International Business Review 19 (2010) 126–139
Contents lists available at ScienceDirect
International Business Review
journal homepage: www.elsevier.com/locate/ibusrev
Global village vs. small town: Understanding networks at the Base
of the Pyramid
Miguel Rivera-Santos a,*, Carlos Rufı́n b,1
a
b
Babson College, Management Division, 231 Forest Street, Babson Park, MA 02457-0310, USA
Sawyer School of Business, Suffolk University, 8 Ashburton Place, Boston, MA 02108, USA
A R T I C L E I N F O
A B S T R A C T
Article history:
Received 25 July 2008
Received in revised form 4 June 2009
Accepted 4 July 2009
We compare and contrast business networks at the Base of the Pyramid (BOP) and at the
Top of the Pyramid (TOP), and analyze their implications for multinational enterprises
(MNEs). We first identify the specificities of BOP environments in terms of competitive
environment and institutions. Building on this analysis, we develop a series of
propositions regarding the impact of these specificities on the structural characteristics
of BOP networks, their boundaries, the characteristics of their ties, the diversity of their
partners, and their evolution over time, as compared to TOP networks. Our analysis
suggests that major differences exist between both types of networks along all dimensions
and that these differences have important implications for MNEs active in BOP
environments.
ß 2009 Elsevier Ltd. All rights reserved.
Keywords:
Base of the Pyramid/Bottom of the Pyramid
Multinational enterprises
Networks
Non-market
The concept of the ‘‘Base of the Pyramid’’ (BOP), which claims that business can help eradicate poverty (Prahalad &
Hammond, 2002; Prahalad & Hart, 2002), has sparked considerable interest in the business community around the world
(Simanis et al., 2005), even though multinational enterprises’ (MNEs) codes of conduct often do not reflect this growing
interest (Kolk & van Tulder, 2006). Most people on the planet live on low incomes relative to the expenses they incur to meet
basic needs, such as nutrition or health care. The premise of the BOP approach is that, by turning these people into customers,
multinational enterprises (MNEs) can bring prosperity to the poor, while, at the same time, finding new revenue sources
(Prahalad & Hart, 2002). Not only is the number of people at the BOP much larger than the number of people outside the BOP
(which, for simplicity, we call ‘‘Top of the Pyramid’’ or TOP in this paper), but BOP populations are still experiencing
significantly higher demographic and income growth than TOP populations (UNCTAD, 2006). As a consequence, even if
higher TOP incomes result in larger TOP market sizes at present, market growth rates can be expected to be significantly
higher at the BOP over the course of this century. At the same time, BOP initiatives can have an important impact on
economic development and poverty reduction. By integrating economically isolated people into the global economy, MNEs
can not only provide products and services at reduced prices but also create entrepreneurial opportunities for people at the
BOP (Prahalad & Hart, 2002). In other words, there is both a business and a moral case for the BOP.
BOP scholars have been much less successful at articulating a strong motivation for academic research on the BOP than for
applied research. One of the premises of the BOP literature is that conditions at the BOP make these environments
significantly different from TOP environments, and thus require major innovation on the part of companies—especially
MNEs—in order to develop successful strategies for BOP markets (Anderson & Markides, 2007; Dawar & Chattopadhyay,
2002; Hart & London, 2005; Ricart, Enright, Ghemawat, Hart, & Khanna, 2004). Yet, to the best of our knowledge, few scholars
* Corresponding author. Tel.: +1 781 239 5325; fax: +1 781 239 5272.
E-mail addresses: mrivera@babson.edu (M. Rivera-Santos), crufin@suffolk.edu (C. Rufı́n).
1
Tel.: +1 617 570 4897; fax: +1 617 994 6840.
0969-5931/$ – see front matter ß 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.ibusrev.2009.07.001
M. Rivera-Santos, C. Rufı´n / International Business Review 19 (2010) 126–139
127
have systematically applied the major analytical frameworks of international business and strategic management to explore
in what specific ways BOP environments and strategies contrast with their TOP counterparts and why these differences
should lead to innovative strategies at the BOP.
This paper takes a step in this direction by focusing on one of the major claims of the BOP literature, namely that business
and social networks at the BOP differ in important ways from their counterparts at the TOP, and that this has important
consequences for firms entering BOP markets (Chesbrourgh, Ahern, Finn, & Guerraz, 2006; Seelos & Mair, 2007; Simanis
et al., 2005; Wheeler et al., 2005). We build on and extend studies that call attention to the role of networks at the BOP,
particularly London and Hart (2004). Using an inductive approach to generate theory about the BOP based on the analysis of
24 in-depth case studies, London and Hart find that MNEs that are successful in BOP markets possess, among others, specific
capabilities in social embeddedness and network building with non-traditional, local partners. Because their approach is
exploratory and inductive, however, they do not fully analyze the causes and dynamics of BOP-specific network building and
how these processes may differ from network building in TOP markets. Applying the major analytical frameworks of
international business and strategic management and extending previous case-based and inductive studies, we
systematically explore the differences between TOP and BOP networks, the reasons for these differences, and their
implications for MNE activity at the BOP, including the processes and capability needs identified by London and Hart.
The aim of this paper is therefore to answer two interrelated research questions: In what systematic ways, if any, are
networks at the BOP different from networks at the TOP? What are the implications of these differences, if any, for MNEs? By
grounding our reasoning in analytical frameworks widely used in international business and strategic management, we hope
to provide greater conceptual rigor to the analysis of BOP networks, and, more broadly, to systematically identify key
strategic factors that MNEs will need to address in order to grow during the next decades, given the relative growth and sheer
demographic weight of BOP populations worldwide. Our work also contributes to research on business networks, by
characterizing a largely unexplored type of network.
We proceed in two steps. First, in order to identify the factors that may lead to differences in BOP and TOP networks, we
explore the specificities of BOP environments, grounding our reasoning in industrial organization (I/O) and institutional
theory. We then derive from this analysis the implications for BOP networks and for MNEs active at the BOP, grounding our
reasoning in network theory. Specifically, we analyze the differences between BOP and TOP networks in terms of structural
characteristics, boundaries, ties, partner diversity, and dynamics, following the literature on networks (e.g., Provan, Fish, &
Sydow, 2007; Wasserman & Faust, 1994) as our framework of analysis.
1. Specificities of the BOP: competitive and institutional environments
To date, there has been no systematic and theory-driven attempt at explaining the link between BOP structural
characteristics and BOP business models, in spite of much anecdotal evidence suggesting that BOP business models are
different because of specific characteristics at the BOP (Simanis et al., 2005; Wheeler et al., 2005). In other terms, while BOP
scholars have made a strong case regarding the specificities of BOP business models, there has been no attempt, to the best of
our knowledge, to use IB theories to systematically examine what characteristics of BOP environments lead to specific
aspects of BOP business models. In this section, we draw on I/O and institutional theories to explore how the BOP and TOP
markets differ in their competitive structure and their institutional environment. We then derive a list of specificities of BOP
markets relevant to BOP networks. It is important to note that this represents an early attempt at applying theory to the
study of the BOP, and that, while we acknowledge that important variations exist across both BOP markets (see Simanis et al.,
2005) and TOP markets (see, for instance, Porter, 1980), we focus on theoretical attributes common to all BOP and TOP
environments, respectively, to develop our propositions. A graphical representation of our analysis of BOP environments can
be found in Fig. 1.
1.1. The competitive environment at the BOP
BOP scholars typically distinguish BOP from TOP markets by customer incomes and geographic location (Prahalad, 2005).
While the exact definition varies, all BOP scholars view the extremely low level of income, usually set around $2/day
(Karnani, 2007; Prahalad, 2005; Whitney & Kelkar, 2004), as the major characteristic of BOP customers. Equally important,
although less often discussed, however, is the irregularity of BOP incomes, as people living at the BOP typically cannot predict
their revenue, even in the short run (Dawar & Chattopadhyay, 2002; Johnson, 2007). A third characteristic identified in the
literature is that BOP populations may be either geographically dispersed—the rural poor (e.g., Anderson & Markides, 2007)—
or live in densely populated areas—slum dwellers in major conurbations (e.g., Johnson, 2007). In both cases, BOP markets
tend to be relatively isolated (Arnould & Mohr, 2005), typically leading to strong local cultures and less contact with national
or international consumer habits.
