Faculty of Business and Law SCHOOL OF ACCOUNTING, ECONOMICS AND FINANCE School Working Paper - Economic Series 2007 SWP 2006/17 Exchange Rate Pass-Through and Relative Prices: An Industry-Level Empirical Investigation Prasad S. Bhattacharya, Cem A. Karayalcin and Dimitrios D. Thomakos The working papers are a series of manuscripts in their draft form. Please do not quote without obtaining the author’s consent as these works are in their draft form. The views expressed in this paper are those of the author and not necessarily endorsed by the School. Exchange Rate Pass-Through and Relative Prices: An Industry-Level Empirical Investigation♣ by Prasad S. Bhattacharya+, Cem A. Karayalcin∗ and Dimitrios D. Thomakos♠ Abstract In this paper we explore the extent of exchange rate pass-through for the USA, UK and Japan using a post-Bretton Woods industry-level dataset. We investigate how different channels of exchange rate pass-through affect domestic and import prices. Our analysis is suggestive of two channels of transmission and we find considerable variation in the extent of pass-through across industries and countries. JEL Classifications: F0, F31, F32, F41. Keywords: Exchange Rates, Pass-Through Effect, Expenditure-Switching. ♣ An earlier version of this paper was presented at the 9th International Conference on Computing in Economics and Finance in Seattle and at Florida International University. We would like to thank two anonymous referees and the conference and seminar participants for useful comments and suggestions. The programs and the data used in the paper are available upon request. All remaining errors are our responsibility. + School of Accounting, Economics and Finance; Faculty of Business and Law; Deakin University, Melbourne Campus at Burwood; Burwood, Victoria 3125; Australia. E-mail: prasbhat@deakin.edu.au ∗ Corresponding author. Department of Economics; Florida International University, University Park, Miami, Florida 33199, USA. E-mail: karayalc@fiu.edu; Phone #: (+1 305)348-3285; Fax #: (+1 305)348- 1524. ♠ Department of Economics; School of Management and Economics; University of Peloponnese; Greece. E-mail: thomakos@uop.gr 1 1. Introduction In this paper, we address the issue of exchange rate pass-through at industry-level prices. The exchange rate pass-through is the percentage change in local currency import prices resulting from a one percent change in the nominal exchange rate. The ongoing debate in the literature over the extent of pass-through has important policy implications.1 One strand of the literature argues that exchange rate pass-through has, at best, a minimal effect on local currency prices. This happens due to either pricing to market (PTM, see Krugman (1987)) or local currency pricing (LCP, see Devereux (1997)). Under these pricing mechanisms, firms either price discriminate (PTM) or set a different price in foreign currency for sales to foreign households (LCP), leading to low pass-through. As a result, changes in the nominal exchange rate may not be fully passed through to goods prices, and, therefore, consumer prices may not be very responsive to such changes. This implies less “expenditure-switching”, i.e., a change in the exchange rate might not lead to much substitution between domestically-produced goods and imports. From a domestic monetary policy perspective, a small nominal exchange rate transmission to import prices may mean a lower expenditure-switching in response to changes in domestic monetary policy, implying that monetary policy is more effective in dealing with real shocks. The low pass-through evidence also makes a fixed exchange rate regime preferable, because a sudden shortage in the supply of foreign goods due to some exogenous shock will lead to large and undesirable currency depreciation under a flexible-rate regime. Previous empirical research reported that prices are indeed sticky in consumers’ currencies.2 New open economy macroeconomics models use the low pass-through evidence from earlier studies in supporting the assumption of nominal price rigidity in buyers’ or consumers’ currency. Another strand of the literature, however, argues that exchange rate pass-through has considerable effects on local currency prices. The main argument here is based on producer currency pricing (PCP) models, which assume that importables are priced in exporters’ or producers’ or sellers’ currencies. Given this, currency depreciation in the destination country will lead to higher import prices in that country. Under producer currency pricing, with only the prices of imported goods changing because of high pass- through, there is a change in relative prices. As a result, we would expect high “expenditure-switching” to follow from a nominal exchange rate change, contrary to the local currency pricing or pricing to market-type models. From a domestic monetary 1 Lane (2001) provides an excellent survey of the growing literature on new open economy macro models involving monetary policy. 2 Existing studies provide evidence of low pass-through towards consumer prices. New theoretical models that support a fixed exchange rate regime based on the assumption of domestic price rigidity are surveyed by Engel (2002). We selectively mention Mussa (1986), Rogers and Jenkins (1995), Engel (1993, 1999, 2000), Engel and Rogers (1996, 2001) and Parsley and Wei (2001). A number of past studies also investigate the extent of pass-through at industry-level prices. Dornbusch (1987) uses disaggregated industry-level prices to investigate the strategic interactions among domestic producers and producers of import-competing goods to explain low pass-through to import prices, in accordance with the pricing to market argument of Krugman (1987). See Yang (1997) and references therein for a review of these studies. For pass-through studies showing evidence of pricing to market in import prices, see Goldberg and Knetter (1997) and Campa and Goldberg (2002). 2 policy perspective, evidence that supports a producer currency pricing mechanism would imply that monetary policy is ineffective in dealing with real shocks. Apart from its role in expenditure-switching, exchange rate pass-through can also affect the prediction and control of inflation. Ball (1999) incorporates the degree of pass-through in the monetary policy rule to control inflation. McCarthy (2000) addresses this issue by incorporating a distribution chain of pricing and finds that exchange rates have a modest effect on domestic inflation while import prices have a stronger effect. In a number of influential new open economy macroeconomics papers, Obstfeld and Rogoff (1995, 1998) show that changes in consumer prices in the short-run can be explained with changes in nominal exchange rates assuming producer currency pricing- type models. Obstfeld (2002) and Obstfeld and Rogoff (2000) also use producer currency pricing and provide some evidence showing that a nominal exchange rate depreciation drives down relative export prices and increases relative import prices. These studies suggest that exporters largely invoice in home currency and, therefore, changes in the nominal exchange rate have significant short-run effects on international competitiveness. In their investigation of German, Japanese and US automobile exports to seven industrial-country destinations, Gagnon and Knetter (1995) find that pricing to market is greater in the long-run than in short-run, which is consistent with invoicing in the exporters currency. Therefore, the pass-through to export prices is not zero, implying that local currency pricing towards export prices may not be true. Proponents of producer currency pricing use similar evidence to suggest price stickiness in sellers’ or exporters’ currency. To address issues regarding price stickiness in terms of buyers’ or sellers’ currency, Obstfeld (2001) and Obstfeld and Rogoff (2000) build hybrid theoretical models incorporating producer currency pricing to import prices and local currency pricing or pricing to market to domestic prices from the perspective of a firm. These models generate expenditure-switching similar to the traditional Mundell-Fleming- Dornbusch models, contradicting the local currency pricing or pricing to market oriented studies. Engel (2002), Obstfeld (2002) and Lane (2001) call for a detailed empirical investigation of the extent of pass-through. In this paper we do this by investigating the extent of pass- through at the industry-level. Our contribution to the relevant empirical literature can be summarized as follows. First, unlike the existing studies which focus exclusively on either long-run or short-run pass-through effects, we investigate the extent of pass- through in both the short and the long-run using industry-level monthly data for import prices, domestic prices (which includes consumer and producer prices) and export prices simultaneously. Second, we identify different channels of pass-through involving prices of both imported and domestically-produced goods. One channel of exchange rate pass- through goes through prices of imported goods at the point of entry. The other channel involves the prices of domestically-produced goods through adjustments in foreign and domestic markups. Third, we provide additional evidence in support of the results reported in Obstfeld and Rogoff (1995, 1998) that suggest that changes in consumer prices in the short-run can be explained by changes in nominal exchange rates as in producer currency pricing models. 3 We carry out the empirical analysis in two parts. In the first part of the study, following the suggestion by Obstfeld (2001, 2002), we empirically explore different pass-through channels to prices of imported and domestically-produced goods using a monthly industry-level dataset for USA, UK and Japan. Based on markup adjustments by domestic importers and foreign exporters, we show that there is a difference in pass- through to different prices in the short-run. Our empirical approach is based on a “triangular” system of equations. This system clearly identifies different pass-through channels and estimates a number of effects: direct short and long-run responses of prices of domestically-produced (consumer and producer) goods and prices of imports to exchange rate changes; the “carry-over” effect from import prices to domestic prices, as well as the “carry-over” effect from import prices to producer prices. The triangularity of the system is based on the observation that one expects no immediate short-run feedback from prices of domestically-produced (consumer and producer) goods to import prices and no feedback from consumer prices to producer prices. Using a comprehensive econometric methodology that starts by investigating probabilistic properties of the data, we build a vector-autoregressive-with-exogenous-variables (VARX) model. We test the proposed “no-feedback” effect and obtain results that support the triangular arrangement for nineteen out of thirty-four industries in the sample. We also find that there is considerable similarity across industries regarding the no past-feedback. For example, in USA and UK, results from industrial prices of beverages and tobacco, chemicals and machinery and transport equipments show that there is no lagged feedback from consumer price indexes (CPI) or domestically-produced goods prices to current prices of imported goods. Similarly, results from subcategories of manufactures industry from both of these countries support this claim. According to Goldberg and Knetter (1997), the markup adjustment following an exchange rate change generally occurs within a year. Thus, in the short-run, there may be a change in relative prices. We find evidence of relative price changes in industry-level import and consumer prices in the short-run. Out of thirty-four industries from three countries (USA, UK and Japan) in the sample, relative prices change for forty-two percent of the cases.3 In the second part of the study, we analyze industry-level export price movements after an unexpected exchange rate shock and check whether there is evidence to indicate export price stickiness in terms of either the sellers’ or buyers’ currency. Though the exchange rate pass-through effect is primarily associated with import and consumer prices, it is useful to consider the pricing practices of exporters whose products enter as imports to the destination country.4 Accordingly, we explore movements in export prices by first investigating the correlation between industry-level relative export prices and the nominal exchange rate (as in Obstfeld, 2002). The results show high correlations in most 3 In a recent study Shambaugh (2003) relies on the methodology of Blanchard and Quah (1989) to generate long-run shocks in the exchange rate and prices. He finds that import prices are set in the producers’ currency with lower CPI pass-through reflecting margin changes in the supply chain. This provides some support to the idea of pass-through separation as proposed by Obstfeld (2002). However, unlike ours, Shambaugh’s study does not focus on a detailed industry-level analysis. It also does not consider changes in export prices in response to changes in exchange rates ignoring a possibly important channel. 