Content

Social Capital

Building on a Network-Based Approach

October 2003

This draft discussion paper was prepared for a PRI-SSHRC Policy Research Roundtable that explored the potential of a network-based approach to social capital and its implications for federal public policy. The Roundtable took place in Ottawa on October 20, 2003. This paper is intended for discussion purposes only, and does not necessarily reflect an official position of the Government of Canada.

Download the PDF version

Contents

Introduction: A Choice of Approaches

The Parallel to Human Capital

Social Capital as Function

Social Capital as Network

A Proposed Conceptual Framework

Implications for Public Policy Development

References

Please note: this draft discussion paper was prepared for a PRI-SSHRC Policy Research Roundtable that will explore the potential of a network-based approach to social capital and its implications for federal public policy. The Roundtable takes place in Ottawa on October 20, 2003. This paper is intended for discussion purposes only, and does not necessarily reflect an official position of the Government of Canada.

Introduction: A Choice of Approaches

Will social capital prove a robust concept for research and policy?

In recent years the concept of social capital has excited much attention and come to worldwide prominence within academic and government circles alike. But will it prove to be a robust concept for research and policy development purposes over the longer term? The answer to this question may well depend on some key choices we make today about what we fundamentally understand by ‘social capital'.

Conceptual confusion and over-versatility undermine its utility.

The current range and ambiguity in the meanings attached to the concept do not help in making a case for its practical value for policy and program development. Indeed, as Portes (1998) has warned, we may be approaching a point where the term has been applied in so many different contexts and to such a range of events as to mean everything and nothing. In part this current wide variation in understandings of social capital is arguably simply a natural reflection of a concept in an early stage of development. While the idea that the nature and quality of social relations have important implications for the well-being of individuals, communities and societies has a long history, viewing them as a form of capital resource potentially productive for particular outcomes is of more recent vintage. Given the rapid spread and popularization of this concept following, in particular, the seminal work of Pierre Bourdieu, James Coleman and Robert Putnam in the late eighties and early nineties, it should not be surprising that some have conveniently latched on to the high-profile label to advance their own particular projects with little concern for theoretical precision. One might hope that with time the rigorous and substantive core of research will prevail and a clearer understanding will emerge.

The core field is divided between functional and network-based approaches.

Yet even if we focus on this substantive core, we can find a significant division in how the concept is envisioned. On the one hand, some researchers see social capital primarily in terms of particular structures (social networks) and the resources that they convey. On the other hand, many prefer a functional definition of social capital whereby it consists of those social resources that enable cooperation and collective action. In practice, the two approaches have much in common. Those who focus on social networks frequently do so precisely because of their potential to encourage cooperative behaviour, and those who take a functional approach to social capital often identify social networks as a key mechanism for pursuing common objectives. Nevertheless, the differences between the two approaches are not mere semantics. This discussion paper attempts to outline some of the ways in which this basic difference in conceptualization, between structure and function, matters to how we operationalize the term and the research agendas we might choose to pursue in studying the productive potential of this capital resource. At the same time, it will attempt to propose a framework for analyzing social capital that draws on the strengths of both approaches.

While defining social capital in terms of networks, a broader framework for analysis must not neglect issues raised by the functional approach.

Through recognizing issues raised by a functional approach in defining the proposed broader framework for undertaking social capital analysis, while maintaining a narrower, network-based definition for the core social capital concept, it may indeed be possible to draw on both approaches. That is, the study of social capital should be multi-dimensional and dependent on the needs of the particular application in question, including analysis of the investments that individuals make in setting up and maintaining social networks, associated norms and values, institutional dimensions, as well as the benefits received by individuals and society. Studies using such a framework can be quite different from each other, but they will always require a common, measurable, core if they are to result in the development of a body of empirical knowledge that will support future policy analysis in areas that cannot easily be predicted in advance.

Such multi-dimensional analysis can be consistently undertaken over a wide range of policy and research applications provided that – in all these functional applications – we maintain a strict core definition of social capital that is based on networks. A narrow definition of the core concept is at least potentially measurable in a consistent way and parallels the tight definitions used in studies of physical or financial capital as well as human capital, thereby facilitating cross-cutting social and economic empirical analysis.

The following tentative definitions of social capital and of a social capital framework for analysis are offered as points of departure:

Social capital refers to the networks of social relations that provide access to needed resources and supports.

A social capital framework to support research and policy analysis uses the core network concept but is multi-dimensional. Depending on the particular research and policy application, social capital studies should encompass, for example, the investments that individuals and collectivitities make in the establishment and maintenance of social networks, the various characteristics of the networks and of transactions, the norms and institutional frameworks in which such networks operate, the resources that can be potentially accessed through participation in the networks, and the returns to those investments in the form of economic, social and health outcomes for individuals, communities and societies.

The Parallel to Human Capital

An analogy to human capital may be helpful in evaluating the two approaches.

In order to evaluate the comparative advantages of the two basic approaches to social capital, it may be helpful to begin by first thinking through what would make it a form of capital generally, and how it might compare to human capital in particular (see, for example, Lin 2001, ch. 2). In general, capital is a resource in which one might invest and develop, and which may be then employed to generate a future flow of benefits. In economics, capital has classically referred to various physical assets, both private and public, that may be developed and used to produce goods and services, including such things as machinery, buildings, transportation networks and the like.

Human capital was introduced as a concept forty years ago to recognize the productive value of knowledge and skills.

Some forty odd years ago, the notion of human capital was introduced in the work of Theodore Schultz and Gary Becker. Theodore Schultz famously made a case for the adoption of the concept of human capital in his widely-cited 1960 Presidential Address to the American Economic Association (Schultz 1961). His central thesis was that human knowledge and skills formed a capital asset, and that the “failure to treat human resources explicitly as a form of capital, as a produced means of production, as the product of investment, … was wrong in the classical period and patently wrong now” (p.3). Indeed he suggested that this failure to conceive of the acquisition of knowledge and skills as a capital investment and to evaluate their productive potential as such had blinded researchers and public policy practitioners to their full importance. For example, reflecting on his work together with other economists following the Second World War, Schultz noted that they had badly underestimated the resilience and efficiency with which Western Europe would recover from the war. This he attributed to their overemphasis on the war-time destruction of physical capital and a failure to understand the role of human capital in production in an industrialized economy. Schultz contended that the increase in national outputs in Western economies stemmed in large part from investments made in human capital, and that more generally investments in education and training could yield significant returns over a long period for both individuals and economies. He further suggested that the adoption of a human capital lens had a number of public policy implications. For example, he pointed to then-existing tax laws that discriminated against investment in human capital in the United States, and called for greater efforts to provide the funds for students to invest in education and thereby their future.