These characteristics have important implications for business at the BOP. Specifically, they lead to smaller product sizes
and local adaptation. Smaller sizes make products more affordable by providing low price points relative to average incomes
(Karnani, 2007), so that customers do not have to disburse a large sum relative to their income. BOP markets also require a
distribution network that is not only designed for very frequent purchases in difficult-to-reach areas, but also strongly
adapted to local specificities in response to the relative cultural isolation of BOP markets (Anderson & Markides, 2007;
Hammond & Prahalad, 2004).
128
M. Rivera-Santos, C. Rufı´n / International Business Review 19 (2010) 126–139
Fig. 1. The BOP environment: competitive and institutional environments.
Beyond these well-known customer characteristics, differences between BOP and TOP markets can also be found in the
competitive environment. Local firms often play an important role at the BOP. While mostly informal, few in number, small,
and offering low-quality products at high price points relative to customer incomes (London & Hart, 2004; Prahalad &
Hammond, 2002), local BOP firms may actually be very strong if they are embedded in the informal environment (discussed
in more detail below) and linked to local powers (De Soto, 2000). Thus, firms entering BOP markets do not actually ‘‘compete
against non-consumption’’ (Hart & Christensen, 2002) but have to face local players embedded in the informal institutions of
the BOP. Likewise, BOP entrants may find that the suppliers, distributors, or complementors they take for granted in TOP
markets, do not exist at the BOP. In other terms, the competitive environment at the BOP is characterized by gaps in the value
chain (Anderson & Markides, 2007; Wheeler et al., 2005), including gaps in the economic infrastructure, such as electricity or
water supply, in support activities, such as financing or distribution, and in the information infrastructure.
1.2. The institutional environment at the BOP
Beginning at least with North (1990), many economists blame weak or incomplete institutional environments for low
levels of development. Weak institutional environments do not provide the support necessary to promote economic activity
(UNCTAD, 2005, 2006), so firms in these environments have to adapt their structure and strategies to fill the institutional
gaps (Khanna & Palepu, 2000). Weak institutions create particularly acute problems in BOP markets. Enforcement of laws
and regulations at the BOP is typically low, leading to informal dispute resolution (Ricart et al., 2004) and non-existent
protection for workers and consumers (Karnani, 2007). Tax evasion, corruption, and clientelism are prevalent (Transparency
International, 2007). Finally, property rights may not exist, may depend on traditional community norms, or may not be
enforced (De Soto, 2000; London & Hart, 2004).
Formal institutions, which refer to the existence of legally valid and enforceable norms, statutes, or regulations, including
legally enforceable private agreements (contracts), or to the compliance with such norms and agreements, are only one
component of the institutional environment, though, and prior research suggests that weak formal institutions lead to
reliance on informal ones in BOP environments (De Soto, 2000; London & Hart, 2004). Informal institutions refer to norms
that have no legal validity (although they may have customary validity), or to activities that do not comply with formal rules
(e.g., tax evasion) (London & Hart, 2004; Portes, 1994). Strong traditional ties within communities (such as kinship, religion,
or race), replace more formalized institutions (Arnould & Mohr, 2005; London & Hart, 2004), even when informal institutions
contradict formal ones (Arnould & Mohr, 2005; Johnson, 2007). As a result, transactions are governed by relationships and
networks rather than by contracts. For an outsider, building legitimacy and trust, and becoming a network member is both a
necessity and a challenge (Wheeler et al., 2005), as the strength of ties within communities is paralleled by deep-rooted
divisions, mistrust, and potential conflicts between communities. The link between communities and the outside is often
created by intermediaries such as local NGOs or influential members of the community (Arnould & Mohr, 2005). Hence, local
firms may be competitively weak but institutionally strong, as they are embedded in local informal networks.
The preceding comparison shows that the characteristics of the BOP competitive and institutional environments have
important implications for new entrants and, more generally, for the way business is conducted at the BOP. In the next
M. Rivera-Santos, C. Rufı´n / International Business Review 19 (2010) 126–139
129
section, we build on this discussion to explore the impact of these specificities on the networks that can be developed at the
BOP and on MNEs that want to participate in them.
2. BOP and TOP networks
The analysis of the specificities of BOP settings presented above gives us a basis to compare BOP and TOP networks, and to
develop propositions about expected differences between the two types of networks. Our arguments, together with the
implications for MNEs operating at the BOP, are summarized in Table 1. Following the literature, we define a network as ‘‘a
group of three or more organizations connected in ways that facilitate achievement of a common goal’’ (Provan et al., 2007:
482). We structure our exploration of the differences between BOP and TOP networks around the dimensions suggested by
network theorists (Provan et al., 2007; Scott, 1991; Uzzi, 1997; Wasserman & Faust, 1994) and group them in five categories:
structural characteristics of the network, which include centralization, linearity, density, and structural holes; boundaries of
the network, which include scope, domains, and size; network tie characteristics, which include the prevalence of direct vs.
indirect ties, the formality of ties, and the frequency of interactions; the diversity of partners; and the dynamics of the
network over time. Since our goal is to understand the differences between BOP and TOP networks, our unit of analysis, and
the corresponding dimensions we explore, is the network, rather than the member of the network (Provan et al., 2007). For
each of these dimensions, we discuss how BOP networks may differ from TOP networks based on our discussion of the
specificities of the competitive and institutional environments at the BOP.
2.1. Structural characteristics of networks
There are several characteristics of the overall structure of a network that influence the way the network functions and
the profitability of its member firms. Among those characteristics is the network’s degree of centralization. A highly
centralized network is one where a few firms have direct ties to most other network members, while the majority of firms
have few ties to other network members (Wasserman & Faust, 1994: 95). The more central a member is in a network, the
more it can benefit from the network (Ahuja, 2000; Soh, 2003; Uzzi & Gillespie, 2002), by bringing disparate actors together
in a value creation process. Although variation across networks exists, networks at the TOP tend to be organized around a few
leading players (Lorenzoni & Lipparini, 1999; McGuire & Dow, 2003).
Another structural characteristic of networks is their degree of linearity. Linearity is the reflection of a sequential
production process, as in Porter’s (1980) value chain. Many TOP networks are relatively linear in the sense that there is little
direct intervention by non-market actors in the value creation process, meaning that transactions are largely commercial in
nature, mostly involving supplier-customer relationships. Social capital and trust, which involve non-market dimensions of
network ties, certainly facilitate transactions in many TOP networks (Tsai, 2000; Uzzi, 1997), but the existence of efficient
formal institutions, such as a developed legal system, weakens the overall importance and use of social capital and trust in
the governance of transactions (Nooteboom, 2007).
A third major structural characteristic of a network is its density, as measured by the number of redundant ties relative to
network size. TOP networks tend to have a relatively high density of ties (Williams, 2005), as low transaction costs, extensive
flows of information, and the existence of specialized intermediaries that bridge gaps among actors, facilitate the
establishment of dense networks. Although related, density and linearity of ties are two different network dimensions,
which are not incompatible with each other, as actors can be linked by a multiplicity of ties with high redundancy, but only
regularly activate connections that are mostly commercial and sequential in nature.
Finally, network structure is also characterized by the number and location of structural holes within the network (Ahuja,
1997, 2000). Structural holes are ties in the network that bridge two otherwise unconnected sections of the network. Because
information flows between two unconnected sections of the network necessarily pass through the structural hole, a firm
bridging the structural hole benefits from the network more than others, and TOP networks tend to be built around a few
structural holes (Ahuja, 2000; Sharma, Vredenburg, & Westley, 1994).
2.1.1. Structural characteristics of BOP networks: centralization
The specificities of BOP environments that we discussed in the previous section are likely to strongly impact the
structural characteristics of BOP networks. First, the importance of the non-market environment and the prevalence of
value chain gaps impact the degree of centralization of BOP networks, as local NGOs, local communities, and, in some cases,
the government, become crucial members of the network. These non-market actors all have their own specific network
reflecting their core activities, which no firm is likely to have, as non-market actors are typically more deeply and
informally embedded in communities (Reed & Reed, 2009; Teegen, 2003). An NGO’s network, for instance, is likely to
include local community members and donors, while a government official’s network is likely to include local decisionmakers and members of his/her political party. As they become members of the BOP network, these non-market actors
contribute their specific connections to the new network, resulting in a network in which no member has connections to a
very large proportion of the network due to the limited overlap between the connections that they each contribute. As a
result, BOP networks will have at least several centers, such as the MNE with ties to suppliers and international markets, a
local NGO with ties to other NGOs and local community members, and the local government officials with ties to regional
decision-makers, rather than one or very few centers, as is typically the case in TOP environments (e.g., Sako, 2004). We
130
Table 1
TOP vs. BOP networks and implications for MNEs.
