Strength of Weak Ties in Microfinance

Strength of Weak Ties in Microfinance

Cornell Jackson

Ana Marr

University of Greenwich

Abstract

This paper introduces social network analysis as a new tool to add to the social intermediation toolset microfinance institutions can used to improve their client’s businesses. The key idea is that weak bridging ties give these clients access to new information that can be used to improve their businesses. These weak bridging ties may also give rise to brokerage opportunities which will also give access to new information. This is important because, as the paper discusses, poor people often find themselves on the periphery of the larger networks of their societies. A case study mapping the social networks of sari sellers who are microfinance clients in Tamil Nadu, India showed that they relied mainly on strong ties. As discussed in the paper, strong ties have the tendency to encapsulate the poor and result in poor connections to the larger network. In order to test this theory,an upcoming intervention will seek to introduce weak bridging ties to these overwhelmingly low caste sari sellers. It is hoped that this intervention will give these sari sellers access to new information that will improve their businesses and also help to connect them to the larger network.

Introduction

Microfinance has been traditionally used to provide credit to help the poor to establish businesses that will help to reduce their poverty. It is now widely recognised that the poor need more than just access to credit to be successful at business but also need other support. However, little research has been done in microfinance to see what support or lack of support the social networks of microfinance clients provide for their businesses. The type and strength of connections could be an important indicator of the success of these businesses. Social network analysis is the scientific discipline that studies social networks. This research is part of a research project funded by the Leverhulme Trust in the United Kingdom looking at how networks can be used in microfinance to optimise the twin goals of poverty reduction and financial sustainability. This research focuses on the impact of social networks of microfinance clients and is a case study that uses social network analysis to map the social networks of 111 sari sellers who are all clients of a large MFI in Tamil Nadu in India. It also used the Progress out of Poverty Index to calculate their probable poverty level and looked at the revenues and profits of their businesses.

Social Network Analysis

Social networks are defined and measured as connections among people, organisations, political entities (states and nations) and/or other units. Social network analysis is a theoretical perspective and a set of techniques used to understand these relationships (Valente 2010, pg. 3). Christakis and Fowler (2010, pg. 32) say that the science of social networks provides a distinct ways of seeing the world because it is about individuals and groups and how the former becomes latter.

Valente (2010, pgs. 3 – 7) says that relationships matter because relationships influence a person’s behaviour above and beyond the influence of his or her attributes. A person’s attributes does influence who people know and spend time with: their social network. Valente quotes Borgatti et al (2009), “one of the most potent ideas in the social sciences is the notion that individuals are embedded in thick webs of social relations and interactions”. The reason that social networks are so important is because human beings are ultra-social animals that create social networks(Haidt, 2006).Christakis and Fowler (2010, pg. 214) add that human beings just don’t live in groups, they live in networks. Valente argues the traditional social science approach of using random sampling is not adequate for measuring network concepts becauserandom sampling removes individuals from the social context that may influence their behaviour. Valente explains that one primary reason social network research has grown in recent decades is that scholars have become dissatisfied with attributes theories of behaviour. Many attribute theories have not explained why some people do things (e.g. quit smoking) while others do not. Social network explanations have provided good explanations in these cases. Social network analysis concepts and techniques have found wide application across a number of scientific disciplines including anthropology, business, communication, computer science, economics, education, marketing, medicine, public health, political science, psychology and sociology to name a few.

Valente (2010, pgs. 9 – 19) identifies the following major advances made by network research in recent decades:

