?"

It's great to hear. I'm glad he said it. If he actually acts upon it, it will be wonderful. When I see it funded, I'll be delighted and will say very nice things about Bush.

Until then, I'm filing it with the "No Child Left Behind" Act, the Homeland Security Act, welfare-to-work, veterans' health care, assistance to New York City, the SEC, election reform...

William Saletan:

In the few minutes Bush spent on what he had “accomplished,” he spoke of processes, not results. “To lift the standards of our public schools, we achieved historic education reform,” he said. “To protect our country, we reorganized our government. ... To bring our economy out of recession, we delivered the largest tax relief in a generation. To insist on integrity in American business, we passed tough reforms. … Some might call this a good record. I call it a good start.”

Record? That isn’t a record. It’s an agenda. An agenda is the measures you enact: education reform, a Homeland Security department, tax relief, corporate oversight reform. A record is what those measures are supposed to accomplish: lifting public schools, protecting the country, ending the recession, improving corporate integrity. By inserting these hypothetical achievements at the beginning of each sentence about his agenda, Bush made them sound real. They aren’t. His education bill remains unfunded. The corporate reforms he signed were watered down. The first Secretary of Homeland Security was sworn in four days ago. And the economy is still a wreck.

I'm pretty uninspired at the prospect of blogging about the SOTU, but it sounds like there's a great discussion going on at the Slate Fray. Geoff's point seems so obvious:

If you're a sixty-year old conservative now, you can have your tax break and your 4% spending increase, and your five simultaneous wars abroad all at the same time, because you're going to be dead by the time I get hit with the bill for your profligacy. The decision to fund America's priorities on layaway is selfish, and it's a betrayal of younger Americans who will inherit a government with staggering debts.

Winston Churchill said, "Democracy is the worst form of government, except for all the other ones." One problem with democracy is that deficit spending is not nearly as unpopular as spending programs and tax cuts are popular. So when you have irresponsible leadership, like the Bush Administration, they'll try to slash and spend themselves into enduring popularity. He did it in Texas, and now he's doing it to the nation. Love him or hate him, Paul Krugman predicted this, and he's been right over and over again.

And when you have a conservative movement full of enablers willing to overlook the gaping deficit, or to pretend that deficit spending is actually a clever plan to hold back the size of government, or ignore the relationship between deficits and long-term interest rates... well, it's a recipe for disaster. We get to watch it all happen. Whee.

Have I thanked Ralph Nader recently?

Dwight Merideth has a long post about how a right-wing group has spent hundreds of thousands of dollars trying to prevent money from being used to provide legal representation for the poor. This is not tax money- it doesn't cost you or me a dime. It's interest from legal trust accounts, and it would be free money for the banks if it wasn't being used to provide legal defense for people who couldn't otherwise afford it.

Why would they do this?

"We are finally in a position we've fought more than a decade to reach -- a position where we can deal a death blow to the single most important source of income for radical legal groups all across the country," wrote WLF Chairman Daniel Popeo. Among the foundation's adversaries in the litigation, Popeo continues, are "groups dedicated to the homeless, to minorities, to gay and lesbian causes, and any other group that has drawn money from hard-working Americans like you and me to support its radical cause!"

Yes, people like this actually exist, and they're able to raise hundreds of thousands of dollars. Go read P.L.A. for the details.

Seanbaby, the funniest man on the internet, has come back from an extended break with a new article: The Top 10 Naughtiest Games of All Time. It's probably not work-safe.

If you only have time for one, make it this one.

I had heard that Glenn Reynolds has worked with some interesting characters, but I have to admit that this one surprised me.

Sometimes I lose faith in blogging. But it's extremely gratifying to pose a complicated issue and have intelligent people respond to it and contribute. Such a thing happened when I posted about the black-white wealth gap.

- Rob Lyman approaches the wealth gap from a conservative perspective, but he and I come to the same conclusion that the wealth gap is a result of historic discrimination against blacks*.

