The great trade collapse and the Spanish export miracle: Firm‐level evidence from the crisis

• Author: Marc‐Manuel Sindlinger,Nicole Meythaler,Peter S. Eppinger,Marcel Smolka
• Published date: 01 February 2018
• DOI: http://doi.org/10.1111/twec.12530
• Date: 01 February 2018

ORIGINAL ARTICLE

The great trade collapse and the Spanish export

miracle: Firm-level evidence from the crisis

Peter S. Eppinger

1

|

Nicole Meythaler

2

|

Marc-Manuel Sindlinger

3

|

Marcel Smolka

4

1

University of T€

ubingen, T€

ubingen, Germany

2

Institute for Applied Economic Research (IAW), University of T€

ubingen, T€

ubingen, Germany

3

University of Bonn, Bonn, Germany

4

Aarhus University, Aarhus, Denmark

1

|

INTRODUCTION

The global recession that followed the 2008 financial crisis continues to place a heavy burden on

the world economy. One important aspect of the crisis that has caught a lot of attention among

both policymakers and economists was the sudden, synchronised and more than proportional

decline in global trade relative to global production—the so-called “great trade collapse”(Baldwin,

2009). While the causes and consequences of this event have been subject to extensive debate, the

available evidence derives largely from aggregate data rather than from detailed firm-level data.

1

This is somewhat surprising, since the issue of firm heterogeneity and the fact that only a fraction

of firms access foreign markets have become cornerstones of modern trade literature. To what

extent have firms decided to leave foreign markets in response to the crisis? Are firms today rely-

ing less on imports and exports than before the crisis? And did firms perform better or worse dur-

ing the crisis if they were active on foreign markets? A fine-grained analysis of the microstructu re

of international trade in the years surrounding the financial crisis can provide answers to these

questions by uncovering patterns in the data that would go unnoticed in an analysis based on

aggregate data alone.

The objective of this paper is to provide such a fine-grained analysis using a representative

sample of Spanish manufacturing firms over the period 2005–12. Spain is a particularly interesting

case to look into. On the one hand, the country was deeply affected by the financial crisis and sub-

sequent recession. In the first half of 2009, real industrial production contracted by 21.4% relative

to the first half of 2008.

2

Importantly, Spain went through very difficult times also after the finan-

cial crisis. Following zero growth in 2010, total production contracted again in 2011–13, reflecting

1

Important contributions using aggregate or sector-level trade data include Chor and Manova (2012) and Eaton, Kortum,

Neiman, and Romalis (2016). We discuss the existing micro-level evidence further below.

2

Annual industrial production in 2009 declined by 16.2% relative to 2008. For real manufacturing exports, the same number

is 21.2%. The data come from the Spanish Instituto Nacional de Estad



ıstica (INE).

DOI: 10.1111/twec.12530

World Econ. 2018;41:457–493. wileyonlinelibrary.com/journal/twec ©2017 John Wiley & Sons Ltd

|

457

what is sometimes called a “double-dip”recession. On the other hand, and perhaps surprisingly,

the country showed a relatively strong export performance over the crisis period. Figure 1 demon-

strates that the Spanish economy was able to improve its competitive position on international mar-

kets compared to other economies in Europe. For example, between 2007 and 2013 export s from

Italy and France decreased by 10% and 7%, respectively. In contrast, exports from Spain incre ased

by 13% over the same period. This development (celebrated by some as the “Spanish export mira-

cle”

3

) put Spain ahead of not only other countries in economic turmoil, but also countries that

quickly returned to economic growth after 2008, such as Germany and the UK.

In this paper, we zoom in on the Spanish crisis experience. Adopting a micro-level perspective

on Spanish firms allows us to investigate two important issues related to the crisis episode that

remain obscure in aggregate data. First, we can disentangle the effects of the crisis at the extensive

and the intensive firm-level margins of trade, that is, we can separate a firm’s decision to access

foreign markets at all from the volume of a firm’s exports and imports (as shares of its total sales

and purchases, respectively). This distinction is crucial for the purpose of our analysis and allows

us to address two interesting questions: Did aggregate trade decline because of firms exiting for-

eign markets, or because of a contraction in firm-level trade volumes? And was the subsequent

recovery and export boom due to firms scaling up their exports, or due to new firms entering for-

eign markets? These questions are important because a destruction of cross-border trade linkages

at the firm level can have long-lasting adverse effects on the economy (Monarch & Schmidt-Eisen-

lohr, 2016), and these effects are not expected for adjustments at the intensive margin. Conversely,

the entry of new exporters might soften the adverse effects of the crisis by increasing the potential

for future economic growth, because new exporters in Spain are more likely to engage in produc-

tivity-enhancing technology upgrading than non-exporters (Hanley & P

erez, 2012). In terms of

methodology, we follow the literature estimating firm-level models of exporting and importing

based on panel data (e.g. Bernard & Jensen, 1997, 1999). However, this literature is typically inter-

ested in the evolution of firm-specific characteristics (e.g., productivity, management, or labour

force composition) and how these influence firms’export and import decisions, respectively. In

contrast, our focus is on the direct effects of the financial crisis and subsequent recession, that is,

changes in macro conditions that are beyond the control of individual firms.

