Words and phrases, as they say, can do a lot of work. Sometimes, evocative terms can be useful even if they paper over imprecise concepts or obscure definitional disagreement. But vagueness often overwhelms the utility of an idea, as, for instance, with the term “systemic racism.” Imprecision can be a feature but also a bug.

“Social capital” is a similarly diffuse term. However, while it’s an extremely useful one—everyone knows what social capital is . . . sort of—its imprecision has handicapped theoretical and empirical work on the underlying concepts about which people care.[1] Perhaps most importantly, while we have no shortage of economic indicators that the federal government tracks annually (or even more frequently), there are far fewer ways to assess trends in the various components of social capital (or to consider their distribution across Americans at a point in time).

Since Robert Putnam first popularized the concept of social capital nearly three decades ago, alternate definitions have proliferated.[2] One review of 2019 research, for example, found 112 definitions of social capital in 100 randomly selected peer-reviewed articles on the topic.[3] So before we get ahead of ourselves on this blog, it’s worth being precise about what we mean—and what we think others should mean—when referring to “social capital.”

Start with “capital.” This term is well-defined. Capital refers to something that is productive for its possessor and therefore valuable. Financial capital generates wealth, as when someone owns corporate stock. Physical capital does the same: Owning a factory or a computer increases the value of the goods or services someone can produce. Human capital involves skills, habits, values, and personality traits that increase the pay someone can command on the labor market or otherwise increase a person’s ability to generate income.

The term “social” is also fairly precise: it refers to human interactions and relationships. (Allow us to sidestep any discussion of bees or other animals.)

“Social capital,” then, is about the relationships individuals possess or may access, which have value to them as they pursue the things in life that are important to them. Social capital is an individual-level attribute that is a function of the number of relationships they have and of the features of their specific relationships.

Relationships may be with people or with institutions. The term “institution” is also often poorly defined, but we can think of it as a group of people interacting in patterned ways to pursue shared ends. We might think of a school as a brick building or a neighborhood as a collection of houses aligned to a grid of streets. But rather than viewing them as physical entities, we can consider them as social institutions. We can think of them as being collective efforts to guide interaction in such a way as to transfer skills and values to children (in the case of schools) or to attach people emotionally and instrumentally to each other based on geographic proximity (in the case of neighborhoods). Having a relationship with an institution really involves direct relationships with people in those institutions serving patterned roles within them, but the institutional patterning gives these individual relationships value they would not otherwise possess.

Since individuals possess social capital, we can also think about groups of individuals (or communities) as having social capital. We can think about the kinds of relationships that constitute social capital—for instance, friendship networks, professional networks, mentors, acquaintances, and (above all) families. We can think of the kinds of value these relationships provide: information and assistance, financial support, emotional support, role modeling, self-conception, value formation, and the like.

We can also think of the various ways in which social capital may be productive, such as by affecting economic and educational outcomes, social outcomes, and physical and mental well-being.

We will often refer to “higher” or “lower” social capital, by which we implicitly mean “more valuable” or “less valuable” social capital. We do not mean to substitute our own judgment for what constitutes value; it is the value to the individuals themselves that matters. Social capital can be more valuable because of the number of relationships involved or because of the qualities of those relationships. Important qualities include attributes of the people in one’s network and attributes of the network itself.

For instance, Raj Chetty and his coauthors at Opportunity Insights consider three types of social capital.[4] Two of their types are qualities of individuals’ networks: “cross-type connectedness,” which describes the diversity of an individual’s relationships, and “network cohesiveness,” which describes the openness and porousness of an individual’s network. It makes sense to think of these as types of social capital, as they constitute aspects of relationships that produce value. At the same time, it is not necessarily the case that more cross-type connectedness or network cohesiveness should constitute more valuable social capital.

Sometimes concepts are deemed types of social capital unhelpfully. The Chetty team includes “civic engagement,” measured in terms of volunteering rates and participation in membership organizations, as a third type of social capital. But civic engagement is a behavior rather than an attribute of an individual’s networks or of the people in them.

The relationships of other concepts to social capital are even more complicated. Civic engagement results from trust, civic-mindedness, and norms of reciprocity. Do those constitute social capital? In fact, they simultaneously facilitate the building of social capital (relationships form more easily when trust is widespread), make social capital more valuable (relationships imbued with trust are more valuable than those that are not), and constitute spillover products of social capital (the more prevalent trustful relationships are, the more trust there will be generally).

We will tend to address these conceptual issues by using indicators that we assume correspond (on average) with more or less valuable social capital, such as growing up with two parents or being a member of civic groups or religious congregations. Fortunately for analytic purposes, these indicators tend to correlate with one another, so we will sometimes use indexes of social capital to compare places or groups.

The review of 2019 research already mentioned is a great place to dive deeper into the various ways other researchers and theorists have defined social capital. Since we are primarily interested in getting a handle on trends in social capital and its distribution, health, preconditions, and real-world importance—all in service of considering policies aimed at building social capital—we will not dwell further on definitional and conceptual questions in this blog but will proceed pragmatically from this grounding.