Network characteristics
Structural characteristics (P1)
Tie characteristics (P3)
BOP networks
Implications for MNEs at the BOP
Often centralized around ‘‘lead firms’’
Less control over network
Linearity
Structure most often relatively linear (value chain)
Density
Relatively dense networks
Decentralized due to importance of non-market
members (P1a)
Non-linear due to importance of non-market
members (P1b)
Very high density in isolated clusters, but few
connections between clusters (‘‘villages’’) (P1c)
Structural holes
Relatively few
Many, as specialized intermediaries are
scarce (P1d)
Scope
Relatively narrow business-driven scope
Wider due to lack of complementors and
prevalence of value chain gaps (P2a)
Tie domains
Few, mainly business-related
Size
Large (often global in the case of MNEs)
Multiple domains because of institutional
gaps and of the demands of diverse network
members, such as NGOs or the government (P2b)
Smaller, centered around local communities (P2c)
Directness
Mostly indirect ties (complex transactional and
logistical chains)
Formality
Mainly formal (networks of alliances)
Frequency of
interactions
Mostly low to medium frequency
Mostly direct, because of the need for ‘‘deep’’
knowledge of counterparts to create trust
(embeddedness) (P3a)
Mostly informal due to the weakness of
formal institutions and a smaller network
size, leading to personalized contact,
extensive bargaining (P3b)
High frequency due to informal environment
and, in the case of interactions with customers,
irregular income
More complex networks to manage
Need to develop local legitimacy to enter
clusters, but potential advantage from
own external links
Need to internalize activities into the
network, either by internalizing the
activity within the MNE or by inciting
local entrepreneurs to bridge the holes
More complex networks to manage, with
a wider variety of actors and a greater
vertical/horizontal diversification
Need to emphasize non-market ties and
activities, and to develop local legitimacy
Easier management of network relations,
but, also, a smaller pool of potential
network partners
Higher costs to develop and maintain ties
Need to develop alternative governance
to compensate for lack of contracts,
with associated risks
More complex logistics
Member diversity (P4)
Small (mostly business members, in spite
of presence of research institutions or
government in research consortia)
Large, because of the presence of many
non-market actors due to a scarcity of
business actors and of the multiplicity
of domains (P4)
Prominence of interactions with
non-market actors, leading to more
complex networks to manage and a
wider set of activities covered
Dynamics (P5)
Relatively stable, in spite of changes in
network membership
More unstable, unpredictable formal
networks, more resilient informal
networks (P5)
Higher risk when dealing with formal
networks; higher resilience of the
network if the MNE is embedded in
the community
M. Rivera-Santos, C. Rufı´n / International Business Review 19 (2010) 126–139
Network boundaries (P2)
TOP networks
Centralization
M. Rivera-Santos, C. Rufı´n / International Business Review 19 (2010) 126–139
131
thus expect BOP networks to be less centralized than TOP networks. For instance, when Celtel Nigeria, a cell phone
company, expanded into rural BOP markets, it had to negotiate with village chiefs and local community leaders not only to
get their approval to build towers, but also to recruit local entrepreneurs as distributors and to ensure safe passage for its
staff (Anderson & Kupp, 2008), leading to local BOP networks with at least two major centers, Celtel and the local village
chief.
Less centralization in BOP networks means that MNEs, like any other network member, will have to share major roles
and decision power with the other market and non-market actors at the BOP. With regard to centralization, therefore, we
depart from Prahalad’s view that MNEs must occupy a central position in BOP networks (Prahalad, 2005), and argue that
they may not be able to. While MNEs may need to play a central role in the construction of BOP networks that create
profitable opportunities for these companies, MNEs may not be able to occupy a central position in the operation of these
networks. This distinction between network development and network operation roles may explain our discrepancy with
Prahalad, who does not distinguish between these two roles. Thus, in the 50 BOP enterprise networks studied by Wheeler
and co-authors, MNEs often play a minor role (Wheeler et al., 2005). Similarly, companies like DuPont’s subsidiary Solae
and SC Johnson, which developed BOP initiatives in India and Kenya respectively, both emphasize the need to ‘co-create’
BOP initiatives with local communities, suggesting a less central role for MNEs as compared with TOP environments
(Simanis & Hart, 2008).
2.1.2. Structural characteristics of BOP networks: linearity
The importance of non-market actors also makes linearity inadequate to describe BOP networks. Rather than a sequence
of commercial transactions, production at the BOP will be a complex mixture of market and non-market interactions, based
as much on trust and even political considerations as on commercial principles. The fact that BOP environments are also
characterized by value chain gaps further reinforces the tendency towards less centralization and less linearity, as BOP
networks will often need to internalize parts of the production process that would not be internalized in TOP markets, due to
the lack of other firms that can provide specialized inputs or infrastructure, or assume certain support activities. For instance,
Honey Care Africa, a company that buys honey from poor farmers in Kenya, Tanzania, and Uganda, and sells it
internationally, does not stop at the traditional role of a buyer/seller of agricultural products, but also sells the beehives to
the farmers, provides access to loans that allow impoverished farmers to buy the beehives, and trains farmers on how to use
the beehives (Branzei & Valente, 2007).
Less linearity also has important implications for MNEs. As we argued earlier, value chain gaps will lead BOP
networks to internalize activities, such as financing or distribution, which would typically be outsourced in TOP
networks. From the MNE’s standpoint, internalizing a necessary activity or complementary product into the network
can be achieved in two ways. The MNE can either find (or support the creation of) a partner to fill the value chain gap, or
it may integrate the activity into its own operations, leading to greater vertical and horizontal integration at the BOP.
For instance, to compensate for the lack of electricity distribution infrastructure in the poorer parts of Uganda, cell
phone company MTN had to enter a partnership to install solar-powered generators in its wireless payphones, thereby
integrating an activity (power supply) that would typically be sourced from electric utilities in TOP environments
(African Business, 2001).
2.1.3. Structural characteristics of BOP networks: density
The impact of BOP environments on network density is less straightforward. BOP environments are characterized by poor
information infrastructures, value chain gaps, a lack of market intermediaries, and relative isolation. This is likely to impact
BOP network density in two opposite ways. At the local level, BOP specificities will lead to very high tie density. For example,
in a rural village, as much as in an urban slum, everybody knows and deals with each other. The density of ties locally is
reinforced by the importance of informal institutions at the BOP, in which informal ties compensate for the difficulty to
enforce formal arm’s-length relationships (De Soto, 2000; Wheeler et al., 2005). At the same time, the typical isolation of BOP
markets greatly limits tie density outside the local area, and while urban BOP environments, such as slums, are likely to be
less isolated from the outside world than their rural counterparts, urban poor also typically lack social ties in mainstream
society, even in developed countries (London & Hart, 2004; Tigges, Browne, & Green, 1998). Hence, the contrast is between
high-density TOP networks and ‘‘clustered’’ BOP networks with higher densities than TOP networks within the local clusters,
as business links are reinforced by informal links, but lower densities among clusters, reflecting, in the network structure, the
overall relative isolation typical of BOP environments.
The implication of the above is that the local density of ties is likely to become, for the MNE, one of the main tools through
which it can overcome weak institutional environments if it can develop enough local legitimacy to become part of local
networks. Egypt’s Sekem, for instance, developed personal long-term relationships with the farmers from whom it buys
grain, herbs, fruits and vegetables, which Sekem then sells globally. These long-term relationships between Sekem and its
farmers, which involved the development of trust-based ties as well as access to health care or education for the community
provided by the company, allowed Sekem to build legitimacy as a reliable buyer in an institutionally weak environment. By
becoming part of the farming community’s local network, Sekem ensured a supply of high-quality grain and produce, often
at prices below market level (Elkington & Hartigan, 2008).