• Small World Networks – This is defined as a network in which most people have few connections yet the overall distance between any two people in the network is shorter than expected by chance (Watts, 1999). Small world networks are characterised by local clustering which indicates dense pockets of interconnectivity. There are bridges, however, that connected these subgroups and these bridges enable people to connect to seemingly distant others by fewer steps than would occur in a random network.
• Scale-Free Networks – Barabási (2003) meant that there is a predictable distribution in the number of connections to and from each person and this distribution is not normal but instead is highly skewed. Most people will have only a few connections but there will be a few that will have hundreds or even thousands of connections. Barabási hypothesised that that the scale-free distribution occurred because people preferred to connect to a network at its most central locations. Thus, as the network grows, people who are most central retain their central position. This is the rich get richer dynamic.
• Networks Dynamics – Networks are dynamic over time. Network dynamics have been important because research has shown that networks can provide access to resources and information. Network changes can occur at two levels:
• Individual – Individuals add and lose connections. Individual indicators of centrality, personal network density and reciprocity can change over time.
• Network – At the network level, overall network density, centralisation and transitivity can change over time
• Social Capital – What is important is not what you know but who you know. Furthermore it is not just who you know but how well one uses his or her social resources. These social resources are referred to as social capital which is the quality and quantity of resources available in a person’s social network (Lin, 2001). Social capital can be measured as trust (Moore et al, 2005). Even the poor have social capital and know who have power and resources and this can be used ameliorate some of the problems they face. However, focus on social capital may burden the poor without acknowledging the institutional and historical causes of their poverty.
• Homophily – This the tendency for people to affiliate and associate with others like themselves. As a result, a person’s social network tends to be a reflection of him or herself because people feel more comfortable being with people like themselves rather than with people who are different. Homophily helps to explain why the small world effect occurs. The set of people from which contacts are drawn are narrowed by homophily and the probability that two people have an acquaintance in common is much higher than random chance alone would dictate. Homophily also explains whynew ideas and practices have difficulty in getting a foothold within most social networks because most people talk to others like themselves and usually hold similar attitudes, beliefs and practices and as a result avoid those who do not share their views slowing the spread of new ideas. However, homophily can also speed diffusion. Once a new idea does gain a foothold in the social network, the trust generated by homophily causes it to spread quickly.
• Diffusion/Contagion – Diffusion of innovations is the process via which new ideas and practices spread within and between communities (Rogers, 2003). There is considerable evidence to suggest that a person’s adoption of a new idea, attitude, opinion or practice is strongly influenced by the behaviour of their social network (Valente, 1995). It is also true that the similarity of behaviours among people who are connected to one another arises because of selection. Selection us the tendency for people to seek out friends whose behaviours are consistent with their own. It seems that individuals have varying thresholds to adoption such that some people adopt an idea when no or few others have, while other people wait until a majority of others have adopted. A population composed of primarily of people with high thresholds will be resistant to diffusion and hence diffusion will be slowed. One’s position in the network affects diffusion (Becker, 1970). Central members of the network both reflect and drive the diffusion process. Having a central position provides and advantaged viewpoint of seeing what the others are doing and it provides advantage in terms of influencing others (Valente and Davis, 1999; Valente and Pumpuang, 2007).
• Centrality – Centrality is the extent to which a person inhabits a prestigious or critical position in the network. Numerous measures of centrality have been developed by social network analysis (Borgatti and Everett, 2006). Centrality can be measured simply as the number of choices one receives from others in the network or betweenness which is the extent a person lies on the shortest path connecting others in the network or closeness which is the average distance that a person is located from everyone else in the network (Freeman, 1979). Central members often provide bridges between different parts of a network to make them small world networks having a short overall path length for a network of a given size.
• Efficient Network Forms – There is not one efficient network form. Valente (1995) argued that innovations diffuse more quickly in dense rather than in sparse networks. However, empirical evidence suggests that this is not always the case. Too much density creates redundant communications and reduces the ability of people in the network to access outside sources of information and influence (Valente et al, 2007). As Granovetter (1973) noted in his strength of weak ties argument, connections to outside sources of information and resources can be very valuable. Optimal network density will vary over time with higher levels of density being needed early in the network’s development and lower levels of density needed as the network matures.
• Interventions – Network interventions can be successful. For example, it is possible to deliberately change networks so that bridges are created or strengthened. Network datain organisations can be collected so holes or gaps in the networked can be spanned or divisions that become revealed can be mended together via deliberate relationship building.