In subsequent decades, the concept of human capital has proven very useful in orienting research and public policy development (for an overview see Healy and Côté 2001, Chapter 2). Researchers have examined the degree to which various specific investments in knowledge and skill development have paid off both for individuals in terms of improved employment earnings and personal well-being, and more generally for the overall health of the economy and society. Similarly, they have been able to examine the dynamics and costs of investments in knowledge and skill development by individuals and firms, and the role of various factors in shaping these investment choices. So too have they been able to explore the role of public policy and expenditure programs, both in shaping the decisions and opportunities to invest in education and training, and in affecting the way the ways in which these investments can translate into labour market opportunities.

The concept of human capital has helped to identify investment choices, returns on investment, and a clearer role for public policy.

Despite its limits, the concept has proven useful for research and policy.

Of course human capital theory is not without its shortcomings (again see Healy and Côté 2001, Chapter 2). Efforts to measure stocks of human capital remain limited, with emphasis typically placed on educational credentials as a simple but weak proxy measure. Poor data quality and a still rudimentary understanding of the complex interactions of human capital and economic growth have hampered efforts to measure the impact on growth. Nevertheless, the introduction of the concept has proven highly useful for allowing researchers to explore the returns from human capital, the dynamics of investing and developing in the same, and the role for public policy in this area.

Social capital also seems to be a concept that points to an important asset.

Will social capital prove to be similarly helpful? Certainly the idea that social relations can form a capital asset has received considerable attention recently. Part of the readiness of so many scholars and researchers to explore the concept may well lie in the intuitive sense this seems to make for most people. As Woolcock and Narayan (2000:3) put it, “… the basic idea of social capital is that a person's family, friends and associates constitute an important asset, one that can be called on in a crisis, enjoyed for its own sake, and leveraged for gain.” Whether you seeking support in hard times, looking for a night on the town with friends, or searching for a new job opportunity, who you know matters. Moreover, the authors argue that what is true for individuals is also true for communities: those with a stronger stock of social capital are able to more effectively negotiate the various challenges they may face.

There is broad interest in its productive potential.

When one looks to the major approaches to social capital, one finds a broad common interest in how the dynamics of social relations might constitute an important asset and be productive of various outcomes. For example, Bourdieu (1986) was interested in how elites could call upon their social networks to reinforce and reproduce their privileged status. James Coleman (1988) examined how social capital in tightly bonded communities helped support family expectations for their children's education and thereby reduce high-school dropout rates. Putnam (1993, 2000) has argued that networks of community engagement can engender norms of reciprocity and trusting dispositions, which in turn may generate greater social collaboration and more effective democratic institutions.

Network-based approaches are more clearly analogous to human capital.

Yet if there is a common interest in exploring how social relations may form a capital asset, there is a wide disparity in how social capital is conceptualized more precisely. In particular, as noted above, the leading approaches may be broadly divided between those who focus on social capital as a structure in the form of a social network, and those who take a more functional approach and view social capital as being those social resources that enable effective cooperation. In practice, this distinction has an enormous impact on the ability to identify what might constitute social capital and how it functions, including being able to distinguish social capital from its determinants and its outcomes. If we are to think of social capital as a form of capital, and if this is to be useful for research and policy development, then ideally we should be able to clearly identify what it is, explore its productive potential and identify the means by which it can be invested in and accumulated. Network-based approaches to social capital may more readily meet this test. Before exploring the usefulness of such an approach, this article will first consider the case for a functional conception of social capital.

Social Capital as Function

The influential functional approach can be traced to James Coleman.

At present, the functional conceptualization of social capital may be the most widely adopted and influential approach. This is especially the case in particular disciplines, such as political science. The seminal work in proposing a functional understanding of social capital is that of a sociologist, however: James Coleman (1988, 1990). Coleman (1990: 302) argued that:

Social capital is defined by its function. It is not a single entity, but a variety of entities having two characteristics in common: they all consist of some aspect of a social structure, and they facilitate certain actions of individuals who are within the structure. Like other forms of capital, social capital is productive, making possible the achievement of certain ends that would not be attainable in its absence. … Unlike other forms of capital, social capital inheres in the structure of the relations between persons and among persons. It is lodged neither in individuals nor in physical implements of production.

The forms of social capital he identified include obligations and expectations, trust, information potential, norms and effective sanctions, authority relations, appropriable social organizations, and social networks.

If Coleman was the key figure in introducing a functional definition of social capital, its widespread impact on research agendas must be principally credited in particular to the early works on social capital by the Harvard political scientist, Robert Putnam (1993, 1995). Although Coleman (1990: 302) explicitly conceptualized social capital as an asset of individuals, Putnam has been more interested to explore the ways in which it represents a collective asset. In his 1993 study of the comparative effectiveness of regional government in Italy, Putnam adapted Coleman's approach and defined social capital as “features of social organization, such as trust, norms, and networks, that can improve the efficiency of society by facilitating coordinated actions” (p.167). A similar definition was offered in the hugely influential 1995 article entitled “Bowling Alone,” in which he suggested that a decline in social capital in the United States in the form of an erosion of civic engagement was undermining the effectiveness of its public institutions.

A functional approach refers to features of social organization that enable cooperation.

Putnam's early work on social capital helped to spread the functional approach.

Why has this approach to defining social capital been so influential? For many, quite simply, Robert Putnam's initial analyses of the concept represent the source for defining social capital. Putnam is reputed to have been the most widely cited social scientist in the 1990s. His oratorical skill combined with his knack for linking the concept to a number of major public policy concerns, have played a pivotal role in the popularization of social capital. It is probably fair to say that today most people have come to know the concept by way of Putnam.

Identifying the factors that enable cooperation and collective action is a compelling research problem.

Clearly, though, a functional conceptualization of social capital as offered by Coleman and Putnam would not have found a sustained audience if the basic ideas did not hold considerable merit. Certainly the attention to the role that social relations play in enabling and sustaining various outcomes is viewed by many as a welcome correction to approaches to economics and politics that have neglected the importance of the social sphere. Further, the core problematique this approach attempts to resolve, either for individuals or for collectivities, is indeed a compelling one. Groups, families, neighbourhoods, and societies in which people are willing to cooperate, in which people are predisposed to trust one another, and in which collective action is welcome, encouraged and enabled, will be able to accomplish much more than where this is lacking. Yet the wide variation in levels of such cooperation and support for collective action underline the difficulty in its achievement. A functional approach seeks to identify the factors that enable the productive potential of socialbility, and thereby open them to possible development through public policy intervention. Indeed, given the spectre of a decline in social capital in the United States first raised by Putnam in 1995, the case for pursuing this line of inquiry would seem to be especially urgent, particularly for its public policy implications.