Rob suggests that a tax deduction for renters would be a good start to addressing this problem; it would help people without family wealth save up their own wealth, and maybe buy a house when they've saved up enough money. Upon reflection, I can't agree. I remember reading an interview with a libertarian law professor a few years ago (can't remember who it was), who said something that still sticks with me. He said that he doesn't even read the preambles to bills anymore, because he doesn't care what lawmakers think they're doing. Rather, he looks at what the bill rewards, and what the bill punishes. Whatever those are, you'll get more of the former and less of the latter.

A tax deduction on rent would help some people save more money towards the deposit on their house, but it would have the perverse effect of encouraging more people to rent. Part of the reason that home-buying is so attractive is that it offers the only big tax deduction most individuals can get. Change the incentives, decrease the relative attractiveness of buying vs. renting, and you'd have x% of prospective homebuyers who wouldn't see the advantage of buying anymore. Furthermore, if renters got a tax break, the pool of renters would increase, and a lot of people wouldn't think, " Great, now I have $100 a month to save towards a deposit." Instead, they'd think, "Great, now I have $100 more a month in my rent budget to spend on a better place." Wouldn't that lead to rents being bid up?

(So what's your idea? Good question. Hopefully, I'll get to that.)

- Ampersand posted an extremely helpful table of wealth and income by racial category, with both mean and median. Helpfully, he's also got an extremely helpful table of comparative wealth of blacks and whites in the same income band.

For example, whites with an income of $15,000- $24,999 have an average net worth of $108,696. Blacks in the same income band have an average net worth of $31,913. Goodish news- the middle income bracket has the smallest difference ("only" 2:1- $136,455 vs. $62,635).

(Ampersand also has some great lightbulb jokes about how many anti-feminists it takes to change a lightbulb. I'm going to finish that lightbulb joke page, I promise. My favorite:

Q: How many anti-feminists does it take to change a light bulb?

A: 51. One to change the light bulb, and fifty to bitch that if it wasn't for those damned feminazis, it wouldn't be dark in the first place.)

- Finally, Mark Kleiman lets a little air out of my bubble.

First, the size of the wealth gap, even controlling for income, suggests black families as wealthy as white families with similar incomes are extreme outliers. That means you have to worry that wealth is proxying for other unmeasured characteristics, rather than acting directly to improve performance. If so, doing things to make black people wealthier shouldn't be expected to bring their kids' performance into line with those black families who are wealthy now.

He then points out that Conley measured educational attainment by last grade completed. One of Mark's colleagues, Meredith Phillips, has measured the link between wealth and test scores and found that it's fairly weak. This is Phillips:

After controlling for income, wealth can explain some additional percentage of the gap in math scores (an additional 15% of the gap if I recall correctly) but barely any of the gap in reading/vocabulary scores (at least among young children). Estimates I've run based on 5-6 year olds' vocab scores show that wealth explains no additional chunk of the gap above and beyond causally prior variables. I doubt that wealth is the full causal culprit here.

The point about outliers is a good one. I'm imagining the aquisition of wealth as a footrace in which the average white competitor starts out with a 50 yard head start over the average black competitor. (I don't think that's a bad analogy, actually.) In this race, blacks who finish with the top tier whites aren't as good as white competitors- they're much better, and their results can't really be generalized to everyone.

Damn, Mark. Why'd you have to go and make things so complicated?

* It would be a methodological nightmare, but I'd be interested to see if black families whose ancestors chose to immigrate to the US have different household wealth than black families whose ancestors were captured as slaves. I suspect that if it could be measured, it would be a significant difference. I mean, black slaves didn't even get to keep their names, let alone any wealth.

Ampersand has a striking post about exactly how much impact affirmative action has on the chances of whites to be admitted to highly selective universities. The discussion in the comments is also first-rate. Hie ye hence.

This month's Atlantic had a short excerpt from a study about black NFL coaches. It's amazing- there have only been five black coaches in the past fifteen years, and they've been much more successful than white coaches. Specifically:

· Black coaches won 1.1 more wins per year than white coaches.

· Black coaches made the playoffs 67% of the time versus 39% for white coaches.

· In their first season, black coaches average 2.7 wins more than first year white coaches.

· Terminated black coaches won an average of 1.3 more games than terminated white coaches.

