Friday, May 25, 2007

q-Theory of Investment



The above picture plots real investment in private fixed assets in the oil and gas extraction industry and the average real price per barrel of oil in the U.S. over time. The spike in oil prices in the late seventies is accompanied by a large spike in investment in the oil and gas industry.

Here are some casual explanations:

1) The change in value of the oil these firms hold in reserves gets valued as investment. This leads to a correlation between oil prices and investment, although the fact that firms held onto these reserves at a time of such high prices suggests they thought prices were going to stay that high. Basically you'd expect them to disinvest in reserves at this point.

2) The q-theory of investment says that when the ratio of the market value of assets goes above their replacement cost firms will invest.

What's strange about this is that more fashionable models of investment under uncertainty suggest that firms wouldn't react so strongly to a large move like this in oil prices because their future path is uncertain. I would have expected a more muted response, although it's possible that firms believed that oil prices had hit a new permanently high plateau.

Thursday, May 24, 2007

More Pop Economics

Pop songs favored along the American coasts and in hip European cities often feature dense lyrics tailored to connoisseurs of world-weariness and subtle relationship problems. By contrast, the average Asian pop song has rather simplistic lyrics, rarely going beyond declarations of mutual infatuation. Genres like grunge and goth have no equivalents in these cultures. Based on this evidence, a Martian student of our pop culture would conclude that there is far more misery in developed countries.

Could pop lyrics be a simple function of per capita income, as opposed to complex cultural factors? Consider Japan, a rich Asian country where the latest in pop despair is celebrated like there's no tomorrow.

Wednesday, May 23, 2007

Speaking of which....

...does anyone actually know what is the return to education?! There must be a million papers out there trying to find clever ways to estimate it, but I doubt anyone has even a ballpark figure in his mind about the actual return. So I'm perfectly sympathetic to Economagic's suggestion to estimate the impact of development research on actual development. I'm all for exposing the academic research racket.

Tuesday, May 22, 2007

Impact Analysis of Development Research

I propose running a simple cross-country regression with yearly growth as the dependent variable and a measure of annual development research about the country as the explanatory variable. It can be measured as the number of researchers working on "development" or the number of articles published in high quality journals about "development" in the country. Lags and cross-country spillover effects can also be analyzed.

It's common knowledge that development research isn't focused on countries with the biggest problems so endogeneity shouldn't be an issue.

Monday, May 21, 2007

Why there are more labor economists than laborers

This is a good time to dust off an old concept of mine: the academic ratio. Roughly speaking, this is the inverse of the universal limiting ratio (rigidly adhered to in nature) between a subject and the number of people studying it. The limiting ratio is very large - consider for example, that there was only one Joyce but there are already a million Joyceans, one Jesus Christ but a million Christian theologians, and of course, more labor economists than there are laborers! Of course, it's all mountains of uninspired hackwork but that's what academia's about, baby - if you ain't got game, write a paper about it instead.

The forces of mass psychology control the dynamics of the actual ratio at any point in time. In particular, the dynamics follow an inverse law - an academic discipline grows at a rate inversely proportional to the current ratio. That explains why labor economics is slowing down but development economics is not - there are many more poor people in the world than there are laborers in developed countries.

Sunday, May 20, 2007

Iraq War and Stock Markets

I posted an interesting result last night that finds a positive correlation between Egyptian stock market returns and U.S. casualties in Iraq. If we believe that higher U.S. casualties would increase the probability that the U.S. withdraws from Iraq, the result suggests that Egypt may want the U.S. out of Mesopotamia. Buy why? One reason could be that U.S. presence in Iraq is a destabilizing to the region (or at least perceived that way by Egyptian investors). Another could be that Egypt is eyeing for influence in Iraq and is waiting for the US to leave. It would be interesting to estimate correlations for other Mideast markets and Pakistan, Russia and India.

Saturday, May 19, 2007

The Second Law of Petropolitics?


According to globalsecurity.org, there have been 3,347 U.S. casualties since the beginning of the Iraq war in March 2003. I decided to merge the casual empiricisms from my previous post and Zero's to analyze the relationship between stock performance and U.S. casualties in Iraq. When I regress monthly US and Israeli stock returns on the number of casualties in Iraq, I find no statistically significant relationship. I think its nice to know that military casualties don't make Wall Street fat cats fatter.

However, when I look at the Egyptian market, I find that a 10% increase in fallen GIs correlates with a 0.2 percentage point rise in Egyptian stocks. Casual empiricism leads me to conclude that I've found the Second Law of Petropolitics (for the First Law, see Friedman's article in Foreign Policy).