In turn, the lower density of ties outside the clusters can provide the MNE with a competitive advantage. The MNE
possesses its own global ties through which it can connect the BOP network to global networks and, for instance, sell local
132
M. Rivera-Santos, C. Rufı´n / International Business Review 19 (2010) 126–139
products globally (Sharma et al., 1994; Wheeler et al., 2005). An MNE can thus become the link between isolated local BOP
networks and the outside world, as the Sekem example illustrates.
2.1.4. Structural characteristics of BOP networks: structural holes
Finally, TOP networks, with their highly developed market intermediaries and strong entrepreneurial forces, are likely to
have fewer structural holes than BOP networks. We are not, of course, claiming that TOP networks have no structural holes,
as scholars have found extensive evidence of the existence of structural holes in TOP networks (e.g., Ahuja, 2000; Burt, 1997).
Rather, we argue that BOP networks, typically suffering from a lack of specialized intermediaries and obstacles to
entrepreneurial activity, will display more structural holes than TOP networks (London & Hart, 2004). Like value chain gaps,
structural holes may force the MNE to integrate activities that it would typically outsource. As we mentioned above, Honey
Care Africa provides access to loans to the farmers from whom it buys honey, in order for them to be able to buy high-quality
beehives, which are five times more expensive than traditional hives. Honey Care initially provided the loans itself,
internalizing the financing function, but today only facilitates access to loans from donor agencies or NGOs, thereby bridging
the structural hole that exists between loan providers and farmers (Branzei & Valente, 2007).
This leads to the following set of propositions regarding the structural characteristics of BOP networks as contrasted with
TOP networks:
Proposition 1a. Relative to TOP networks, BOP networks will be characterized by less centralization.
Proposition 1b. Relative to TOP networks, BOP networks will be characterized by less linearity.
Proposition 1c. In contrast to the relatively high and uniform density of TOP networks, BOP networks will have localized clusters
with higher densities within the clusters but lower densities among the clusters.
Proposition 1d. Relative to TOP networks, BOP networks will be characterized by more structural holes.
2.2. Network boundaries
Networks are also defined by their boundaries. The scope of a network refers to the range of activities that are carried out
within a network. While broader than the scope of a single firm, TOP networks tend to be relatively narrow in scope because
strong competition between networks leads to network specialization and the pursuit of efficiency through the integration
of only essential players into the network (Koza & Lewin, 1998; Lorenzoni & Lipparini, 1999; Rothaermel & Deeds, 2004).
Narrow scopes are also fostered by the widespread presence of complementary products, which allow companies within the
network to focus on core activities.
The set of tie domains is a second characteristic of the boundaries of a network, distinct from network scope. The domains
of a tie refer to the number of dimensions (social, political, economic, environmental, etc.) included in ties among actors
(Wasserman & Faust, 1994). Where network scope refers to the range of activities conducted within a network as a whole,
the domains of ties concern the range of activities conducted within dyadic ties in the network. TOP networks tend to have
few tie domains, as they are focused on business transactions. High levels of institutional development allow companies to
operate with limited attention to non-market domains, including the political domain, in most industries. Although some
companies in TOP networks do engage very extensively in non-market and political networks, their primary focus is on
commercial interactions, and in fact most companies do not engage in political activity, or only indirectly through business
associations (Hansen & Mitchell, 2000).
Finally, the boundaries of a network are also characterized by the overall size of the network. Unlike scope and tie
domains, which refer to activities, the size of a network is the number of members of the network. TOP networks can be small
or large in size, depending on the goal of the network, although MNE networks for the TOP are typically complex and global in
scale (Burgers, Hill, & Kim, 1993), comprising large numbers of members engaged in the worldwide production and
distribution of MNE products.
2.2.1. Network boundaries of BOP networks: scope
The scope of BOP networks, with more limited competition, more prevalent value chain gaps, and fewer complements than
at the TOP, is likely to be wider than that of TOP networks, with important implications for MNEs. At the BOP, specialization will
be less attractive to MNEs; in fact, to ensure the availability of complementary products to make their own products valuable to
BOP consumers, MNEs will have to widen the scope of their networks to adapt to the specific holes existing in the local value
chain, resulting in more complex network to manage. BOP networks, for instance, will typically include activities such as
consumer financing, which are usually excluded from TOP networks because at the TOP they can be readily obtained from
specialized firms (Hammond, Kramer, Katz, Tran, & Walker, 2007; Wheeler et al., 2005), but the specificities of the local BOP
environment vary and ultimately determine what specific activities the MNE will need to integrate into the network. Codensa,
an electricity distribution subsidiary of Spanish utility Endesa, developed a highly successful consumer credit program for its
customers in the poor barrios of Bogotá, Colombia, when market research showed that many of these customers were excluded
from credit markets, and yet their payment record with Codensa demonstrated their creditworthiness. This program brought
both profits and greater legitimacy for the company (Millán, Caballero, & Millán, 2007).
M. Rivera-Santos, C. Rufı´n / International Business Review 19 (2010) 126–139
133
2.2.2. Network boundaries of BOP networks: tie domains
Second, the neat separation of domains prevalent in TOP networks is unthinkable at the BOP. The high prominence of
social and political factors at the BOP, due to the importance of informal institutions, means that BOP network members
cannot afford to downplay the political and social dimensions of their interactions with other members. Not only are NGOs
and local communities essential members of BOP networks, but they are also likely to demand that firms expand their
domains of activity to include social or environmental dimensions, in exchange for helping firms gain local connections
(London & Hart, 2004). US household product company SC Johnson thus reports spending a lot of time discussing with local
NGOs and community members to understand what their expectations were in terms of what SC Johnson would bring to the
community and to develop local legitimacy, when it launched its BOP initiative in Kibera, a major slum of Nairobi, Kenya
(Johnson, 2007). Similarly, in order to develop legitimacy in two adjacent villages of Parvathagiri Mandal in Hyderabad, India,
DuPont subsidiary Solae sent three employees to participate in a range of activities with local villagers before launching its
BOP initiative there. Activities included harvesting rice, operating a small kiosk and a village pay phone, and preparing a meal
for children at a local child care facility (Simanis & Hart, 2008). MNEs may also have to be careful about competing with local
firms connected to local politicians that rely on these firms for patronage purposes, and about entering BOP markets with
commercial objectives alone. In this sense, the BOP is but a specific instance of ‘‘high context’’ interaction among network
members (Hall, 1977) due to the lack of formal institutions, where the trustworthiness of potential counterparts is assessed
not only on business merits but also more broadly on the counterparts’ social and personal characteristics. In fact, informal
institutions often result in the fragmentation of BOP networks along an easily recognizable characteristic of network
members—such as ethnic, religious, kinship, or geographic affiliation—that makes it possible to create high levels of social
capital within each grouping and hence facilitate transactions within the group, even at the expense of transactions across
groups (Greif, 1993). MNEs, in response, need to develop local legitimacy within the different BOP networks in which it is
present, knowing that the investment in legitimacy is mostly local and not easily transferable across groups. In spite of a
history of BOP initiatives in different African countries, such as Ghana and South Africa, SC Johnson recognized that local
legitimacy developed in Kibera would not help the company much outside this specific slum (Johnson, 2007).
2.2.3. Network boundaries of BOP networks: size
Finally, BOP networks will be typically smaller than TOP networks, even if they involve MNEs. Due to their relative
isolation, BOP networks are substantially local in nature, as transactions need to be embedded within a community for
informal institutions to be enforced, and we can expect the global nature of TOP networks to be decisive in making these
networks larger than BOP ones. Furthermore, although BOP networks will incorporate non-market actors such as local
communities and NGOs, the local nature of these networks will limit the total number of network members relative to global
TOP networks. The smaller size of BOP networks means, in turn, that MNEs cannot rely, like in many TOP networks, on an
extensive pool of network partners. As both SC Johnson, in Kibera, and Solae, in Parvathagiri Mandal, discovered, locating
‘‘embedded community partners is not easy, as they are, almost by definition, small in size and operate intensely within a
narrow geographical range.’’ (Simanis & Hart, 2008: 12)
Our conclusions can be stated in the following set of propositions:
Proposition 2a. Relative to TOP networks, BOP networks will be wider in scope.