But the approach has been criticized.

And yet this functional approach to social capital has also been heavily criticized. For example, Coleman's rather vague definition and the “laundry list” of forms — a list that arguably conflates determinants, sources, and outcomes of social capital — has been lamented by Portes (1998) for having opened the way to confusion and contradiction in the wider social capital literature. Similarly, Portes suggests that the functional approach of Putnam is tautological and that (at least in the case of his 1993 work on Italy) fails to adequately distinguish the definition of the concept both theoretically and empirically from its attributed effects.

Functional definitions are tautological.

How much merit do these criticisms hold? Certainly, functionally defined concepts are frequently criticized as being tautological, and this has been the case for such approaches to social capital as well (see Lin 2001 and Lévesque and White 1999). In discussing Coleman's approach, for example, Lin (2001: 28) argues that his functional view may well be tautological in that for Coleman

Social capital is identified when and if it works; the potential causal explanation of social capital can be captured only by its effect … This is not to deny that a functional relationship may be hypothesized (e.g., resources embedded in social networks may make it easier to obtain better jobs). But the two concepts must be treated as separate entities with independent measurements (e.g., social capital is the investment in social relations, and better jobs are represented by occupational status or supervisory position). It would be incorrect to allow the outcome variables to dictate the specification of the causal variable (e.g., for actor X, kin ties are social capital because they channel X to get a better job, and for actor Y, kin ties are not social capital because they do not channel Y to get a better job).

Thus critics of a functional approach suggest that it fails to adequately distinguish what social capital is from it supposedly does. And yet only if you can isolate the supposed cause from the putative effect can you adequately test the degree of validity of this causal relationship.

The explanatory utility of the functional approach is undermined by a lack of theoretical parsimony.

Moreover, as Lin (2001) notes, a single theory that needs to account for all elements of social resources that lie behind every instance of social cooperation quickly loses its parsimony and thereby its explanatory utility. Of course, it is precisely the lack of parsimony in this theory that has enabled so many different researchers from different backgrounds to include their particular interests under a social capital banner. Social networks, norms and trust are perhaps the most frequently cited forms of social capital. Many are also interested in how public institutions as well as other political and legal arrangements are forms of social capital when and where they encourage and coordinate cooperative collective action. Indeed, a functional approach leads to an ever-broadening list of those elements that may be considered to be social capital. However, this continually expanding list makes it hard at times to isolate social capital from other forms of capital while grouping together many distinct phenomena with different, perhaps even contradictory, dynamics. For example, a strong economy and abundance of equitably distributed resources might be considered a form of social capital to the degree to which it facilitates greater cooperative behaviour. Similarly, knowledge of interpersonal dynamics and developed social skills may similarly contribute toward cooperation. Yet labeling them as social capital may obscure more than enlighten.

Trust and social networks may have quite separate dynamics that may be obscured in a functional approach.

Similarly, many view trust as a key form of social capital. Yet while trust is indeed a moral and cultural attribute of individuals that conditions or constrains the ways in which people behave towards one another, it is clearly theoretically possible to have very high levels of trust while only engaging in a bare minimum of social interaction. A shy person with few social connections may nevertheless have high levels of trust in public institutions to provide necessary services or may confidently trust strangers in the neighbourhood not to do them harm as they pass by on the street. Indeed, trust itself is a complex phenomenon that may have a very different and separate dynamic from other supposed forms of social capital such as social networks. As important as it undoubtedly is to study the interactions between forms of trust and social networks, an approach that too quickly lumps all these elements together as being part of one overall dynamic may do a disservice in failing to adequately identify and isolate their independent characteristics and effects.

Putnam's recent work is more clearly network-based, but he still views social trust as a proxy indicator.

Although Putnam's use of a functional definitions of social capital in his earlier works on the topic (1993, 1995) may be in large part responsible for the influence of this approach, by the time of his book-length treatment of the decline of social capital in 2000 (also called Bowling Alone ) his definition has shifted to one more centred on social networks; he defines social capital as “connections among individuals – social networks and the norms of reciprocity and trustworthiness that arise from them” (2000: 19). Putnam (2000, 2001) now seems to argue that norms and trust, which the earlier functional definitions had seemed to suggest could possibly be treated as distinct forms of social capital, are only part of social capital to the degree to which they arise from social networks. He argues that social networks, particularly networks of civic engagement, promote and sustain norms of generalized reciprocity which in turn encourage greater dispositions to trust one another, thereby enabling greater social cooperation for mutual benefit. Thus, while Putnam now more clearly grounds his understanding of social capital in a network-based approach, he continues to be interested in norms and trust which are presumed to be important outcomes of social networks, which in turn still address the central functional concern of enabling collective action. To be fair, Putnam does at one point caution that “the causal arrows among civic involvement, reciprocity, honesty, and social trust are as tangled as well-tossed spaghetti. Only careful, even experimental, research will be able to sort them apart definitively. For present purposes, however, we need to recognize that they form a coherent syndrome” (2000: 137). Nevertheless, the basic assumption behind much of Putnam's book is that social networks of civic engagement, particularly in the form of participation in voluntary associations, will produce generalized norms and increased levels of trust. Consequently, this allows Putnam to suggest that “social trust is not part of the definition of social capital but it is certainly a close consequence, and could be easily thought of as a proxy” (2001: 45).

Empirical studies suggest some grounds to be cautious about using trust as a proxy, however.

However, the assumed connection between these elements has been called into question, which means that we should be cautious of using trust as a proxy. For example, Stolle (2003) states that “… the assumption of most social capital theorists as to the efficacy of voluntary associations in producing generalized norms and values such as trust should be taken with caution at best. There is no strong empirical evidence to confirm the microrelationship between membership and trust” (2003: 35). Indeed, Stolle offers some evidence that suggests that, in fact, participation in voluntary associations may well not produce greater levels of generalized trust. Given the importance that Putnam attaches to network connections based in association membership, this suggests we should be wary of using measures of social trust as proxy indicators for the strength of social networks.

More empirical study of the relationship between trust and social networks is needed.

So, while questions of general social trust may be deeply important to investigate in their own right, one cannot assume that social trust is necessarily a product of social networks. The relationship between the two phenomena remains a question for empirical testing. To the degree that a functional approach lumps multiple phenomena under one label of “social capital” and thereby obscures their independent dynamics, this does a disservice to developing a sound basis for rigorous research and ultimately for evidence-based policy.

