Social Capital and Economic Growth in Europe: Nonlinear Trends and Heterogeneous Regional Effects
| Date | 01 October 2016 |
| Author | Jesús Peiró‐Palomino |
| DOI | http://doi.org/10.1111/obes.12131 |
| Published date | 01 October 2016 |
717
©2016 The Department of Economics, University of Oxford and JohnWiley & Sons Ltd.
OXFORD BULLETIN OF ECONOMICSAND STATISTICS, 78, 5 (2016) 0305–9049
doi: 10.1111/obes.12131
Social Capital and Economic Growth in Europe:
Nonlinear Trends and Heterogeneous Regional
Effects*
Jes ´us Peir ´o-Palomino†
†Department of Applied Economics II, University of Valencia, Av. dels Tarongers, s/n
Eastern Department Building E-46022, Valencia, Spain (e-mail: jesus.peiro@uv.es)
Abstract
Most of the literature focused on the social capital-growth nexus has followedthe one-size-
fits-all approach, neglecting that regional disparities might modify this relationship. This
article analyses the role of two social capital indicators on the growth of 237 European
regions in the period 1995–2007 byimplementing non-parametric regression, which relaxes
the linearity assumption and allows the data to give the underlying functional form. The
results show that social capital effects on growth are nonlinear. Parameter heterogeneity
could also be examined, showing heterogenous effects across regions and over time. In
particular, social capital is mostly negative in regions from Eastern and Central Europe
during the first years of transition from socialism to market economies. However, this
pattern has changed in the more recent period.
I. Introduction
The theory of social capital has received a great deal of attention in the economic growth
literature in relativelyrecent times (see, for instance, the seminal contributions by Coleman,
1988; Putnam, 1993; Knack and Keefer, 1997) and more recent analysis such as Beugels-
dijk andVan Schaik (2005), Bjørnskov (2012) or Forte,Peir ´o-Palomino and Tortosa-Ausina
(2015). Many definitions of social capital have been proposed, most of them closely re-
lated and sharing the same intuitions. In a broad sense, however, social capital might be
understood as a set of informal forms of institutions and organizations based on social
relationships, networks and associations that create shared knowledge, mutual trust, social
norms and unwritten rules (Durlauf and Fafchamps, 2005). Therefore, social capital is not
a unitary concept, but a combination of multiple coexisting facets that capture different
aspects (see Bjørnskov, 2006). Consequently, the indicators used as proxy variables for so-
JEL Classification numbers: C14; R11; Z13
*I am grateful to Jeffrey Racine for kindly providingsuppor t with the technical part and to EmiliTortosa-Ausina,
four anonymous referees and the editor, James Fenske, for their helpful comments and suggestions. The financial
support of the Ministerio de Ciencia e Innovaci´on (ECO2011-27227 and ECO2014/55221-P) as wellas the Generalitat
Valenciana (VALi+d ACIF 2080/2011 and PROMETEO II/2014/053) is also gratefully acknowledged. The usual
disclaimer applies.
718 Bulletin
cial capital are manifold. While the most common proxy is a measure of social trust,1the
list of proxies includes other indicators such as indices of associational activity in society,
the quality of social norms and measures of political and social participation.
Theory suggests that social capital stimulates economic growth through a variety of
channels. Forinstance, Putnam (1993) argued that social capital facilitates coordination and
cooperation for mutual benefit and it helps in solving problems of collective action. Social
capital reduces monitoring costs (Knack and Keefer, 1997) and facilitates complex agree-
ments by mitigating information asymmetries (Dearmon and Grier, 2009). These might
be considered as direct effects of social capital which reduce transaction costs (Bjørnskov
and M´eon, 2013). Following these authors, apart from the direct effects, there are two main
mechanisms through which social capital (in particular social trust) effects on economic
development are channelled. These are (i) education and (ii) the quality of the legal and
bureaucratic institutions.2Considering the first channel, Coleman (1988) suggested that
students endowed with more social capital had lower drop-out risk. Similarly, Bjørnskov
(2009) claims that trust might also affect the demand for higher education, since high-trust
employees are better at cooperating and require less monitoring, and therefore companies
will prefer hiring a more educated labour force.
Regarding the links between social capital and institutional quality, it is argued that in
high-trust societies, politicians and government officials are likely to be more trustworthy
and less prone to take advantage of their positions for personal benefit. Moreover, accord-
ing to Boix and Posner (1998), social capital generates societies better able and willing to
monitor public officials. They also suggest that social capital might help in striking deals,
as trust allows political promises that include payoffsto be effective in future periods. In the
same vein, more recent contributions such as Heinemann and Tanz (2008) hold that social
capital facilitates institutional reforms. In addition, the environment of mutual trust, partic-
ipation and cooperation created where social capital is abundant stimulates other activities
which, in turn, are related to higher growth. These are, for instance, financial development
(Guiso, Sapienza and Zingales, 2004), investment (Zak and Knack, 2001; Dearmon and
Grier, 2011; Peir´o-Palomino and Tortosa-Ausina, 2015), innovation processes (Ak¸comak
and Ter Weel, 2009) or trade (Guiso, Sapienza and Zingales, 2009).
