The Washington Consensus emphasizes the economic costs of real exchange rate distortions. However, a sizable recent empirical literature finds that undervalued real exchange rates help countries to achieve faster economic growth. This paper shows that recent findings are driven by inappropriate homogeneity assumptions on cross-country long-run real exchange rate behavior and/or growth regression misspecification. When these problems are redressed, the empirical results for a sample of 63 developing countries suggest that deviations of the real exchange rate in either direction from the value that is consistent with external and internal equilibrium reduces economic growth. Deviations from Balassa-Samuelson adjusted purchasing power parity on the other hand do not seem to matter for growth performance. The real exchange rate should thus be consistent with external and internal balance irrespective of implied purchasing power parity benchmarks.