Thursday, December 29, 2005

Human Capital

This remarkable interview by African Economist, James Shikwati, says that the best way to help Africa is to stop all foreign aid. Africa is poor, he argues, because the people there lack skills and countries lack institutions, not because it does not have stuff. Foreign aid undermines the ability for either of these to develop, thus condeming the country to poverty.

I think he has a good point, but I'm not sure that cutting of aid will enable that capital to develop. Capital accumulation requires strong individual property rights, and I don't think anyone knows how to create institutions that support that.

The urge to redistribute is very strong. Brad DeLong reviews a book that argues economic growth is a good thing (the review was published in Harvard Magazine, where I suppose such a statement is still controversial) but in the end he concludes that the US should be more redistributional. This assumes that people in the US as poor because they don't have enough money, and that the skills and human capital destroyed by transfer payments is costless. I don't think that reflects the situation of the poor underclass accurately.

Wednesday, December 28, 2005

Daughters create lefties?

Some researches claim that families with daughters tend to vote left-wing, while families with more sons tend to lean more to the right. Interesting. This follows a bit of tortured hang-wringing (isn't it sexist to claim that women are linked to soft politics) but that sorts itself quickly by noting that maleness is one step away from mental retardation and psychopathic.
The researchers have been accused of propagating gender stereotypes, and of perpetuating the idea that women go in for softer politics than men. The feminist ire, however, is misdirected. Some of the most exciting scientific research points to a similar, intuitive conclusion? that, on the whole, men and women think and behave differently. One emerging theory of autism,? a condition that affects three times as many men than women is that autistic behaviors are extreme versions of typical male traits
.The reasons given for why women may prefer nurturing politics is also muddled:
But how and why? He frames his theory in utilitarian terms: because women tend to be more group-oriented, and to be paid less, we may expect them favorour a political system that taxes heavily and spends the taxes on communal improvements, such as creches or police patrols.
The reason women are paid less is *because* they are more group oriented. In a study of MBA graduates, they found that, controlling for occupation, women were paid ~5% less than men. They also found that men were much more likely to try and negotiate for a better offer than women, and it was this willingness to ask for more than explained the entire salary difference. Women don't like to negotiate because they feel it puts the relationship at risk. Men do not feel the same way.

If you live in a family of daughters, then the best incentives are probably emotional -- being hurt by a child's misbehavior may get you farther than punishing them. With boys though, being hurt will not do the trick (they probably will not notice).

In a large diverse population, communal ties and social pressure become less effective ways of rewarding good behavior and punishing bad. The larger, more complex, and more diverse a population is, the more you need clear disciplinary incentives to elicit the behavior you want. Less mommy, more daddy.

Sunday, December 25, 2005

Merry Christmas!

And Happy Quaid-i-Azam's birthday to those of you from Pakistan.

Wednesday, December 21, 2005

Houses not moving in the burbs

Here's an example of someone who wanted to sell her suburban house and move to the city, but got stuck because the house isn't selling.

Tuesday, December 20, 2005

Economists and money

I liked this observation from Cafe Hayek:
The misunderstanding in these letters is evidence that economists are the last people who are obsessed with prices -- the last people who believe that the only things that matter are money and costs expressed in money. Rather, it's non-economists who are obsessed with money measures and money prices; it's non-economists who are most prone to overlook costs that aren't pecuniary -- to assume that only when people make monetary outlays do people incur costs.
The economic perspective forces you to be much more clearheaded about costs than people seem to be otherwise -- nominal vs real, out-of-pocket vs opportunity, time vs money are all considered.

I had a long email exchange with a reader (TY) regarding this post on torture. Ultimately, I asked what I should have asked right at the start and got the following response:
Q: If it could be proved that the benefit from torture exceeded the cost, would you change your mind an support it.

A: [No]. I think torture inherently debases human dignity, which I hold as a fundamental human right. But I also appreciate that my morality may not be my neighbor's, and that many people such as yourself are persuaded by a costs-benefit analysis. I think a policy of torture can be defeated on those grounds as well.
I don't think this is an uncommon position, but it is pointless to discuss. While TY argues that cost-benefit is not the right way to think about this, his position is essentially that he will bear any cost and forgo any benefit a priori. Costs and benefits do not go away just because you don't want to think about them.

In a seperate post, TY's position was summed up well in the comments "The question is how you justify torture without assuming that the end justifies the means. I don't know that that can be done." And that is exactly right.

Mechanical Turk

I recently heard about Amazon's "Mechanical Turk" project and decided to try it out. Basically, the project takes a number of tasks that are simple for humans but difficult for computers (like picking out a band's name from an album cover) and distributing it to people online. You get paid a few pennies per task you complete.

Here is their opening page


Here is an example of the sorts of tasks (HITS) you can do:


Unfortunately, there were too many people want trying to do the tasks, and I found that I could not pick one fast enough to be selected to do it.


I put in about 20 minutes of effort and made 3 cents. Better keep my day job.

