ECON 337 Lecture Notes - Lecture 8: Instrumental Variable, Yellow Fever, Ethnolinguistics
Document Summary
It is true that countries with good institutions enjoy high incomes per capita. It is just as possible that rich countries are those that can afford good institutions. It is possible for causality to run both ways. Solving endogeneity problems requires some detective work, good smarts, and a bit of econometrics. Institutions (north a / economic growth (g. clark) Suppose we want to estimate the effect of x on y. We know there is an effect of x on y, but it is also possible that y has an effect on x as well. A solution to this problem is to find an instrumental variable. This is a variable that is correlated with x, but uncorrelated with y except through its effect on x. let"s call it z. Z x / y (e. g. x is institutional quality and y is gdppc) Our problem is that, when x and y change simultaneously, we don"t know who"s affecting who.