0.70

0.80

0.90

1.00

1.10

1.20

1.30

2007 2008 2009 2010 2011 2012 2013

USA

Spain

UK

Germany

France

Italy

Export Volume Index (2007 = 1)

FIGURE 1 Export volumes, 2007–13

Note: The data are taken from the World Development Indicators (WDI) provided by the World Bank.

3

See, e.g., the article “El milagro del sector exterior de Espa~

na: admirable, pero con algunos claroscuros”, published on 10

May 2013 in the Spanish daily newspaper elEconomista.es.

458

|

EPPINGER ET AL.

The second issue we investigate are differences in firm performance and crisis resilience between

exporting and non-exporting firms. Since exporting firms are known to be larger and more productive,

on average, than non-exporting firms, their behaviour can be important for aggregate outcomes. It is

thus crucial to understand the performance of these firms in times of exceptional economic distress.

Does exporting to foreign markets make firms more immune to shocks, or does it make them more

vulnerable? This is an interesting question that should be settled empirically, as there exist theoretical

arguments supporting either view. While allocating sales across various markets, domestically and

abroad, insures the firm against an adverse demand shock in one market, there is also a substantial risk

involved in exporting (e.g., currency risk, non-payment risk, transport risk), and relying on foreign

markets in times of a globally synchronised crisis might prove particularly harmful to firm perfor-

mance. To answer this question, we estimate differences between exporters and non-exporters in terms

of size, productivity and survival, so-called exporter premia (Bernard & Jensen, 1999), and we study

the evolution of these premia over the crisis years. Importantly, increasing exporter premia during the

crisis could be taken as an indication that economies become less vulnerable to economic shocks

through exporting. Furthermore, if it is primarily non-exporting firms that are forced to exit the market

due to the crisis, then this might (in the medium to long run) induce a reallocation of resources away

from non-exporting firms towards exporting firms, where they are put to more efficient use. The same

logic applies if for non-exporters the evolution of productivity through the crisis and afterwards is less

favourable than for exporters. Hence, differences in crisis performance of exporters vs. non-exporters

are relevant also for the long-run growth perspective of the Spanish economy.

The main results of our empirical analysis can be summarised as follows. First, the sharp drop

in international trade that the Spanish manufacturing sector experienced in 2009 took place at the

intensive margin, not the extensive margin. This means that, while the financial crisis caused a

strong reduction in firm-level imports and exports, it did not prompt firms to exit foreign markets

altogether. In the years after the financial crisis, we do see changes at the extensive margin, but

we see more, rather than less, firms starting to enter foreign markets. As a result, there is now a

larger share of firms involved in international trade than before the crisis. Furthermore, firms have

diversified their export portfolios to include more distant destinations outside the European Union.

Second, while firms active in the export market saw their export volumes plummet in the finan-

cial crisis, this decline was not limited to exports, but rather, it was visible to the same extent in

their domestic sales. This observation might seem surprising in the light of the discu ssion about

the great trade collapse. Moreover, the decline in exports was fully made up for (and even over-

compensated) already by 2011. Those firms that entered the financial crisis as export ers have in

fact been allocating ever larger shares of their production to foreign markets over the past few

years. It seems that these firms have effectively compensated for the lack of domestic demand by

expanding their sales abroad. In this sense, firms in the Spanish manufacturing sector are on aver-

age more, not less, “globalised”today than they were before the financial crisis.

Third, we find that it made a significant difference for key economic performance indicators

(such as jobs, productivity and survival) whether or not firms were active on export markets when

the crisis hit the Spanish economy. While all firms strongly reduced their output and laid off large

numbers of workers during and after the financial crisis, firms that entered the crisis as exporters

(and continued to export throughout the crisis years) saved more jobs, stayed more productive and

were more likely to survive. One of the more alarming findings is that from 2007 to 2009 firms’

average total factor productivity (TFP) deteriorated by around 15%. For non-exporters, TFP contin-

ued to decline by another 15% from 2009 to 2011. Exporters, in contrast, maintained about the

same level of productivity in 2011 as they had in 2009. Our analysis also shows that aggregate

TFP in the Spanish manufacturing sector declined as a result of the crisis.

EPPINGER ET AL.

|

459