A functional approach raises a vital issue, but its definition of social capital may too often obscure more than enlighten.

This said, the broad question of identifying the sources of social cooperation and collective action remains both a valid and vital concern for research. And good and careful scholarship may be able to employ a functional definition while avoiding the potential pitfalls of this approach. Nevertheless, a functional definition can too often support confused research that obscures more than it enlightens. Moreover, efforts to group all phenomena that produce collective action into a single theory may prove insurmountably complex and, over the longer term, limit the utility of the concept for research and policy development. Finding a means to more clearly define social capital without losing sight of the fundamental problematique raised by functional approaches may help to avoid this predicament.

Social Capital As Network

Network-based approaches, such as that of Bourdieu, offer cleaner definitions.

In contrast to functional conceptualizations, network-based approaches to social capital may offer a much cleaner definition. To this end, many scholars have come to ‘rediscover' the work of Pierre Bourdieu on social capital. Bourdieu defined social capital as “the aggregate of the actual or potential resources which are linked to the possession of a durable network of more or less institutionalized relationships of mutual acquaintance and recognition—or, in other words, to membership in a group”. For Bourdieu, “the volume of social capital possessed by a given agent … depends on the size of network connections he can effectively mobilize and on the volume of the capital (economic, cultural or symbolic) possessed in his own right by each of those to whom he is connected” (1986: 249).

There are a wide variety of network-based approaches.

Though there exists a considerable variation in network-based approaches to social capital (see Adler and Kwan 2002: 22-23), most approaches share with Bourdieu the view that such approaches must be concerned to study both the structure of the network and the resources contained within the network that may be drawn by its members. This said, various authors associated with a network-based approach differ somewhat in their definitions of the concept. For example, there are those who focus primarily on the networks themselves (e.g., this paper), those who include equally the networks and the resources they convey (e.g. Bourdieu), and those whose definition stresses the resources accessible through networks (e.g., Lévesque and White, 2001). More generally, Adler and Kwan (2002) argue that network-based approaches are split between those that are principally concerned with the effects generated by the pattern or structure of relations within a network, and those that insist that the content of those relations, such as particular norms or levels of trust, must also be investigated.

Critics fear the approach is too narrow...

Although the different authors who have taken network-based approach have been subject to a variety of individual critiques, in general many critics fear that network-based approaches are too narrow, and that they leave out too many important dimensions, such institutional and legal arrangements, that may be important in understanding and explaining various social phenomena, particularly cooperative collective action. Given the compelling case made by many within the functional approach for understanding these broader issues, an inability of a network-based approach to explore such concerns would be serious indeed.

... but greater parsimony improves explanatory power ...

This said, such concerns may be misplaced. While network-based approaches to social capital may be more modest and parsimonious than functional definitions, this may in fact greatly increase the potential explanatory power over the longer term. Rather than opening the door to an ever-expanding list of social resources that are purported to function as enablers of collective action, defining social capital in terms of social networks allows one to better define the concept, distinguishing it both from other forms of capital and from its purported effects. This in turn allows for more careful empirical testing of the theorized connections between the determinants of social capital, its outcomes, and social capital itself. Moreover, it does not force one to conclude that social capital is absent if its theorized effects are not perceptible.

(Lévesque and White 1999: 28).

... and a wide variety of phenomena and research problems may still be studied.

Defining social capital in terms of social networks does not mean that one cannot investigate its possible role in promoting collective action, or the ways in which it interacts with norms, trust, institutions, the wider socio-political environment and the like. Indeed, by carefully distinguishing between these phenomena rather than lumping them into a single dynamic one may more carefully study and test the supposed connections between the various elements. Which of these additional elements one wishes to study will vary with the particular problem one wishes to understand. Nevertheless, a common conceptual focus on social capital as social networks provides a means of ensuring consistent measurement across a variety of research and policy applications.

Network-based approaches to social capital may prove useful in the same way as human capital.

Potentially, one may more clearly understand and empirically study social capital as a form of capital through a network-based approach. Human capital has proven a robust concept over the past forty years because its core elements have more or less been clearly identified as knowledge and skills, and this has allowed researchers to organize their efforts to evaluate the productive potential of this form of capital and to study the dynamics by which individuals invest in and accumulate human capital. This in turn has allowed for analyzing the potential role of public policy in encouraging the most beneficial forms of investment in human capital and for maximizing the returns on this investment, particularly in the labour market.

The next section lays out an analytical framework for a network-based approach to social capital.

In the social capital framework proposed in the following section of this paper , the ability to outline more clearly what social capital is, how it is invested in and developed, and its potential returns are the strengths of its network-based approach to the concept. At the same time, it still offers the opportunity to explore and examine the broad concerns raised by those working with the functional approach. The paper then concludes with an overview of some policy implications suggested by this framework.

A Proposed Conceptual Framework

In order to better organize and coordinate social capital research efforts within the Government of Canada, the development of a network-based conceptual framework for social capital may prove fruitful and it is to this task that this paper now turns.

We have an opportunity to adopt a more specific approach to social capital.

Our intention is to steer clear of a multi-faceted approach that ends up embracing and confounding a disparate variety of concepts, as this would risk rendering the concept of social capital less meaningful and less useful. Instead, there is an opportunity to adopt a much more specific approach that "gets back to basics” and uses something closer to Bourdieu's notion of social capital (focused on social networks and the resources they enable) to address a range of policy questions. It may or may not be important, for example, to discuss norms of reciprocity and levels of trust specific to networks of social relations, depending on their importance for the problem in question. This is distinct from approaches to social capital that see trust and norms as forms of social capital, independent of any necessary connections to social networks.

Nan Lin (2001: 243) has argued that a comprehensive, network-based model of social capital must allow for the investigation of three elements:

• the determinants of investment in social capital;
• the ways in which social capital may be accessed and mobilized; and
• the returns on investment in social capital.

A leaner definition offers greater constancy when applying the concept of social capital across a variety of policy problems.

The findings from an interdepartmental workshop on social capital as a public policy tool held in June 2003 point to a number of further key considerations in developing an analytical framework for analyzing the concept. First, the basic definition of social capital should be lean and focus on social capital as social networks, thereby allowing for greater consistency in applying the concept to a variety of policy problems. That said, the model should also allow investigators to examine various attributes and characteristics of social networks depending upon their relevance to the research or policy problem in question, as well as to potentially investigate its interaction with other forms of capital resources and the wider social, political and economic environment in contributing to particular outcomes. Again, this is in contrast to approaches that see trust and norms as forms of social capital, independent of any necessary connections to social networks, or group other forms of capital and environmental factors under the label of social capital itself.

