Higher integration into GVCs mitigates negative effects of a strong currency on exports

Dario Fauceglia, Andrea Lassmann, Anirudh Shingal, Martin Wermelinger

18 February 2015



An indicator of a country’s integration in global value chains (GVCs) is the extent to which its exports rely on the share of imported intermediate inputs in foreign value added (backward participation) and the extent to which its exports serve as inputs in value added in the exports of other countries (forward participation). Switzerland was ranked 16th in GVC participation amongst OECD and BRICS economies in the year 2009, with a higher share of backward (28%) than forward (23%) participation. This high share of backward participation was especially true of manufacturing industries such as chemicals, machinery and electrical equipment. In fact, 35% of the final output of manufactured goods and market services in Switzerland in 2009 represented value added created abroad, with foreign value added shares for textiles and transport equipment being close to 100%.

This significant use of intermediate inputs by Swiss manufacturing industries has implications for their economic resilience to short and long-term changes in macroeconomic fundamentals, in particular exchange rates. Thus, adverse effects on Swiss manufacturing exporters resulting from an appreciation of the Swiss franc would be expected to be mitigated at both margins of trade (the probability of trading and the value of exports) by decreasing the relative prices of imported intermediate inputs, thereby reducing the need for export price increases (in foreign currencies) due to reduced profit margins (for instance see Drechsel et. al, 2015). This would result in a higher resilience of export demand to exchange rate fluctuations. This mechanism is referred to as "natural hedging". Its effectiveness is determined by the extent to which exchange rate changes are transmitted to imported input prices (exchange rate pass-through). The recognition of the role of integration into GVCs on responses to exchange rate adjustments has also led to research on alternative measures of real effective exchange rates (for instance see Bems and Johnson, 2012). In fact, there is now a body of work on this in the ECB’s Competitive Research Network (CompNet) project.   

In a previous study (Fauceglia et.al, 2014) on the Swiss manufacturing sector, also published on VoxEU on 19 October 2012, we studied exchange-rate pass-through into imported input prices and export prices using bilateral and disaggregated unit values as proxies for import and export prices using quarterly product-level trade data over 2004-2012. The findings from the study indicated high pass-through into imported input prices in a majority of sectors, providing evidence for the effectiveness of natural hedging. On the export side, the results indicated strong sectoral pass-through heterogeneity across sectors. Moreover, the cost-adjustment effects were found to be overwhelmingly insignificant implying that exporters do not pass on imported input price changes to foreign consumers. Thus, through the import channel, the appreciation of the franc in the wake of the Eurozone crisis was found to mitigate the negative effect on profit margins and thus imported inputs were found to act as a natural means for hedging exchange-rate risks.

In a recent study (Fauceglia et.al, 2015) for the Swiss State Secretariat of Economic Affairs (SECO), we examine exchange rate-driven adjustments of the Swiss manufacturing industry (both on the probability of trading and the value of exports) given the latter’s pronounced reliance on the use of imported inputs. We also study the extent to which the exchange rate sensitivity of exports depends on backward participation in global value chains. Finally, we simultaneously examine the extent to which export propensities in the current period depend on those in the preceding period to examine the export hysteresis (Baldwin, 1988) hypothesis. If past export status has a positive effect on the export probability, then this is an indication that temporary exchange rate appreciations can have a lasting damaging effect on the export structure.

We employ two different yet complimentary datasets to examine our research questions, i.e. HS 6-digit product-level data from the Swiss Federal Customs Administration (Eidgenössische Zollverwaltung) from 2004-2013 and firm-level data from the KOF innovation survey covering a sample of manufacturing firms in 7 different years in time between 1996 and 2013. Our twofold approach offers the unique possibility to study heterogeneous patterns in firm reactions to exchange rate changes while providing the ability to control for a rich number of characteristics that are unobserved in aggregate data over a largely overlapping time period.


Our results find validity in both product- and firm-level analyses and are robust to the use of different estimation strategies. They suggest that an appreciation of the Swiss franc has a negative impact on both the propensity and the value of Swiss exports, but that this negative effect is mitigated in sectors where the Swiss import share of intermediate inputs is high.

