We show that the use of the real effective exchange rate to test for purchasing power parity, as in Astorga (2012) and other studies, is subject to a problem that biases tests against finding evidence of PPP. The problem is illustrated using Astorga´s data on six Latin American countries.