We study the differences in currency misalignment estimates obtained from datasets derived from two different International Comparison Programme (ICP) surveys. A decomposition exercise reveals that year 2005 misalignment estimates are substantially affected by the ICP price revision. Furthermore, we find that differences in misalignment estimates are systematically affected by a country's participation status in the ICP survey and its data quality—a finding that casts doubt on the economic and policy relevance of these misalignment estimates. The findings are robust to the use of alternative datasets and specifications. The patterns of changes in estimated degrees of misalignment across individual countries, as exemplified by the Brazil, Russia, India and China economies, are highly variable. Copyright © 2013 John Wiley & Sons, Ltd.