Dornbusch Was Wrong: There is no Convincing Evidence of Overshooting, Delayed or Otherwise

Several articles claim that Eichenbaum and Evans (1995) shows that nominal exchange rates experience a delayed version of Dornbusch overshooting. These same articles usually claim that impulse responses similar to those in Eichenbaum and Evans are evidence of such overshooting. But Eichenbaum and Evans never claim that their evidence implies overshooting, delayed or otherwise. More importantly, impulse response functions like those in Eichenbaum and Evans do not support overshooting. Three recent articles repeat this misinterpretation of the evidence. My objective is to use those articles to illustrate how the evidence about overshooting is widely misinterpreted. What is interpreted as supporting overshooting is at least as consistent with an efficient market as it is with overshooting.