Handbook of Exchange Rates

11.2 Models/Taxonomy

Over the years, a number of models of fair value have been developed. Although some approaches are more theoretical, all of these models have been used in applied work, some only occasionally for ad hoc studies, while others are being updated and modified continuously. The latter is particularly true for those models used in the private sector or by international institutions for policy advice. In this category, we include the three fair value models used by the IMF Consultative Group on Exchange Rate Issues (CGER), as well as the Goldman Sachs fair value model.

11.2.1 “Adjusted PPP”: Harrod-Balassa-Samuelson and Penn Effects

Many studies have tried to measure exchange rate misalignment by exploiting the positive relation between real per capita income and relative prices across countries, that is, the fact that rich countries tend to have higher price levels than poor countries. This empirical relation is also known as the Penn effect after the Penn World Table of Summers and Heston (1991), and has been explained usually by appealing to productivity differentials between the tradeable and nontradeable sectors—the Harrod-Balassa-Samuelson effect (Rogoff, 1996). This effect provides a structural interpretation of long-run deviations from PPP (hence the “adjusted PPP” terminology used in many studies) based on real factors that can be exploited for measuring exchange rate misalignments.

The exchange rate misalignment can therefore be measured as the residual ...