For the purposes of our framework, social capital refers to the networks of social relations that provide access to needed resources and supports. Any study of social capital should encompass the investments that people make in establishing and maintaining these networks, the norms and institutional frameworks in which such networks operate, and the returns to those investments in the form of economic, social and health outcomes for individuals, communities and societies.

The framework proposed in this paper is basically a systems approach that allows us to see how social capital is developed, how it operates, and how it may contribute to the achievement of various outcomes. The system does not operate in a vacuum. Rather its components are embedded and are influenced by a broader context, and these very interactions can be subjects of interest and policy relevance in themselves. The systems framework in Figure 1 is presented very much as a linear process. Like any model, of course, it is a highly abstracted simplification of reality. In practice, many more interactive arrows and feedback loops would need to be identified to reflect the particular dynamics of any given social network. For the sake of clarity, however, the diagram has been kept in as simple a form as possible.

The framework is basically a systems approach embedded in a broader context.

The framework presents the various elements that may help to investigate the ways in which social networks constitute a form of capital. More generally it provides an approach to taking a particular policy research question and thinking through how social capital can be an important factor, combined with others, in influencing various kinds of outcomes. Illustrative examples of such outcomes might include:

• educational attainment of youth (and of youth at risk in particular)
• the labour market participation of new immigrants (securing and retaining appropriate employment, earnings and career advancement)
• successful transfer of scientific knowledge from universities to businesses leading to innovation and product development
• health status of Aboriginal communities
• resilience of remote communities to economic challenges (such as those posed by structural economic changes)
• improved neighbourhood safety (reduction of personal and property crime rates)

The framework emphasizes explanations and policy implications at the micro-level (individuals, groups and goals in specific contexts). As we will see, it also provides the flexibility and the frame of reference to examine other related issues at various levels of analysis. It is important, however, to ensure that any examination of social capital is grounded in a specified research question or policy problem, and that the model is applied in light of that particular context.

We will generally be interested in the positive social, economic and health outcomes that social capital may facilitate.

Another point worth reinforcing is that, for policy purposes, we will normally be interested in “desired” outcomes – such as positive social, economic or health outcomes for individuals, groups or communities – that would generally be regarded as benevolent. This is not to suggest that social capital cannot be put to negative purpose; indeed it can, as can other forms of capital (e.g., financial, physical or human capital). It is still highly worthwhile to examine the role various forms of capital may play in the achievement of positive outcomes or how public policy might incorporate them as tools to achieve various goals.

We will now examine each of the components of the framework. It is important to note that, depending on the policy question at hand, various subsets or combinations of these components, and aspects of their context, can be worthy of attention – and they may also suggest the use of different kinds of policy levers.

Focus on Determinants

Determinants of social capital formation can operate both at the individual and at the community level.

The framework gives us the option of examining not only how social capital can be exploited, but also how people invest in it and how it is developed. Determinants of social capital formation are factors that influence the extent to which people are predisposed and able to join or form networks. Two main categories of factors are important here: individual-level and community-level determinants. Individual-level determinants of social capital formation may include such factors as demographics, employment history, education, ethno-cultural background and various perceptions and attitudes. Examples of community-level determinants include: the presence of existing networks, the existence of infrastructure conducive to supporting network development, and prevailing norms and values.

The determinants of interest will be driven by the policy question at hand.

Depending upon the particular problem we are interested in investigating, varying subsets of these determinants may prove to be relevant. Their potential role as determinants must be empirically determined, but this means that one must be careful to distinguish the determinants of investment of time, energy and resources into developing social networks from the networks themselves. In addition, the opportunity costs of network formation and maintenance may also be an important determinant.

Inside the “Black Box” of Social Capital

We also need a way of systematically examining the dynamics of what is at the heart of social capital – social networks themselves – to understand how the resources they convey can be accessed and mobilized. To this end, there are a number of factors on which we may wish to direct our attention. Again, a particular study will not examine each and every one of these factors. Instead, the policy question at hand will suggest the factors that are most relevant in a given context.

Network Characteristics

Key dimensions include:

… network structure and composition

First, we may wish to examine various characteristics of the network. Here we might focus on the structure and composition of social networks. This has traditionally been the focus of much of network analysis. It includes considerations of size, density, strength of ties, degree of heterogeneity, etc. Understanding whether the network is primarily composed of family members, business colleagues or hobby enthusiasts may also be important. As discussed above, the distinctions between various structural characteristics of social capital (as exemplified by the distinctions between bonding, bridging and linking social capital, or how “opened” or “closed” specific networks are) may have diverse and important implications for particular outcomes.

… resources and abilities embodied within the network

Further, an examination of the resources and abilities of members of a social network may be critical in determining the productive potential of a given network. Here it is useful to underline that certain networks may convey a number of resources that are useful for one particular outcome, but be completely irrelevant in other situations. Their relative value will depend upon the particular context and desired outcome in question.

… normative characteristics

… and network dynamic.

he normative characteristics of the network may also be of interest. In the context of a specific network, values and norms (including those of trust and reciprocity) can be critical to understanding the functioning of that network. Finally, it is important to appreciate that networks are not static entities, but rather they may shift and evolve or even disappear over time. Network dynamics , therefore, often may be a topic of analytical interest and policy relevance.

Network Transactions and Outputs

We may also frequently be keenly interested in the transactions that take place within a network structure as the resources it conveys are accessed. Examples of such transactions may include processes of negotiation, the sharing of information, encouragement and support, or the application of sanctions.

Network transactions and their immediate outputs won't necessarily produce positive outcomes.

The framework model also allows us to examine the potential returns of the investment in the network. Network outputs are the immediate results of the network transactions. For example, an individual might seek to draw on the information resources of members in their social network in order to get a new job. The immediate result of a given network transaction may take the form of information as to a suitable vacancy. This immediate output, however, will not necessarily translate into a positive labour market outcome for the individual – other factors may influence the final outcome.

Other Complementary Resources and the “Opportunity Structure”

Complementary resources, including other forms of capital, will often be necessary.

Other intervening factors that may feature in a given analysis include the presence (or not) of complementary resources and the prevailing opportunity structure. Social capital will frequently prove insufficient on its own to realize a particular outcome. The presence of complementary resources including physical, financial or human capital may be important factors or even necessary conditions. Indeed, ensuring the presence of sufficient complementary resources might be one key site of public policy intervention in achieving various outcomes.

The opportunity structure involves the relevant aspects of the broader context.

