Copyright © 2010 John Wiley & Sons, Ltd and ERP Environment
* Correspondence to: Peter Lund-Thomsen, Associate Professor, Centre for Business and Development Studies, Centre for Corporate Social
Responsibility, Copenhagen Business School, Porcelænshaven 18A, 2000 F, Denmark. E-mail: plt.ikl@cbs.dk
Business Strategy and the Environment
Bus. Strat. Env. 19, 1–13 (2010)
Published online in Wiley InterScience
(www.interscience.wiley.com) DOI: 10.1002/bse.670
Global Value Chains, Local Collective Action and
Corporate Social Responsibility: a Review of
Empirical Evidence
Peter Lund-Thomsen
1
* and Khalid Nadvi
2
1
Centre for Business and Development Studies, Centre for Corporate Social Responsibility,
Copenhagen Business School, Denmark
2
Institute for Development Policy and Management, School of Environment and Development,
University of Manchester, Manchester, UK
ABSTRACT
A key debate in the corporate social responsibility (CSR) literature is the tension between
global pressures and local responses. Developing country suppliers often grumble that CSR
compliance adds costs. Yet, local collective action, articulated through industry associa-
tions, can potentially reduce costs and promote local embeddedness of CSR initiatives.
Through case study analysis, this paper considers how demands for CSR compliance
prompted collective action responses in selected developing country export industries. We
argue that differences in collective responses can be partially explained by how local export
industries are inserted into global value chains. We distinguish between ‘highly visible’
value chains, led by internationally well known brands as lead fi rms, and relatively ‘less
visible’ chains, where external CSR pressures come from a variety of sources, including less
dominant lead fi rms, international/national regulatory frameworks and national media. This
differentiation suggests a possible trade-off between the independence and the embedded-
ness of collective CSR initiatives. Copyright © 2010 John Wiley & Sons, Ltd and ERP
Environment.
Received 6 February 2009; revised 7 July 2009; accepted 16 November 2009
Keywords: global value chains; collective action; industrial associations; industrial clusters; corporate social responsibility;
developing countries
Introduction
A
N IMPORTANT TREND IN THE LITERATURE ON CORPORATE SOCIAL RESPONSIBILITY (CSR) AND INTERNATIONAL
development is the focus on how small and medium-sized enterprises (SMEs) in the developing world
are increasingly confronted with pressures to comply with international labour standards and CSR norms
as a pre-condition to access developed world markets. Proponents of CSR codes emphasize that compli-
ance can improve business processes, increase transparency in global supply chain operations and ratchet up social
2 P. Lund-Thomsen and K. Nadvi
Copyright © 2010 John Wiley & Sons, Ltd and ERP Environment Bus. Strat. Env. 19, 1–13 (2010)
DOI: 10.1002/bse
and environmental standards in the developing world (Jenkins et al., 2002). Yet, these codes, which are usually
based upon internationally agreed labour standards, are also criticized for being formulated in corporate headquar-
ters in Europe or North America without any signifi cant input from the intended benefi ciaries, suppliers and
workers, in the Global South (De Neve, 2009). Individual global brands insist that their suppliers follow their
particular codes of conduct. This often results in local producers facing multiple, costly, and at times contradictory,
CSR requirements and audits (Lund-Thomsen, 2008). Consequently, there have been moves towards greater
convergence in CSR codes within and across industries (Nadvi, 2008). However, it is unsurprising that many
developing country producers feel little ownership for the CSR standards that their Northern buyers press upon
them. This raises doubts about the long-term sustainability of social compliance, especially when sustained pres-
sure from international buyers is absent (Barrientos, 2008).
In the context of the debate on global pressures and local responses around CSR codes and international labour,
environmental and ethical standards, there have been greater efforts at empirically assessing the gains from com-
pliance, especially on working conditions and workers’ lives (see Barrientos and Smith, 2007; Locke et al., 2007;
Locke and Romis, 2007). At the same time, calls for a more South-centred CSR agenda that maps out existing
CSR practices in developing countries as they are defi ned by Southern fi rms, workers and communities are also
growing (see Vives, 2006; Singh et al., 2008). A relatively important but under-researched subject is the role that
local collective action, often articulated through local industry associations, can play in ensuring meaningful com-
pliance with international labour standards and codes of conduct in the developing world (Nadvi, 2008; Lund-
Thomsen and Nadvi, 2009). Local collective action could potentially lower costs of compliance for local producers,
improve collective monitoring of codes and more effectively embed the social and environmental goals associated
with CSR codes into local practices and norms.
