Selection Bias Chart of the Day

Unobservable individual heterogeneity?

Liberal Media Bias

I've blogged about this before, and it seems at first the answer was yes. Is it more complicated? A new Vox post suggests that on average it is 'liberal' but not much moreso on average than the average citizen. Furthermore, it appears (from print media) that the bias is more leftward on social issues, and more rightward on economic issues.

The solution is growth (Daron Acemoglu - HBR Blog)? Will growth curb the debt problem or will debt reduction stimulate growth?

A neat randomization technique to facilitate truth-telling (Freakonomics).

As Greenspan says - we'll never be "forced" to default. Greece didn't have this option. Whither Euro? (Financial Times)

The downside of self-regulating occupational licensing boards (Jay Parkinson)

More policy options that would be dominated by a carbon tax. (Freakonomics)

Another way of looking at the value-added problem in trade statistics. Not all of the "Made in China" product's value should be counted as China's exports? But where do the profits really end up? (NYTimes)

I'm guessing the blockquote here is toungue-in-cheek. (Caplan) I worry about students who only learn a particular algorithm for applying knowledge to specific situations. Those students will be easily replaced by computers. Learn to think.

Some links

Increasing taxes a little and funding health reform may do more good (at least in the short run) than cutting spending. Someone remind folks that tax revenues are already at 50 year lows as a percentage of GDP.

A similar (but maybe a bit less firm) conclusion here.

More similar conclusions from Krugman on modern Keynesianism (on VOX not his hackish blog).

A couple of cool statistics links:

Paydays are a killer?

Billion Prices Project

Some institutional links:

Gender bias and agricultural technology

Cursed by resources or institutions?

Why continuing to be a Royals fan (among other things) is good for me.

Exports are overrated?

More Random Links

Can democracy precede (human, and hence economic) development? (VOX)
Just scroll down to the last graph. How long can the "big government" narrative be sold? Government isn't that big; transfers are just high because of the recession and taxes are horrendously low for the same reason, as well as the 2001 & 2003 tax cuts. (Menzie Chinn)
Some "odd"ities from InTrade. (Freakonomics)
"[H]ow many students struggle with elasticity formulas because they struggle to manipulate and understand fractions?" (Worthwhile Canadian Initiative)

Some links on immigration and trade (and a couple on statistics)

So, what we can get from this is: (1) immigrants are favorably selected (relative to education levels in their home countries - nothing new); (2) this is not always a bad thing for the home countries in the long run. It does not lead to the conclusion that immigrants should be deliberately cherry picked by some artificial policy mechanism.
Banking in the US is finally catching up with resources already available to Africa.
Finance and trade: A historical perspective. (Vox)
Gambling: chance or skill? (Freakonomics)
Some interesting polling data on perceptions about homosexuality. (Economist/Democracy in America) Makes me think the "gay marriage trick" won't work in 2012 like it did in 2004.
Semi-random link on self-publishing a book. (Freakonomics)

Another Interpretation of Payroll Numbers

There's been a lot of chatter about the new unemployment numbers. This Free Exchange post talks about some of the methodology, and why, when it's said and done, the unemployment drop is for real.

3 Links on Bayesian Probability; 2 Links on Teaching

Another reason Bayesian statistics might be intriguing. (Gelman)
Conditional expected productivity and benching players with fouls. (ESPN)
A Bayesian rational for a situation where high variance might be a good thing. (MR)
Paying for grades isn't easy. (Freakonomics)
Ugly fonts and material retention. (Wired) This gives me my quote of the day:

making material harder to learn — what the researchers call disfluency — can actually improve long-term learning and retention

...and worsen teaching evaluations.

A Couple of Links about Children

Steven Landsburg's Big Answer about preferences for boy children and the proportion of boys and girls. (The Big Questions)
A game theoretic solution to child punishment. (Freakonomics)

Some links

Poverty-reduction matters globally. Chalk one up for redistribution, a.k.a. "socialism." It's not just about money. 
American public opinion, then and now.
Happy old people. It beats the alternative.

Sometimes the biggest factor is the least important

This WSJ editorial tries to make the case that we shouldn't care about distracted driving in a strange way. The author argues that even though distracted driving fatalities are up, overall fatalities are down. But then he seems to argue that  makes is that growth in fatal motorcycle accidents lead all categories for increases in road deaths and that many of these deaths involve inexperienced cyclists.

So what? I don't say that to seem insensitive, but someone who chooses to ride on a motorcycle, and eventually gets in a wreck is often only hurting (sometimes fatally) himself. When motorcycle fights car, car wins. In economic jargon, there is seldom an externality involved with this sort of accident. The driver of the motorcycle knows riding is riskier, and chooses to do so. On the other hand, when someone is dicking around on his cell phone (let's say while driving a Yukon or comparable grocery-and-kiddie wagon SUV) he puts innocent, responsible drivers at risk. Policy makers should only intervene if an argument can be made that there is a social cost (or externality) to certain types of behavior. I don't see that argument in the case of motorcycles.