4 In this we follow Knetter (1989) (who studies exchange-rate pass-through and pricing-to-market by looking at the prices charged by US and German exporters to importers of a given country) and the literature that followed this seminal contribution. 4 of the industry-level export prices, providing some support for invoicing in terms of producers’ or sellers’ or exporters’ currency. Second, we use the changes in the logged ratio of export prices and the trade-weighted exchange rate to further corroborate our findings from the first part of the export price analysis. The results from both parts of our study indicate that exchange rate pass-through for industries in the USA ranges from thirty percent to fifty percent, a finding that is consistent with those of Goldberg and Knetter (1997) and Campa and Goldberg (2002). Within manufacturing and food industries, Japanese food, metal and textiles industries, UK’s iron and steel industry and US rubber and furniture industries support earlier results by Campa and Goldberg (2002) in finding negligible pass-through in these industrial categories. However, our results for the US food and machinery and transport equipment industry as well as UK tobacco and non-ferrous metal industry show evidence of a higher extent of pass-through. Results from the Japanese wood industry and the UK pulp and metal ores industries show greater evidence of producer currency pricing, again supporting Campa and Goldberg (2002) findings that lead to the rejection of local currency pricing for these two industry categories.5 The rest of the paper is organized as follows. In section 2, we discuss a simple theoretical model and our empirical methodology. In section 3 we present the dataset. In section 4 we present and discuss the results from our empirical analysis. Some concluding remarks are given in section 5. All empirical results are reported in tabular form in the first appendix, data details and information are given in the second appendix. We present some figures in the third appendix. 2. Empirical analysis from industry-level import and export prices 2.1 A simple model for import prices To analyze different channels of exchange rate pass-through (as proposed by Obstfeld (2001, 2002)) in a simple framework, we focus on the following four prices: 1. Prices of imported goods at the point of entry denominated in local currency (denoted by imp) 2. Domestic prices of imported goods as paid by the end-users (denoted by dig) 3. Domestic prices of nontradable goods (denoted by ddg) 4. Domestic prices, as measured by the CPI (the final price consumers pay). The dig and ddg are the constituent parts of CPI. To investigate changes in CPI, we look for changes in either ddg or dig or in both. 5 Campa and Goldberg (2002) show that for OECD countries, the Manufacturing and Food import prices reject both PCP and LCP in the short-run. They present, however, evidence on long-run pass-through for Non-manufacturing and Raw Materials import prices. Goldberg and Knetter (1997) report that the median pass-through for OECD Manufacturing import prices hovers around fifty percent over a year period (taken to be the long-run). 5 Under the local currency pricing (LCP) mechanism, neither the dig nor the CPI may change after an exchange rate change. With pricing to market (PTM), there are proportionate and opposite adjustments in the markups charged by sellers after exchange rate changes, resulting in zero or low pass-through. However, with producer currency pricing (PCP), both prices (dig and CPI) respond proportionately to changes in the exchange rate. Different channels of exchange rate pass-through can now be hypothesized in the following way: (1) With an unexpected change in the exchange rate there is pass-through to imp. There may be a potential markup adjustment by foreign exporters at this point (in accordance with the PTM argument proposed by Krugman (1987)). The end result will be either higher or lower pass-through to imp. This is the first and direct pass-through channel at the point of entry. If the pass-through to imp is fast and of greater magnitude as compared to pass-through to dig and CPI, then there would be changes in relative prices. With the transaction cost of the imported good added to the domestic price of the imported good, there may be a difference between imp and dig. In addition, domestic importers may charge a markup on the imported good (the price of which is denoted by dig). With different domestic markup adjustment in dig vis-à-vis markup adjustment in imp (as described in (1) above), there may be a relative price difference between dig and imp. Whether CPI changes or not depends on the domestic markup adjustment reflected in dig. If there is domestic wage rigidity (or price rigidity, as proposed in a number of models, see the related discussion in Bergin (2003) and the references cited)6, ddg will not change very fast in response to a domestic monetary shock that affects the exchange rate. The exchange rate change will only affect imp.7 (2) If there is no markup adjustment by foreign exporters at the entry point, the imp will change proportionately with a change in the exchange rate. If the ddg responds sluggishly, there would be a difference between imp and ddg. To maintain the domestic sale of imported goods, domestic importers may adjust the domestic markup in proportion to the imp hike. This will be reflected in the dig. This is the mechanism emphasized by LCP or PTM. Thus the ddg and dig remain at the level before the unanticipated shock. As a result, CPI will not change. This is the second and indirect channel of exchange rate pass-through to dig and CPI. After all the adjustments have taken place as above, there may still remain a difference between imp and dig as well as CPI. This may trigger expenditure-switching as argued in Obstfeld (2001, 2002) and Obstfeld and Rogoff (2000). As this discussion suggests, the speed and extent of markup adjustment drives all the results of high and low pass-through to imp and dig as well as CPI. 6 Domestic wage rigidity leads to domestic price sluggishness as there may be wage contracts signed before the unexpected monetary shock. 7 This can happen if invoicing contract is signed a period ahead. 6 The second and indirect channel of pass-through to CPI, as described earlier, calls for looking at the extent of domestic markup adjustments. Since we do not have data for the domestic cost of production at the industry-level at monthly frequency, we devise an indirect way of looking at this effect. Changes in CPI show changes in either ddg or dig or both. The dig reflects adjustment at the second stage when the imported good enters the domestic distribution chain. So the exchange rate change at the entry stage (reflected in imp) can indirectly capture the change in dig because the second stage adjustment happens only if there is a change in markups at the first stage. Therefore, we take the exchange rate coefficient as a proxy to the adjustment in dig based on markups. In addition, if there is any change in the PPI or CPI due to a change in the dig (assuming that there is tradable component in production), it is possible to separate out that effect. We denote this potential effect as the “carry-over effect”. For some industries, the data allows us to look for this effect. The above discussion calls for suitably disaggregated data. We do not have access to industry-level domestic prices of the imported goods, yet the data we have include (1) consumer prices (CPI) (which reflect prices of final products or final consumer goods), (2) import prices (IMP) for final goods, intermediate goods as well as crude materials and (3) prices for domestic producers (PPI), which can be categorized in terms of final goods, intermediate goods and crude materials. With this information, we are able to distinguish in terms of end uses, and this allows us a better understanding of the channels of pass- through under investigation. In our setup, we take the exchange rate to be exogenously determined and focus on the effect of unexpected changes in the exchange rate.8 Within this framework, we build three different reduced-form systems to capture the effect of changes in the exchange rate on import prices, producer prices, and consumer prices. The setup we have is designed to analyze various channels that transmit a change in the exchange rate to the prices of final goods as measured by the CPI. The CPI reflects tradable (denoted by dig) and nontradable goods (denoted by ddg) prices. With a higher proportion of tradables, any change in the exchange rate that affects their prices would lead to a significant change in CPI. Assuming PCP, the effect will be full. On the other hand, under LCP/PTM mechanism, the extent of the CPI change will be closer to zero. System I: Here we focus on the prices of imported final goods. At the point of entry, retailers pay the price, for imports which possibly include a markup charged by the exporters. Thus retailers pay the price ri P * 0 i m ≥ ( ) * 1 ri i i P m E = + *P (1) 8 We do so since we are only interested in the transmission of exchange rate shock and not where and how the shocks are generated. See Adolfson (2001) for a discussion regarding the treatment of the exchange rate as exogenous and omitting “controls” for disaggregated industry-level work. 7 where denotes the exchange rate and is the foreign currency price of the i-th good imported. Depending on whether changes or not, a change in the exchange rate may or may not be reflected in . As pointed out earlier, this is the first channel of short-run exchange rate pass-through. Here foreign exporters can price discriminate among destination markets. E * iP * i m ri P In the home market, retailers have to bear the transport and distribution costs of final goods before selling those to the consumers. Denoting these costs for the i-th good as , the total marginal cost that the retailers face is: riΩ ( ) 1 ri ri ri MC P = +Ω (2) Letting the domestic markup charged by the retailer to be iγ , we have the final price of the i-th good as: ( ) 1 ci i ri P γ = + MC (3) Given (2) and (1), this yields ( )( )( ) * 1 1 1 ci i ri i i P γ = + +Ω + * m EP (4) Equation (4) shows the link between and import prices as mediated by the various markups clearly. An import price increase driven by an exchange rate depreciation (assuming no change in the foreign mark-up) may or may not lead to a proportional increase in consumer prices. This is the second channel of pass-through. The response of domestic prices to changes in the price of imports can be explained by changes in the existing domestic markup, E ci P iγ in the way we have described before. The foreign and domestic markup adjustments may dampen the transmission of the changes in the exchange rate to import prices and domestic prices. These adjustments can be a reflection of sluggish nominal wage adjustments in the foreign country as well as in the home country. System II: Here, we concentrate on the case where final consumption goods are produced domestically using tradable and possibly nontradable intermediate inputs. We allow for the possibility that these intermediate inputs themselves have some imported components. Therefore, any change in producer prices of final consumption goods can be explained indirectly by changes in the exchange rate through changes in the prices of domestically produced intermediate inputs as well as changes in the prices of imported intermediate inputs. To capture the underlying structure of System II in terms of the price adjustments, we look at the following expression. In the case where domestic producers use imported 8 intermediate goods, the price ( ) they pay at the point of entry once the markup charged by the exporters’ is taken into account, will be given by: ni P * 0 ni m ≥ ( ) * 1 ni ni i P m E = + *P (5) In addition to the transaction cost, niΩ , associated with these imported inputs, the producer also has to cover the labor costs. Let be the wage rate and let denote the unit labor requirement in the production of i-th good. The marginal cost is: i w il ( ) 1 ni ni ni i i MC P w = +Ω + l (6) If markup charged by the domestic producer is iμ , we have the producer price of the i-th good as: ( ) 1 pi i ni P μ = + MC (7) Using (6) this yields ( ) ( )( ) * * 1 1 1 pi i i i ni ni i P w l m μ EP ⎡ ⎤ = + + +Ω + ⎣ ⎦ (8) We have a similar interpretation in terms of markup adjustments and producer prices change as before. System III: Here we focus on the prices of producer goods that are produced using imported intermediates. As before, denoting the transaction cost for the i-th good by , foreign markup by and the domestic markup charged by the producer by piΩ * pi m iν , the producer price of the i-th good will be: ( )( )( ) * 1 1 1 pi i pi pi i P ν = + +Ω + * m EP (9) The various channels through which change in exchange rate is passed through in this system are analogous to the ones described above. What differentiates the three systems is that the pass-through is towards consumer prices in system I and towards producer prices in systems II (final consumption goods) and III (producer goods). With the above three systems approach, we can identify two different channels of exchange rate pass-through. One channel is at the point of entry and captured by fluctuations in import prices at the point of entry. The other channel is one which transmits changes in the exchange rate to domestic prices (can be consumer or producer prices) through markup adjustments by domestic and foreign agents. To capture these effects empirically, we need properly matched industry-level import prices, consumer 9 prices and producer prices. In addition, as we try to determine the unidirectional causality running from exchange rate fluctuations to different prices, we need to devise a framework that reflects this directionality. Since we propose that different pass-through channels can be differentiated in the short-run, data at the monthly frequency will be an appropriate choice to carry out our empirical analysis. 2.2 Empirical framework: import prices The conceptual framework expressed above needs to be converted into an econometric framework, which then can be used both for testing the implied triangularity of the system we propose and for estimating short and long-run effects. Our analysis is conducted in percentage changes of prices. Before estimating the three systems described above, we performed standard unit root and cointegration tests for the price levels (in logarithmic forms), but found no strong evidence of cointegration.9 In what follows we describe the empirical implementation of system I. Consider a ( vector with CPI inflation, PPI inflation and percentage change in import prices, say ) ) 3 1 × ( 1 2 , , t t t t y x x ′ = z , and redefine the percentage change in the exchange rate as 3 t t x w ≡ .10 We assume that can be adequately modeled by a vector autoregression with an exogenous input variable (VARX) as: tz 1 1 p q t i t i j t j i j w − − = = t = + ∑ ∑ z Π z β u + ) (10) where, are ( parameter matrices and { } { } 1 p i i= Π 3 3 × 1 q j= β are ( ) 3 1 × parameter vectors. show the autoregressive or lagged values of CPI inflation, PPI inflation and the percentage change in import prices. Similarly, t i− z t j w − contain the lagged values of exchange rate changes in percentage terms. The error vector is assumed to be multivariate white noise with variance-covariance matrix . The model in the above equation will be our broadest, unrestricted model (U-model). tu Σ 11 The implied triangularity of the conceptual model of the previous section can now be tested using this model. Consider the restrictions implied by the following null hypothesis and corresponding to our first restricted model (R1-model): { } 0 : = 0 | for and , = 1,2,3 i ab H a b a b π > i∀ (11) 9 These unit root and cointegration test results are not presented here but are available on request from the corresponding author. 10 All variables are taken as deviations from their respective sample means. 11 The U-model was estimated using conditional least squares with the orders chosen by the Schwarz (BIC) criterion. 10 where, i ab π is the coefficient of . These restrictions imply absence of feedback from CPI inflation to PPI inflation and from CPI and PPI inflation to growth of import prices; they are immediately testable using a Wald-type test applied to the U-model. ( ,a b) iΠ If the above null hypothesis is rejected, we proceed by eliminating the insignificant coefficients from the U-model and by re-estimating the remaining parameters by seemingly unrelated regression (SUR). This is our second restricted model (R2-model), which we then compare to the U-model using a likelihood ratio (LR) test. If the R2- model is rejected in favor of the U-model we use the estimates from the U-model to compute the long-run effects; if the R2-model is not rejected we use its estimates to calculate long-run effects. Similarly, if the null hypothesis of triangularity is not rejected, we proceed by eliminating the insignificant coefficients from the R1-model and re- estimate the remaining parameters using SUR. This constitutes our third restricted model (R3-model), which we then compare to the R1-model using a LR test. Depending on whether the R3-model is rejected or not we use the estimates from either the R1-model or the R3-model to calculate the long-run effects. To illustrate the computation of the long-run effects, consider the U-model and re-write it using lag operator notation as: ( ) ( ) t t L L w t = + Π z β u (12) where, and . When the system is in long-run equilibrium we expect that the variables do not deviate substantially from some fixed values, say , and ( ) 3 1 p i i i L L = = −∑ Π I Π *z *w ( ) 1 q j j j L = =∑ β β L [ ] * = Ε 0 = t u u . Therefore, we have the representation: ( ) ( ) * 1 1 w = Π z β * (13) from which all long-run effects can be easily computed by summing the estimates of the ’s and the ’s. For example, the long-run effects of the exchange rate growth on CPI inflation, PPI inflation and growth of import prices are given by the estimate of the vector iΠ iβ ( ) Π ( ) 1 * * 1 1 w − ∂ ∂ = ⎡ ⎤ ⎣ ⎦ β z . Standard errors for the long-run effects were obtained using the Delta method. In the first appendix we report the total short and long-run effects, as well as the carry-over effect. 2.3 Empirical framework: export prices Let t p denote the price of export of the home country, * t p denote the export price of the foreign country and E denote the nominal exchange rate (in units of home’s currency per unit of foreign currency). For the first part of the analysis, we construct home’s relative export prices (denoted by REP ) vis-à-vis foreign’s export prices in the following way: 11 t t p p E REP * . = (14) We then look at the correlation of monthly changes in logarithm of relative export prices and the nominal exchange rate. If export prices remain fixed in the currencies of the originating countries, the ratio of two prices on the right-hand side of (14) will not vary. Thus, relative export prices REP and the exchange rate E will be highly correlated. This would lead to high pass-through and count as evidence against the LCP mechanism. For the second part of the analysis, the dependent variable is the monthly growth in the ratio of export prices . The explanatory variable is the monthly growth of the trade-weighted exchange rate ( ) 1 t y L Δ = − 3 t t y x . The model we consider is a regression in monthly percentage changes, namely: ( ) ( ) ( ) 3 t t L y c L x z π δ Δ = + + ν t (15) where, ( ) 1 1 r i i i L L π π = = −∑ and ( ) 0 1 s i i i L L δ δ δ = = +∑ are polynomials in the lag operator and where the term captures the regression error dynamics of the equation and depends on the auxiliary parameter vector . For example, if the equation includes the first lag of ( t z ν t ) ν y Δ 3 t and x and the regression error follows a seasonal autoregression of orders one and twelve then ( ) 1 1 L L π π = − , ( ) 0 1 L L δ δ δ = + and , with ( ) t z = ν t u ( )( ) 12 1 12 1 1 t t L L u φ φ ε − − = , ( ) 2 ~ 0, iid t ε ε σ and ( 2 1 12 , , ε ) φ φ σ ′ = ν . If no lagged dynamics of t y Δ and 3 t x are explicitly included then * 0 δ δ = gives us the long-run “equilibrium” effect of a change in monthly relative export prices from a change in the monthly trade- weighted exchange rate. If lagged terms are present, then the long-run effect is computed as ( ) ( ) * 1 δ δ = 1 π , i.e., as the equilibrium solution of the dynamic part of the model. As above, standard errors for the long-run effects are obtained using the Delta method. 