Terrorism as instrumental variable
Perhaps my econometrics is a bit rusty, but I found this summary of new research on Vox kind of odd:
There is both casual and formal evidence from the post-war period that abrupt movements in the price of oil have significant effects on the macroeconomy...However, the price of oil and the state of the world economy are endogenously determined and the links and interactions between the two are far from simple. The authors of DP 6937 experiment with terrorist acts as the instrumental variable, in order to examine the relationship between the price of oil, terrorist incidents and the resultant effects on profitability and margins.
How is it that terrorist acts are exogenous to the functioning of an economy? This book might suggest otherwise.
Comments (1)
Delicious
E-mail
Facebook
Follow us on Twitter
Easy, there. Of course, in the real world, everything is endogenous to everything else. However, as random variables go, terrorist "acts" (as opposed to, say, endemic violence) are endogenous to no-nonsense good ole macroeconomic stats in horribly nonlinear, weird and scary ways. Modeling them as endogenous variables would probably be both useless and impossibly complicated (we do not have the necessary thickness in the dataset, for one).
I'm not sure that even trying to do econometrics on that stuff makes any sense at all. In fact, I would be really surprised if it did.
Posted by: Alberto Cottica | Aug 23, 2008 3:20:48 PM