For all the money invested in optical character recognition technology and artificial intelligence, I am amused that the most efficient way to identify the name of an album is by distributing it across the internet and having hundreds of online penny-wage workers (with Internet connectivity!) read and click.

Monday, December 19, 2005

End of the condo bubble in Boston?

I sat next to a real estate guy on the plane today, and we discussed the condo market in Boston. He told me that condos in the city are bought by empty nesters who have sold their suburban home and are moving back, and by young people just starting out.

In Boston, suburban houses are no longer selling. Since they can't sell in the burbs, they cannot move to the city. And young people are not moving to the city. But since volume slows in the winter, I don't think people will start cutting prices until at least the spring/summer.

Thursday, December 15, 2005

Google Adwords

Google adwords is a phenomena. It's made life safe for the banner by taking all the direct marketing people (who spam your mail box and count reponse rates) and giving them something they can count, leaving the "creative" types free to ignore actual customer responses entirely.

Exactly how Google adwords works is opaque, as is much of Google. This Cringely article describes an experiment where a small eRetailer created a second site that mirrored his first one, and then used this second site to see what impact different pay rates for adwords has. Adwords work by the retailer paying for placement -- the more he pays, the higher the placement is.

On the original site, the eRetailer paid 10cents a word and got 15,000 clickthroughs. On his mirror site he paid 100cents a word and got more clickthroughs, but Cringely does not say how many more. He then dropped his price on the mirror site and got 1,200 clickthroughs.

Cringely has all sorts of conspiracy theories for why this happens, but I'm guessing that Google Adwords tries to reduce how much the system can be gamed by trying to weed out non-serious retailers. After all, if someone clicks through an adword and then gets ripped-off, they may not click through an adword again. A new website (which is what this mirror was) that changes how much it pays dramatically in a short period of time is certainly less reliable than an old website that's been paying the same for months. It does not seem unreasonable for Google to degrade the ranking such a company would get. (via /.)

Wednesday, December 14, 2005

House prices drop in Boston

According to this Globe article, house prices are dropping dramatically in Boston.
[R]eductions in asking prices of 10 percent or 20 percent are now common in both high and moderately priced neighborhoods, according to real estate agents and listings of homes for sale. In Cambridge, price cuts averaged $300,000 in a sampling of a dozen houses listed in the $1.25 million to $4.3 million price range. In suburbs like Tewksbury and Hopkinton, homes originally listed for around $500,000 have been slashed to the low $400,000s.
The article does not detail what these prices were selling for 5 years ago.

The pricing in Boston defies me. It does not have great weather, unlike CA, and is not a cultural mecca, like New York. It should be priced at the level of Chicago, which I think is a great city.

Tuesday, December 13, 2005

Quality of Service

On the current internet, all traffic is treated the same, so carriers cannot offer higher service even if customers are willing to pay for it. AT&T and Bell South are lobbying to change this -- presumably in their new plan, some IP packets will get higher priority over others.

This does not seem obviously bad to me, but I instinctively oppose anything that messes with the basic end-to-end design of the Internet.

iTunes tapping out

Growth in sales on the iTunes music store seem to be flattening out. The article is trollbait, but I'm wondering why iTMS has done as well as it has. After all, a $400 60 GB iPod can hold about 30,000 songs (I'm guessing), and so would cost $30,000 to fill up legally. That's simply not going to happen.

Why not torture

Slate's Michael Kinsley, in arguing against legalizing (and regulating) torture says
The trouble with salami-slicing is that it doesn't stop just because you do. A judicious trade-off of competing considerations is vulnerable to salami-slicing from both directions. You can calibrate the viciousness of the torture as finely as you like to make sure that it matches the urgency of the situation. But you can't calibrate the torture candidate strapped down before you. Once you're in the torture business, what justification is there for banning (as Krauthammer would) the torture of official prisoners of war, no matter how many innocent lives this might cost? If you are willing to torture a "high level" terrorist in order to save innocent lives, why should you spare a low-level terrorist at the same awful cost? What about a minor accomplice?
I must admit, I find the moral soft-headedness Kinsley displays on this point (along with other so called "Geneva Convention" adherents") disappointing.

The reason you might OK torturing unlawful combatants but ban torturing official prisoners of war is to encourage combatants to fight lawfully--ie. in a way which minimizes civilian deaths--not unlawfully. Soldiers wear uniforms, stay away from civilians, and have a chain of command who can order a surrender. Terrorists hide amongst civilians, using them as shields putting them in danger, and have no authority who can accept a surrender (or give one) on their behalf. Therefore, anything which makes fighting as a soldier instead of a terrorist desirable is good -- good for civilians.

Kinsley apparantly does not consider the effect that blurring the line between soldiers and terrorists has on civilian populations. He correctly identifies that "Torture is like almost every other issue: It involves trade-offs between the rights of individuals and the needs of society" and then refuses to even contemplate those very principles - both moral and practical.