Using product-level data, the negative effect of an appreciation on exports was estimated to be about -0.7, i.e. a 1% appreciation of the Swiss franc was associated with a 0.7% fall in exports, ceteris paribus and on average. An increase of the franc by 1% also reduced the likelihood that the product was exported by approximately 0.075 to 0.1 percentage points. We then investigated the effect of imported inputs on the overall exchange rate effect, taking "natural hedging" into account. Our analyses revealed that a 1% appreciation of the imported-inputs-weighted exchange rate increased the probability of exporting by just over one tenth of a percentage point, thereby offsetting the adverse direct exchange rate effect. However, no compensating effect of the import-weighted exchange rate could be detected on the value of exports. In further estimates, the ratio of sectoral imported inputs from the destination of an exported good in that sector to total imported inputs within the same sector (across all export destinations) was used as a destination-specific measure of natural hedging within an industry. These estimates also showed that an increasing proportion of imported inputs from the destination country for Swiss exports significantly reduced the negative effects of currency appreciation on both export probability (see left side panel in Figure 1) and export revenues (see right side panel in Figure 1)

Figure 1. Exchange rate effects on export status

Firm-level results suggest that a 1% increase in the exchange rate index is associated with a 0.3% reduction in the volume of exports, ceteris paribus and on average. However, once the degree of international integration approximated by the overall share of intermediate inputs in sales is considered, this negative effect is found to be considerably mitigated and – with increasing intermediate input shares – even offset in various empirical specifications.

Furthermore, we find strong evidence for export hysteresis. This suggests that products that are not exported in the previous year require larger exchange rate depreciations to achieve positive export profits and to be exported in the following year than products that are already present in an export market. The previous export experience is found to be the most important determinant of export probability with the magnitude of the effect ranging from 0.10 (10 percentage points) in the product-level results to 0.38 (38 percentage points) in our firm-level results. This suggests the existence of significant entry costs and implies that companies no longer exporting due to the strong Swiss franc require a comparatively disproportionate devaluation to export again profitably. It is therefore possible that temporary exchange rate fluctuations have permanent negative effects on the export structure of Switzerland.

Policy implications

Results obtained from both our studies reveal a significant overall extent of "natural hedging" of exchange rate fluctuations. Sectoral integration into GVCs is a rough indicator of a given industry’s exposure. Going by the results of our two studies, major Swiss export sectors such as chemicals and engineering that have high foreign share of value added in exports of 42% and 33%, respectively, are most likely to be less adversely affected by a strong franc. In contrast, the food and paper industry (backward participation of 24%) are likely to be more exposed to the vagaries of exchange rate fluctuations. Overall, our results imply that firms and sectors with a higher degree of international integration are likely to be less affected by the negative effects of a stronger Swiss franc.  


Baldwin, R (1988), "Hysteresis In Import Prices: The Beachhead Effect", The American Economic Review 78(4): 773-785.

Bems, R and R C Johnson (2012a), "Value-Added Exchange Rates", NBER Working Paper, 18498.

Bems, R and R C Johnson (2012b), "Value-added exchange rates", VoxEU.org, 6 December.

Drechsel, D, H Mikosch, S Sarferaz and M Bannert (2015) "How are firms affected by exchange rate shocks? Evidence from survey based impulse responses", KOF Working Papers No. 371.

Fauceglia, D., A. Shingal and M. Wermelinger (2014), "Natural hedging of exchange rate risk: The role of imported input prices", Swiss Journal of Economics and Statistics 150(4): 261-296.

Fauceglia, D, A Lassmann, A Shingal, and M Wermelinger (2015), "Backward Participation in Global Value Chains and Exchange Rate Driven Adjustments of Swiss Exports, Study commissioned by the Swiss State Secretariat of Economic Affairs (SECO).



Topics:  Exchange rates International trade

Tags:  Switzerland, GVCs, pass-through

Senior Lecturer in Economics, Zurich University of Applied Sciences

Postdoctoral Researcher the Chair of Applied Economics: Innovation and Internationalization at KOF Swiss Economic Institute, ETH Zurich

Senior Research Fellow at the World Trade Institute (WTI), University of Bern

Research Fellow at the Swiss Institute for International Economics and Applied Economic Research, University of St. Gallen