In recent years, the use of multi-stakeholder partnerships (that bring together public and private actors) have
been seen as a possible way forward in promoting greater local ownership and participation of local stakeholders
in the formulation, implementation and monitoring of CSR standards within global value chains (Dolan and
Opondo, 2005; Henkle, 2005; ETI, 2006). This might provide greater scope for coordination and harmonization
of codes amongst buyers, thereby reducing the need for individual brands to undertake their own audits. It might
also facilitate greater local embeddedness through the involvement of local industry associations in the industry-wide
formulation, implementation and monitoring of CSR standards (O’Rourke, 2006; Nadvi, 2008).
Industry and business associations can have an important function in formulating local collective action by
SMEs, and thereby promoting local economic development (Doner and Schneider, 2000; Nadvi, 1999a). However,
the involvement of local industry associations in facilitating compliance with international CSR standards has not
been the subject of much debate in the CSR and development-oriented literature. One source of inspiration is the
now extant body of empirical studies on developing country industrial clusters. Since the 1990s there has been a
growing literature on industrial clusters, demonstrating how local industrial competitiveness in the developing
world often came from small- and medium-sized fi rms agglomerated in the same geographical region (Schmitz
and Nadvi, 1999). In short, the literature highlights the ability of small fi rms to overcome the barriers they face
individually in relation to accessing international markets (Schmitz and Nadvi, 1999; Humphrey and Schmitz,
2004). Agglomeration economies and the potential for joint action are the critical elements of collective advantage
within clusters. Collective action can improve the capacity of local producers to confront external challenges and
enhance competitiveness (Schmitz, 1999; Nadvi, 1999b; Knorringa, 1999; Bair and Gereffi , 2001). Local industrial
associations are often at the heart of such collective action (Nadvi, 1999a). Yet, even in this literature, few studies
explicitly address the question of how external pressures in the form of demands for CSR compliance might trigger
collective action responses from industry associations in the developing world.
1
1
There are some partial exceptions to this. Nadvi (1999b) details the case of collective action responses by clustered producers to external
pressures on compliance with international quality assurance standards, but the study does not provide much detail on CSR standards.
Mezzadri (2008) studies CSR in the Delhi Garment cluster. However, she does not explicitly address the question of how external pressures
on CSR compliance provoked collective action responses. There are a number of recent policy studies. UNIDO, for example, produced a range
of studies that map local CSR practices in clusters in India, Senegal and Ecuador (Accountability (w. UNIDO), 2006; ASK, 2007; Krishnadas
et al., 2007; Sodhi, 2007; Sachdeva and Panfi l, 2008; Rana and Singh, 2008). Similarly, the ILO describes collective responses to CSR pres-
sures in the cocoa sector (ILO-IPEC, 2007a, 2007b). However, this policy-oriented literature does not contain in-depth analysis of external
CSR pressures and local collective action cluster responses.
Global Value Chain, Local Collective Action and Corporate Social Responsibility 3
Copyright © 2010 John Wiley & Sons, Ltd and ERP Environment Bus. Strat. Env. 19, 1–13 (2010)
DOI: 10.1002/bse
In this paper, we make an attempt at assessing the limited empirical evidence to date for the potential role that
developing country industry associations may play in facilitating collective action responses by export oriented
SME producers to CSR pressures originating from the global value chains into which local fi rms are inserted. In
our view, the nature of CSR pressures emanating from the value chain can be varied. We distinguish between
what we term ‘highly visible’ and ‘less visible’ chains. Highly visible chains are those where scrutiny by the global
media and non-governmental organizations (NGOs) has been extensive, often because the chain’s lead fi rms are
internationally well recognized global brands. For such fi rms, brand identity, brand value and brand risks are of
paramount concern for both its global marketing and global production organization strategies. Such global brands
are often the targets of international and local advocacy groups and the media. Thus, they place great emphasis,
and costs, on the ability of local suppliers to comply with their CSR codes in order to mitigate risks to brand
integrity. In highly visible chains compliance with social and environmental concerns are central to the organiza-
tion and governance of the chain and the relationships between local suppliers and global lead fi rms. In contrast,
‘less visible’ chains are those where external CSR pressures are driven by a wider set of actors/institutions and/or
less dominant global lead fi rms. On the one hand, less visible chains may include smaller or medium-sized branded
buyers less capable of enforcing their CSR demands throughout the chain. On the other hand, external CSR pres-
sures in less visible chains may also refl ect a mixture of international/national regulatory frameworks and national
media as important CSR drivers in the chain.
