Lagged Explanatory Variables and the Estimation of Causal Effects

…without careful arguments on substantive grounds, lagged explanatory variables should never be used for identification
purposes.

This is from a new working paper of mine, coauthored with Cornell’s Tom Pepinsky, an associate professor of government, and Taka Masaki, a PhD student in government, on “lag identification.” This is a very common strategy to avoid problems of endogeneity or reverse causality, especially in political science, although table 1 in our paper shows that this is not exactly uncommon in economics.

We expose the assumption that underlies this strategy, which we term “no dynamics among unobservables.” We also argue that this assumption is almost impossible to defend on substantive grounds, because it requires knowledge about the time series properties of a variable which is unobserved.

We are fairly certain that the germ of the idea for our paper was this post by Phil Arena.

(This post is being simultaneously published today on Tom’s blog.)

No related content found.