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...

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...