Proposition 2b. Relative to TOP networks, BOP network ties will involve a multiplicity of domains and will not be limited to the
business or professional domain.
Proposition 2c. Relative to TOP networks, BOP networks will have a smaller size as measured by the number of network members.
2.3. Tie characteristics
A third dimension along which network theory suggests that networks be analyzed is the nature of the ties within the
network. The proportion of direct vs. indirect ties within a network is an important tie characteristic. Indirect ties tend to be
dominant in TOP networks. Value chains, with their linear structures, link chain members indirectly except for the
immediate supplier and customer for each member in the chain. Firms building networks at the TOP typically try to
minimize the number of direct ties to reduce the costs of redundancy (i.e., the cost of receiving the same information from
the same source multiple times through different channels), and complexity (Baum, Calabrese, & Silverman, 2000).
Tie formality, the second characteristic of network ties, refers to the degree of flexibility and variability in the interactions
among network members. Although not always included in network analyses, the degree of formality of relationships among
connected network members constitutes an important element of networks. At the TOP, a variety of institutional
mechanisms facilitate formal interaction among actors. Arms’ length relationships are possible because judicial institutions
enforce contracts and extensive information flows allow evaluation of transaction counterparts, while limited corruption
and clientelism keep relations with public officials formal too. Fixed prices (no haggling) further diminish personal
interaction between TOP network members. As a result, ties within TOP networks tend to be formalized through alliance
agreements, either based on a contract or on equity sharing, even though informal networks of firms exist as well (Koza &
Lewin, 1998; Lorenzoni & Lipparini, 1999; Schrader, 1991). In other terms, while informal aspects of the network are often
important (Uzzi, 1997), TOP network members typically strongly rely on the formal aspects of the network.
134
M. Rivera-Santos, C. Rufı´n / International Business Review 19 (2010) 126–139
Finally, TOP network ties do not necessarily need frequent actualization, and the relative stability of TOP networks
(discussed in more depth below) allows the frequency of TOP network ties to have regularity. In other terms, tie actualization
tends to follow relatively low-frequency, regular patterns in most TOP networks.
2.3.1. Tie characteristics of BOP networks: directness, formality, and actualization
Along this dimension, too, the specificities of BOP environments lead BOP networks to exhibit important peculiarities
when compared to TOP networks. At the BOP, ties are mainly direct. Informal transactions and poor information flows,
especially with the outside world, force actors to deal directly with each other through highly personalized and scarcely
formalized interactions to prevent opportunism and assess the trustworthiness of counterparts (De Soto, 2000). Actor
embeddedness is needed for effective participation in the network; corruption and clientelism give relationships with
officials a personalized component as well and create specific risks for MNEs. The use of authority is negotiated with officials
in exchange for bribes, or structured within patron/client relationships that are often highly personalized (Scott, 1972).
Haggling often turns everyday commercial transactions into extended conversations between a buyer and a seller. In other
terms, BOP networks members will typically rely on the informal, rather than formal, aspects of the network to organize
transactions (De Soto, 2000; Hart & London, 2005). As a result, the frequency of interactions in networks at the BOP is also
higher, with important logistical implications for MNEs. The successful experience of Spanish company Iberdrola in Salvador
da Bahia, Brazil, is illustrative. Iberdrola’s local electricity distribution subsidiary, COELBA, relies on ‘‘community agents’’ to
reach slum dwellers in Salvador. The agents are young people from the same slum communities as the customers they visit.
They provide information about electricity use, pricing, billing, and company initiatives to save energy and better manage
energy consumption, such as company-funded refrigerator replacement programs, all of which helps customers avoid
disconnections for non-payment and the need to rely on unsafe, poor quality illegal connections. As the customers’
neighbors, the agents maintain direct, informal, personal, and frequent ties with the customers, greatly facilitating
communication between the utility and local residents (Pinhel, 2005). Of course, these types of ties have to be consistently
maintained by the company throughout its interactions with various network members. The use of community agents,
which was pioneered by Électricité de France in Rio de Janeiro, failed in that city precisely because the company separated
initial contact—made by community agents—from commercial interaction—run by the regular company commercial offices.
When commercial staff followed routine procedures to disconnect delinquent customers in the slums, community agents
began to be threatened and some had to leave their communities (personal communication, Rio Light staff, July 2007).
The following propositions state our conclusions regarding BOP network ties:
Proposition 3a. Relative to TOP network ties, direct ties are more prevalent in BOP networks.
Proposition 3b. Relative to TOP network ties, BOP network ties, including those involving public officials, will be less formal and
involve greater personal interaction and bargaining.
Proposition 3c. Relative to TOP networks, tie frequency in BOP networks will be higher.
2.4. Partners
One of the major benefits for a firm that derives from being embedded in a network is the access to resources and
knowledge which it cannot or does not wish to develop internally (Uzzi, 1996). For this reason, firms active in TOP networks
try to maximize the diversity of partners active in their networks in order to expand the scope of knowledge or resources they
can access through the network (Baum et al., 2000). Many TOP networks thus typically include competitors, suppliers,
distributors and complementors.
2.4.1. Member diversity in BOP networks
While TOP networks may exhibit a wide diversity of partners, BOP networks can be expected to include a far more diverse
set of members. Contrasting with TOP networks, non-market actors will be prevalent in a BOP network, independently from
the scope of activities conducted within the network. The reason is the lack of specialized and suitable market participants in
BOP markets, which means that, as London and Hart (2004) find in their study of 24 BOP initiatives, MNEs must rely on nonmarket actors to carry out commercial functions. A commonly found example is that of NGOs which establish alliances with
MNEs that entail some degree of involvement in the MNEs’ value chain (Chesbrourgh et al., 2006). Solae, for instance,
partnered with Modern Architects for Rural India (MARI), an NGO that promotes strong community based organization of
the poor and introduced Solae to the local community, to launch its BOP initiatives in India. Similarly, SC Johnson, partnered
with a local NGO in Kibera, Kenya, to distribute its products in the slum. The local NGO, Carolina for Kibera, whose goal is to
fight poverty and prevent violence through sports, young women’s empowerment, and community development, recruits
groups of unemployed youth, who offer cleaning services using SC Johnson’s products to community members for a small fee
(Johnson, 2007). For SC Johnson, the partnership with Carolina for Kibera thus provides a de facto distribution network. For
the NGO, it creates income earning opportunities for impoverished youth, and helps improve living conditions and
cleanliness in the community. Finally, governments, whether local, national, or in the form of international institutions are
also often involved in BOP initiatives, providing funds, advice or legitimacy (UN Global Compact, 2007; Wheeler et al., 2005).
This contrasts strongly with TOP networks, where most activities in the network are much more likely to be carried out by
M. Rivera-Santos, C. Rufı´n / International Business Review 19 (2010) 126–139
135
business firms, even though non-profit institutions such as research centers may also be involved in some cases (e.g.,
Mathews, 2002). We therefore derive the following proposition:
Proposition 4. Relative to TOP networks, BOP networks will be characterized by greater diversity among network members, with
less prevalence of business firms.
2.5. Network dynamics
The stability of a network over time is much less studied as a characteristic of networks by network theorists, as networks,
like firms, evolve over time, with members entering and others leaving the network (Olk & Young, 1997). Studies suggest that
network members typically strive to keep a balance between the adaptability that stems from loosely interconnected firms,
and network instability, which typically results from member isolation within the network, member migration to a
competing network, the creation of cliques within the network, and attrition (Dhanaraj & Parkhe, 2006).