In light of this distinction, we analyse seven case studies that are selected using three criteria.
(i) Each of the case studies involves joint industry association CSR responses to external CSR pressures at
national, regional or local (cluster) levels.
(ii) They refl ect the differential integration of developing country export industries into global value chains. Some
of the case studies involve insertion in the ‘highly visible’ value chains of leading global brands, while others
are in comparatively ‘less visible’ chains.
(iii) The case studies are suffi ciently well described in the academic or policy literatures for us to analyse
them.
Table 1 provides an overview of these case studies.
In Table 1, we observe that the four case studies of highly visible chains (garments, Bangladesh; cut fl owers,
Kenya; garments, Cambodia; soccer balls, Pakistan) refl ect a largely similar pattern in terms of the initial external
pressure and the subsequent collective action response. In all these cases, international media exposure of child
labour or labour rights violations in the value chains of internationally branded companies threatened access to
global markets. The fi nal three case studies, the Tiruppur dyeing and bleaching sub-cluster, the Palar Valley tannery
cluster and the Jalandhar soccer ball cluster, are examples of joint action responses to external pressures in com-
paratively less visible chains. In these cases, CSR pressures came from a variety of sources including GVCs,
international/national regulatory frameworks and national media.
In the next two sections of the paper, we analyse the case studies from both ‘highly visible’ and ‘less visible’
chains, to assess how local export industries have formulated through their industry associations similar and dif-
ferent collective action responses to external CSR pressures. We are keen to observe whether the nature of the
chain in which local producers are inserted is likely to impact on the types of local collective action response. We
show that the differences observed can be partially explained with reference to how local producers are tied into
GVCs. Our fi ndings also suggest that there may be a trade-off between maintaining the independence of local
collective action CSR initiatives and ensuring their local embeddedness.
External Pressures and Joint CSR Responses: Highly Visible Chains
The recent literature on standards in global value chains has emphasized the power of lead fi rms in defi ning
standards and codes of conduct, as well as the ways in which CSR pressures can alter the nature of governance
within the value chain (Altenburg, 2006; Dolan and Humphrey, 2004; Gereffi et al., 2005; Hughes, 2005; Hum-
phrey and Schmitz, 2004; Humphrey, 2008; Nadvi, 2008; Ponte and Gibbon, 2005; Ponte, 2007). The interplay
between local – national, regional and/or cluster-based – governance and global – value chain based – governance
4 P. Lund-Thomsen and K. Nadvi
Copyright © 2010 John Wiley & Sons, Ltd and ERP Environment Bus. Strat. Env. 19, 1–13 (2010)
DOI: 10.1002/bse
Highly visible chains Less visible chains
Garment sector,
Bangladesh
Garment sector,
Cambodia
Cut fl ower sector,
Kenya
Pakistan soccer
ball cluster
Palar Valley
tannery
cluster, India
Tiruppur
bleaching and
dyeing cluster,
India
Jalandhar soccer
ball cluster,
India
External
pressure
1993
International media
reports, Harkin
Bill
1998
Threat of boycott,
reports on
poor work
conditions
2002 Reports on
ETI code
violations, poor
work conditions
1995 International
media reports,
threat of US
boycott
1996 Indian
Supreme
Court
ruling
1996 Madras
High Court
ruling
1997
International
media reports
Industry
response
1996
Memorandum of
Understanding
2000
Better Factories
Programme
2002
Horticultural
and Ethical
Business
Initiative
1997
Atlanta
Agreement
1996
Common
Effl uent
Treatment
Plants
1996
Common
Effl uent
Treatment
Plants
1999
‘Indian’ Atlanta
Agreement
Table 1. Collective industry responses to external pressures: CSR case studies
Global Value Chain, Local Collective Action and Corporate Social Responsibility 5
Copyright © 2010 John Wiley & Sons, Ltd and ERP Environment Bus. Strat. Env. 19, 1–13 (2010)
DOI: 10.1002/bse
can be critical in determining the ability of local fi rms to collectively respond to external CSR pressures (Nadvi,
2008).
