Tuesday 9 August 2011

On having a powerful instrument

Frances Woolley's onto the problem:
But any exogeneous phenomenon that is closely related to an economic outcome of interest is usually thoroughly understood. Researchers will have had numerous opportunities to observe cause-and-effect relationships. Hail destroys crops and causes the price of wheat to rise. Cold winters increase the demand for fuel and cause the price of heating oil to rise. The low-hanging fruit have been picked, processed, and made into jam. 
This mean that, in practice, researchers face an exogeneity-plausibility trade-off -- hoping to find truly exogeneous explanatory factors, they have turned to less and less plausible explanations.
One line of research that exemplifies this trend is the use of biological or ancient historical factors to explain economic growth. True, some of this literature is fascinating and thought-provoking, for example Alesina, Giuliano and Nunn's research suggesting that modern attitudes towards gender can be explained by historical use of ploughs.
...
The problem is that exogeneity/endogeneity is easier to assess than plausibility. So a paper with an exogeneous but wacky explanatory variable has a reasonable chance of being published. A paper with a potentially endogeneous but sensible explanatory variable faces more challenges. 

When I was a grad student, I saw a paper presented that tried to estimate the effects of political efficacy on public spending. The author argued that counties with more radio listeners saw more funds dispersed during the New Deal. Seems plausible. The obvious endogeneity problem is that richer places get radio first; parsing out the effects of radio is then tough. The author used ground conductivity as an instrument: it affected AM radio reception but wasn't correlated with income. He presented all the usual econometric tests for exogeneity of instruments. I asked whether he'd tried just checking whether ground conductivity affected spending before radio could have hit the area. Maybe the data wasn't good enough for the obvious instrument plausibility test, but the paper hit the QJE so I suppose it wasn't necessary.


No comments:

Post a Comment