2.5.1. Network dynamics in BOP networks
At the BOP, we expect network dynamics to differ considerably from the dynamics in TOP networks. Overall, we expect
BOP networks to evolve more rapidly in certain aspects and more slowly in others relative to TOP networks. These complex
dynamics stem from a combination of a greater instability and unpredictability of formal, i.e. contract-based and
institutionally backed, ties at the BOP, on the one hand, and of a greater resilience of informal, i.e. trust-based and socially
backed, ties, on the other hand. First, we expect the weakness of formal institutions at the BOP to make the formal aspect of
ties more unstable and unpredictable. In the context of weak formal institutions, the enforceability of formal ties that
involve, for instance, contracts, becomes problematic, leading to higher instability. Furthermore, poor governance arising
from deficient political institutions typically leads to greater economic instability and, in turn, greater unpredictability of
economic conditions, as one of the dimensions of governance is the ability to manage changes in the economic environment
(Acemoglu, Johnson, Robinson, & Thaicharoen, 2003). Economic instability can further disrupt formal networks by leading to
the rapid appearance and disappearance of economic actors, such as firms, as economic conditions fluctuate. Poor
institutions add to unpredictability by leading to wide swings in political outcomes and, in extreme cases, to disruption and
even violence in the transfer of political control over time (Gates, Hegre, Jones, & Strand, 2006). Political regime changes can
result not only in changes in office holders, but also in the reorganization of governmental agencies, while violence can
destroy the capacity of formal organizations to operate, as in the case of armed rebellions that cut a region off from access by
government officials or humanitarian assistance. Finally, poor institutions and the lack of proper infrastructure exacerbate
the vulnerability of formal ties to diseases or natural disasters (World Bank, 2000). It would, of course, be an exaggeration to
claim that extreme violence or natural disasters are systematically present in BOP environments, and some BOP
communities have been very stable and protected from disasters over long periods of time. However, economic and physical
vulnerability, defined by the World Bank as the ‘‘probability of being exposed to a number of [. . .] risks (violence, crime,
natural disasters, being pulled out of school)’’ (2000: 19) is an important characteristic of such environments and most
scholars now consider that vulnerability is an integral dimension of poverty (Banerjee & Duflo, 2007; World Bank, 2000).
Research on poverty thus suggests that poor communities are economically and physically typically more vulnerable than
their TOP counterparts, due to weak formal institutional environments. SC Johnson faced the issue of vulnerability and
instability in its Kibera BOP initiative, as the network suffered major disruptions from not only political unrest and gang
violence, but also natural disasters such as flooding (Johnson, 2007).
Yet at the same time, SC Johnson executives were also surprised by the resilience of the community in Kibera, even in the face
of such hardships. Contrasting with the formal aspects of BOP networks that we just discussed, we expect the informal aspects
of BOP networks to be stable and resilient, as Johnson discovered in Kibera. Uzzi (1997) argues that networks that are developed
for reasons beyond instrumental motives are more likely to be long-lasting, highly valued, and stable. As we discussed earlier,
one of the characteristics of BOP networks is precisely that they are developed for reasons beyond purely commercial motives
and that they are strongly embedded in the pre-existing social structure of the community. Actors are bound not only by
commercial ties but more importantly by strong traditional and social ties within communities, such as kinship, religion, or
race. Because of their embeddedness within a wider informal structure, the informal aspects of BOP networks are thus likely to
resist external shocks, at least as long as the community itself is not disrupted. In fact, we expect these informal bonds to be
further reinforced by the instability of the formal ties in BOP networks discussed earlier, as the difficulty to rely on less reliable
formal networks is likely to increase the value of informal linkages in BOP environments. De Soto (2000) found, for instance, that
slum communities in the metropolitan area of Lima in Peru developed strong informal networks and rules, as a response to a
state that, after severe macroeconomic and political instability during the 1990s, still showed little capacity to help these
communities. Similarly, Banerjee and Duflo (2007) found that informal networks in poverty-stricken communities replace, at
least to a certain extent, unavailable formal insurance mechanisms against external shocks.
These complex dynamics have important implications for MNEs. The instability of the formal aspect of the network
comes with increased costs for MNEs, especially if they are used to heavily relying on formal mechanisms in their TOP
environments to minimize the vulnerability of these ties. At the same time, the stability and resilience of the informal aspect
of the network suggests that, as long as the MNE is embedded in the local social structure, its BOP network is likely to survive
external shocks.
136
M. Rivera-Santos, C. Rufı´n / International Business Review 19 (2010) 126–139
This reasoning leads us to argue that the formal aspects of BOP networks will be less stable, while the informal aspects of
BOP networks will be more stable, relative to their TOP counterparts. We therefore derive the following proposition:
Proposition 5. Relative to TOP networks, the formal aspects of BOP networks will display more instability and unpredictability,
whereas the informal aspects of BOP networks will display greater resilience.
3. Discussion and conclusions
One of the fundamental elements of the business model paradigm emerging from the BOP literature is the necessary
creation of a network that allows MNEs to develop appropriate products and provide them to BOP populations at a profit. In
this paper, we have sought to explore systematically what BOP networks can be expected to look like, and especially in what
specific ways, if any, BOP networks differ from TOP ones. Our starting point has been the characterization of BOP
environments established by the BOP literature, but contrasting with the mostly case-based inductive approach common in
the BOP literature (e.g., London & Hart, 2004), we apply key conceptual frameworks from the fields of international business,
I/O and institutionalism, to analyze BOP environments and derive theory-driven implications for BOP networks.
Our analysis builds on and expands the claims of the BOP literature to reveal important and systematic differences
between BOP and TOP networks: relative to TOP networks, BOP networks are likely to be less centralized, wider in scope, less
dense overall (but also containing high-density clusterings), and contain more structural holes; network ties are more
commonly direct and informal, are actualized more frequently, and involve a multiplicity of domains of interaction among
network members; the diversity of network members is greater, and, lastly, BOP networks are more unstable and
unpredictable in their formal aspects but also more stable and resilient in their informal aspects.
Our work makes several contributions to different streams of research. First and foremost, we contribute to the emerging
BOP literature within the field of international business, by providing theoretical rigor to a stream that has been largely
characterized by induction based on a small number of cases, and by a strong applied focus rather than theory building.
Second, we contribute to international business strategy research by systematically identifying key strategic factors that
MNEs will have to address in order to grow during the next decades, given the relative growth and sheer demographic weight
of BOP populations worldwide. In particular, we highlight the identification of non-market actors and development of crosssector relationships as key strategic capabilities for MNEs at the BOP, thus extending and systematizing London and Hart’s
(2004) findings. Third and last, our work contributes to scholarship on business networks by characterizing a largely
unexplored type of network. BOP networks are important to understand because, like high-tech networks, they are
laboratories of product and business model innovation, but to a much greater extent than networks at the TOP, they include a
variety of non-market actors and interactions over multiple domains.
As we discuss throughout the paper and as we summarize in Table 1, the implications of our analysis for MNEs are many.
To create value successfully at the BOP, MNEs must be prepared to share control over their activities with other actors, and
particularly with a variety of non-market actors, either because these actors help address institutional gaps, or because they
can take on roles occupied by firms in TOP networks where firms are missing at the BOP (London & Hart, 2004). These
conditions will put a premium on the ability of MNEs to collaborate across organizational boundaries as well as across
sectors, and to coordinate disparate actors through persuasion rather than control. At the same time, the lack of firms capable
of providing inputs in the value creation process will require MNEs not only to engage non-market actors, but also to either
foster local entrepreneurs or internalize some activities, including the supply of complementary products and the creation of
distribution channels. Hence MNEs must accept greater vertical integration and horizontal diversification than in TOP
markets.
The fragmentation of BOP networks along geographic, ethnic, kinship, or religious lines poses more familiar challenges to
MNEs, which face these kinds of differences on an everyday basis. If anything, operating at the BOP will underscore the need
for sensitivity to understand these differences and flexibility to adapt to them. Earning trust and political goodwill in
communities tied by strong informal links will require a significant degree of local embeddedness on the part of MNEs,
underscoring the need for intermediaries that can bridge the gap between the MNE and BOP communities. Again, with
specialized firms scarce at the BOP, MNEs will need to search for non-traditional partners such as NGOs, community
organizations, or government agencies. These conditions will also slow rollout and scaling efforts, requiring longer financial
and operating horizons.
These implications pose significant challenges for MNEs. BOP initiatives are fraught with risk, like any other form of major
innovation. MNEs will have to rely on a scarce set of non-traditional counterparts, increasing risks stemming from
collaboration failures and opportunism. Political risk will also be higher at the BOP because of the multiple domains of
interaction and the limited development of political institutions. Furthermore, the links between local informal powers and
firms that may see MNEs as competitors will increase the level of political risk, as the success of BOP initiatives may be
dependent on eliminating existing intermediaries (Hart & London, 2005). Dealing with governmental organizations, and
interacting with other actors across multiple domains, can also increase the risk of fraud and corruption. More extended
payback periods may clash with the demands of international financial markets; yet at the same time, greater instability and
unpredictability at the BOP may create more unstable and unpredictable profits and hence require a higher risk premium.