Debt and interest rates

When interest rates are low (too low?) what is the rational way to take advantage of that (mis)pricing? In this note on JaneGalt, I argued that governments should run deficits in times of subsidized lending.

But what should households do? This excellent post by Calculated Risk details what households *have* done:
It appears households initially took on debt and kept their payments steady as interest rates dropped. However, recently the portion of disposable personal income dedicated to debt service has risen steadily. As interest rates rise, and with the new credit card minimum payments, the DSR and FOR will probably continue to rise, putting more pressure on household finances.
As debt got cheaper, individuals took on more and overall payments stayed the same. However, it seems that people are now allocating more of their household income to debt servicing.

One might expect that, as debt gets cheaper, households would buy what they bought before and pocket the saves (reducing the % they spend on debt servicing overall). What seems to have happened is that people took the savings from lower interest rates and spent it competing up the price of housing. But now they seem to be just plain-ole spending more on housing as both mortgage financial obligations and household debt overall rise in lockstep.

This, I might add, is a stupid action to take in the face of low rates. Moving consumption forward is sensible (and hard to do) but just consuming more, or consuming overprices assets, is dumb.

And on the subject of dumb, this article on Slate is really bogus, btw. It argues that mortgage lenders are afraid to foreclose on delinquent homeowners because it's onerous.

If it's onerous today it's been just as onerous in the past, so this is not a credible reason for housing lenders to be soft on deilnquency. It concludes
All of which means the housing boom is being fueled by the willingness of lenders to let borrowers get behind—and stay behind—on their payments. Homeowners go deeper and deeper in debt and become less and less home "owners," but they get to keep the roof over their heads. It used to be that only gigantic banks and corporations like Citigroup and Chrysler were regarded as too big to fail. Today, the humble homeowner enjoys that status as well.
This is additionally bogus. Firstly, there is nothing mystical about someone becoming less and less a home "owner" and getting a roof over their head -- it's called renting. Secondly, a shift in financing to ignore the principle and focus on financing means that banks are moving from being financiers to being landlords, and therefore being extremely exposed to the price of real estate.

When prices fall and become a drag on lenders balance sheets, they will work hard to get them off. To date, housing has not fallen in real or nominal terms, nor have rates moves that much, so home default rates remain low.

Friday, December 09, 2005

Needs an editor

I started reading George RR Martin's "Song of Fire and Ice" whatever-ogy several years ago, and really enjoyed the first few books but felt things had begun to drag near the end. Years later, he's finally produced the next (part) installment of this behemoth (Act II Scene I), with the promise of subsequent books getting larger and larger.

This man needs an editor.

Fantasy, for some reason, seems to be particularly prone to this sort of narcissistic excess, and I'm not sure why. Forgiving readers, perhaps?

Wednesday, December 07, 2005

More Dubai Madness

They just opened their ski slope.

Tuesday, December 06, 2005

Unintended consequences

I very much enjoyed the first comment in this post about unintended consequences of government regulation
1970 - Massachusetts becomes the first state to pass "no fault" car insurance regulations. By the mid 1970's fully regulated rates are in place for auto coverage.

But wait...

2005 - Massachusetts continues to be among the states with the lowest number of insurance providers. It's citizens also pay the highest rates in the nation for their demographic category.

Now why didn't that work?? Well, if you ask them, which I have, they plead ignorance to the cause, but will then assure you that if the market is deregulated and allowed to be competitive their rates will go up. (I'm not making this up.)

If you show them charts and graphs of their rates versus other similar states who enjoy lower rates, which I have, they are baffled.

Would it make sense to require one year of macroeconomics for high school graduation?
One of the many reasons that I've chosen to leave MA is the asanine car insurance regulations in this state. My wife and I share one small car, and the insurance we pay on it is far in excess of the depreciation, car loan financing, and gasoline costs combined. If the state was 30 degrees warmer they may be able to get away with regulation this dumb, but it's December and I'm not feeling charitable.

My car insurance company, the ludicrously overpriced AMICA, sent me a letter saying they were agitating for a more competitive car insurance market and that I should call my Congressman and complain. It was like the twilight zone.

The stupidity of one-size-fits-all

I am sympathetic to the record industries plight -- people are massively infringing on copyright online and it is reasonable to assume that this infringement will hurt their business.

That said, I think current copyright laws are crazy, the harm to music sales from P2P seems empirically to be rather small, and it is impossible to outlaw P2P without also outlawing the Internet.

Live365 seems to have cut a deal where they can stream tracks so long as they don't play more than 2 consecutive tracks from the same CD. Sadly, some classical music does not fall into tihs category. Longtime reader CD sent in this link detailing how a stupid law is having a bad effect.

Great read

I very much enjoyed this article by Michael Lewis on how the Texas Tech football team manages to punch above their weight through inspired coaching. Moneyball for college football.

Thursday, December 01, 2005

Funny

I think this is really really funny. It's short, and I won't give away the punchline here.