In this framework, the term “opportunity structure” is used to describe those aspects and configurations of the broader environmental context that may influence whether desired outcomes will be realized. Assessing the compatibility of the broader context to the translation of network outputs into desired outcomes may at times prove to be essential in understanding the potential returns on investment in social networks. A downturn in the business cycle or the existence of systemic racism, for example, may limit the potential of social networks to respond to the particular needs of specific individuals or communities.

The major point is that a successful network output will not necessarily result in a desired outcome. Various complementary resources and environmental conditions need to be in accord for this to happen.

Focus on Outcomes

The final element of the framework is the final outcome of this model. In fact, this may well be the first consideration in deciding which aspects of social capital we will operationalize in a given study. Indeed, the degree to which social capital is productive of particular outcomes goes to the heart of its potential importance: we are not interested in social capital for its own sake, but because of the extent to which it plays a role in achieving various desirable outcomes for individuals and communities.

The social, economic and health outcomes we would like to see achieved will often serve as a starting point for a given social capital analysis.

The specific outcomes in which federal public policy researchers are interested will depend upon the policy objectives in question. Not all factors will be equally relevant for all outcomes. Nor will they all need to be studied in all cases. Thus, although the outcomes are labeled at the end of the diagram, they will be typically the place where federal researchers and program developers will begin their analysis. The specific types of outcomes we are interested in (e.g., social, economic and health) will determine the aspects of the framework that we bring into our analysis, as well as the levels (e.g., individual, community, society) at which that analysis takes place.

Benefits of a Common Analytical Framework

A coordinated approach to social capital could offer a powerful analytical and policy tool.

The potential is there to harness the concept of social capital as a powerful analytical and policy tool. But much research and policy discourse is still quite imprecise about what social capital is, and fails to distinguish between its core components, its sources, and outcomes. So we need to be rigorous in its conceptualization and measurement since much "fuzziness" remains. We also need to resist the temptation to fall back on a "grab-bag" approach to social capital that embraces and confounds a diverse array of social phenomena under a single but confused banner of “social capital.”

The opportunity exists to develop a coordinated approach and an analytical framework that can be adapted to various policy areas in the Canadian context – and this is precisely the purpose of this proposed framework. By focusing on network determinants, characteristics and transactions, the proposed framework offers a consistent language that can be applied across a variety of policy concerns.

Implications for Public Policy Development

Policy Applications

What, in practice, are the implications of using a social capital framework (with a focus on social networks) for developing new policy responses? What are some of the main areas for policy application?

Using social networks (and their various structural and qualitative attributes) as the main basis for registering the presence or assessing the effects of social capital, can be helpful in various ways. Some of these include:

• Allowing policymakers and program developers to concentrate on social capital as a resource for action in practical, local-level situations, from the point of view of individuals and communities;
• Considering social capital as a “value-neutral” instrument for achieving policy outcomes, rather than a policy objective in itself;
• Making a distinction between what social capital is from what it does; and
• Building on the explanatory power of the structural and normative characteristics of networks found in the “bonding, bridging and linking” typology of social capital.

A Focus on Individuals and Communities

A network-based approach is particularly helpful in micro-level applications.

A conceptualization of social capital that focuses on networks is particularly helpful for understanding practical, local level issues from the point of view of individuals and communities. Very often, other approaches to social capital have concentrated on large-scale (or macro) issues that investigate such things as trust in government or voter participation in elections as proxy measures of the strength of social capital at a national level. Under a network-based approach, the policy potential of social capital reflects more the realities of individuals, organizations or groups in relation to a specific problematique and context.

The role of social capital in labour market outcomes is a good example.

ne useful micro-level application of social capital lies with the ways in which social connections and networks factor into labour market outcomes for individuals (particularly those individuals with employment-related difficulties, such as social welfare recipients, or members of “at-risk” groups in the population). Investigating the role of social capital through an analysis of the characteristics and use of social networks in relationship to employment can reveal important information on:

• the importance of social capital as a determining factor for achieving and maintaining attachment to the labour market (in relation to other well known predictors of labour market outcomes, such as human capital);
• the conditions that are important for enabling individuals to access and benefit from the potential of certain diversified connections and social relations; and
• the value of including measures aimed at building diversified social networks as integral components of active labour market programming.

At the group or community level, mapping and analyzing the characteristics and functioning of social networks can provide useful information on the conditions that contribute to improved health, social and economic outcomes. A social capital lens can be used, for instance, in better understanding the factors that influence the capacity of isolated communities to make effective use of scarce financial, natural or physical resources for achieving economic self-sufficiency, compared with other communities with similar characteristics.

Understanding the policy implications of social capital goes beyond looking at network structure, and includes other elements identified in the proposed conceptual framework.

The idea here is not simply to register the presence of social capital by examining the number and type of resident associations, community organizations or social groups in the community. It also involves a better understanding of the level and nature of the ties that are built over time between disparate groups, individuals and institutions that make up communities, neighborhoods or localities. For example, Saegart et al. (2002) have investigated the role of social capital in the prevention of crime in New York City's low-income housing developments. They have found that understanding the dynamics of networks in the form of tenants associations, including the presence of tenant pro-social norms and links with a building's formal organization, have been found to be particularly relevant for explaining how certain housing developments are more effective at preventing crime than others. Understanding the factors that allow tenant networks to address challenges related to crime and juvenile delinquency has provided valuable information for public agencies intervening in these neighborhoods. It also has provided important lessons learned for policymakers involved in the design and implementation of housing policy, alternative home ownership structures and building characteristics.

Social Capital as an Instrument for Action

Social capital is best viewed as an instrument in achieving a variety of outcomes, for better and for worse.

Social capital is best understood as a potential means or instrument for achieving policy objectives, rather than an objective in itself. Indeed, social capital in and of itself is value neutral. As indicated by Briggs (1997), “… social capital has no right or wrong to it until some judgment is made about the ends to which we put it.” Social capital can as easily be used to exclude as to include, thereby reinforcing inequalities. Policymakers must be careful, therefore, in promoting any specific social relationships. They must also recognize that social capital is only one of several potential factors that may be key in the achievement of broader socially desirable objectives, whether they be higher school completion rates, safer neighbourhoods, more responsive government or healthier babies.

Identifying strategic opportunities to support social capital is a challenge for policy and program development.

The challenge lies in identifying strategic ways of supporting investments in those forms of social capital that may prove a useful resource in specific contexts. This applies in particular to members of more vulnerable populations, who have the opportunity to develop existing sources of social capital and treat these as assets. These forms of social capital may help compensate for, or provide access to, other types of capital (e.g., financial, human, physical) that these groups may be lacking. As is the case with other assets, however, social capital is unlikely to be equally distributed. Policies might have a role in this context to preserve existing stocks of social capital in communities and to ensure more equal access to this resource. This can be done by influencing certain determinants of social capital formation, or by facilitating the mobilization of social capital that is sometimes left unused in the absence of sufficient public support.

