Exploring the role of imported inputs in exchange-rate adjustments of exports has a relatively long tradition in the empirical trade literature (see for example sector-level studies by Athukorala & Menon 1994, Goldberg & Campa 2010, or more recently, firm-level studies by Greenaway et.al. 2010; Berman et.al.2012). The rationale for studying this channel is the potential role of exchange-rate appreciation not just in raising foreign export prices of traded goods and services, but also in lowering the prices of imported inputs. Domestic firms exporting abroad may thus have the means to offset some of the adverse effect of exchange-rate appreciations on profit margins through cheaper imported inputs ('naturally hedge' exchange-rate risks). However, this rationale only holds if the exchange-rate pass-through to imported input prices and/or export prices is non-zero.
It turns out that the recent empirical literature mainly focuses on (semi-)final goods price adjustments and investigates the cost effect due to imported inputs only indirectly using measures such as the share of imported intermediate inputs in total intermediate inputs (Greenaway et.al. 2010), or in studies with firm data, the ratio of total imports to total sales (Berman, et.al. 2012). These studies however do not look at actual price developments of imported inputs as a result of exchange-rate shocks. Thus, they implicitly assume full exchange-rate pass-through into imported input prices, which is a rather strong assumption.
In a significant departure from this literature, we study exchange-rate pass-through into imported input prices using bilateral and disaggregated unit values as proxies for import prices. We use these unit values to calculate indices of average imported input prices that are faced by each sector over time and investigate their role in the price-setting behaviour of exporters (Fauceglia et al. 2012).
Analysing imported input prices in Switzerland is particularly interesting as the Swiss economy has high ratios of imported intermediate inputs relative to total intermediate inputs, especially in the manufacturing sector, and about half of total imports are processed and re-exported (SECO 2011). Table 1 uses data and sector classifications from the 2001 input-output table for Switzerland published by the OECD to illustrate this. The table reports ratios of imported inputs relative to the sum of total inputs and total compensation to employees (i.e. total production costs) as well as ratios of imported inputs relative to total inputs. As Table 1 highlights, imported inputs make up more than 10% of total production costs in all Swiss sectors and are particularly high in some manufacturing sectors (e.g. textiles 27%, electrical machinery 25%). These figures are even higher when looking at the simple ratios of imported relative to total intermediate inputs (e.g. textiles 38%, iron & steel 35%).
Table 1. Share of imported inputs of total production costs in Switzerland by sector
To gain more insight into price and exchange-rate developments, Figure 1 considers the constructed indices of average imported input prices faced by each sector and the nominal-effective-exchange-rate index (from the Bank of International Settlement) from January 2005 to September 2011. We calculate imported-input price indices using import-weighted unit values at the HS eight-digit level and for each trading partner separately1. Trade data is obtained from the Swiss Federal Customs Administration. As energy prices are likely to make up a significant amount of production costs, Figure 1 also includes a line for a crude oil price index (calculated as the simple average of three spot crude oil prices - Dated Brent, West Texas Intermediate and the Dubai Fateh). All indices are set to 100 in January 2005. To eliminate seasonal fluctuations, all reported figures correspond to averages of the last 12 months (e.g. the oil price index for March 2005 corresponds to the average oil price index between April 2004 and March 2005).
The figure is divided into three panels. Each panel looks at imported-input price developments for sectors facing a similar pattern. The time axis is roughly divided into five phases: boom, commodity crisis, economic crisis, economic recovery, strong franc. Panel A sectors import intermediates with the least price fluctuations and are at first sight the least responsive to oil price shocks, in particular from January 2008 to May 2009. During the commodity crisis, imported-input prices even decreased slightly while crude-oil prices almost doubled. Panel B and panel C sectors clearly show the expected positive relationship between oil prices and imported input prices. Panel C sector prices are relatively more volatile (in both directions) than panel-b sectors. For some panel C sectors (e.g. iron & steel) imported input prices increased by a factor of four between January 2005 and September 2008, which is a considerably larger price hike compared to the oil shock during the same period.
