Exporting and Productivity: The Effects of Multi‐Product and Multi‐Market Export Entry
| Published date | 01 September 2015 |
| DOI | http://doi.org/10.1111/sjpe.12077 |
| Date | 01 September 2015 |
EXPORTING AND PRODUCTIVITY: THE
EFFECTS OF MULTI-PRODUCT AND
MULTI-MARKET EXPORT ENTRY
Jaan Masso* and Priit Vahter*
ABSTRACT
Empirical studies on the micro-level effects of exporting on productivity pay usu-
ally little attention to the potentially heterogeneous effects of the different modes
of export market entry. We show that multi-product export entry is associated
with higher post-entry productivity compared to other firms. This can imply sig-
nificant benefits from experimentation with different products. Our analysis is
based on detailed export data from full population of firms in Estonia, disaggre-
gated for each firm by export markets and individual products.
II
NTRODUCTION
Most of the empirical papers on learning-by-exporting at firm level have
focused on the effects of exporting in general rather than on the potentially
heterogeneous effects of different modes of entering export markets. These
papers seldom find evidence that exporting affects productivity of firms.
Instead, they almost always show selection into exporting based on productiv-
ity: that only firms with relatively high productivity start exporting (Bernard
and Jensen, 1999, 2004). See, for example, Wagner (2007) or Greenaway and
Kneller (2007) for an overview of the literature.
In this paper we argue that the shortage of evidence on learning-by-export-
ing may reflect the fact that empirical studies on the micro-level effects of
exporting on productivity pay usually little attention to firms’ different modes
of export market entry and that a closer examination of the mechanisms
underpinning the learning-by-exporting process may shed a different light on
this phenomenon. Specifically, in this paper we contend that multi-market and
multi-product export entry may be associated with stronger learning-by-
exporting effects. To examine this hypothesis we investigate the role of exten-
sive margins of trade at product and market level in the learning effects of
exporting.
Simultaneous entry into several markets and/or with several export prod-
ucts may have a greater effect on a firm’s productivity compared with entry
*University of Tartu
Scottish Journal of Political Economy, DOI: 10.1111/sjpe.12077, Vol. 62, No. 4, September 2015
©2015 Scottish Economic Society.
325
into a single foreign market and/or with one export product. This is, partly,
because there is more scope for learning and the transfer of knowledge when
several foreign partners are involved.
In our empirical analysis the distinction between multi- and single product
is made according to the number of products at CN (Combined Nomencla-
ture) 8-digit product code level exported during the first year of exporting. We
find evidence in accordance with potential learning effects of multi-product
export entry. Our empirical study is based on detailed product-, market- and
firm-level data on foreign trade of Estonia’s manufacturing industry from
1995 to 2003. The advantage of our dataset compared with data sets from
other countries –for example, those from US and Mexico in Bernard et al.
(2010) or Iacovone and Javorcik (2010) –is that it covers the full population
of exporters. Our empirical approach relies on the propensity score matching
(PSM).
II MULTI-PRODUCT,MULTI-MARKET EXPORT ENTRY
AND FIRM PERFORMANCE
The empirical literature that investigates the relationship between productivity
and exporting supports the implications from Melitz (2003) or other heteroge-
neous firm trade models: that there are sunk costs associated with entry, and
only the most productive firms are able to cover these sunk costs and export
their goods (see, for example, Clerides et al., 1998; Bernard and Jensen, 1999,
2004; Bernard et al., 2003; Wagner, 2007). Also, papers like Roberts and Tyb-
out (1997) and Aitken et al. (1997) show that various other key characteristics
of producers matter for exporting. For example, Roberts and Tybout (1997)
find that size, age, and the ownership structure are positively associated with
export propensity. Aitken et al. (1997) present evidence that size, wages and
foreign ownership are strong determinants of the decision to export.
The causal relationship between productivity and exporting can of course
run from exporting to the subsequent increase in performance: firms may
learn as a result of exporting (Bernard and Jensen, 2004; De Loecker, 2007).
Exporters could benefit from the transfer of technology from their foreign cli-
ents and they might need to upgrade the quality
1
of their product to better
match the demands of their international clients. Improvements in productiv-
ity may also simply stem from the scale effect of a larger market (Falvey and
Yu, 2005). The effects may also be the result of exposure to tougher interna-
tional competition, which can increase the incentives of exporters to innovate,
to reduce managerial slack and in general to reduce X-inefficiency at the
firm.
2
1
This quality change may be reflected in the value added and consequently also in the
standard proxies for TFP of the firm.
2
For example, the effects of competition on the performance of firms and on their incen-
tives to innovate are discussed in detail in Aghion and Griffith (2005). The effects of competi-
tion on the work effort of employees and on managerial or worker slack are discussed in
Vickers (1995). The effects on X-inefficiency of firms in general are discussed in Leibenstein
(1966).
326 JAAN MASSO AND PRIIT VAHTER
Scottish Journal of Political Economy
©2015 Scottish Economic Society
To continue reading
Request your trial