The World Bank’s new World Development Indicators were released a bit over a month ago. The impact on the estimates of RMB misalignment are substantial. (This is an elaboration on a RGEMonitor post by Yin-Wong Cheung from a week and half ago, and is based on preliminary results from a presentation made yesterday at a Deutsche Bundesbank and Center for Financial Studies/Goethe University Frankfurt Workshop on Panel Methods and Open Economies”.)
Recall, in Cheung, Chinn and Fujii (2007) [pdf], we used an empirical relationship between price levels and per capita income called the “Penn Effect” to establish an “equilibrium” level of the real exchange rate. Using this estimated relationship (based on 160 countries’ worth of data over the 1970-2004 period), we found that the RMB was some 40% or more undervalued in 2006. Despite this enormous estimated misalignment, we also found that this deviation was not statistically significant, when using the 5% level of significance. (Additional discussion: ).
Figure 1: Price level-per capita income in PPP terms relationship (blue line), +/- 1 std error band (long dashed lines), +/- 2 std error band (short dashed lines); OLS estimates for 1970-2004 period. Source: Cheung et al. (2007).
Using the same methodology, but the latest revision of the WDI (see discussion , ), we obtain drastically different results. These results can be summarized in the below graph, analogous to Figure 1.
Figure 2: Price level-per capita income in PPP terms relationship (blue line), +/- 1 std error band (long dashed lines), +/- 2 std error band (short dashed lines); OLS estimates for 1980-2006 period. Source: authors’ calculations.
The new estimated misalignment is on the order of 10%. The substantial change in outcomes (slope coefficients, estimated misalignment) suggest that uncertainty regarding the data, in addition to uncertainty about the coefficient estimates, should be kept in mind when considering misalignment estimates, even if the extent of uncertainty is assessed in an informal manner.