Service Trade and Occupational Tasks: An Empirical Investigation

• Published date: 01 September 2017
• DOI: http://doi.org/10.1111/twec.12440
• Author: Giordano Mion,Andrea Ariu
• Date: 01 September 2017

Service Trade and Occupational Tasks:

An Empirical Investigation

Andrea Ariu

1,2,3

and Giordano Mion

4,5,6

1

Department of Economics, University of Geneva, Geneva, Switzerland,

2

McDonough Business School,

Georgetown University, Washington, DC,USA,

3

CRENOS, Sassari, Italy,

4

Department of Economics,

University of Sussex, Brighton, UK,

5

CEPR, London, UK and

6

CEP, London, UK

1. INTRODUCTION

TRADE in services increased tremendously in the past 15 years evolving from 15 per cent

to almost 30 per cent of world trade (World Trade Organisation, 2008). This incredible

performance has been seen as a consequence of the new opportunities created by information

technology (IT) (Freund and Weinhold, 2002; Blinder, 2009). Many service providers started

exploiting the potentiality of computers and the Internet to offer their services abroad. This is

particularly true for services that do not require the physical proximity of the customer and

the supplier, like call-centres and standardised financial services. However, the effect of new

technologies is less clear for those services requiring physical proximity and/or human inter-

action. On the one hand, communication of information between distant locations has become

easier, thus facilitating the remote execution of services like bookkeeping and accounting. On

the other hand, some services have become more and more tailored and complex thanks to

the opportunities provided by increasingly powerful computers and softwares. Despite the fact

that higher sophistication and tailoring might have augmented the appeal of services in the

export markets, the increasing complexity of the process involved in producing and delivering

services like consultancy has made face-to-face communication even more important than

before thus rendering, everything else equal, export activities more difficult.

Using microdata for Belgium, we investigate how these changes in the tasks used in the

production are linked to the rise of service trade. A suitable framework to tackle this question

is provided by the ‘task approach’ developed by both labour economics and international

trade. Both strands consider the production process as a mix of different tasks that are com-

bined together to deliver a final product. These can be classified in several categories depend-

ing on how repetitive is their nature and whether they imply manual, cognitive or interactive

activities. Autor et al. (2003), Spitz-Oener (2006), and Autor and Acemoglu (2011) document

the remarkable change occurred in workers’ tasks, both within and across occupations, during

the last two decades and argue that IT has been a key driving force in this process. Therefore,

this framework allows us to study how the evolution of complexity and human interaction are

This work was produced for the 2010 bi-annual conference of the National Bank of Belgium (NBB).

The views expressed in this paper are our own and do not necessarily reflect those of the NBB. The

authors thank Andrew Bernard, Gianmarco Ottaviano and Lindsay Oldenski for providing us with help-

ful insights, Daniela Rohrbach-Schmidt for assistance with the BIBB data, Christian Viegelahn for help

with translation from German and Alexandra Spitz-Oener for providing the STATA classification code

for tasks. We also thank seminar participants at the 2010 NBB bi-annual conference meetings, 2nd GIST

conference, the RES 2011 conference and 26th EEA conference for helpful comments and suggestions.

All remaining errors are ours. Financial help under the Globalisation Investment and Trade in Services

(GIST) project, funded by the EU 7th Framework Programme (ITN-2008-211429), is gratefully acknowl-

edged by Andrea Ariu.

©2016 John Wiley & Sons Ltd

1866

The World Economy (2017)

doi: 10.1111/twec.12440

The World Economy

associated with the service trade participation. Moreover, occupational tasks measures offer a

much richer portrait of changes occurred in the production process as compared to standard

measures of the IT impact like the use of computers or the degree of Internet diffusion.

Our results uncover a rich pattern linking occupational tasks’ changes and the increase in

firms’ participation to service trade while at the same time questioning the common view

about IT diffusion and the service trade boom. In sectors in which the production has shifted

more towards the use of interactive tasks, we observe a relatively lower probability of firm-

level entry in exports markets. At the same time, the probability of exporting increased more

in sectors where cognitive tasks increased. Therefore, in sectors where services have become

more elaborate, firms have been able to better leverage IT and succeed in the export market,

while the opposite holds in sectors in which face-to-face communication has become more

prevalent. Our estimations further suggest that the change in IT use per se does not translate

into a significantly higher or lower firms’ participation to service export. Complexity and need

of personal communication play in opposite directions, and the overall balance is such that

technological change does not strike us as being a key underlying force behind the increas e in

the extensive margin of service exports. Of course, other alternative forces like offshoring

(Baumgarten et al., 2013; Becker et al., 2013), demand shifts, trade liberalisation and compar-

ative advantage might also be competing drivers of the changes in the production structure

and in the participation to export. However, our analysis reveals that while they might have

some importance, our results remain robust when controlling for them.

Most previous analyses have used aggregate service trade data. Freund and Weinhold

(2002), who are no exception to the rule, study the impact of Internet diffusion on the

increase in the value of trade in services by focusing on cross-country data. Their research

topic is closely related to ours, some of the key differences are that we focus on changes in

occupational tasks, use firm-level trade to look at the extensive margin and concentrate on a

single country (Belgium). The link between trade in services and the change in the task con-

tent of jobs has been previously analysed by Oldenski (2012), albeit in a different setting.

Using US sector-level data, Oldenski (2012) analyses the determinants of the FDI versus

export decision in the context of services. She shows that the usual trade-off between econo-

mies of scale and proximity to the final consumer, which is recognised to be a key element in

the exporting versus FDI strategy for manufacturing goods, does not apply to services. We

share the same occupational tasks approach, but we use firm-level trade data and focus on the

determinants of entry and exit into the export and import of services activities.

Our research is also related to recent descriptive studies of trade in services at the firm

level started with Breinlich and Criscuolo (2011) for the UK and then extended by Kelle and

Kleinert (2010) for Germany, Gaulier et al. (2011) for France, Federico and Tosti (2012) for

Italy, Ariu (2016) for Belgium and Walter and Dell’mour (2010) for Austria. All of these

studies concur that service traders share many common features with goods traders in terms

of export participation patterns, exports distribution and firm characteristics. In our analysis,

we make use of similar firm-level data for Belgium and build on these studies in the choice

of firm-level control variables. Finally, by considering the production process as a combina-

tion of different tasks, our paper is related to the frameworks developed by Grossman and

Rossi-Hansberg (2008) and Baldwin and Robert-Nicoud (2014) for international trade and

Levy and Murnane (1996), Autor et al. (2003), Spitz-Oener (2006) and Autor and Acemoglu

(2011) for labour.

The structure of the paper is as follows. In Section 2, we describe the data and the main

variables we use. Section 3 provides some key facts about trade in services in Belgium. In

©2016 John Wiley & Sons Ltd

SERVICE TRADE AND OCCUPATIONAL TASKS 1867