Avoiding the Pitfalls of Instrumental Variables Estimation with Few or Many Instruments

Instrumental variables estimation can, in principle, avoid biases that ordinary least squares estimation suffers when explanatory variables are correlated with the disturbances. Finding appropriate instruments is a challenge. Ten strategies for avoiding invalid instruments (those correlated with the disturbances), as well as recently developed best practices for coping with weak instruments (those little correlated with the troublesome explanatory variable), are summarized and illustrated using ten published empirical papers. Recently developed methods for grappling with heteroskedasticity and with large numbers of instruments are also discussed and illustrated. The interpretive perils posed by heterogeneity in agents' behavioral responses to a troublesome explanator are also described and illustrated. Almost all of the procedures recommended in the paper can be implemented using existing commands for one or more standard econometric packages