Just so you know, Johnnie Cochran was one of the people involved in the preparation of this study, although the actual number-crunching was done by labor economist Dr. Janice Madden of the University of Pennsylvania. Hopefully, this will not be turned into a lawsuit, as I can already hear the arguments, and they're already giving me a headache.

I probably shouldn't make too big a deal about this. The small sample size is so small, and of course it's conceivable that there's a big freakin' drop-off in coaching talent after the fifth black coach. (Isn't the sample size part of the problem in the first place? ed.) But these results really are incredible. It's just, some people would suspect that this one of those things that the genius of the market is supposed to take care of here in color-blind America, especially in the not-exactly-lily-white football talent pool. I'm just saying.

Rob Lyman has a job!

John Cole has a spiffy new home!

Gary Farber has a great blog that deserves more attention!

Dave Barry started a blog!

And I'm doing something I've wanted to do for a while. Inspired by Leftbanker, I rented a cello this weekend and start lessons tonight.

Good news all around. Let's celebrate!

Quick post on John Lott:

James Lindgren has not asserted that the survey wasn't conducted, which speaks well of his caution in the difficult task of proving a negative. (James Lindgren also makes the important point that if Lott had gotten IRB approval, that could be taken as positive proof that the survey had been conducted, but the lack of IRB approval doesn't mean that it wasn't conducted. IRB approval is often overlooked at the U. of C. law school and in survey research. Point taken.) Having said that, there are a long list of reasons to doubt that the 1997 survey was ever conducted, and I've listed some in previous posts. There's only one piece of corrororating evidence that the survey was done- the recollection of a pro-gun activist, David Gross. The case is far from closed.

I can't improve on Tim Lambert's Lott blog, or ArchPundit's posts. I might have a post on this later, but I'm finding it more depressing than enlightening. You know why I'm agnostic about guns? Because I can't believe a goddamn word that either side says, and it doesn't make me feel any better to see smart people circling the wagons around Lott. Just as some conservatives seem to see criticism of Bjorn Lomborg's The Skeptical Environmentalist as proof of an international conspiracy of left-wing climatologist$, some people seem to think that everyone but James Lindgren who's interested in this question has a Bellesiles poster in their rec room. It's very frustrating.

Let me just say that you don't need to be a statistician to understand the statistical flaws in the 98% brandishing figure. I'll put it in terms anyone can understand. Let's pretend, as Glenn does, that the main claim (Lott never did a survey) has been put to rest. Now imagine this:

- A: 98% of Americans don't want an attack on Iraq.
- B: Where did you get that from?
- A: Published surveys.
- B: What are you talking about? There aren't any surveys that show a number even close to 98%
- A: Well, actually, I conducted my own survey.
- B: And you just forgot about it?* Well, never mind that. How many people did you talk to?
- A: 2,424.
- B: And how many responses are you basing the 98% figure on?
- A: 25.
- B: Based on 25 responses, you're going on television and saying that 98% of the public opposed a war on Iraq.

Can you imagine an explanation of weighting that would make you think that A is in the right? Especially if the weighting data has been lost?

Let me ask this another way, at the risk of repeating myself. If Lott is right, 98% of the time, you can defend yourself from a criminal with a rubber gun.

Does anyone believe that?

* It really is that bad. From Tim Lambert:

Here's the sentence in the first edition:

"If national surveys are correct, 98 percent of the time that people use guns defensively, they merely have to brandish a weapon to break off an attack."

and here it is in the second:

"If a national survey that I conducted is correct, 98 percent of the time that people use guns defensively, they merely have to brandish a weapon to break off an attack."

He didn't add a cautionary phrase at all. He changed the attribution of the 98% figure. And he won't admit to doing it.

UPDATE: Jane Galt has a reasonable reponse in the comments to this post. The bottom line is that she's very hesitant to believe that Lott didn't conduct the study without incontrovertable proof. He's pretty far from "vindicated" (my read), but she wants to be careful about the most serious accusation. All right. It'll be interesting to see how people react when James Lindgren releases his report.

Has anyone finished Yann Martel's Life of Pi? I'm really eager to discuss it, but I don't want to give away anything, so I'll put it in the comments. WARNING: Spoilers a'plenty.