Exchange rate effects: A case study of the export performance of the Swiss agriculture and food sector

• DOI: http://doi.org/10.1111/twec.12611
• Author: Andreas Kohler,Ali Ferjani
• Published date: 01 February 2018
• Date: 01 February 2018

ORIGINAL ARTICLE

Exchange rate effects: A case study of the

export performance of the Swiss agriculture

and food sector

Andreas Kohler

|

Ali Ferjani

Federal Department of Economic Affairs, Education and Research, Agroscope, Ettenhausen, Switzerland

1

|

INTRODUCTION

Since the global financial crisis in 2008, the Swiss franc has appreciated strongly against the cur-

rencies of Switzerland’s major trading partners. During 2 years, from the onset of the global finan-

cial crisis in 2008 until the introduction of the exchange rate peg against the Euro by the Swiss

National Bank in September 2011, the Swiss franc appreciated about 25 per cent (in real terms)

against the currencies of Switzerland’s most important export markets for agriculture and food

products, thus potentially depressing foreign demand for Swiss products. At the same time, the

Swiss agriculture and food sector has become more integrated into the world market. Against this

backdrop, there has been renewed interest in the question of how changes in the exchange rate

affect exports in general, and exports of the Swiss agriculture and food sector in particular.

The main contribution of this paper is to estimate the exchange rate elasticity of Swiss agro-food

exports using a wide range of models to assess the robustness of the estimates. The case study of the

Swiss agriculture and food sector has the following advantages. On the one hand, Switzerland is a

small-open economy, with an independent economic and monetary policy, which has lately experi-

enced sharp movements of its currency (safe haven currency). On the other hand, the agro-food sector

is relatively small compared to the rest of the Swiss economy such that a causal interpretation of the

results becomes more plausible (in particular, we do not have to worry about reverse causality).

Hence, for policymakers this case study could yield valuable insight into the reaction of exporters to

exchange rate fluctuations under a particular set of economic policies. We will see below that the

behaviour of exporters in other sectors of the Swiss economy is relatively similar. Thus, we think that

the lessons learned can be generalised to some extent to other sectors, as well as to other economies

similar to Switzerland. Furthermore, we use time series as well as panel data models to estimate

exchange rate effects. Time series analysis has the advantage of a clear identification strategy since

only variation over time is used to identify parameters but might suffer from bias due to small sample

size. Panel data analysis allows us to increase sample size and exploit information contained in the

cross-section, which might ameliorate potential bias. The downside is that (dynamic) panel data mod-

els are sensitive to model specification (and the set of instruments). The analysis of both time series

and panel data helps us to assess the sensitivity of our results with respect to model specification, esti-

mation methods and data structure—something that has been neglected in the literature.

DOI: 10.1111/twec.12611

494

|

©2018 John Wiley & Sons Ltd wileyonlinelibrary.com/journal/twec World Econ. 2018;41:494–518.

We find that the estimated exchange rate elasticities are remarkably similar across all model speci-

fications, estimation methods and data structure. The short-run exchange rate elasticity of Swiss agro-

food exports is between 0.7 and 0.8, whereas the long-run exchange rate elasticity is between

0.8 and 0.9. Our results are similar to the results found in the existing literature. Interpreting the

exchange rate elasticity as the price elasticity of foreign demand implies a relatively inelasti c foreign

demand for Swiss agro-food products. This suggests that on average, no close substitutes for Swiss

agro-food products are available. One explanation could be that Swiss producers are able to success-

fully differentiate their products based on quality (differences) and thus, avoid price competition

abroad. The focus of Swiss agricultural policy on a quality strategy (labels and financial support of

innovative projects) could also play an important role by setting the right incentives for domestic pro-

ducers of agricultural commodities. This may be a valuable lesson for policymakers in other small-

open economies with similar agro-food sectors (e.g., Norway).

We are not aware of comparable studies that focus on exchange rate effects on the exports of

the agro-food sector in a small-open economy like Switzerland using time series and panel data

models. In general, studies use either time series models or panel data models to estimate exchange

rate effects but not both, making it difficult to assess the robustness of the results. The existing lit-

erature can broadly be divided into three broad categories. First, there is a recent literature looking

into exchange rate effects on (aggregate) exports of Switzerland. This literature generally finds sur-

prisingly small effects of changes in the exchange rate on (aggregate) Swiss exports. Estimates of

long-run elasticities from time series models are between 0.7 and 1.1 (F€

urer, 2013, Hanslin

Grossmann, Lein, & Schmidt, 2016; SECO 2010; Tressel & Arda, 2011). Estimates of long-run

elasticities from panel data models are slightly lower, and in the range of 0.5 and 0.7 (Auer &

Saure, 2012; Gaillard, 2013). Second, there is a relatively large but older literature concerned with

exchange rate effects on agricultural exports of large countries or regions, like the United States,

Canada or the European Union. Kristinek and Anderson (2002) give an excellent survey of this lit-

erature. In general, it seems that the evidence is mixed; some studies find exchange rate effects on

agricultural exports while others do not. For example, estimating a vector error-correction model,

Kim, Cho, and Koo (2004) find that the exchange rate has a significant impact on US agricultural

trade with Canada, whereas Vellianitis-Fidas (1976) does not find evidence that the exchange rate

of the United States does affect its agricultural exports using cross-section and time series data.

Third, there is a literature analysing the reaction of exporters to exchange rate changes in terms of

pass-through into exports and prices based on firm-level data (Berman, Martin, & Mayer, 2012;

Fabling & Sanderson, 2014; Greenaway, Kneller, & Zhang, 2010).

The remainder of this paper is organised as follows. In Section 2, we give a brief introduction

to the Swiss agriculture and food sector. Section 3 discusses the data in detail. Sections 4 and 5

introduce and discuss the time series and panel data models, respectively. Section 6 conclu des.

2

|

THE SWISS AGRICULTURE AND FOOD SECTOR

To facilitate the interpretation of the results, this section provides a very short introduction to the Swiss

agriculture and food sector focusing on its structureand exports, as well as agricultural policy.

2.1

|

Structure and exports of the sector

The Swiss agriculture and food sector is relatively small in terms of its share in GDP as well as

employment, accounting for about 3 per cent of GDP and employing around 5 per cent of the total

KOHLER AND FERJANI

|

495