For all these reasons, MNEs will need to develop the ability to manage a different and more significant set of risks relative to
M. Rivera-Santos, C. Rufı´n / International Business Review 19 (2010) 126–139
137
TOP-related risks. And, contrary to the ‘‘moral imperative’’ to accept lower profitability at the BOP that we find in much of the
BOP literature (Prahalad, 2005; Simanis et al., 2005), we argue that BOP initiatives will have to earn higher average profits
than TOP ones to compensate for the risks involved.
An important limitation of our work is the fact that it is not based on additional empirical research, but relies mainly on
the limited number of cases from which the BOP literature has drawn its conclusions. However, we believe that the
application of valuable theoretical tools to case data is an important and useful exercise because it can generate hypotheses
that can be tested with large-N samples and statistical tools (Eisenhardt, 1989). Our characterization of the BOP, building on
the BOP literature, may also introduce subjectivity in our results. For example, we pay little heed to the issue of affordability
and of price/unit ratio, as we believe that it is of limited use for our purposes. Furthermore, our characterization of the BOP,
following the literature, is based on stylized facts, and we do not fully take into account variations across BOP environments.
In particular, the distinction between urban and rural poor (World Bank, 2008) is likely to bring important nuances to general
BOP characterizations. The urban poor typically have, for instance, relatively easier access to infrastructure than the rural
poor because of their geographic proximity to TOP environments (Banerjee & Duflo, 2007). Similarly, networks may be more
centralized in rural than in urban BOP environments, as rural environments are often characterized by local patrons, such as
village chiefs, ‘caciques’, or landowners, who establish strong patron-client relationships and thereby increase overall
centralization in the village (Bhaduri, 1973; Blair, 2005; Scott, 1972). We hope that our work will encourage other scholars to
examine the degree to which alternative and more nuanced characterizations of different BOP environments would lead to
different conclusions about BOP networks.
The most important extension of our work, however, is the empirical testing of our propositions. We believe that BOPrelated research must go beyond the case-based approach followed so far in order to gain rigor and credibility. In this sense,
our propositions may be useful in terms of guiding data collection and analysis efforts. Specifically, a research design
consistent with the preceding analysis would be structured around the collection and analysis of network data for a TOP
network and a BOP network. The first question that this design has to answer is the choice of networks for data collection, so
as to minimize the possibility that observed differences in the structural characteristics of the two networks are due to
factors other than income and poverty differences. Because network data collection is labor- and time-intensive, it is
probable that controls will have to be approximated by choosing settings that minimize variation other than by income and
prevalence of poverty. The obvious choice is a pair of TOP and BOP networks in close proximity in a developing country. Such
settings exist, particularly in Latin America, where high inequality levels create sharper differentiation between TOP and
BOP, and where there is a long-standing MNE presence at the TOP.
The second question concerning the empirical research design for testing our propositions is the actual collection and
analysis of data. Empirical network analysis is a well-established field. The analysis of network characteristics is well
supported by a variety of analytical tools for network data (for a comprehensive treatment of these tools, see for instance
Wasserman & Faust, 1994). For example, widely accepted measures of network centralization are available, based on the
structure of bilateral ties among actors in a network. The type of data required to carry out this analysis is also well
documented, and supported by many published network analysis studies. The main challenge for this component of the
research design is therefore the collection of network data—the actual fieldwork involving the collection of information from
network members through questionnaires or personal interviews. This challenge is intensified by the need to collect BOP
network information in environments where data of any kind is very scarce relative to the settings of extant empirical
network research, and where data collection may itself be complicated by mistrust of outsiders, poor physical conditions,
and other difficulties. For these reasons, just as MNEs need alliances with local, non-traditional actors to operate successfully
at the BOP (London & Hart, 2004; Simanis et al., 2005), we envisage that BOP network data could be collected in collaboration
with an NGO or community organization. Such an organization would assist in the identification of network members and
the solicitation of information through questionnaire distribution or personal interviews, as well as the hiring of personnel to
carry out these tasks where necessary. In fact, similar approaches are often followed for TOP networks, although in this
instance local university students or research organizations typically suffice, given fewer barriers in terms of physical access,
trust, security, and other obstacles.
In sum, in this paper, we take a first step in the direction of greater conceptual thoroughness and greater reliance on
existing theoretical frameworks, as contrasted with case-based inductive approaches, to understand the specificities of Base
of the Pyramid environments. The propositions that we develop in the paper, based on well-established network analysis
constructs, provide a strong basis for empirical research on BOP environments and have important implications for MNEs
entering BOP markets. We hope that this work will encourage other scholars to pursue a more systematic and theory-driven
exploration of the Base of the Pyramid.
References
Acemoglu, D., Johnson, S., Robinson, J., & Thaicharoen, Y. (2003). Institutional causes, macroeconomic symptoms: Volatility, crises and growth. Journal of Monetary
Economics, 50(1), 49–123.
African Business. (2001). Sun powered pay phones. African Business, March: 29–32.
Ahuja, G. (2000). Collaboration networks, structural holes, and innovation: A longitudinal study. Administrative Science Quarterly, 45(3), 425–457.
Anderson, J., & Kupp, M. (2008). Celtel Nigeria: Serving the rural poor. Tilburg: TiasNimbas Case Study.
Anderson, J., & Markides, C. (2007). Strategic innovation at the Base of the Pyramid. MIT Sloan Management Review, 49(1), 83–88.
Arnould, E. J., & Mohr, J. J. (2005). Dynamic transformations for Base-of-the-Pyramid Market Clusters. Academy of Marketing Science Review, 33(3), 254.
Banerjee, A. V., & Duflo, E. (2007). The economic lives of the poor. Journal of Economic Perspectives, 21(1), 141–167.
138
M. Rivera-Santos, C. Rufı´n / International Business Review 19 (2010) 126–139
Baum, J. A. C., Calabrese, T., & Silverman, B. (2000). Don’t go it alone: Alliance network composition and startups’ performance in Canadian biotechnology. Strategic
Management Journal, 21(3), 261–294.
Bhaduri, A. (1973). A study in agricultural backwardness under semi-feudalism. The Economic Journal, 83(329), 120–137.
Blair, H. (2005). Civil society and propoor initiatives in rural Bangladesh: Finding a workable strategy. World Development, 33(6), 921–936.
Branzei, O., & Valente, M. (2007). Honey Care Africa: A tripartite model for sustainable beekeeping. Richard Ivey School of Business Case Studies (907M60).
Burgers, W. P., Hill, C. W. L., & Kim, W. C. (1993). A theory of global strategic alliances: The case of the global auto industry. Strategic Management Journal, 14(6),
419–432.
Burt, R. (1997). The contingent value of social capital. Administrative Science Quarterly, 42(2), 339–365.
Chesbrourgh, H., Ahern, S., Finn, M., & Guerraz, S. (2006). Business models for technology in the developing world: The role of non-governmental organizations.
California Management Review, 48(3), 48–61.
Dawar, N., & Chattopadhyay, A. (2002). Rethinking marketing programs for emerging markets. Long Range Planning, 35(5), 457.
De Soto, H. (2000). The mystery of capital: Why capitalism triumphs in the west and fails everywhere else. New York: Basic Books.
Dhanaraj, C., & Parkhe, A. (2006). Orchestrating innovation networks. Academy of Management Review, 31(3), 659–669.
Eisenhardt, K. M. (1989). Building theories from case study research. Academy of Management Review, 14(4), 532–550.
Elkington, J., & Hartigan, P. (2008). The power of unreasonable people. Boston, MA: Harvard Publishing Press.
Gates, S., Hegre, H., Jones, M. P., & Strand, H. (2006). Institutional inconsistency and political instability: Polity duration, 1800–2000. American Journal of Political
Science, 50(4), 893–908.
Greif, A. (1993). Contract enforceability and economic institutions in early trade: The Maghribi traders’ coalition. The American Economic Review, 83(3), 525–548.
Hall, E. T. (1977). Beyond culture. Garden City, NY: Doubleday.
