Migration and servicification: Do immigrant employees spur firm exports of services?
| Author | Andreas Hatzigeorgiou,Magnus Lodefalk |
| Published date | 01 November 2019 |
| DOI | http://doi.org/10.1111/twec.12838 |
| Date | 01 November 2019 |
3368
|
wileyonlinelibrary.com/journal/twec World Econ. 2019;42:3368–3401.
© 2019 John Wiley & Sons Ltd
Received: 14 July 2018
|
Revised: 29 March 2019
|
Accepted: 11 June 2019
DOI: 10.1111/twec.12838
ORIGINAL ARTICLE
Migration and servicification: Do immigrant
employees spur firm exports of services?
AndreasHatzigeorgiou1
|
MagnusLodefalk1,2
1The Ratio Institute, Stockholm, Sweden
2Department of Economics,Örebro University, Örebro, Sweden
Funding information
Jan Wallander and Tom Hedelius Research Foundation; Swedish Entrepreneurship Forum
KEYWORDS
trade, firms, migration, services
1
|
INTRODUCTION
Services account for nearly a quarter of total world exports when measured by conventional statistics
and as high as 53 per cent in value‐added terms (OECD, 2015; World Bank, 2019). Services account
for the major share of employment and value added, and this share is continually growing (World
Bank, 2019).1 However, exports of services are subject to substantially higher trade costs than are
merchandise exports (Lodefalk, 2016; Miroudot, Sauvage, & Shepherd, 2013). Thus, services exports
are increasing despite the current trade environment, not because of it.
Servicification is resulting in an increasingly important economic role of services. As countries’
dependence on services for production and employment increases, the need for policies to facilitate
services exports increases in tandem. Trade liberalisation and trade facilitation, however, have tradi-
tionally focused on measures that promote trade in manufactured goods. This, in combination with
recent pressures on the multilateral trade system and the multilateral or plurilateral approach to lower
trade barriers, increases the need for knowledge of how to find new and alternative ways to promote
services exports.
In this paper, we analyse the role of immigrant employees as facilitators of firm exports of services.
For this aim, we have acquired access to granular, exhaustive and longitudinal Swedish register data
at the employer–employee level. Sweden is the paragon of a small and open economy. However, the
country is also characterised by a high degree of servicification and by its ranking as one of the top
1 Firms purchase more services, produce more services and sell more services both domestically and abroad; that is, they
“servicify” (Lodefalk 2013; Lodefalk 2016). Several studies point to the importance of services for manufacturing firms and
their exports (e.g. Crozet and Milet, 2017; Lodefalk, 2014; Nordås, 2008; Pilat, 2005; Pilat et al., 2008).
|
3369
HATZIGEORGIOU And LOdEFALK
EU immigration countries, in terms of both the share of immigrant residents and asylum applications
per capita.2
Our focus on immigrant employees as facilitators of exports of services is motivated by the following
conjecture: an immigrant's effect on trade costs is country‐specific and likely to be strongest at the em-
ployer–employee level. Immigrants have information about and experience in specific markets, usually
their country of birth, which may facilitate trade in regard to those markets (e.g. Gould, 1994; Rauch,
2001). The intensity of the interaction between immigrant workers and their managers is governed by
geographical proximity through employment. This proximity enhances the volume and quality of knowl-
edge that immigrants transmit to their managers about other markets (Herander & Saavedra, 2005). We
test whether the transmission of such information through employment raises the probability that the firm
will export to the migrant's country of birth and affects the intensity of existing trading relationships.
To the best of our knowledge, this is the first study of the effect of immigrant employees on firms’
exports of services. In addition to exploring this dimension of international trade, this paper contrib-
utes to a careful investigation of the relationship between immigrant employees and services exports.
The findings have important policy implications. In a time when the multilateral approach to trade
liberalisation and facilitation is being challenged – in combination with the trend towards more restric-
tive migration – policymakers may need to find new ways to promote exports of services. This study
has the potential to contribute insights in this regard.
Calls for limiting migration held centre stage in the 2016 UK referendum on EU membership,
as well as in the 2016 US presidential election. In these circumstances, and as has been the case tra-
ditionally, discussions of migration have largely been about implications for the labour market and
public finances. However, as we will discuss in this paper, migration is relevant for trade in general
and services exports in specific.
To answer the questions of whether immigrant employees affect firms’ exports of services and
how this potential impact relates to the influence on merchandise exports, we exploit detailed em-
ployer–employee data from Sweden. Our data include approximately 1.8million full‐time employees
in nearly 30,000 Swedish private sector firms and allow us to directly connect information about the
countries to which firms export with workers’ countries of birth.
