The instrumental variables strategy is commonly employed in empirical research. For correct inference using this econometric technique, the instruments must be perfectly exogenous and relevant. In fact, the standard t-ratio test statistic used in this context yields unreliable and often...

Currently, the commonly employed instrumental variables strategy relies on the knife-edge assumption of perfect exogeneity for valid inference. To make reliable inferences on the structural parameters under violations of exogeneity one must know the true correlation between the structural error...