3. Data 3.1 Import prices Our choice of data is guided by two considerations: (1) we need to have industry-level price data for imported goods, as well as matching industry-level price data for intermediate goods (if available, as pointed out in System II) and matching price data for final goods (can be for producers goods, as pointed out in System III or for final consumers goods, as pointed out in System I); (2) we concentrate on three of the largest active economies, USA, UK and Japan, that are involved in heavy trading across three different regions. It is important that these countries do have the required data. We have access to perfectly matching industry-level price data for fourteen US industries, thirteen UK industries and seven Japanese industries (and not for all SITC industries). The data frequency we use is guided by the question we try to answer in this study: whether there 12 is evidence of short-run pass through at the SITC level industrial manufacturing goods data. The data for the US comes from the Bureau of Labor Statistics web site. Import prices are r the UK, the data for SITC level import prices and producer prices (PPI) come from r Japan, the wholesale price index (WPI)15, domestic producer price index (PPI), .2 Export prices or the USA, SITC level data for export prices are again obtained from the web site of or the UK, we take SITC level export prices, including prices for some subcategories, panese export prices are taken from the Bank of Japan web site. The data correspond to six industries from January 1983 to December 2002. As in the case of the US, we have also converted the base year from 2000 to 1995. taken using the SITC classification with the corresponding producer prices (PPI) being taken both from industry-level and commodity-level classifications. We use the CPI for urban consumers.12 For import prices from Food and Beverages, Mineral Fuels and Lubricants and Textiles industries, we use the corresponding CPI for these industries. In every case of our data for the USA, the end point is December, 2002. However, the starting point varies across industries.13 The data for trade-weighted exchange rate for the USA come from the Federal Reserve Bank of St. Louise web site. For import prices, the base year is 2000 and for consumer and producer prices the base year is between 1982 to1984; for conformity and comparability we change the base year to 1995. The analysis is carried out for fourteen industries in US, including broad as well as subcategories of SITC level industries.14 Fo the National Statistics Online. The trade-weighted exchange rate data is available from the Bank of England’s web site. Fo import prices (IMP) and trade-weighted exchange rates are obtained from the Bank of Japan web site. Our sample ranges from January, 1971 to December, 2002 and we have seven broad SITC industries for our study. 3 F the Bureau of Labor Statistics. We examine sixteen industries including subcategories of broad SITC level industries. Most of the monthly data start from January, 1993 and we include observations until December 2002 in our analysis. The reported base year for all prices is 2000 but for conformity we changed the base year to 1995. F for sixteen industries. The reported base in this case is 1995 = 100. The sample span is from January 1983 to December 2002. Ja 12 We use CPI as a measure of retail prices for these industries in order to match the corresponding import price data for these industries. 13 This is due to the monthly industry-level data availability for all the matching imported goods, intermediate goods as well as final goods. 14 Details about the import data are provided in tabular form in Appendix 2. 15 We use this as a measure of retail prices as described in our System I earlier. Now this index is replaced by the corporate goods index in the source web site. 13 In examining relative price ratios across countries we require exact industry matching. However, this is not possible for all industries in all three countries. As a result, we have a smaller number of export price ratios, especially when Japan is involved. For the USA .1 Summary of results involving import prices he empirical evidence for the extent of pass-through transmission and relative price We cannot offer conclusive support ither for absolute LCP/PTM (where the extent of pass-through is precisely nil or zero) or o provide some additional, ex-ante, check for the presence of lagged feedback, we s (Pesaran and Shin (1998)) involving the ndogenous variables in our systems (the commodity prices). Some figures with impulse and Japan and UK and Japan country pairs we only have five relative prices, whereas, for the USA and UK pair we have fifteen relative export prices. 4. Empirical results 4 T changes across countries and industries is mixed. e for absolute PCP (where the pass-through extent is exactly one). For the present analysis, we consider a cut-off value of twenty percent evidence in support of PCP. This cutoff value is guided by two considerations. First, it is a plausible value given the average monthly change across industries and countries. Second, a cutoff of twenty percent is consistent for a short-run analysis: Campa and Goldberg (1997) and Obstfeld (2002), among others, suggest that the pass-through extent hovers around fifty percent in the long-run. Therefore, from a short-run perspective, choosing a less than fifty percent cutoff appears economically plausible.16 Overall, about sixty-four percent of the sampled industries appear to lend support to PCP. We choose less than fifteen percent exchange rate pass-through effect for the industries that lend support to LCP/PTM. As a result, almost sixty-two percent of the sampled industries show LCP/PTM evidence. In addition, fifty-three percent of all industries support the no feedback assumption (“triangularity” of the systems), indicating that there is no lagged effect of previous producer or consumer prices on the current import prices. The results from our systems estimation show that, in the short-run, about eighty-three percent of the industries show evidence in favor of PCP to import prices and sixty-four percent provide support for LCP in the domestic prices. Overall, there are high and significant relative prices changes in forty-two percent of the sampled cases. For the long-run results, there is empirical support for relative price changes for forty-two percent of the sampled industries. Taking all industries for all countries, tobacco prices in UK and Machinery and equipment prices in Japan show the highest and lowest extent of short-run and long-run pass-through and relative price change respectively. We find that there is no significant extent of pass-through to the US industries, a result that conforms to earlier studies for the US. 4.1.1 Results from impulse response analysis T performed generalized impulse response analysi e responses are given in Appendix 3. Our results show considerable support for the proposed no-feedback hypothesis for a number of industries in three countries. In the UK, 16 Replication of a part of our analysis with a fifty percent cutoff yields results mostly consistent with LCP evidence and not PCP. It is apparent that the choice of the cutoff value is data and country specific. 14 nine out of thirteen industries show that an innovation in import prices will generate responses in producer prices only, implying that past producer prices and wholesale prices have no significant impact on import prices. For the US, the producer price data do not reflect imported components prices, so the innovation shocks in imports have no impact on producer prices. A majority of US industrial prices (measured by producer price index), except for the Chemicals industry, also support ex-ante “triangularity”. But, consumer prices for three industries in the US, CPI for Food and Beverages, CPI for Mineral fuel and CPI for Apparels, show evidence against the no-feedback argument, as they affect the import prices contemporaneously. For Japan, the impulse responses do not support absence of feedback for all industries. The figures in Appendix 3 show impulse response for three representative industries, Chemical in Japan, Metal-working machinery in the US and Organic chemical in the UK. The first one does not support “triangularity” from the unrestricted VAR while the later two provide evidence of no- feedback, thus supporting “triangularity” from the unrestricted VAR. 4.1.2 Results from System I: industries in Japan and USA Our estimation results are given in Table 1 while the specification testing results are dustries in Japan and three dustries in USA, we reject the hypothesis of no-feedback from CPI inflation to PPI given in Table 2. Looking at Table 2 first, for all the seven in in inflation as well as from CPI and PPI inflation to growth of import prices. As a result, estimation for these ten industries is carried using the R2-model. Short-run results give us the direct effects of exchange rate depreciation on import prices (denoted by SR IMP ), the indirect effect on WPI for Japan or on CPI for USA, (denoted by SR WPI ) and the carry- over effect from import prices to either wholesale prices or CPI (denoted by C . SR OE ) 17 In the short-run, for three out of ten industries, there is no evidence of carry-over effect. For the rest of the industries, however, there are significant carry-over effec idenc ts. Ev e f higher level of relative price changes is supported by at least sixty percent of the rice change is found for these latter industries. Negative coefficients in the tables can be o tabulated cases, with Mineral fuel and lubricants industry in USA providing the highest magnitude of relative price change. As described earlier, this happens due to PCP for import prices in all of these industries and LCP for wholesale prices or CPI for fifty percent of the results corroborated by low carry-over effect in forty percent of the cases. In the long-run we find evidence in favor of PCP for sixty percent of industries while forty percent of them support LCP towards domestic prices. A higher level of relative p explained by the higher extent of export price declines compared to exchange rate depreciation. We have verified this from the data on export prices and these results are available on request. 17 Therefore, for these two countries, results under and SR WPI LR WPI denote short-run and long-run calculations from retail industry-level prices interaction with trade-weighted exchange rate and import prices. 15 4.1.3 Results from System II: industries in UK Tables 3 and 4 report the estimation and specification testing results from system II nvolving intermediate imported inputs). Due to data availability we have done the d short-run coefficients are calculated om the unrestricted model (U-model) for Chemicals and Organic chemicals industry n Tables 5 to 8. This is two equation system as described earlier. The first equation in the system looks at the ort-run (denoted by (i analysis for the UK. In this case, the long-run an fr and from the R2-model for Machinery and Electrical equipment industry. Results from both the industries generate low level of exchange rate pass-through to domestic producer prices for intermediate goods and low carry-over effect in the short as well as in the long run. These findings, therefore, suggest a potentially high relative price change conditional on the extent of PCP for imported goods prices. However, both industries show negative coefficients for the imported goods prices. As a result, the magnitudes of export price declines in both of these industries are much higher than the extent of exchange rate depreciation. Therefore, we do not get the relative price change. 