Network Rules

Christakis and Fowler (2010, pgs. 17 – 26) described the following rules on networks discovered through research:

• Rule 1:Individuals shape theirnetwork – One example they give of how individuals shape their networks is homophily which was described above. Individualsalso decide the structure of the network by deciding how many people they are connected to, influencing how densely interconnected their family and friends are and by controlling how central they are to the social network. Individuals also shape their networks through transitivity which is the tendency wherean individual has strong ties to two separate people; those two people will know each other thus forming a triangle.
• Rule 2: Thenetwork shapes us – The network shapesindividuals because the number of social contacts can affect people, transitivity, or the lack of it, can affect individuals and how many contacts an individual’s friends and family have can affect them.
• Rule 3: Friends affect individuals –Due to the human tendency to influence and copy one another, friends help determine the content that flows across the network which affects the individual.
• Rule 4: Our friends’ friends’ friends affects individuals – Two examples of this rule are described. First ishyperdyadic spread which is the tendency of effects to spread from person to person to person beyond an individual’s direct social ties.The second example is Milgram’s famous sidewalk experiment (Milgram et al, 1969). In this experiment, researchers would stop and look up at a window and record how many other passersby also looked up. The more researchers that looked up, the more passersby that looked up. This illustrated the importance of a threshold in influencing a network.
• Rule 5: The network has a life of its own –Christakis and Fowler give two reasons why the network has a life of its own. First, networks combine properties and functions that are neither controlled nor perceived by its members. They can only be understood by studying the whole network. Second, networks also have emergent properties. Emergent properties are new attributes of a whole that arise from the interaction and interconnection of the parts.

The Reach of Connection and Influence

One question that comes up is how far does an individual’s connections and influence reach into the social network? Christakis and Fowler (2010, pgs. 26 – 30) give different answers to the question for connections and influence.

For connections, they point to Milgram’s famous six degrees of separation experiment. In this experiment (Travers and Milgram, 1969), Milgram gave a few hundred people who lived in Nebraska in the USA a letter addressed to a businessman in Boston more than 2300 kilometres to the east in the USA. These people were asked to send the letter to someone they knew personally. The goal was to get the letter to someone they thought would be more likely to have a personal relationship with theBoston businessman. The number of times the letter changed hands was tracked and it was found that on average it changed hands six times. Dodds, Muhamad and Watts (2003) repeated Milgram’s experiment using e-mail instead of letters. This time 98,000 subjects were recruited.Each subject was randomly assigned a target from a list of eighteen targets in thirteen countries.The subjects sent an e-mail to someone the subject knew who might in turn know the targeted person. Again, it took roughly six steps to get the e-mail to the targeted person replicating Milgram’s results. Therefore, Christakis and Fowler conclude that an individual reach extended six steps or degrees into their networks.

For influence, Christakis and Fowler conclude that an individual’s reach is much shorter. They promulgate the three degrees of influence rule. This rule states that an individual’s influence through the network gradually dissipates and ceases to have a noticeable effect on people beyond the social frontier that lies at three degrees of separation. They give three reasons for this:

1. Intrinsic Decay Explanation – Influence eventually peters out as information loses it fidelity
2. Network Instability Explanation – Links beyond three degrees have a tendency to become unstable as the network evolves
3. Evolutionary Purpose Explanation – Humans evolved in small groups in which everyone was connected to everyone else by 3 degrees.

Networks and the Poor

So, how do the dynamics of the network affect the poor? Barabási argued that scale-free networks occurred because people want to connect to networks at its most central locations. This would result in those in the most central locations in the network retaining their central positions. Barabási called this the rich getting richer dynamic. This describes the situation that many of the poor find themselves as the rich maintain themselves at the centre of the most important networks in the nation and the poor find themselves on the periphery. Christakis and Fowler (2010, pgs 31 – 32) argue that the rich get richer dynamic of social networks can reinforce two different kinds of inequality. First is situational inequality where some are better off in socioeconomic terms. Second is positional inequality where some are better off in where they are located in the networks. Christakis and Fowler (2010, pg. 167) also argue that the rich get richer dynamic means that the positive feedback loop between social connections and success could create a social magnifier that concentrates even more power and wealth in the hands of those who already had it.

For Christakis and Fowler (2010, pgs. 301 – 302), the ability of network inequality to create and reinforce inequality of opportunity comes from the tendency of people with many connections to be connected to other people with many connections. This distinguishes social networks from neural, metabolic, mechanical and other nonhuman networks.The reverse holds true as well: those who are poorly connected usually have friends and family who are themselves disconnected from the larger network. They argue that to address poverty, the personal connections of the poor must be addressed. To reduce poverty, the focus should not merely on monetary transfers or even technical training; the poor should be helped to form new relationships with other members of society. When the poor on the periphery of the network arereconnected, the whole fabric of society benefits and not just any disadvantaged individuals at the fringe.