Figure 1 also shows that the nomimal-effective-exchange-rate index is relatively stable from January 2005 to January 2009, and is followed by a steady appreciation of the Swiss franc over 2009 and a sharp appreciation in 2010 and 2011. Interestingly, during 2009 input prices show a decline during the period of steady appreciation but a rise in the 'strong' franc phase up until May 2011; this suggests that these prices were more correlated with oil prices during this period (with approximately a six-month lag). It was only after May 2011 that the price decreasing impulse of the strong tranc seemed to overcompensate for the price increasing tendencies of the oil price hike. Thus, in the course of continued appreciation, prices of imported inputs started to fall, which is likely to have decreased the exposure of Swiss exporters to the adverse exchange rate.
Figure 1. Development of imported input prices faced by output sectors: 2005-2011
Our empirical results, that are impervious to various robustness checks, firstly indicate high exchange-rate pass-through into imported input prices in all sectors. However, contrary to assumptions made in the recent empirical literature, we do not find evidence of full pass-through for all sectors either in the short or long run, though we are able to reject zero pass-through in a majority of sectors. The high magnitudes of pass-through coefficients into imported input prices are in line with related literature, but depart from Stulz (2007) and SECO (2011) who study pass-through into import prices more generally (not only intermediate imports). This difference could be due to low input demand elasticities with respect to local prices and/or a low share of distribution costs for inputs.
On the export side, our results indicate strong sectoral pass-through heterogeneity in both the short and long run. Moreover, differentiated and customised products such as machinery & equipment or rubber & plastics products generally have higher pass-through rates. This is consistent with Yang's (1997) argument that sectors with differentiated goods should attain higher pass-through rates. Our results also hold remarkably well with recent explanations based on distribution costs (Corsetti & Dedola, 2005). Sectors with high distribution cost shares (incurred in the local currency) such as mineral products and textiles, tend to have low pass-through rates and also engage more in local currency pricing.
Finally, the cost-adjustment effects in our results are found to be overwhelmingly insignificant implying that exporters do not pass on imported input price changes to foreign consumers. Thus, an appreciation of the franc leads to higher profit margins through the import channel and imported inputs act as a natural means for hedging exchange-rate risks.
The appreciation of the Swiss currency began in 2009 and progressed steadily until the middle of 2010 after which it accelerated in response to the ensuing EZ crisis. In the last year the Swiss National Bank has intervened to assuage Swiss exporters of the adverse effects of this appreciation. However, our final empirical result suggests that the pricing strategies of Swiss exporters may not have changed in response to the strong franc in wake of the EZ crisis. Significantly, a similar result at the extensive margin would strongly question the Swiss National Bank’s intervention during this period. Examining this is the subject of our next paper.
Athukorala, P and J Menon (1994), "Pricing to market behaviour and exchange rate pass-through in Japanese exports", Economic Journal, 104(423):271-81.
Berman, N, P Martin and T Mayer (2012), "How do different exporters react to exchange rate changes?", Quarterly Journal of Economics, 127(1):437-492.
Corsetti, G and L Dedola (2005), "A macroeconomic model of international price discrimination", Journal of International Economics, 67(1):129-155.
Fauceglia, D, A Shingal and M Wermelinger (2012), "Natural hedging of exchange rate risk: the role of imported input prices", MPRA Paper 39438.
Goldberg, L and J Campa (2010). "The sensitivity of the CPI to exchange rates: Distribution margins, imported inputs, and trade exposure", Review of Economics and Statistics, 92(2):392-407.
Greenaway, D, R Kneller and X Zhang (2010), "The effect of exchange rates on firm exports: The role of imported intermediate inputs", World Economy, 33(8):961-986.
SECO (2011), "Weitergabe von Einkaufsvorteilen aufgrund der Frankenstärke", Working paper, Staatssekretariat für Wirtschaft.
Stulz, J (2007), "Exchange rate pass-through in Switzerland", Swiss National Bank Economic Studies 4, Swiss National Bank.
Yang, J (1997), "Exchange rate pass-through in US manufacturing industries", Review of Economics and Statistics, 79(1):95-104.
1 We use the WTO classification of intermediate goods published by UN Comtrade; thus, solely imported intermediate HS eight-digit goods are considered in these calculations. Import-weighted averages are constructed from these unit values for each two-digit ISIC product group, which are further averaged for each sector used in Swiss input-output tables. Each table sector consists of one up to five two-digit ISIC product groups. To calculate the average imported intermediate input prices (or unit values) faced by Swiss industries, the constructed sector price averages are re-weighted according to the share of imports from each input sector in each output sector. The weights are taken from the 2001 input-output table for Switzerland.