Economic Effects of Yen Appreciation: a Sectoral Approach (in Japanese)

1. Background The Japanese economy has experienced a drastic currency appreciation since 1985. Price changes in the Japanese economy in response to the exchange rate changes, however, have not been as large as expected. One of the indication of the inresponsiveness of prices is the aggravation of so-called internal-external price differential problem. This study investigates changes in export, import, and domestic output price at the sectoral level after 1985 and tries to identify for what commodities prices are distorted and to analyze what causes such distortion. 2. Analytical framework This study starts from an intuition that the yen appreciation was the most significant factor to affect prices in the Japanese economy in the past ten years. The analysis focuses on long-run deviations from the law of one price for tradable goods rather than short-run price responses to exchange rate fluctuation. The benchmark of the analysis is a long-run equilibrium where commodity arbitrage internationally equates prices of tradable goods. By comparing the actual price changes with those which are theoretically expected in the long run, anomalies in pricing are identified. The major statistical data used here are sectoral deflators of exports, imports, and domestic outputs obtained from the Japanese SNA-based input-output tables. The study mainly relies on the data disaggregated into 89 sectors. We first measure proportional price changes in the period of 1985 to 1994 and then calculate four indicators: export pass-through rates, export price discrimination ratios, import pass-through rates, and import price penetration ratios. By analyzing the values of these indicators, anomalies in pricing are identified. Second, cross-sectional correlations between four indicators and a number of economic variables are checked to investigate the causes of pricing anomalies. Third, the domestic price changes of each good are decomposed into tradable / nontradable input elements and value added element in order to characterize price changes from the input side. 3. Major findings The Major findings are twofolds. First, from the export side analysis, the pricing behavior of material industries such as iron&steel and chemical industry is found to be peculiar. In these industries, export prices get lower along the yen appreciation while domestic prices do not decrease much, which suggests the existence of pricing-to-market behavior. In theory, such price discrimination is possible only if the domestic market is under imperfect competition and there exists some mechanism to impede international commodity arbitrage. Although we need to investigate what sort of mechanism makes domestic and export markets segmented even in the long run, our finding suggests the possible existence of trade barriers such as anticompetitive behavior of producers and distributors in these industries. Because the industries are in the upstream of industrial linkage, such pricing behavior may negatively affect downstream industries such as machinery industries by raising their production costs. Second, we found that in the period of 1985 to 1994, domestic output prices had not decreased as much as import prices had in most of the commodities. Such tendency is salient in the commodities of industries in which Japan has comparative disadvantage, namely, agricultural products, mining products, and light industrial products. If domestic products become high-grade while imported products shift to low-grade, the expanding price gaps between domestic goods and imported goods may be partially explained. However, the results strongly suggest the existence of some "trade barriers" in the wide sense which impede import price penetration into domestic prices. Moreover, the "trade barriers" seem to become higher for many products. The "trade barriers" could be either the one imposed by the government or not, either the one accompanied with anticompetitive business practices or not, and either "fair" or "unfair." It may be therefore misleading to interpret the result as indicating the existence of "trade barriers" in the narrow sense to be removed. However, the systematic expansion of the gap between import prices and domestic prices must be investigated in further research.