4.1.4 Results from System III: industries in UK and USA Estimation and model testing results for system III are presented i a exchange rate pass-through effect to import prices in the sh SR IMP ange to ). he second equation reflects the direct transmission from import prices ch ” is not rejected. Therefore, the short and long- n coefficients of pass-through for these industries are calculated from R3-model, except aining industries, owever, report significant carry-over effect in the short-run. We get the highest carry- T producer prices in the short-run, measured in terms of SR COE and indirect transmission from import-competing goods prices change in the short-run, captured by the markup adjustment, and measured by SR PPI . Tables 5 and 6 summarize UK findings and USA results18 are reported in Tables 7 and 8. Looking at Table 6 in four out of thirteen industries, the R2-model is used to calculate both short and long-run effects. The remaining industries support the R3-model, showing that the null hypothesis of “triangularity ru for Organic chemicals’ industry as the LR test supports the R1-model. From the estimation results at Table 5, evidence from three industrial prices (Pulp, Wood and Medicinal products) show that there is no carry-over effect from import prices to producer prices both in the short and long-run. Seventy percent of rem h over effect in Iron and steel industry with a magnitude of fifty seven percent. In the short- run, forty-six percent of the sampled industries support high exchange rate pass-through to import prices. Tobacco reports the highest (more than one hundred percent) and Iron and steel shows the lowest (close to fifteen percent) pass-through effects in the short-run. As noted earlier, negative coefficients can be explained through the greater increase in export prices of these industrial products than the extent of exchange rate depreciation. As compared to pass-through towards import prices, the pass-through magnitude to 18 We also compared our results using industry-specific exchange rates (provided by Goldberg (2004)) and trade-weighted exchange rates for USA, but found no significant differences in the outcomes. Therefore, we do not report these results in the text. 16 producer prices is less in four out of thirteen industries in the short-run. These four industries, Tobacco, Metal ores, Plastics and Iron and steel, therefore, support the higher extent of pass-through in the short-run, though the effect declines for Iron and steel and Plastics’ prices because of higher magnitude of carry-over effect. The results are similar for the long-run calculations also. In all, we get mixed support for high relative price change. In case of the USA, “triangularity” is accepted from the R1-model and then R3-model is used for estimation in ten out of eleven industries in our sample (see Table 8 for reference). For Chemicals, however, the R2-model is taken as the null hypothesis of triangularity” is rejected from the first step. Long and short-run coefficients are producer prices for Fruit, Inorganic hemicals and Electrical machinery industry either. This is perfectly consistent with the ompetitiveness for Japanese, US and UK industry-level products. Looking at the petitiveness of Japan vis-à-vis USA (denoted y Japan-US), General machinery and equipment price generates the highest and “ estimated from R2-model for Chemicals and from R3-model for all other industries, except for the Metal-working machinery industry. The estimation results from Table 7 show no carry-over effects in case of Meat, Rubber and Non-metallic minerals industries both in the short and long-run. There are no pass- through effects (both in short-run and long-run) to c data for US PPI, as the imported components are excluded in the US PPI calculation. In our analysis, for forty-six percent of cases in the short-run, we get positive pass-through towards import prices with the highest magnitude of fifty-three percentage points (for Fruit). The producer price pass-through coefficients are very low in the short-run for almost thirty-six percent of sampled industries with the extent varying between two to nine percent. To generate rapid relative price change, a higher level of pass-through to import prices is needed. This effect is present for almost forty-six percentage of sampled industries with the effect significant and most prominent in the Metal-working machinery industry (thirty percent). Relatively low pass-through to Furnitures, which is a part of broad SITC category, Miscellaneous manufactured articles, provide some support to Yang (1997)’s earlier findings for manufacturing industry in USA. Across industry comparisons from the table show considerable variations between manufacturing industries and food and beverages industries. These results are in line with Knetter (1993), which shows substantial variations in pass-through coefficients across industries. 4.2 Results from analysis of export prices In Table 9 we compare the correlations of exchange rates and relative export c correlation results on the relative export com b Chemicals the lowest correlation (see Figures 4 and 5 after Appendix 2). The results also point to the fact that there is considerable heterogeneity regarding currency invoicing across industries. These findings are consistent with earlier conclusions from Klitgaard (1999), which reports that for electrical machinery and transportation equipment industry, there is considerable pass-through effect. For these two industries, the effect of yen depreciation pass-through to the foreign consumers is higher, which indirectly support the argument of invoicing in domestic currency. For the five industry comparison for these 17 two countries, the average correlation is 0.76, which is almost equal to the overall aggregate finding for the Japan and USA case in Obstfeld and Rogoff (2000). Comparing US exports vis-à-vis UK exports (denoted by US-UK), we find considerable differences in industry-level prices. For instance, results from Meat and Fruit industries (part of SITC category, Food and live animals) and Textile fibers, Crude materials except el and Metalliferous ores industries (categories of SITC classification, Crude materials, nt industry (refer to Figures 8 and 9). For this two-country result, the dustry average is 0.67, which is comparable with the aggregate case of Obstfeld and pirical support for the theoretical argument on the bsence of zero or low extent of exchange rate pass-through towards relative export fu inedible, except fuels) show low correlations. This may support LCP/PTM type pricing, or invoicing in destination currencies. However, for the remaining industries, the correlations are high, supporting the fact that export prices may be largely invoiced in the exporters’ currency. Higher correlations support the fact that, across industry-level export prices, the PCP type of pricing seems prevalent for these two countries. There are significant variations across industries with Miscellaneous Manufactures and Road vehicles showing the highest extent of PCP. We have presented the figures for Miscellaneous Manufactures and Road vehicles industry prices in Figures 6 and 7 in Appendix 3. For the relative export competitiveness between Japan and the UK (denoted by Japan- UK), the lowest correlation is for Road vehicles and the highest is for General machinery and equipme in Rogoff (2000) finding.19 Results across industries identify segments where the PCP is prevalent. Except for Japan and the UK, the Miscellaneous manufactured articles’ and Road vehicles industry show some evidence of PCP. For Japan and USA, the correlation is the highest for General machinery and equipment, which is again consistent with earlier findings of Klitgaard (1999). Estimation results from ratios of export prices are presented in Table 10. The last three columns of the table show coefficient estimates associated with trade-weighted exchange rates. Negative coefficients provide em a prices. Out of the total seventeen industries, we get the desired effect in fifteen industries, thus having almost ninety percent of cases provide some evidence of apparent pass- through. Within the USA and UK comparison, the pass-through extent is the largest in Textile fiber industry (forty-seven percent) and lowest in Road vehicles industry (ten percent). From this two-country analysis, in sixty percent of industries, there is high pass- through towards export prices. Comparison between US-Japan export prices show pass- through for eighty percent of industries with Road vehicles industry experiencing the largest amount (almost sixty-seven percent) of pass-through among all the industries in the sample. The relative price comparison for UK and Japan reports the highest level of pass-through in Miscellaneous Manufacturing industry (nearly thirty-five percent). Eighty percent of the sampled industries here provide significant pass-through. These results, therefore, support Keynesian arguments and establish the validity of PCP mechanism. Our results, overall, are quite similar to that of Obstfeld and Rogoff (2000) and Obstfeld (2002). Within cross-country estimates, Chemicals industry’s relative 19 They find a correlation of 0.76 for the aggregate export price index. 18 export prices pass-through effects are relatively comparable, as the estimates range from eighteen percentages to twenty-nine percentage points. It shows that producer currency pricing type mechanism is prevalent within these countries as far as setting the export price of this industry. Miscellaneous manufactures’ industry estimates also point out the same with the exception of the USA and UK estimate. Except for the US-Japan estimate, the Road vehicles’ industry prices from other bilateral country comparisons report the lowest evidence of exchange rate pass-through among all the industries and countries in the analysis. This may be due to higher extent of local currency pricing that is being practiced in the US-UK automobile markets. 4.3 Results from both import and export prices analysis Our results indicate that the producer currency pricing mechanism is supported by a large umber of industrial categories. For the three countries in our study, Japan, USA and UK, as well as Miscellaneous anufactures industry and sub-categories of miscellaneous manufactures industry show ugh at the industry-level. We do so by cusing on the extent of pass-through in both the short and the long-run using industry- mport prices, domestic prices (which includes consumer and roducer prices) and export prices simultaneously. We identify different channels of oldberg (2002). Within the manufacturing nd food sector, Japanese food, metal and textiles industries, UK’s iron and steel industry n Chemicals and sub-categories of chemicals industry M evidence of PCP. In the USA and UK, Beverages and Tobacco industry, Organic Chemicals industry, sub-categories of General Machinery and Equipment industry and Non-ferrous Metals industry also lend partial support for the PCP. In Japan, the Textile Fabrics industry yields evidence in support of PCP as well. This is in line with Obstfeld (2002) that also compares co-movements of the USA and Canadian SITC level export prices and import prices and finds similar results. 5. Concluding remarks In this paper we investigate the extent of pass-thro fo level monthly data for i p pass-through involving prices of both imported and domestically-produced goods. One channel of exchange rate pass-through goes through prices of imported goods at the point of entry. The other channel involves the prices of domestically-produced goods through adjustments in foreign and domestic markups. Our empirical results indicate that exchange rate pass-through for industries in the USA ranges from thirty percent to fifty percent, a finding that is consistent with those of Goldberg and Knetter (1997) and Campa and G a and US rubber and furniture industries support earlier results by Campa and Goldberg (2002) in finding negligible pass-through. However, our results for the US food and machinery and transport equipment industry as well as UK tobacco and non-ferrous metal industry show evidence of a higher extent of pass-through. Results from the Japanese wood industry and the UK pulp and metal ores industries show greater evidence of PCP, again supporting Campa and Goldberg (2002) findings that lead to the rejection of LCP for these two industry categories. 