The fi rst case of a highly visible chain that we review is the Bangladesh garments industry. Leading global apparel
brands as well as branded retailers and multiples have been directly sourcing garments from fi rst tier suppliers
in Bangladesh since the mid-1980s (Bhattacharya, 2000; Mostafa, 2008). Bangladesh has also been a focus for
many advocacy groups concerned with working conditions in labour intensive garment production. In 1993, the
Harkin Bill passed in the United States Senate called for a ban on imports into the United States (US) of goods
manufactured or fabricated by child workers. The Bangladeshi garment industry responded to this by dismissing
child workers en masse. Subsequently, a memorandum of understanding was signed between the International
Labour Organisation (ILO), the United Nations Children Fund (UNICEF) and the Bangladesh Garment Manufac-
turers and Exporters Association (BGMEA) aiming to eradicate child labour from the industry by introducing a
child labour monitoring system, and providing social protection for affected children (Nielsen, 2005; ILO and
UNICEF, 2004).
Our second case study of a highly visible chain is the Kenyan cut fl ower industry, which supplies leading inter-
national retailers, including major branded supermarkets in Europe. Allegations of labour rights abuses on farms
prompted members of the UK’s Ethical Trading Initiative (ETI) to visit Kenya in 2002. At the same time, local
Kenyan stakeholders, fearing the loss of their most important market, organized themselves with the support of
the ETI and UK-based NGO Women Workers Worldwide (WWW) to form the Horticultural and Ethical Business
Initiative (HEBI) (Dolan and Opondo, 2005). HEBI is an independent non-profi t organization, which seeks to
improve ethical and social behaviour in the Kenyan horticulture and fl oriculture industry. In 2002 HEBI set out
to investigate the claims of labour rights abuses. It initially undertook a review of other codes. It then developed
a base code for its members that refl ected Kenyan law, formulated a methodology that could prove malpractices,
and trained social auditors in the use of participatory audit methodology of the new code (http://www.hebi.or.ke/,
accessed 9 August 2008).
Our third case is the Cambodian garments industry, which supplies major global brand retailers in the United
States under the US–Cambodia Free Trade Agreement (USBTA). A stipulation of the USBTA was that local pro-
ducers met international labour standards as a basis for preferential trade access to the US market. This prompted
the Cambodian government and the Cambodian Garment Manufacturers Association (CGMA) to initiate negotia-
tions with the US Department of Labor and the ILO. It is argued that this led to better labour rights monitoring
and improved labour standards in the Cambodian garments industry, thereby ensuring market access to the US
and enhancing the competitiveness of the Cambodian garment sector (Polaski, 2004, 2006).
Our fourth case is the Pakistani soccer ball industry, clustered in the town of Sialkot. It is the leading global
manufacturer of premium quality hand stitched footballs for the major sports goods brands, including Nike and
Adidas. In 1995 it faced international media allegations of extensive child labour in football stitching activities. A
complex negotiation ensued between international brands, UN organizations, NGOs and the local industry, rep-
resented through the Sialkot Chamber of Commerce (SCCI). In 1997, these parties signed the so-called ‘Atlanta
Agreement’. The agreement envisaged the development of an industry-wide monitoring mechanism that could
identify child workers and transfer them from football stitching to schools while creating social safety nets for
their families to make up for the income lost as children were withdrawn from work (Schrage, 2004; Khan, 2007;
Rafi Khan, 2004, 2007; Nadvi, 2004, 2008; Seigmann, 2008).
A number of common themes emerge from the four case studies, as shown in Table 2. First, in all four cases
the pressures on compliance came from outside the local export industries, usually GVCs that are led by interna-
tionally dominating lead fi rms. These companies are vulnerable to NGO and trade union criticism (regarding child
labour or labour rights violations) as this affects their brand value. Second, in all four case studies, the exposure
of these brand buyers to media criticism, the brands’ demands for CSR compliance and the threat of US regula-
tion (in the Bangladesh, Cambodia and Pakistan cases) prompted a complex multi-stakeholder response involving
international and local actors. This was necessary to ensure access to markets in the United States and the EU.
Third, central to the responses in all four cases was collective action led by, and organized through, the respective
industry associations. Fourth, these responses emerged fairly rapidly in the four case studies. It took from one
year to two and a half years from when the initial pressure was experienced until joint action CSR responses were
formulated. Fifth, in all of the case studies, funding for local industry responses was provided by international