Table 1.
Long run pricing-to-market coefficients for Japanese transportation equipment industries.
| (1) | (2) | (3) | (4) | (5) | (6) | (7) | (8) |
|---|---|---|---|---|---|---|---|
| Sector | Long Run Pricing-to- Market Coefficient | Long Run Pricing-to- Market Coefficient in Exchange Rate Depreciation Periods | Long Run Pricing- to- Market Coefficient in Exchange Rate Appreciation Periods | Probability Value of Null Hypothesis that the PTM Coefficients Are Equal in Appreciation and Depreciation Period | Share of Yen-invoicing | Share of Yen-invoicing During Yen Depreci-ation Periods |
Share of Yen-invoicing During Yen Appreci-ation Periods |
| Small cars | 0.994*** (0.108) | 1.05*** (0.088) | 0.638*** (0.122) | 0.002 | 0.04 | 0.06 | 0.01 |
| Standard cars | 0.887*** (0.211) | 0.996*** (0.256) | 0.696*** (0.197) | 0.336 | 0.09 | 0.09 | 0.07 |
| Motorcycles | 0.516*** (0.150) | 0.594*** (0.129) | 0.217 (0.189) | 0.064 | 0.16 | 0.15 | 0.17 |
| Standard trucks | 0.394*** (0.029) | 0.410*** (0.049) | 0.393*** (0.056) | 0.832 | 0.36 | 0.42 | 0.27 |
| Marine engines | 0.395*** (0.030) | 0.394*** (0.039) | 0.440*** (0.080) | 0.492 | 0.37 | 0.36 | 0.39 |
| Small truck | 0.454*** (0.072) | 0.439*** (0.082) | 0.454*** (0.106) | 0.904 | 0.45 | 0.50 | 0.38 |
| Auto parts | 0.330*** (0.066) | 0.379*** (0.042) | 0.072 (0.081) | 0.000 | 0.54 | 0.53 | .56 |
| Bus | 0.219*** (0.041) | 0.188*** (0.036) | 0.290*** (0.054) | 0.000 | 0.64 | .68 | .58 |
| Bicycle parts | 0.022 (0.046) | −0.013 (0.056) | 0.096** (0.046) | 0.03 | 0.91 | .91 | .92 |
Note: The long run pricing-to-market (PTM) coefficient in column (2) comes from an autoregressive distributed lag (ARDL) regression of the change in the yen export price on the lagged value of the yen export price, the lagged value of the yen/dollar exchange rate, the lagged value of the producer price index, the lagged value of industrial production in OECD countries, and lagged values of the first differences of these variables. The long run PTM coefficients in columns (3) and (4) come from a nonlinear ARDL regression of the change in the yen export price on the lagged value of the yen export price, the lagged value of the producer price index, the lagged value of industrial production in OECD countries, the sum of the change in the yen/dollar rate for all periods up to t-1 when the actual exchange rate was weaker than the value predicted by the Tankan survey, the sum of the change in the yen/dollar rate for all periods up to t-1 when the actual exchange rate was weaker than the value predicted by the Tankan survey, and lagged values of the first differences of all of these variables. Monthly data over the 2000M01-2018M12 period are used. The share of yen invoicing in column (6) represents the share over the January 2000 - December 2015 period and comes from Ito et al. (2016b) who obtained them from the Bank of Japan. The share of yen invoicing during depreciation periods in column (7) is a weighted average of the values from Ito et al where the weights depend on the fraction of time the yen is in a depreciation period. Yen depreciation periods are times when the yen is weaker than forecasted by the BoJ Tankan survey. The shares during appreciation periods in column (8) are calculated in an analogous manner. Heteroskedasticity and autocorrelation consistent standard errors are in parentheses.
*** (**) Denotes significance at the 1 % (5 %) levels.
Source: Bank of Japan, Organization for Economic Cooperation and Development, Federal Reserve Bank of St. Louis FRED database, Ito et al. (2016b), and calculations by the author.