Understanding How People Invest in, Accumulate and Use Social Capital

Developing social capital as a public policy tool requires a good understanding of how it works.

In order to turn social capital into a productive resource for individuals, communities and organizations, we need to be able to understand the conditions that affect how individuals and groups have access to, invest in, accumulate and use social capital. The possibilities for developing forms of social capital that can be productive in the labour market, or in school or for living a healthier lifestyle, can vary from one individual to an other, depending on characteristics such as race, gender, culture, religion and education. These possibilities also vary according to the broader social, cultural and institutional environment in which people live and make these connections.

For example, researchers in New Zealand (King et al. 2003) have developed a research design that focuses on the determinants of social capital formation for explaining the employment experiences of people in disadvantaged populations, including members of Aboriginal communities. In this case, the researchers are interested in looking at the effects of family status (single men and women) and culture on the capacity to establish and maintain productive social networks. If conclusive, these study results could offer insights for the design and implementation of labour market interventions geared at these populations.

Understanding how to support investments in the potential value of social capital as a resource for addressing particular challenges might also require that we take a closer look at how social networks operate to create productive resources. It might require that we look at how social capital manifests itself, either through greater trust, exchange of information, or cooperation, to produce various outcomes.

In a case study (Dominguez and Watkins 2003) using a social capital framework that differentiates the sources of access to social networks from the effects of participating in those networks, researchers have been able to draw important insights into the changing role of social capital on the mobility opportunities of low-income mothers. Creating and maintaining support relationships for these women can often be challenging work that involves learning how, when and whom to ask for assistance. In describing the conditions under which strategies for using different types of social networks change over time, these studies emphasized the importance of viewing the production and use of social capital as a process rather than a static phenomenon. Findings reveal that beyond economic resources, several factors (including family tensions and terms of reciprocity) constrain, complicate or promote social capital for low- income mothers. This, in turn, was found to have important implications (positive and negative) for their daily survival and socio-economic mobility.

Bonding, Bridging and Linking Social Capital:
Understanding the Different Roles of Social Networks

Bonding, bridging and linking social capital play different roles.

A focus on network characteristics allows us to take advantage of the explanatory force behind the “bonding, bridging and linking” typology of social capital. This typology has proven to be very useful for discerning the different roles played by social capital in different contexts and individual circumstances, and for identifying very concrete implications for public policy.

For example, the idea of “linking” social capital has particular practical importance for community development policies and other anti-poverty strategies. Linking networks that reach all parts of the community across ethnic, racial and linguistic boundaries to institutions and structures of power and influence is a key challenge for policy makers. Developing and strengthening social capital by bringing in the point of view of diverse players in the community and finding mechanisms for involving them in decision-making has been found to be a key element for promoting sustainable changes in living conditions and the health of vulnerable populations (Davies 2001).

Programs can sometimes reinforce the wrong kinds of networks.

Several authors provide evidence that public programs can sometimes reinforce the “wrong kind of networks” (Field 2003). Increasing the density of homogeneous networks is not necessarily a solution that will result in productive outcomes for all populations of interest. This is often what is found, for example, in job-training programs where the unemployed meet only other unemployed persons with similar life experiences and socio-economic characteristics. On the other hand, ignoring the essential role of bonding social capital when implementing changes in communities that rely on strong social ties to maintain a high degree of cohesion, identity and support can have disastrous consequences. This has been the case in the city of Yonkers where a program aimed at relocating families in more mixed residential areas away from their high-crime, high-poverty neighborhood has not resulted in expected outcomes. Residents continued to feel a strong attachment to their old community, felt isolated and cut-off from their social supports and resources, and failed to adapt to their new living environment (Briggs 1997).

Differentiating between forms is key.

Differentiating between various forms of social capital is also key for understanding the processes of immigrant integration, ethnicity and inter-group relations. Cultural and linguistic identification offered by close ethnic ties can be crucial in the first stages of orientation and settlement in a new country. At the same time, bridging social capital appears to be attractive in allowing immigrants who are isolated in low-resource communities to connect to opportunities outside (Li, 2003). Various forms of micro-projects aimed at linking isolated and sometimes fractured ethnic communities, often living side by side, have provided members with the resources and information they needed to identify solutions to their own problems. Evaluation results from social capital-building experiences in deprived ethnic neighbourhoods in Spain, France, Belgium and the UK all point to self and collective improvements arising from an increase in self-confidence and sense of empowerment (ECOTEC 2001).

The importance of different forms varies over the life course.

Finally, interventions aimed at creating favorable conditions for the creation of beneficial social capital need to be sensitive to the fact that different forms of social capital may be important at different times within the individual's lifecourse. Researchers have suggested that bonding social capital is vital in providing physical and mental support during early childhood and old age. On the other hand, bridging social capital seems to provide useful resources in youth and adult life when negotiating the risks, changes and uncertainties associated with educational achievement, employment and career advancement (Pahl and Spencer, 1997).

Promising Venues and Potential Next Steps

These examples highlight a number of promising venues for the integration of social capital into public policy and point to a series of potential new directions for policy research in this area.

Better connections to program partners and participants is increasingly important.

Interventions to tackle poverty and social exclusion increasingly require the involvement of those who are affected as partners and participants in identifying and delivering solutions. Similarly, interventions aimed at facilitating the social, economic and political inclusion of immigrants, require a good understanding of the different social processes of identification and affiliation that interact with other integration mechanisms (i.e. human capital, language training). These are the areas where the impact of social capital might be most significant.

Better understanding makes for better policy.

More generally, the relevance of social capital for public policy rests mainly in understanding how social capital is formed, how precisely it operates in various contexts, and the role it plays in contributing to desired outcomes. Social capital research into these aspects can continue to provide the evidence that is needed to support interest in the development of social capital in relation to the achievement of various policy or program objectives.

Applying a social capital lens to existing policies and programs may be helpful.

Beyond the advancement of knowledge and development of research evidence on the various positive or negative effects of social capital, there may be more direct action that government can take. This implies, for example, the application of a social capital lens to policies and programs, to ensure that the work is conducive to productive social network formation, builds on and works with existing sources of social capital or at least does not inhibit existing networks. It can also imply the integration of more explicit social capital development indicators in the design of government programs, to ensure that these are tracked and measured once project outcomes are evaluated.