Hammond, A. L., Kramer, W. J., Katz, R. S., Tran, J. T., & Walker, C. (2007). The next 4 billion: Market size and business strategy at the Base of the Pyramid. Washington,
DC: World Resources Institute/International Finance Corporation.
Hammond, A. L., & Prahalad, C. K. (2004). Selling to the poor. Foreign Policy, 142(May/June), 30–37.
Hansen, W. L., & Mitchell, N. J. (2000). Disaggregating and explaining corporate political activity: Domestic and foreign corporations in national politics. The
American Political Science Review, 94(4), 891–903.
Hart, S. L., & Christensen, C. (2002). The great leap: Driving innovation from the Base of the Pyramid. Sloan Management Review, 44(1), 51–56.
Hart, S. L., & London, T. (2005). Developing native capability. Stanford Social Innovation Review, 3(2), 28–33.
Johnson, S. (2007). SC Johnson builds business at the base of the pyramid. Global Business and Organizational Excellence, 26(6), 6–17.
Karnani, A. (2007). The mirage of marketing to the Bottom of the Pyramid: How the private sector can alleviate poverty. California Management Review, 49(4), 90–
111.
Khanna, T., & Palepu, K. (2000). The future of business groups in emerging markets: Longrun evidence from Chile. Academy of Management Journal, 34, 268–285.
Kolk, A., & van Tulder, R. (2006). Poverty alleviation as business strategy? Evaluating commitments of frontrunner Multinational Corporations. World
Development, 34(5), 789–801.
Koza, M. P., & Lewin, A. Y. (1998). The co-evolution of strategic alliances. Organization Science, 9(3), 225–264.
London, T., & Hart, S. L. (2004). Reinventing strategies for emerging markets: Beyond the transnational model. Journal of International Business Studies, 35(5), 350.
Lorenzoni, G., & Lipparini, A. (1999). The leveraging of interfirm relationships as a distinctive organizational capability: A longitudinal study. Strategic Management
Journal, 20(4), 317–337.
Mathews, J. A. (2002). The origins and dynamics of Taiwan’s R&D consortia. Research Policy, 31(4), 633–651.
McGuire, J., & Dow, S. (2003). The persistence and implications of Japanese keiretsu organization. Journal of International Business Studies, 34(4), 374–389.
Millán, J., Caballero, C., & Millán, N. (2007). CODENSA 10 años. Bogotá: Fedesarrollo.
Nooteboom, B. (2007). Social capital, institutions and trust. Review of Social Economy, 65(1), 29–53.
North, D. (1990). Institutions, institutional change, and economic performance. Cambridge: Cambridge University Press.
Olk, P., & Young, C. (1997). Why members stay in or leave an R&D consortium: Performance and conditions of membership as determinants of continuity. Strategic
Management Journal, 18(11), 855–877.
Pinhel, A. (2005). COELBA Agent Project. Salvador da Bahia: Presentation—September 13, 2005.
Porter, M. E. (1980). Competitive strategy. New York: Free Press.
Portes, A. (1994). The informal economy and its paradoxes. In N. J. Smelser & R. Swedberg (Eds.), The handbook of economic sociology (pp. 426–452). Princeton, NJ:
Princeton University Press.
Prahalad, C. K. (2005). Fortune at the Bottom of the Pyramid. Upper Saddle River, NJ: Wharton School Publishing/Pearson Education.
Prahalad, C. K., & Hammond, A. L. (2002). Serving the world’s poor, profitably. Harvard Business Review, September: 48–57.
Prahalad, C. K., & Hart, S. L. (2002). The fortune at the bottom of the pyramid. Strategy+Business, 20, 1–13.
Provan, K. G., Fish, A., & Sydow, J. (2007). Interorganizational networks at the network level: Empirical literature on whole networks. Journal of Management, 33(3),
479–516.
Reed, A., & Reed, D. (2009). Partnerships for development: Four models of business involvement. Journal of Business Ethics, 90, 3–37.
Ricart, J. E., Enright, M. J., Ghemawat, P., Hart, S. L., & Khanna, T. (2004). New frontiers in international strategy. Journal of International Business Studies, 35(3), 175–
200.
Rothaermel, F. T., & Deeds, D. L. (2004). Exploration and exploitation alliances in biotechnology: A system of new product development. Strategic Management
Journal, 25(3), 201–222.
Sako, M. (2004). Supplier development at Honda, Nissan and Toyota: Comparative case studies of organizational capability enhancement. Industrial and Corporate
Change, 13(2), 281–308.
Schrader, S. (1991). Informal technology transfer between firms: Cooperation through information trading. Research Policy, 20(2), 153–171.
Scott, J. C. (1972). Patron-client politics and political change in Southeast Asia. American Political Science Review, 66(1), 91–113.
Scott, J. T. (1991). Social network analysis: A handbook. Newbury Park, CA: Sage Publications.
Seelos, C., & Mair, J. (2007). Profitable business models and market creation in the context of deep poverty: A strategic view. Academy of Management Perspectives,
21(4), 49–63.
Sharma, S., Vredenburg, H., & Westley, F. (1994). Strategic bridging: A role for the multinational corporation in Third World development. Journal of Applied
Behavioral Science, 30(4), 458–476.
Simanis, E., & Hart, S. L. (2008). The Base of the Pyramid Protocol: Toward Next Generation BoP Strategy (Version 2.0). Ithaca, NY: Center for Sustainable Global
Enterprise.
Simanis, E., Hart, S. L., Enk, G., Duke, D., Gordon, M., & Lippert, A. (2005). Strategic Initiatives at the Base of the Pyramid: A protocol for mutual value creation (Version
1.0). Racine, WI. www.bop-protocol.org.
Soh, P.-H. (2003). The role of networking alliances in information acquisition and its implications for new product performance. Journal of Business Venturing,
18(6), 727–744.
Teegen, H. (2003). International NGOs as Global Institutions: Using social capital to impact multinational enterprises and governments. Journal of International
Management, 9, 271–285.
Tigges, L. M., Browne, I., & Green, G. P. (1998). Social isolation of the urban poor: Race, class, and neighborhood effects on social resources. Sociological Quarterly,
39(1), 53–77.
Transparency International. (2007). Report on the Transparency International Global Corruption Barometer 2007. Berlin: Transparency International.
Tsai, W. (2000). Social capital, strategic relatedness and the formation of intraorganizational linkages. Strategic Management Journal, 21, 925–939.
UN Global Compact. (2007). The ten principles. http://www.unglobalcompact.org/AboutTheGC/TheTenPrinciples/index.html (accessed November 21).
UNCTAD, U. N. C. o. T. a. D. (2005). Trade and Development Report. New York and Geneva: United Nations.
M. Rivera-Santos, C. Rufı´n / International Business Review 19 (2010) 126–139
139
UNCTAD, U. N. C. o. T. a. D. (2006). The Least Developed Countries Report 2006. New York and Geneva: United Nations.
Uzzi, B. (1996). The sources and consequences of embeddedness for the economic performance of organizations: The network effect. American Sociological Review,
61(4), 674–698.
Uzzi, B. (1997). Social structure and competition in interfirm networks: The paradox of embeddedness. Administrative Science Quarterly, 42(1), 35–67.
Uzzi, B., & Gillespie, J. J. (2002). Knowledge spillover in corporate financing networks: Embeddedness and the firm’s debt performance. Strategic Management
Journal, 23(7), 595–618.
Wasserman, S., & Faust, K. (1994). Social network analysis: Methods and applications. Cambridge: Cambridge University Press.
Wheeler, D., McKague, K., Thomson, J., Davies, R., Medalye, J., & Prada, M. (2005). Creating sustainable local enterprise networks. MIT Sloan Management Review,
47(1), 33–40.
Whitney, P., & Kelkar, A. (2004). Designing for the Base of the Pyramid. Design Management Review, 15(4), 41–47.
Williams, T. (2005). Cooperation by design: Structure and cooperation in interorganizational networks. Journal of Business Research, 58(2), 223–231.
World Bank. (2000). World Development Report 2000/2001: Attacking poverty. Oxford: Oxford University Press.
World Bank. (2008). World Development Report 2008: Agriculture for development. Washington, DC: The International Bank for Reconstruction and Development/
The World Bank.