So, motorcycle accidents are accounting for the biggest increase in driving fatalities, but there is no theoretical reason that policymakers to give a rat's ass. My guess is that Mr. White and his boss at WSJ, Mr. Murdoch are simply reveling in finding some research that promotes their supposedly pro-liberty (of a certain sort) agenda, and they would probably claim that no policy action is warranted in the case of Motorcycle OR distracted driver traffic deaths. If that's the case, why not just repeal all drunk-driving laws? It seems highly unlikely that Mr. White would support this proposal. Given that the impairment from cell use rivals that of intoxication, that seems to be the better place for policy.

Stunning "Results" about "Entitlement America"

There's a bogus calculation circulating today claiming that poor folks are making out like bandits due to all of the subsidies they're eligible for. I don't buy it, but neither does Tyler Cowen (who runs on the right side of the economic ideology spectrum):

That's after various government benefits and taxes, but the calculation
seems incorrect to me.  For instance, should the Medicaid and CHIP
benefits of the poorer household actually be valued -- to the user -- at
$16,500 a year?  (Is that number coming from some kind of cost basis? 
If so, is it adjusted for the age of the Medicaid recipients to rule out
nursing home expenditures?)  Is the $60,000 per year family receiving
employer-supplied health insurance?  The assumption seems to be that
they do not.

Of course, the "calculation" also sandbags the tax credits: The EIEC is credited to the 14,000 family, but other tax credits are probably not credited to the 60,000 family. Also, what about retirement contribution matching? Doesn't seem like the "total economic benefit" of the family "earning" 60,000 is being counted (and in particular the portions that are tax-exempt!). Most folks who earn 60,000 have a total compensation (with health and retirement benefits) of upwards of 80,000, about a quarter of which they never even think much about!

Andrew Gelman (who runs on the left side of the spectrum) isn't buying it either, for similar reasons as Kaiser Fung. From the latter:

If we concede that the middle-income person would end up with less
disposable income than the lower-income person, then we'd expect that
the middle-income people will take lower-paying jobs so as to increase
their disposable income. But I have not seen reports of such reverse
social mobility. Theory needs to fit reality. This hole in the theory
needs to be covered.

What's funny about this is that the theorist (Cowen) disagrees on data-related grounds, whereas the statisticians (Fung and Gelman) dispute it on theoretical grounds.

Some Links: Sometimes, kids will cooperate.

Careful how you try to introduce game theory to your kids. They may seem competitive with each other, but they will cooperate against a common foe. (Marginal Revolution)
Bill Easterly on the new Human Development Indicators: "The HDR addressed [a criticism of the previous human development indicators] by making the problem much worse. Previously we were all whining about differences in the value of life of 70 times between rich and poor – now it’s a differential of 17,000 to one. Sorry, Zimbabweans, UNDP thinks your lives are worth 50 cents. (Aidwatchers)
Lotteries don't have to have a negative sum for their participants. (Freakonomics)
Bayesian Bees. The author suggests 5 lessons we could learn from bees. Here are two:
1. Minimise the leader's influence on the group. Here we humans have much to learn.
2. Seek diverse solutions to the problem. Humans realised only recently that diversity is good for a group.
(Marginal Revolution)

Links on taxes and incentives, arbitrage, and political momentum.

Looks like there are cases where high taxes actually spur innovation and investment. (Reuters)
Why is there room for arbitrage on used books? Many professors get desk copies of textbooks, and many professors sell these copies to book buyers. Usually the arbitrageur offers about 20-30 bucks for a book they can turn over for 90-100. It makes some sense that there will be enough individual small fish for whom the spread is not big enough to justify the costs of finding the market price and finding a buyer (which, is in fact not too hard with an Amazon Marketplace seller's account). But it amazes me that a used book store isn't researching it's market better. It seems like the bookstore's proprieter would be interested in picking up those extra bills off the sidewalk. (Marginal Revolution)
Finally, an interesting nerdy statistical criticism of seemingly contrarian findings about political momentum. (Freakonomics)

Post hoc, propter hoc

via xkcd

Incentives or Endogeneity?

Via MR and Barry Ritholtz:

It turns out that major league hitters on the verge of a 3 handle batting average — .300 — hit an astounding .463 on their last at bat of the season.

Why? I saw several explanations, but this seems the most plausible (from the MR comments):

If a .299 hitter gets a hit in his first AB of the last game of the season, his manager takes him out. If he makes an out, he stays in the game. If he goes 0-4, he finishes at .295 or so but doesn't count in the sample of how .299 batters do in their last AB.

Seems like Tyler's assignment of the phenomenon was a jump to conclusions by an economist looking under the light instead of looking where the keys were.

More on small-farm organic locavorism (from Freakonomics). Just because there are fewer cases of salmonella reported on from free-range chickens doesn't mean that the risks are lower - free range farms just don't produce enough eggs for their salmonella to get noticed.

Chart of the Day

From the Daily Dish.

Chart of the Day

A rare good use of a pie chart.

GDP vs. Welfare

A new measure of economic well-being?