Opportunities for applying a social capital lens and for ensuring a more equal distribution of bridging and linking forms of social capital can be found in many different areas of public intervention. The following are just a few examples (more specific interventions are described in the Social Capital Workshop Report, June 2003):

• Family policies and parental interventions, including work-family balance and parental support networks
• Urban design, housing and spatial planning, including settlement patterns, mixed residential arrangements and home ownership programs
• Community participation and volunteering, including early youth experiences in associational activity and mentoring
• Community services, including access to libraries, museums, sporting grounds and other easily accessible and locally managed places for interaction
• Access to information technologies and community-based learning, including public internet access and links among schools and communities
• Social responsibility initiatives encouraging partnerships between government, the business sector and civil society

In considering the use of a social capital lens in these existing areas of policy involvement, it is essential that a focus on social capital relate to specified issues and reflect the realities of groups of citizens that we wish to influence.

Applying a social capital perspective to a specific policy problematique and program design can help emphasize the positive resources that communities already have. This is highly relevant in a context where governments are increasingly interested in developing partnerships at the local level as a basis for policy development. Much is already being done to support civic engagement through the promotion of volunteering (see the Social Capital Workshop Report, June 2003). Under a social capital perspective, however, emphasis should be placed on finding the most effective ways in which service delivery agencies and institutions interact and create linkages with these communities for developing sustainable changes in their living conditions and well-being (Field, 2003; Woolcock, 2003).

The June 2003 Social Capital Workshop Report identified a number of potential next steps regarding the development of research plans, indicators and measurement tools for supporting federal policy research on social capital. The proposed conceptual framework outlined in this paper constitutes the critical first step needed for developing a consistent approach for defining, measuring and analyzing social capital. The framework will be tested and refined through consultations and discussions with experts and policy departments. To fruitfully apply this framework, subsequent steps we should consider include:

• Developing specific research plans for the use and analysis of upcoming data from existing surveys, to address specific policy-relevant problematiques;
• Identifying and developing new measures of social capital that reflect a social-network approach, and assessing options for using these indicators in a module of questions that could be adapted in existing survey vehicles;
• Building on existing program evaluation knowledge, by developing, for example, inventories of results and lessons learned from initiatives that have incorporated measures relating to aspects of social networks as part of their evaluation framework; and
• Developing a consistent set of social capital measurement indicators for policy and program development purposes. Empirical evidence on the role of social capital should capture the dynamics of how social capital works in different circumstances. Encouraging the study of social capital in small demonstration projects could help provide a more complete appreciation of how this resource is accessed, mobilized and used in the context of local public interventions.

References

Adler, Paul S. and Seok-Woo Kwon (2002) “Social Capital: Prospects for a New Concept,” Academy of Management Review, 27(1): 17-40.

Bourdieu, Pierre (1986) “The Forms of Capital,” in Handbook of Theory and Research for the Sociology of Education, J.G. Richards, (ed.), New York: Greenwood Press, pp. 241-258.

Briggs, Xavier de Sousa (1997) “Social Capital and the Cities: Advice to Change Agents”, National Civic Review , 86(2): 111-18.

Coleman, James (1988) “Social Capital in the Creation of Human Capital,” American Journal of Sociology , 94: S95-S120.

——— (1990) Foundations of Social Theory, Cambridge, MA: Harvard University Press.

Davies, J.K. (2001) “Partnership Working in Health Promotion: the potential role of social capital in health development” in Ballock, S. and Taylor M. (eds), Partnership Working: Policy and Practice , Bristol: Policy Press, pp.181-200.

Dominguez, Sylvia and Celeste Watkins (2003) “Creating Networks for Survival and Mobility: Social Capital Among African-American and Latin-American Low-Income Mothers”, Social Problems , 59(1): 111-35.

ECOTEC Research and Consulting Ltd (2001) Foundations and Social Capital: A Final Report to the Network of European Foundations for Innovation and Cooperation (NEF). Birmingham, UK.

Field, John (2003) Social Capital . London: Routledge.

Healy, Tom, and Sylvain Côté (2001) The Well-Being of Nations: The Role of Human and Social Capital, Paris: OECD.

King, Peter and Charles Waldegrave (2003) “Social Capital, Social Netwoks, and Access to Employment: Investigating the constraints upon labour market participation experienced by single women and men presently living at or below the poverty threshold”, Presentation, Ministry of Social Development, New Zealand

Li, Peter (2003) “The Potential and Limit of Social Capital: Social Connectedness and Economic Outcomes for Immigrants and Ethnic Minorities”, Unreleased paper prepared for the November 2003 PRI-OECD Conference on the Opportunity and Challenge of Diversity: A Role for Social Capital?

Lévesque, Maurice, and Deena White (1999) “Le concept de capital social et ses usages,” Lien social et Politiques – RIAC , 41: 23-33.

——— (2001) “Captial Social, capital humain et sortie de l'aide sociale pour des prestataires de longue durée, ” Canadian Journal of Sociology , 26(2): 167-192.

Lin, Nan. (2001) Social Capital: a theory of social structure and action , Cambridge: Cambridge University Press.

Pahl, R. and Spencer, L. (1997) “The Politics of Friendship”, Renewal , 5 3-4: 100-7.

Portes, Alejandro (1998) “Social Capital: Its Origins and Applications in Modern Sociology,” Annual Review of Sociology, 24: 1-24.

Putnam, Robert (1993) Making Democracy Work: Civic Traditions in Modern Italy, Princeton NJ: Princeton University Press.

——— (1995) “Bowling Alone: America's Declining Social Capital,” Journal of Democracy, 6: 65-78.

——— (2000) Bowling Alone: The Collapse and Revival of American Community, New York: Simon and Schuster.

——— (2001) “Social Capital: Measurement and Consequences,” Isuma: Canadian Journal of Policy Research , 2(1): 41-52.

Saegert, Susan, Gary Winkel and Charles Swartz (2002) “Social Capital and Crime in New York City's Low-Income Housing” Housing Policy Debate , 13(1): 189-226.

Schultz, Theodore W. (1961) “Investment in Human Capital,” The American Economic Review , 51(1):1-17.

Stolle, Dietlind (2003) “The Sources of Social Capital,” in Marc Hooghe and Dietlind Stolle, eds. Generating Social Capital: Civil Society and Institutions in Comparative Perspective , New York: Palgrave Macmillan; pp. 19-42.

Woolcock, Michael and Deepa Narayan (2000) “Social Capital: Implications for Development Theory, Research and Policy,” The World Bank Research Observer , 15(2)(August): 225-249.