Deviations in Real Exchange Rate Levels in the OECD Countries
Martin Berka and Daan Steenkamp
Why are some countries expensive and others cheap? The standard explanation – the Balassa-Samuelson theory based on sectoral differences in productivity – has long been proved to be irrelevant. Recently, Berka et al (2018) showed that a carefully measured case in the Eurozone shows strong support for the theory when they additionally included the notion that labour markets can set wages at “inefficient” levels (so-called “labour wedges”).
In this paper, we advance that literature by first constructing a dataset that precisely measures sectoral total factor productivity in levels for 17 OECD countries between 1970 and 2012, together with a precise panel measures of real exchange rates, and that of labour market institutional differences. We show that, the augmented Balassa-Samuelson model actually seems to hold empirically rather well in this sample, once the level information is taken into account. The inclusion of the labour market institutional differences is crucial to the extent that they do not cause productivity differences.
Finally, we observe that large unexplained differences in RER remain between the OECD member states even after we account for the differences in TFP and labour market institutions. While these could be understood to measure RER misalignment, we leave that to future research.