One way to learn about what makes some countries richer is to run statistical tests to see which factors are good at predicting a nation’s level of productivity. Sometimes it turns out that a relationship is just a coincidence (like the fact that people in rich countries eat more ice cream), while other statistical tests really can tell you about the ultimate causes of productivity. A statistical test can’t tell you everything, but it might help point you in the right direction. In courses on econometrics and statistics, you can learn about how to run sensible tests.
Let’s look at one well-known set of tests, to see if what you learned in this chapter matches the statistical evidence. Here are 17 variables that turned out to be very strong predictors of a nation’s long-run economic performance in literally millions of statistical tests
They are in rank order, and a “+” means more of that value was good for long-run productivity:
1. Whether a country is in East Asia (+)
2. Level of K-6 schooling (+)
3. Price of capital goods (–)
4. Fraction of land close to the coast (+)
5. Fraction of population close to the coast (+)
6. Malaria prevalence (–)
7. Life expectancy (+)
8. Fraction of population Confucian (+)
9. Whether a country is in Africa (–)
10. Whether a country is in Latin America (–)
11. Fraction of GDP in mining industries (+)
12. Whether a country was a Spanish colony (–)
13. Years open to relatively free trade (+)
14. Fraction of population Muslim (+)
15. Fraction of population Buddhist (+)
16. Number of languages widely spoken (–)
17. Fraction of GDP spent on government purchases (–)
a. Which of these factors sound like the “three factors of production?” Which ones do they sound like?
b. Which of these factors sound like the “five key institutions?” Which ones do they sound like?
c. Which of these factors sound like geography?
d. The western United States was a Spanish colony until 1849. On average, former Spanish colonies have had poor economic performance. Does the western United States fit that pattern? Why or why not?