A number of contributions have focused on the European regions, where evidence is
still comparatively scant and results mixed,since some of them are contrary to both the pre-
dictions of the social capital theory and previous empirical results for other geographical
contexts. In particular, in contrast to most of the literature at the country level Schneider,
Plumper and Baumann (2000) found a negative effectfrom social tr ust to growth.Beugels-
dijk and Van Schaik (2005) considered 54 (NUTS 1)3regions and two indicators of social
capital, namely social trust and associational activity and found that only the latter are
1Social trust is a particular form of social capital but it is so widely used that in most studies the terms ‘social
trust’ and ‘social capital’are used interchangeably.
2Social capital has been related to a wide variety of socioeconomic phenomena. Forinstance, another strand of the
literature has explored the relationship between social capital and crime (see Glaeser and Shleifer, 2002; Ak¸comak
and ter Weel,2012).
3NUTS stands for Nomenclature ofTerritorial Units for Statistics. It is a hierarchical system for dividing up EU eco-
nomic territory for the collection, development and harmonization of EU regional statistics and their socioeconomic
analysis.
©2016 The Department of Economics, University of Oxford and JohnWiley & Sons Ltd
Social capital and economic growth in Europe 719
related to higher growth. Ak¸comak andTer Weel (2009) analysed the effect of social trust
on innovation in 102 regions, finding a significantlypositive link. However,they suggested
that trust and growth are not directly related. In all these studies, however, the underlying
sample is quite homogenous. According to Beugelsdijk, De Groot andVan Schaik (2004),
in samples where the spatial units are heterogeneous in terms of social capital a positive
effect of trust on growth is more likely to be found. In this vein, Forte et al. (2015) test
different social capital indicators on a more heterogeneous sample of 85 NUTS 1 regions
in the period 1995–2008 and found that social trust and active participation in associations
are related to higher growth, whereas no effects werefound for the quality of social nor ms.
In contrast to most of the previous literature, this paper devotes special attention to
Eastern and Central European (ECE) regions.4Their consideration is important for various
reasons. On the one hand, social values in ECE regions differ from those of their Wester n
peers. Some authors such as Rose (2000), Paldam and Svendsen (2001) and _
Zukowski
(2007) suggest this is a consequence of the long communist experience, which modified
social patterns and negatively affectedsocial capital, which today lags behind the European
average. Authors such as Fidrmuc and G¨erxhani (2008) defend that this backwardness is
a result of poor institutions and low economic development in transition economies and
therefore it can be expected that social capital will increase when development comes.
However, others such as _
Zukowski (2007) are more skeptical and claim that cultural values
in ECE regions are well ingrained in society and have remained remarkably stable over
time.5In addition, social capital in these transition economies may adopt negative forms,
as suggested by Paldam and Svendsen (2001) and Rose (2000). Given the impossibility of
the state and public institutions coveringpeople’s basic necessities, corruption and negative
forms of organizations created to fulfill their own interests have proliferated, generating
harmful effects for the general interest. Followingthese authors, these negative effects could
be particularly relevantduring the transition years. Therefore, considering that these regions
are now progressivelycatching-up with Western Europe in severalaspects, a negative effect
of social capital would slow down this process.6
On the other hand, major disparities between ECE and Western European regions
have emerged in the evolution of income per capita in recent years. In particular, the
former have grown faster than the latter (see Crespo-Cuaresma, Doppelhofer and Feld-
kircher, 2012), thus confirming the predictions of the neoclassical growth theory, which
4With the exception of Forte et al. (2015), these regions have traditionally been disregarded in previous social
capital studies for the European context.
5The stability of social values has become a recent focus of debate. Contributions such as Uslaner (2008) and
Nunn and Wantchekon(2011) have shown that social capital (in particular social trust) is inherited from generation
to generation. For instance, whereas social capital is lowin ECE countries, it is historically high in Nordic countries.
This becomes an important argument when establishing causality betweensocial capital and economic growth. While
the former has remained stable at least since the WorldWar II, the latter has undergone remarkable changes.Another
strand of the literature of the literature (see, for instance Rainer and Siedler, 2009; Heineck and S¨ussmuth, 2013)
highlights the stability of social trust against institutional changes using the German reunification after the fall of
the Berlin Wall as a natural experiment. Therefore, the idea that social capital is a result of a given economic or
institutional scenario is difficult to reconcile (Bjørnskov and M´eon, 2013).
6Despite the literature provides both theoretical and empirical argumentshighlighting the stability of social capital
even during turbulent periods, it is worth mentioning that great transition processes such as the one experienced by
the Eastern and Central European regions might have some impact on social values.It is perhaps too early to evaluate
the effect of the transition on social capital but we cannot disregard the possibility that social capital levelsmight be
modified to some extent due to the global and profound changes that ECE regions are undergoing.
©2016 The Department of Economics, University of Oxford and JohnWiley & Sons Ltd
To continue reading
Request your trial