Questions about the effects of immigration on exports inevitably raise concerns about the presence
of confounding factors and alternative explanations for any observed correlation. The presence of
unobservable managerial or owner characteristics is a concern in our setting because these charac-
teristics can make the firm more open to international markets and more open to the employment of
immigrant workers. A more extreme version of this issue could arise if the same characteristics make
both exports and immigrant employment with respect to the same country more likely.
The richness of the data allows us to control for a wide range of confounding factors at the firm
and country levels that may explain a correlation between immigration and exports. In this paper, we
assume that these unobservable managerial or owner characteristics exist at the firm-partner-country
level but are time‐invariant, which allows us to identify the effects of immigration by exploiting the
panel dimension of the data. That is, we are able to ask whether the employment of more immigrant
workers from a particular country makes it more likely that the firm will export services or export more
intensively to that country, controlling for all time‐invariant characteristics of the firm that could affect
its decision to export to that country. While this shortens the list of potential confounding factors that
2 The immigrant stock in Sweden accounted for 12.9 and 15.3 per cent of the population in 2007 and 2014, respectively, and
in the 2014–2017 period, the country received asylum seekers corresponding to 2.7 per cent of the population, ahead of other
large recipients such as Germany and Denmark (EU, 2019; OECD, 2018). In previous eras, Sweden hosted a large immigrant
stock, and the country has received several waves of asylum seekers in the postwar period.
3370
|
HATZIGEORGIOU And LOdEFALK
might explain our main findings and controls for a range of time‐varying firm and country characteris-
tics, we cannot completely rule out their presence. We therefore examine the results using a wide range
of robustness tests and conduct difference‐in‐difference and instrumental variable (IV) analyses to anal-
yse the causal characteristics of the relationship between migration and exports of services.
The data allow us to probe the relationship between migrant employment and firms. If firms’ ser-
vices exports are affected by the employment of foreign‐born workers, it is important to know whether
this relationship rests upon the characteristics of immigrant workers or types of services or whether
the relationship is ubiquitous in character. It has been typical in the broader literature on migration
and international trade to assume that immigrants are a homogeneous group. This is something we can
investigate by adding information about the skill level of immigrants. Exploring the potential role of
skills is especially important in the policy context of the current refugee crisis, as well as in the general
debate about the costs and benefits of immigration, as in the 2016 US presidential election and the
debate leading up to the UK decision to exit from the EU.
A large share of recent refugees arriving in Europe lacks post‐secondary education. In the case of
Sweden, approximately 40 per cent of Syrian‐born immigrants have at most primary education, with
only approximately 28 per cent having secondary education (Statistics Sweden, 2014). How these
education levels alter the overall economic impact of immigration largely remains unresolved.
We furthermore analyse the possible influence of another important immigrant characteristic: the
amount of time since immigration. The theoretical literature on migration and trade identifies two
relevant and opposing mechanisms concerning time since immigration. The first mechanism suggests
that increasing time since immigration enhances the ability of immigrants to disseminate knowledge
and transform contacts into links that are useful to firms, thereby enhancing immigrants’ impact on
firm exports. The second mechanism postulates that over time, immigrants’ knowledge of foreign
markets becomes obsolete and ties with networks weaken, thereby reducing their impact on firm
exports of services.
The results of our analysis indicate that the employment of immigrants increases both the like-
lihood that firms export services to immigrant source countries and the value of services exports.
Our results suggest that firms can hire foreign‐born workers to spur services exports to immigrants’
source countries. Hiring one additional foreign‐born worker can increase services exports by approx-
imately 2.5 per cent on average. Although this is a conservative estimate, it comes with important
caveats for policy.
Our findings are conditioned upon employment in firms, as we find that the overall stock of immi-
grants in the country has a small influence on firms’ services exports. Thus, our results do not support
increasing immigration as such for the purpose of increasing services exports. Policymakers should
note that our findings on the promotion of services exports through immigration are confined to
skilled immigrants. We also find that the trade‐increasing impact of foreign‐born workers diminishes
with time since immigration.
The remainder of this paper is organised as follows. Section 1 describes the motivation for our
paper and the previous research. Section 2 presents the conceptual framework. Section 3 explains the
empirical approach and the estimation strategies. Section 4 describes the data and provides stylised
facts. Section 5 presents the results, and Section 6 concludes.
2
|
PREVIOUS RESEARCH
While trade and migration are viewed as substitutes in neoclassical economics (e.g. Massey et al., 1993;
Mundell, 1957), more recent research suggests that there is a positive relationship between migration
To continue reading
Request your trial