Firm predicted exchange rates and nonlinearities in pricing-to-market

This paper employs the nonlinear autoregressive distributed lag (NARDL) model to investigate possible short- and long-run asymmetry in the pricing-to-market (PTM) behavior of Japanese exporters. In contrast to the conventional threshold specification, this study uses firm predicted exchange rates published by the Bank of Japan, to distinguish between yen appreciation and depreciation periods. Using Japanese export price data at an industry level, we demonstrate that (1) the short-run PTM is almost complete and symmetric in all industries over the entire sample period from 1997 to 2018 and (2) during the latter sub-sample period from 2007 to 2018, Japanese exporters tend to engage in asymmetric PTM behavior in the long run. In the yen appreciation period, most industries choose incomplete but relatively strong long-run PTM. However, in the yen depreciation period, competitive industries tend to conduct complete PTM in the long run, while less-competitive industries tend to raise the degree of exchange rate pass-through (ERPT) in the long run. The results are robust when extended to disaggregated industry level. These empirical findings have important policy implications for the recent unresponsiveness of Japanese real exports to the substantial depreciation of the yen from the end of 2012.