19 References [1] Adolfson, Malin, 2001. “Export Price Responses to Exogenous Exchange Rate Movements”, Economics Letters, vol. 71, pp. 91-96. ] Ball, Laurence, “Policy Rules for Open Economies”, Working Paper of the National ] Bergin, Paul R., 2003. “Putting the ‘New Open Economy Macroeconomics’ to a ] Blanchard, O. and D. Quah, 1989. “The Dynamic Effects of Aggregate Demand and ] Campa, J. M. and L. S. Goldberg, “Exchange Rate Pass-Through into Import Prices: ] Dornbusch, Rudiger, 1987. “Exchange Rates and Prices”, American Economic ] Engel, Charles, 1993. “Real Exchange Rate and Relative Prices: An Empirical ] Engel, Charles, 1999. “Accounting for Real Exchange Rate Changes”, Journal of 0] Engel, Charles, 2000. “Local-Currency Pricing and the Choice of Exchange-Rate 1] Engel, Charles, “The Responsiveness of Consumer Prices to Exchange Rates and au of Economic Research, No. 8725, January, 2002. 3] Engel, C. and J. H. Rogers, 2001. “Deviations From the Purchasing Power Parity: [2 Bureau of Economic Research, No. 6760, October, 1998. [3 Test”, Journal of International Economics, vol. 60, pp. 3-34. [4 Supply Disturbances”, American Economic Review, vol. 79, pp. 655-673. [5 A Macro or Micro Phenomenon”, Working Paper of The National Bureau of Economic Research, No. 8934, May, 2002. [6] Devereux, Michael B., 1997. “Real Exchange Rates and Macroeconomics: Evidence and Theory”, Canadian Journal of Economics, vol. 30, pp. 773-808. [7 Review, vol. 77, pp. 93-106. [8 Investigation”, Journal of Monetary Economics, vol. 32, pp. 35-50. [9 Political Economy, vol. 107, pp. 507-538. [1 Regime”, European Economic Review, vol. 44, pp. 1449-1472. [1 the Implication for Exchange-Rate Policy: A Survey of A Few Recent New Open- Economy Macro Models”, Working Paper of The National Bure [12] Engel, C. and J. H. Rogers, 1996. “How Wide is the Border”, American Economic Review, vol. 86, pp. 1112-1125. [1 Causes and Welfare Costs”, Journal of International Economics, vol. 55, pp. 29-57. 20 [14] Gagnon, J. E. and M. M. Knetter, 1995. “Markup Adjustment and Exchange Rate Fluctuations: Evidence from Panel Data on Automobile Exports”, Journal of International Money and Finance, vol. 14, pp. 289-310. , 6] Goldberg, P. K. and M. M. Knetter, 1997. “Goods Prices and Exchange Rates: What 7] Klitgaard, Thomas, 1999. “Exchange Rates and Profit Margins: The Case of ers”, 9] Knetter, Michael M., 1993. “International Comparisons of Pricing-to-Market rugman, Paul K., 1987. “Pricing to Market When the Exchange Rate Changes”, in . 2] McCarthy, Jonathan, “Pass-Through of Exchange Rates and Import Prices to 4] Mussa, Michael, 1986. “Nominal Exchange Rate Regimes and the Behavior of Real Series 117-213. 6] Obstfeld, Maurice, “Exchange Rates and Adjustment: Perspectives from the New 7] Obstfeld, M. and K. Rogoff, 1995. “Exchange Rates Dynamics Redux”, Journal of [15] Goldberg, Linda S., 2004. “Industry-Specific Exchange Rates for the United States” Federal Reserve Bank of New York Economic Policy Review, vol. 10, pp.1-16. [1 Have We Learned?”, Journal of Economic Literature, vol. 35, pp. 1243-1272. [1 Japanese Exporters”, Federal Reserve Bank of New York Economic Policy Review, vol. 5, pp. 41-54. [18] Knetter, Michael M., 1989. “Price Discrimination by U.S. and German Export American Economic Review, vol. 79, pp. 198-210. [1 Behavior”, American Economic Review, vol. 83, pp. 473-486. [20] K Real Financial Linkages Among Open Economies, ed. by Sven W. Arndt and J. David Richardson., MIT Press, Cambridge, Massachusetts, USA [21] Lane, Philip R., 2001. “The New Open Economy Macroeconomics: A Survey”, Journal of International Economics, vol. 54, pp. 235-266. [2 Domestic Inflation in Some Industrialized Economies”, Staff Reports of Federal Reserve Bank of New York, No. 111, September, 2000. [23] Menon, Jayant, 1995. “Exchange Rate Pass-Through”, Journal of Economic Surveys, vol. 9, pp. 197-231. [2 Exchange Rates: Evidence and Implications”, Carnegie-Rochester Conference on Public Policy, vol. 25, pp. [25] Obstfeld, Maurice, 2001. “International Macroeconomics: Beyond the Mundell- Fleming Model”, IMF Staff Papers: Special Issue, vol. 47, pp. 1-39. [2 Open Economy Macroeconomics”, manuscript, July, 2002. [2 Political Economy, vol. 103, pp. 624-660. 21 [28] Obstfeld, M. and K. Rogoff, “Risk and Exchange Rates”, Working Paper of the 9] Obstfeld, M. and K. Rogoff, 2000. “New Directions for Stochastic Open Economy 0] Obstfeld, M. and K. Rogoff, 2000. “The Six Major Puzzles in International arket lt University, USA, 2001. 4] Rogers, J. H. and M. Jenkins, 1996. “Haircuts or Hysteresis? Sources of Movements 360. 5] Shambaugh, Jay C., “A New Look at Pass-Through”, Working Paper, Economics 6] Yang, Jiawen, 1997. “Exchange Rate Pass-Through in U.S. Manufacturing National Bureau of Economic Research, No. 6694, August, 1998. [2 Models”, Journal of International Economics, vol. 50, pp. 117-153. [3 Macroeconomics: Is There a Common Cause?”, NBER Macroeconomics Annual, The MIT Press, Cambridge, Massachusetts. [31] Obstfeld, M. and K. Rogoff, “Global Implications of Self-Oriented National Monetary Rules”, manuscript, Department of Economics, University of California Berkeley, USA, 2001. [32] Parsley, D. C. and S. Wei, “Limiting Currency Volatility to Stimulate Goods M Integration: A Price Based Approach”, mimeo, Owens Graduate School of Management, Vanderbi [33] Pesaran, M. H. and Y. Shin, 1998. “Generalized Impulse Response Analysis in Linear Multivariate Models”, Economics Letters, vol. 58, pp. 17-29. [3 in Real Exchange Rates”, Journal of International Economics, vol. 38, pp. 339- [3 Department, Dartmouth College, April, 2003. [3 Industries”, The Review of Economics and Statistics, vol. 79, pp. 95-104. 22 Appendix 1 Table 1: Estimation Results from System I For SR : tz 1 1 p q i t i j t j i j w − − = = = + ∑ ∑ Π z β u ; t + { } 0 : = 0 | for and , = 1,2,3 i ab H a b a b π > ∀i ; For LR : ( ) ( ) t t t L L w = + Π z β u ; ( ) ( ) * 1 1 = Π z β *w ntr Industry Cou y SR IMP SR WP I SR COE LR IMP LR WPI LR COE Japan Chem 0.0 ** (0 4) 0.021*** (0.008) na 0.123** (0.619) .351 .755) na i 88 .04 0 (0 Japan Food 0.143*** (0.068) 0.042*** (0.018) 0.171*** (0.032) 0.386* (0.238) 1.251 (6.809) 5.130 (27.860) 0.053* 0.027** 0.036* (0 ) 0.071* 0.022** 0.029** (0 ) 0.085*** -0 - - 0 - - 0 0 0 0 0 1 0 - 0 - 0 0 0 0 - 0 - 0 Japan Mach (0.031) * (0.088) .020 (0.042) * (0.007) .015 .443 Japan Metal 0.025 (0.085) 0.032** (0.016) (0.017) 0.051 (0.172) -0.165 (0.196) (0.477) Japan Petrol 0.221*** (0.100) 0.124*** (0.040) .382*** (0.017) 0.421*** (0.168) 0.076*** (0.025) 0.234*** (0.023) Japan Textiles .071*** (0.028) 0.027* (0.019) 0.048 (0.033) .078*** (0.036) 0.019* (0.011) 0.035 (0.023) Japan Wood .401*** (0.089) .221*** (0.052) .161*** (0.032) .257*** (0.457) 0.904 (1.511) 0.658 (1.010) USA Food. .354*** (0.124) 0.082*** (0.026) na .291*** (0.106) 0.084*** (0.029) na USA Mfl 0.522 (0.385) .450*** (0.215) .154*** (0.039) 0.522 (0.385) .450*** (0.215) .154*** (0.039) USA Appa. 0.254*** (0.055) .158*** (0.056) na 0.256*** (0.062) .060*** (0.022) na Not , ** a hows 5% ev fica ecti ar reported in parentheses; the o o “short-run” e th s to n” e es: *** nd * s 99%, 9 and 90% l el of signi nce resp vely; stand d errors are index c SR rresponds t ffects while e index LR correspond “long-ru ffects. IMP denotes the e estim d coe f th cha rate rowth in the import . notes the sum of the estimated coefficients of th xchange eq f, “M ds fo for les of sum of th ate fficients o e lagged ex e lagged e nge g equation WPI de rate growth in the wholesale price equation. COE denotes the sum of the estimated coefficients of lagged import prices in the wholesale price uation. For Japan, “Chemi” stands for Chemicals, “Food” stands for Foodstuffs and feedstuf ach” stands for Machinery and equipment, “Petrol” stands for Petroleum, coal and natural gas, “Textiles” stan r Textiles and “Wood” stands for Wood, lumber and related products prices. For USA, “Food.” stands Beverages and tobacco, “Mfl” stands for Mineral fuels, lubricants and related materials and “Appa.” stands for Artic apparel and clothing accessories. 23 Table 2: Nested Model Testing for System I ; 1 1 p q t i t i j t j i j w − − = = = + ∑ ∑ z Π z β ut + { } 0 : = 0 | for and , = 1,2,3 i ab H a b a b π > ∀ i Country Industry 3 R 1R is correct 2 R is correct is correct Japan Chemi 52.788 93.992 na p-value 0.002 0.567 na Japan Food 54.057 129.918 na T p-value 0.002 0.002 na Japan Mach 46.112 72.193 na p-value 0.064 0.999 na Japan Metal 70.017 33.745 na p-value 0.000 0.999 na Japan Petrol 55.121 61.439 na p-value 0.001 0.997 na Japan extiles 165.985 115.564 na p-value 0.000 0.969 na Japan Wood 63.177 69.462 na p-value 0.000 0.995 na USA Food. 338.499 6.053 na p-value 0.000 0.999 na USA Mfl 404.010 89.932 na p-value 0.000 0.999 na USA Appa. 67.792 77.684 na p-value 0.000 0.888 na Notes: F r Japan, “Chem for Chemicals, ” stands for F fs and feedstu Mach” stand chinery and e t, “Petrol” stan troleum, coal ral gas, “Textiles” stands for Tex “Wood” s ood, lumber a ed products pr r USA, “Food.” stands for everages and tobacco, “M for Mineral fu ricants and rel terials and “Ap r Articles of apparel and clothing accessories. “p-value” denotes probability values for corresponding test o i” stands “Food oodstuf ff, “ s for Ma tiles and quipmen ds for Pe and natu tands for W fl” stands nd relat els, lub ices. Fo ated ma B fo pa.” stands statistics. 24 Table 3: Estimation Results from System II For : SR 1 1 p q t i t i j t j i j w − − = = = + ∑ ∑ z Π z β u ; t + { } 0 = 0 | for and , = 1,2,3 i ab H a b a b π > ∀i ; : For LR : ( ) ( ) t t t L L w = +u ; Π z β ( ) ( ) * * 1 1 w = Π z β Country Industry SR IMP SR PPI SR COE LR IMP LR PPI LR COE UK Chemi & Org-che -0.344*** (0.172) 0.054*** (0.083) 0.021** (0.011) -0.344*** (0.172) 0.054** (0.083) 0.021** (0.011) UK Mach & El-mach -0.461*** (0.101) 0.018** (0.011) 0.028*** (0.012) -0.571*** (0.140) 0.018** (0.011) 0.029*** (0.012) Notes: ***, ** and * show ively; standard errors are r d i entheses; the index corresponds to “short-run” effects while the index corresponds t g-r ” effects. s 99%, 95% and 90% level of significance respect SR LR IMP d n eporte n par o “lon un eno e sum the estimated coefficients of the lagge change ra growth in the import e . tes th of te quatio P d ex P I de qua notes the sum of the estimated coefficients of the la xchange rate growth in the pro tion. denotes th um of the estimated coe ts of lagged port prices in the producer price equation. In the table, “Chemi & Org-che” stands for Chemicals & O icals industry and “Mac & El-mach” stands for Machinery and transport equipment & E ch y. gged e fficien ducer e e e s pric COE im rganic chem lectrical ma h inery industr Table 4: Nested Model Testing for System II 1 1 p q t i t i j t j t i j w − − = = = + + ∑ ∑ z Π z β u ; { } 0 : = 0 | for and , = 1,2,3 i ab H a b a b i π > ∀ Country Industry 1R is correct 2 R is correct 3 R is correct UK Chemi & Org-che 50.762 139.908 na p-value 0.010 0.015 na UK Mach & El-mach 74.055 101.359 na p-value 0.000 0.236 na N E otes: In the table, “Chemi & Org-che” stands for Chemicals & Organic chemicals industry and “Mach & l-mach” stands for Mac nery industry. “p-value” denotes probability values hinery and transport equipment & Electrical machi for corresponding test statistics. 