How Do Exporters React to Changes in Cost Competitiveness?

• Author: Catherine Fuss,Jozef Konings,Stefaan Decramer
• Date: 01 October 2016
• Published date: 01 October 2016
• DOI: http://doi.org/10.1111/twec.12412

How Do Exporters React to Changes in

Cost Competitiveness?

1

Stefaan Decramer

1

, Catherine Fuss

2

and Jozef Konings

3

1

University of Leuven, Belgium,

2

NBB, Belgium and Universit

e Libre de Bruxelles, Belgium and

3

University of Leuven, Belgium and CEPR, UK

1. INTRODUCTION

THE growing imbalances in the euro area have triggered a debate about the role of cost

competitiveness for growth and how far austerity should go. When competitiveness

improves, countries can ‘grow’ out of the crisis, and hence, austerity measures become less

stringent. Globalisation and its associated increase in international competition have led to the

view that exports have become more sensitive to costs, and hence, competitiveness is often

measured in terms of unit labour costs,

2

defined as the labour cost per unit of output. The

focus on unit labour costs as a measure for competitiveness is based on the idea that increases

in unit labour costs are passed on in the form of higher export prices, resulting in a deteriora-

tion in the balance of payments, hampering economic growth and increasing unemployment.

The unequal evolution of unit labour costs in the euro area has therefore been a major con-

cern in recent years, or as the ECB (2008, p. 69) puts it in a recent report: ‘Cumulative

increases in labor costs across euro area countries can be indicative of growing imbalances

and losses in competitiveness and, as such, are an important early sign of the need for adjust-

ment. Relative developments in labor costs across the euro area countries, together with other

indicators of competitiveness, have therefore to be closely monitored’.

While there is a clear policy concern about the evolution of unit labour costs in many

European countries, stirred by close monitoring by the European Commission, there exists

very little conclusive evidence about the impact of unit labour costs on export performance.

This paper analyses therefore this relationship for Belgium, where nominal labour costs are

among the highest of all OECD countries. Belgium is also a small open economy, and hence,

maintaining export competitiveness is typically a high policy priority.

As early as the 1970s, Kaldor (1978) had argued that the growth in unit labour costs to

measure international competitiveness is at best too simplified. In particular, he demonstrated

that countries with the highest growth rates in GDP tend to have high growth rates in unit

labour costs. This is also known as the ‘Kaldor paradox’.

3

In Figure 1, we plot the evolution

of the aggregate export market share of Belgium and the aggregate relative unit labour costs

(relative to the EU-27). While we can note a negative correlation, it is clearly not a very

strong one. For instance, from 2007 onwards it seems that relative unit labour cost s and

export market shares have been moving together. The simple correlation coefficient between

1

This paper benefited from discussions at the FREIT 2013 conference, Ljubljana and discussions at the

ECB CompNet meeting December 2013. We thank Filip Abraham, Flora Bellone, Koen Breemersch,

Jan De Loecker, Jo Van Biesebroeck, Hylke Vandenbussche and two anonymous referees of ECB-

CompNet for useful suggestions and discussions. The views expressed in this paper are those of the

authors and do not necessarily reflect the views of the National Bank of Belgium.

2

See, for instance, the European Competitiveness Report 2012 of the European Commission.

3

Fagerberg (1988) analyses this in more detail using macrodata on technological competitiveness.

©2016 John Wiley & Sons Ltd

1558

The World Economy (2016)

doi: 10.1111/twec.12412

The World Economy

the growth in relative unit labour cost and the growth in export market share for the entire

period is in fact quite weak, only 0.044.

This apparent paradox has also recently been documented for Spain during the Great

Recession where, despite the unfavourable evolution of Spanish relative export prices, only a

modest decline in Spanish exports took place (Correa-Lopez and Domenech, 2012). This ‘dis-

connect’ between relative costs and exports has also been a widely researched puzzle in inter-

national macro analysing low exchange rate pass-through into export prices. When changes in

relative costs (or real exchange rates) only have a limited impact on relative export prices and

hence on export quantities, the policy focus on wage moderation and convergence of unit

labour costs between countries seems less appropriate. What explains this low aggregate cor-

relation and does it mean that the widely used measure of unit labour costs to measure com-

petitiveness should not be used?

Both the theoretical and empirical work has stressed the importance of taking firm-level

heterogeneity into account. The use of microdata helps to understand the different dimensions

of trade, such as the extensive and intensive margins. Furthermore, it helps to avoid biases

from aggregation (such as ‘the Spanish paradox’) as explained in Altomonte et al. (2013).

Recent work in international trade therefore makes increasing use of detailed disaggregated

data to understand the apparent low correlation between a number of macroaggregates. For

instance, Amiti et al. (2014) used highly disaggregated firm-product data to show that the lar-

gest exporters are also the largest importers. This turns out to be important because when

exporters are hit by an exchange rate shock in their destination market, they typically face a

compensating movement in their marginal costs if they are importing their intermediate

inputs. And since the largest exporters account for most of the exports, they dominate the

aggregate picture. For instance, in our data, 20 per cent of Belgian exporters account for

almost 90 per cent of all exports.

4

We therefore use disaggregated data, at the firm level, to analyse the impact of unit labour

costs on firm export performance. It also enables us to incorporate the heterogeneity of firms

in the analysis and to distinguish between the intensive and extensive margin. In particular,

0.8

0.9

1

1.1

1.2

2000 2002 2004 2006 2008 2010

Export Share Belgium in EU-27 (2000 = 1)

Relative Unit Labour Cost (ULC Belgium Compared to ULC EU) (2000 = 1)

FIGURE 1

Evolution of Relative Unit Labour Costs and Export Market Share in Belgium

Source: Eurostat, Author’s Computations.

4

See also Mu^

uls and Pisu (2009) documenting the highly skewed export distribution in Belgium.

©2016 John Wiley & Sons Ltd

COST COMPETITIVENESS AND EXPORTS 1559