Diversions.

I finally have some time to blog! These past two weeks have been crazy. By the end of this week, I will have been on a bus/car for a total of 27 hours...
I picked out my apartment in Pentagon City and it's going to be awesome! It's pretty much across the street from the Pentagon City mall, Pentagon City metro station, Costco and a Mr. Smoothie. What more could a girl who does not drive ask for? Maybe a puppy...

Anyway, I linked you guys to Jenn's blog earlier. She wrote an awesome post about Chavez's impact on Venezuela. As promised, here is my post on expropriation and social institutions. (I promised Jenn it would be meaty...)

In ECON 333, we studied a paper called, "Why do some countries produce so much more output per worker than others?" by Robert E. Hall and Charles I. Jones. In this paper, Hall and Jones hypothesize that output per worker is affected by differences in social infrastructure across countries. They assume that social infrastructure is determined historically by location and other factors captured in part by language.

So, what exactly is "social infrastructure?" Unfortunately, social infrastructure is not easily quantifiable. However, Hall and Jones think that "the ideal measure of social infrastructure would quantify the wedge between the private return to productive activities and the social return to such activities. A good social infrastructure ensures that these returns are kept closely in line across the range of activities in an economy." Good social infrastructure protects people against "diversion." Diversion includes things like expropriation, corruption, etc.

Not surprisingly, Hall and Jones find that:
1) Differences in social infrastructure across countries cause large differences in capital accumulation, educational attainment, and productivity, and therefore large differences in income across countries.

One might argue that there is a feedback effect from output per worker to social infrastructure (i.e. high output per worker may lead to the creation of better social infrastructure). However, Hall and Jones control for this feedback by using instrumental variables.* This is explained in their last finding:
"The extent to which different countries have adopted different social infrastructures is partially related to the extent to which they have been influenced by Western Europe. Using distance from the equator and language data, we conclude that our finding that differences in social infrastructure cause large differences in income is robust to measurement error and endogeneity concerns."

Basically, Hall and Jones are saying that countries that have been influenced by Western Europe tend to have better social infrastructure and higher output per capita. They study a sample of 127 countries and give a nice table of different countries' productivity levels as ratios to U.S. values.

Here are some figures:
Canada: 0.941
U.K.: 0.727
Japan: 0.587
India: 0.086
China: 0.060
Zaire: 0.033
The average for 127 countries was 0.296 w/a standard deviation of 0.268.

Why is the quality of social infrastructure so different across countries? (UGH, my mom just pulled a white hair out of my head as I was typing...WTF. THAT HURT.)

Anyway, there are several theories that address the above question, but I think I'm going to need another blog post to write about Acemoglu, Johnson & Robinson, and Engerman & Sokoloff...

* Instrumental Variable:
Let's say we want to find the effect of "x" on "y." However, "x" is not quantifiable. So, we create an instrumental variable, "z," which is quantifiable. "z" affects "y" ONLY through "x." A bad instrumental variable affects "y" through other pathways.
For example,
x = total number of hours of TV watched
y = prevalence of autism in a city
z = amount of rainfall in that city

If we had no way of measuring the number of hours of TV watched in a city, but wanted to study its effect on the prevalence of autism, we could create instrumental variable "z," which is the amount of rainfall in that city.

More rain --> More kids stay indoors and watch TV --> higher prevalence of autism(?)

However, "z" is not a very good instrumental variable because you can make a pathway like this:

More rain --> More kids wear raincoats --> higher prevalence of autism(?)

Does all of that make sense?

OMGSHHH. I'm going to die laughing. I'm watching FRIENDS DVDs and my dad says, "Why do these people never look older? They're still so young!" I told him I was watching DVDs, and he said, "Oh okay. So they're old now, right?" Haha. LOVES IT.