25 Table 5: Estimation Results from System III For SR : 1 1 p q t i t i j t j t i j w − − = = = + + ∑ ∑ z Π z β u ; { } 0 : = 0 | for and , = 1,2,3 i ab H a b a b i π > ∀ ; For LR : ( ) ( ) t t t L L w = +u ; Π z β ( ) ( ) * * 1 1 w = Π z β Country Industry SR IMP SR PPI SR OE C LR IMP LR PPI LR COE UK Toba 1.655*** (0.327) 0.522*** (0.135) .354*** .055) 1.117*** (0.261) 0.401*** (0.111) -0.272*** (0.045) -0 (0 UK Pulp 0.276*** (0.121) 0.299*** (0.112) na 0.161*** (0.073) 0.622*** (0.255) na UK Wood -0.722*** -0.262*** na -0.722*** -0.394*** (0.071) na UK Metal (0.314) (0.112 (0.052) (0.152) .487*** (0.180) -0.218*** (0.087) mi -0.536*** (0.082) na 0.036 (0.032) -0.397*** (0.060) na 0.075 (0.061) UK Org-che 0.21 .16 ) 0.332*** (0.090) 0.163 (0.081) 0.231 (0.203) 0.323*** (0.108) 0.159 (0.082) -0.313*** (0.086) -0.332*** (0.117) na -0.241*** (0.064) -0.332*** (0.117) na P T s - 0 (0 ) - 0 (0 ) 0.573*** (0 ) 0.409*** (0 ) N 0 0 0 (0 ) 0 (0 ) 0 - - (0.170) (0.070) (0.170) 0.546* 0.324*** -0.145*** 0.258* 0 ) UK Che 5 (0 2 UK Medi. UK lastics 0.085 (0.065) 0.243*** -0.079 (0.072) .028*** 0.459*** (0.151) .086*** 0.187 (0.151) 0.359*** -0.089 (0.083) .057*** 0.518*** (0.131) .177*** UK ex-fab (0.066) 0 (0.013) .026 (0.113) 0 (0.028) .056 UK Iron 0 (0.071) .145*** 0 (0.063) .131*** .111 0.146** (0.073) 0 (0.046) .093*** .073 UK onfme .251*** (0.079) 0.487*** 0.425*** (0.122) .024*** 0.058 (0.049) .066*** .146*** (0.045) 0.391*** 0.600*** (0.194) 0.024** 0.082 (0.070) .067*** UK Mach - (0.123) .011 (0.022) - (0.092) .012 ( 0.096** 0.023) UK El-mach 0.479*** (0.109) 0.047*** (0.018) 0.050** (0.027) 0.511*** (0.122) 0.091*** (0.040) (0.049) No *, * sh 5 level signi e dard e s are rep n par the corresponds t ” e hile t ds to run tes: ** * and * ow 9 s 99%, SR % % and 90 of fi sp cance re ct an ively; st rror orted i “long- entheses; ” effects. index o “short-run ffects w he index LR correspon IMP denotes the the d co of d exchange rate gro th sum of estimate efficients the lagge wth in e import equation. PPI de at notes the sum im ients of the la rat th i en sum tim ed im ices i rodu qu e a” Tob p” lp and e pap ood” Woo rk, stan etal hemi Ch s, “ n c , “M not al Plast for cs, den tile f ron” or Ir eel, ” stan n- ferrous metals ” ch tra ip El enot al machinery. of the est ated coeffic gg ficient ed exchange e grow in the produ on. COE d cer price equ otes the of the es ated coef s of lagg port pr wast n the p er, “W cer price e ation. In th table, “Tob denotes acco, “Pul denotes Pu denotes d and co “Metal” ds for M ores, “C ” stands for emical Plasti Org-che” de “Tex-fabs” otes Organi chemicals edi.” de stands f es Medicin products, “ ics” stands otes Tex abrics, “I on and st “Nonfme ds for No , “Mach denotes Ma inery and nsport equ ment and “ -mach” d es Electric 26 Table 6: Nested Model Testing for System III 1 1 p q t i t i j t j t i j w − − = = = + + ∑ ∑ z Π z β u ; { } 0 : = 0 | for and , = 1,2,3 i ab H a b a b i π > ∀ Country Industry 1R is correct 2 R is correct 3 R is correct UK Toba 51.146 na 50.149 p-value 0.351 na 0.999 UK Pulp 64.107 46.194 na p-value 0.000 0.464 na UK Wood 19.018 na 10.727 p-value 0.088 na 0.978 UK Metal 24.510 23.351 na na UK Chemi na p-value 0.003 0.258 na UK Org-che 18.275 na 37.008 p-value 0.107 na 0.012 p U p 0. U 26 p U p 0. U U 39 U p p-value 0.017 0.612 38.741 49.647 UK Medi. 18.946 na 19.563 -value 0.395 na 0.994 K Plastics 27.529 na 40.591 -value 0.069 na 237 K Tex-fabs 16.874 na .804 -value 0.531 na 0.838 K Iron 24.871 na 39.093 -value 0.131 na 252 K Nonfme 22.944 20.027 na p-value 0.028 0.829 na K Mach 11.898 na .744 p-value 0.852 na 0.194 K El-mach 32.678 na 45.650 -value 0.111 na 0.528 Notes: I able, “Toba” o, “Pulp p and wast per, “Wood” d ood and cork, “Metal” stands ds for Chemicals, “Org-che” den ganic chemica edi.” denote products, “Plastics” stands for Plastics, “Tex-fabs” de xtile fabrics, stands for teel, “Nonf nds for Non-fe s metals, “M otes Machinery a d transport e and “El-mach” denotes Electrical hinery. “p-va notes probabil es for correspo istics. n the t denotes Tobacc ” denotes Pul e pa enotes W for Metal ores, “Chemi” stan otes Or ls, “M “Ir n” s Medicinal s notes Te a en o Iron and quipment m a e” st rrou mac ch” d lue” de n ity valu nding test stat 27 Table 7: Estimation Results from System III For SR : 1 1 p q t i t i j t j t i j w − − = = = + + ∑ ∑ z Π z β u ; { } 0 : = 0 | for and , = 1,2,3 i ab H a b a b i π > ∀ ; For LR : ( ) ( ) t L L wt t = + Π z β u ; ( ) ( ) * * 1 1 w = Π z β SR IMP SR PPI SR COE LR IMP LR PPI LR COE Country Industry USA Meat -0.316*** (0.174) -0.133 (0.086) na -0.309** (0.173) -0.174 (0.119) na USA Fruit 0.525 (0.362) na 0.019*** (0.007) 0.228 (0.158) na 0.025*** (0.011) USA Beve -0.046* (0.028) -0.164*** (0.049) 0.319*** (0.100) -0.033* (0.019) -0.164*** (0.049) 0.319*** (0.019) USA Chemi 0.105*** 0.094*** 0.237*** 0.105*** 0.121*** (0.048) 0.306*** (0.130) USA Org-che (0.103) (0.1 (0.212) (0.095) -0.298*** (0.122) 0.564*** (0.177) USA Inor-che -0.125 (0.124) na 0.485*** (0.113) -0.171 (0.165) na 0.382*** (0.087) USA Rubber 0. 8 (0.065) -0.005 (0.028) na 0.063 (0.113) 0.014 (0.069) na -0.111*** (0.042 0.028* (0.018 na -0.134*** (0.051 0.035* (0.022 na 0 0.10 (0 ) 1. (0 ) ch - -0 (0.039) (0.033) (0.116) (0.038) -0.174* -0.284*** 0.535*** -0.161* 15) 03 - USA Nm-min ) ) ) ) USA Me- mach 0.303*** (0.098) .077*** (0.035) 1** .047 0.431*** (0.148) 0.811 (0.165) 067 .217 USA El-ma 0.118** (0.062) na 0.078*** (0.023) .189** (0.108) na 0.145*** (0.049) USA Furni 0.100*** (0.043) 0.027*** (0.012) 0.016 (0.027) 0.124*** (0.055) 0.081* (0.050) 0.048 (0.079) N **, * shows 5% a level ifica ctive ard e r n p he i o “short-run” eff th to “long-run” effects. otes: * * and * 99%, 9 nd 90% of sign nce respe ly; stand errors ar eported i arentheses; t ndex SR rresponds to ects while e index LR c corresponds IMP denotes the he e oeff f the c g n the sum of t stimated c i o cients lagged ex hange rate rowth i import equation. PPI de a notes sum o ated ents of the la rate growth in tion denotes t m of the estimated ts of lagged i rices duce uation. In the table, at” sta eat t prepa , “ tand bl nut r dried, eve” s ver hemi” denotes C ls an od sen ed, -che” Org micals, “Inor- che” st ds f rganic c Ru ds f r m , no ially “ ” st Non- ne ctur ac Me ng m , “El-mach” sta Electri ery pme ni” urniture ar the f the estim u coeffici gge en d e exchang the prod in the pro ucer price equ . COE he s “Me coeffici mport p r price eq nds for M and mea rations Fruit” s hemica s for Vegeta d related pr es, fruit and u s s, fresh o “B “Org tands for Be r ages, “C cts, not e tially specifi stands fo anic che an Nm-min or Ino ands for nds for hemicals, “ m i bber” stan ra a or Rubbe e m anufactures h s t essent specified, etallic m l manuf and equi s, “Me- ” denote talworki and p achinery cal machin nt and “Fur denotes F ts thereof. 28 Table 8: Nested Model Testing for System III 1 1 p q t i t i j t j t i j w − − = = = + + ∑ ∑ z Π z β u ; { } 0 : = 0 | for and , = 1,2,3 i ab H a b a b i π > ∀ Country Industry 1R is correct 2 R is correct 3 R is correct USA Meat 21 25 na 21.802 .8 p-value 0. 51 na 0.995 USA Fruit 13 35 na 41.308 p-value 0. 46 na 0.287 USA Beve 12 12 na 23.810 p-value 0. 36 na 0.973 USA Chemi 44 51 66.422 na na USA 47.863 p-value 0.635 na 0.214 USA Inor-che 21.748 na 18.696 p-value 0.243 na 0.992 p U p U M p U 0. USA Furni 42 3 .7 7 .2 8 .8 p-value 0.001 0.059 Org-che 17.283 na USA Rubber 21.686 na 27.366 -value 0.357 na 0.949 SA Nm-min 9.220 na 27.155 -value 0.980 na 0.983 SA e-mach 7.252 na 68.739 -value 0.995 na 0.001 SA El-mach 9.271 na 30.111 p-value 0.979 na 914 8.299 na .688 p-value 0.989 na 0.398 Notes: In the table, “Meat” stands for Meat and m ions, “Fruit” ds for Vegetab t and nuts, fres ed, “Beve” everages, i” denotes Chem ls and related s, not essentia ified, “Org-c for Organi als, “Inor-che” stands for Inorgan icals, “Rubbe or Rubber tures, not es specified, “Nm n” stands for tallic mineral m ures, “Me enotes Metal g machinery, “El-mach” stands for Electrical machin uipment denotes Fur d parts thereo alue” denot ility values for corresponding tes eat preparat stan les, frui h r dri o stands for B “Chem ica product lly spec r” stands f he” stands manufac c chemic sentially ic chem Non-me -mi anufact -mach” d workin ery and eq and “Furni” niture an f. “p-v es probab t statistics. 29 Table 9: Correlations of Exchange Rates and Relative Export Competitiveness t t p p E REP * . = and E Correlation between Industry US-UK JAPAN-US JAPAN-UK Meat 0.289 na na Fruit 0.403 na na Beve. and tobacco 0.765 na na Crude mate., except fuel 0.369 na na Textile fibers 0.264 na na Metalliferous ores 0.195 na na Organic chemicals na na Inorganic chemicals 0.572 na na Medicinal products Uncoated paper 0. N 0.550 na na Manu. of metals 0 Machinery and 0 0. 0 0. etc. Chemicals 0.745 0.602 0.605 0.505 etc 0.658 na na 0.681 na na Textile fabrics o ls na na 753 n-ferrous meta na .692 na trans. Road vehicles na .977 875 0.500 .630 500 Misc. manufactures 0.825 0.910 0.618 Notes: table, “Meat” d eat preparations, “Fruit” denotes V ables, fruit a Beve. and tobacco” denotes Beverages and tobacco, “Crude mate., except fuel” s s, inedible fuels, “Textile fibers notes Textile fibers a eir waste, “Metallifer res etc.” denotes Metalliferous ores and metal scrap, “Chemicals” denotes Chemicals and rela anic chemicals” denotes Organic chem nic chemicals” stands for Inor “Medicinal products etc” stands for Medicinal and pharmaceutical products, “ r Uncoa er or paperboard, and ard, “Textile fabrics” stands for T etals” deno -ferrous metals, “ metals” stands f etals and rel inery and tran neral machin equipment, “Road vehicles’ stands for ransportation ipment and “Misc. factures” denotes Miscellaneous manufact cles/Other manufactu industry products. In the above enotes Meat and m eget nd nuts, fresh or dried, “ tands for Crude material ous o , except ” de nd th ted products, “Org icals, “Inorga ganic chemicals, Uncoated paper” stands fo ex m ted pap linearbo tiles, “Non-ferrous ated products, “Mach tes Non s.” denote Ge Manu. of ery and or M Road vehicles/T ured arti equ ring manu 30 Table 10: Estimation Results from Relative Export Prices ( ) ( ) ( ) 3 t t t L y c L x z π δ Δ = + + ν Industry US-UK US-JAPAN UK-JAPAN Meat -0.637* (0.429) na na Fruit -0.507* (0.334) na na Beve. and tobacco -0.243*** (0.081) na na Crude mate., except fuel 0.328*** (0.158) na na Textile fibers -0.472*** na na Metalliferous etc. 269 (0.244) na na Chemicals -0.259*** -0.283 -0.179*** Organic chemicals Inorganic ch s Uncoated paper -0.304 (0 ) Manu. of metals -0 Machinery and - (0 ) - (0 ) Road vehicles -0.667*** (0 ) -0.090 (0 ) -0.374*** (0 ) -0.347*** (0 ) (0.232) ores 0. (0.089) -0.431*** (0.211) (0.073) (0.147) na na emical -0.168* (0.112) na na Medicinal products etc. -0.208** (0.106) na na -0.327*** (0.136) na na Textile fabrics na na *** .068 Non-ferrous metals -0.203 (0.187) na na na .436*** (0.198) 0.283*** na trans. na .104 0.249*** .060 -0.095** (0.055) .214 .082 Misc. manufactures -0.101* (0.064) .178 .092 Notes: *** nd * shows 99% 90% level of significance respectively; standard errors are rep all the estim espond to the long-r stimates from equat (15) in the met he text. In the a ve table, “Meat” denotes Meat and meat preparations, “Fruit” den it and nuts, fr or dried, “Beve. and tobacco” denotes Bevera tobacco, “ l” stands de materials, inedible, except fuels, “Textile fibers” denotes T waste, “M ous ores etc.” deno Metalliferous ores and etal scrap, “C hemicals and r ted products, “Org als” denotes Organi hemicals, “In tands for Inor c chemicals, “Medic ucts etc” stands for Medicinal and pha ts, “Uncoated paper” stands for Unc or paperboard rd, “Tex r Textiles, “ -ferrous metals” de us metals, “ etals” stands fo and related prod hinery and trans eneral machinery uipment, “Ro s for Road ransportation eq d “Misc. manuf denotes Miscellan nufactured articles anufacturing indu cts. , ** a , 95% and orted in parentheses; ates corr un e * δ ion hodology part of t otes Vegetables, fru bo esh ges and Crude mate., except fue extile fibers and their for Cru etallifer tes m hemicals” denotes C organic chemicals” s ela gani anic chemic c c inal prod rmaceutical produc tile fabrics” stands fo r Metals oated paper n -ferro , and linearboa m Non otes Non Manu. of ucts, “Mac v T .” denote G u n and eq ad vehicles’ stand ehicles/ /Other m ipment a stry produ actures” eous ma 31 Appendix 2 Table 1. SITC Categories for Import Prices in USA Country SITC code number for Import prices Full name of the SITC coded industry Subcategory of SITC category Denoted as (in table nos. 7 and 8 in appendix 1) USA 01 Meat and meat preparations Food and live animals Meat USA 05 Vegetables, fruit and nuts, fresh or dried Food and live animals Fruit USA 1 Beverages and Tobacco Beverages and Tobacco Food. USA Beve USA Mineral fuels, Mineral fuels, lubricants and USA rel ts, n.e.s n.e.s O s rela cts O rel Inorg-che manu .e.s Manufactured Manufactured g N Eq nt USA 82 Furn Furni USA 84 and clothing M s m d Appa. 11 Beverages Beverages and Tobacco 3 5 lubricants and related materials Chemicals and ated produc related materials Chemicals and related products, Mfl Chemicals USA 51 rganic chemical Chemicals and ted produ rg-che USA 52 Inorganic chemicals Chemicals and ated products USA 62 Rubber factures, n goods classified chiefly by materials Rubber USA 66 Non-metallic mineral manufactures oods classified chiefly by materials m-min USA 73 Metalworking machinery Machinery and Transport uipme Me-mach USA 77 Electrical machinery and equipment iture and parts Machinery and Transport Equipment Miscellaneous El-mach thereof Articles of apparel manufactured articles iscellaneou anufacture accessories articles Notes: For each of the above import prices, we rresp ucer prices or consumer prices to carry-out our empirical estimation of exchange rate ffect. Therefore, the corresp producer or consumer prices ei ch tho some as the clo tch with the im ted products. that this ade based tes available from each of these detailed industry-le tatistics ation, please refer to the Bur r Statistics http://w have taken the co onding prod pass-through e se SITC codes or match is m onding sest ma ther exactly mat Please note times taken on the no por vels price s ww.bls.gov . For more inform eau of Labo web site at . 32 Table 2. SITC Cate Country SITC code number for Import prices Subcategory of S Denoted as (in table nos. 3, 4, 5 and 6 in appendix 1) gories for Import Prices in UK Full name of the SITC coded industry ITC category UK 12 Tobacco Bev nd Tobacco Toba erages a UK 25 Pulp and waste paper Crude materials Pulp UK 24 Wood and cork Crude materials Wood UK 27 Metal ores Crude materials Metal UK 5 Chemicals Chemicals Chemi UK 51 Organic chemicals Chemicals 54 Medicinal products Chemicals Org-che UK Medi. UK 57+58 Plastics Chemicals UK UK Non- tals Machinery and e 716+75+76+77 Electrical transport El-mach Plastics Tex-fabs Iron Nonfme 65 67 68 Textile fabrics Iron and steel ferrous me Manufactures Manufactures Manufactures UK UK 7 transport equipment Machinery and transport quipment Mach UK machinery Machinery and equipment Notes: F h of the above i ort prices, we he corresp ucer prices -out our emp imation of exc ge rate pa erefor onding pr ices either ex atch those SIT des or so close h the import ucts. Please note that this match i ased on from each of these detailed in vels price statistics. For more n, ple e Nat Online at http://www.statistics.gov.uk or eac mp have taken t onding prod to carry irical est han ss-through effect. Th e, the corresp oducer pr actly m C co metimes taken as the the n lable st match wit ed prod s made b in matio otes avai dustry-le for ase refer to th ional Statistics web site . 33 Table 3. Industry Categorie port Prices in Japan Country Full name of the industry for which we are taking Import prices Denoted as (in table nos. 1 and 2 in appendix 1) s for Im Japan Chem ls Chemi Japan Foodstuffs a d feedstuff Food Japan Machinery a ent Mach Japan Metals and rel ted products Metal Japan Petroleum, coal and natural gas Petrol Japan Tex es Textiles Japan Wood, lumb and related products Wood ica n nd equipm a til er otes: For each of the above import prices, we have taken the corresponding wholesale prices or producer rices to carry-out our empirical pass-through effect. Therefore, the corresponding wholesale price or pro h th taken as th tch with the im that th notes available from each of these detailed ce statistics. refer to the B pan web site at http://www N p estimation of exchange rate ducer prices either exactly matc ported products. Please note industry-levels pri ose industry codes or sometimes is match is made based on the For more information, please e closest ma ank of Ja .boj.or.jp/en. 34 Table 4. SITC Categories for xport Prices in USA and UK Country SITC code number for Export prices Full name of the SITC coded industry Subcategory of SITC category Denoted as (in table nos. 9 and 10 in appendix 1) E USA+UK 01 Meat and meat preparations Food and live animals Meat USA+UK 05 Vegeta es, fruit and nuts, fresh or d d Food and live animals Fruit USA+UK 1 Bevera es and tobacco Beverages and Tobacco Beve. and Tobacco USA+UK Crude m terials, Crude materials, Crude mate., except fuel USA+UK Textile fibers and Crude materials, inedible, except USA+UK an p Me es USA+UK 5 rela ts, Chemicals 51 Organic icals Ch d O USA+UK pr relate ucts USA+UK 64 pa d Uncoated paper USA+UK 68 N Non-ferrous R s 8 M s Manufactures bl rie g 2 inedible, except fuels inedible, except fuels a 26 28 their waste Metalliferous ores d metal scra fuels Crude materials, inedible, except Textile fibers talliferous or etc. fuels Chemicals and ted produc Chemicals and related products, n.e.s chem n.e.s emicals an USA+UK related products Chemicals and ted produ rganic chemicals Inorganic chemicals USA+UK 52 Inorganic hemicals c Medicinal and pharmaceutical oducts rela cts Chemicals and d prod 54 Medicinal products etc. Uncoated paper or perboard, an linearboard Manufactured goods classified chiefly by materials Manufactured onferrous metals goods classified chiefly by materials metals USA+UK 78 Road vehicles Machinery and Transport Equipment oad vehicle USA+UK iscellaneou manufactured articles Miscellaneous manufactured articles Misc. Notes: By USA+UK, we mean that for these two the SITC t prices almost exactly match each other. Based on this informatio d t prices for USA and UK. countries, codes for expor the relative expor n, we have calculate 35 Table 5. Industry Categories for Export Prices in Japan Country Full na ustry for which we are taking E as (in table nos. 9 in appe ) me of the ind xport prices Denoted and 10 ndix 1 Japan Chemicals a products Chem Japan Textiles Textile fabrics Japan equipment Machinery and Trans. Japan Transportation equipment Road vehicles Japan Other manufacturing industry products Misc. Manufactures nd related icals Japan Metals and related products Manu. of metals General machinery and 36 Appendix 3 Figure 1: Impulse Response for Chemical Industry in Japan -.002 .000 .002 .004 .006 .008 .010 .012 2 4 6 8 10 12 Response of DLWPI_CHE to DLWPI_CHE -.002 .000 .002 .004 .006 .008 .010 .012 2 4 6 8 10 12 Response of DLWPI_CHE to DLPPI_CHE -.002 .012 .000 .002 .004 .006 .008 .010 2 4 6 8 10 12 Response of DLWPI_CHE to DLM_CHE -.002 .000 .002 .004 .006 .008 .010 2 4 6 8 10 12 Response of DLPPI_CHE to DLWPI_CHE -.002 .000 .002 .004 .006 .008 .010 2 4 6 8 10 12 Response of DLP CHE to DLPPI_CHE -.002 PI_ .010 .000 .002 .004 .006 .008 2 4 6 8 10 12 Response of DLPPI_CHE to DLM_CHE -.010 -.005 .000 .005 .010 .015 .020 .025 2 4 6 8 10 12 Response of DLM_CHE to DLWPI_CHE -.010 -.005 .000 .005 .010 .015 .020 .025 2 4 6 8 10 12 Response of DLM_CHE to DLPPI_CHE -.010 .025 -.005 .000 .005 .010 .015 .020 2 4 6 8 10 12 Response of DLM_CHE to DLM_CHE Response to Generalized One S.D. Innovations ± 2 S.E. Notes: and for percentage change in wholesale prices in chemical industry, percentage change in producer prices in chemical industry and percentage change in import prices in chemical industry for Japan, respectively. CHE DLM CHE DLPPI CHE DLWPI _ and _ , _ st 37 Figure 2: Impulse Response for Metal-working Machinery Industry in USA -.0010 -.0005 .0000 .0005 .0010 .0015 .0020 .0025 1 2 3 4 5 6 7 8 9 10 11 12 Response of DLPPI_MEWMA to DLPPI_MEWMA -.0010 -.0005 .0000 .0005 .0010 .0015 .0020 .0025 1 2 3 4 5 6 7 8 9 10 11 12 Response of DLPPI_MEWMA to DLM_MEWMA -.004 -.002 .000 .002 .004 .006 .008 1 2 3 4 5 6 7 8 9 10 11 12 Response of DLM_MEWMA to DLPPI_MEWMA -.004 -.002 .000 .002 .004 .006 .008 1 2 3 4 5 6 7 8 9 10 11 12 Response of DLM_MEWMA to DLM_MEWMA Response to Generalized One S.D. Innovations ± 2 S.E. Notes: stand for percentage change in producer prices in metal orking machinery industry and percentage change in of import prices in metal-working machinery for USA, respectively. MEWMA DLM MEWMA DLPPI _ and _ - w industry 38 Figure 3: Impulse Response for Organic Chemical Industry in UK -.004 .000 .004 .008 .012 .016 1 2 3 4 5 6 7 8 9 10 11 12 Response of DLPPI_ORCHE to DLPPI_ORCHE -.004 .000 .004 .008 .012 .016 1 2 3 4 5 6 7 8 9 10 11 12 Response of DLPPI_ORCHE to DLM_ORCHE -.01 .00 .01 .02 .03 1 2 3 4 5 6 7 8 9 10 11 12 Response of DLM_ORCHE to DLPPI_ORCHE -.01 .03 .00 .01 .02 1 2 3 4 5 6 7 8 9 10 11 12 Response of DLM_ORCHE to DLM_ORCHE Response to Generalized One S.D. Innovations ± 2 S.E. stand for percentage change in producer prices in organic emical industry and percentage change in import prices in organic chemical industry for UK, spectively. otes: ORCHE DLM ORCHE DLPPI _ and _ N ch re 39 Figure 4: Relative Export Price and Nominal Exchange Rate Correlation for Chemical Industry for USA and Japan -.2 -.1 .0 .1 .2 93 94 95 96 97 98 99 00 01 02 DRE X_CHE DNE X Notes: stand for percentage change in relative export prices in chemical industry nd percentage change in nominal exchange for USA and Japan, respectively. DNEX CHE DREX and _ a 40 Figure 5: Relative Export Price and Nominal Exchange Rate Correlation for Machinery and Transport Industry for USA and Japan -.2 -.1 .0 .1 .2 93 94 95 96 97 98 99 00 01 02 DRE X_M T E DNE X Notes: and for percentage change in relative export prices in machinery and transport industry and percentage change in nominal exchange for USA and Japan, respectively. DNEX MTE DREX and _ st 41 Figure 6: Relative Export Price and Nominal Exchange Rate Correlation for Miscellaneous Manufactures Industry for USA and UK -.2 -.1 .0 .1 .2 94 95 96 97 98 99 00 01 02 DRE XP _M ANU DNE X otes: stand for percentage change in relative export prices in miscellaneous anufactures industry and percentage change in nominal exchange for USA and UK, respectively. DNEX MANU DREXP and _ N m 42 Figure 7: Relative Export Price and Nominal Exchange Rate Correlation for Meat Industry for USA and UK -.2 -.1 .0 .1 .2 93 94 95 96 97 98 99 00 01 02 D RE XP _M E AT DNE X otes: stand for growth rate of relative export prices in meat industry and ercentage change in nominal exchange for USA and UK, respectively. DNEX MEAT DREXP and _ N p 43 Figure 8: Relative Export Price and Nominal Exchange Rate Correlation for Machinery and Transport Industry for Japan and UK -.2 -.1 .0 .1 .2 84 86 88 90 92 94 96 98 00 02 DR E X_M AC HINE DNE X otes: stand for growth rate of relative export prices in machinery and ansport industry and percentage change in nominal exchange for Japan and UK, respectively. DNEX MACHINE DREX and _ N tr 44