French President Nicholas Sarkozy, convinced that GDP was not properly measuring French citizens’ well-being, appointed a 25-member blue-chip commission (chaired by Nobel Laureate Joseph Stiglitz, and largely composed of the world’s best economists, including 4 other Nobel Laureates) to advise him on whether GDP adequate mesured well-being and, if not, what else to measure. [The findings are supported by Gallup World poll surveys.]
The Commission has reported back, and strongly recommended, among other things, measuring social capital (social connections, political voice, etc.).
Their descriptions of: 1) what is social capital; 2) why is social capital important, especially to subjective well-being (or happiness); and 3) how to measure social capital are among the best things written on this subject.
On what is social capital, they say the following (pp. 182-183):
Like political voice and the rule of law, social connections and the attendant norms of trust and trustworthiness are important for people’s QoL. These social connections are sometimes subsumed under the heading of “social capital”. Definitions of social capital (as were other forms of “capital” at an equivalent stage in their conceptual development) have been much debated, but there is now convergence towards a “lean and mean” definition: social networks and the associated norms of reciprocity and trustworthiness. Since it is impractical to measure social networks at large geographic levels, researchers generally rely on proxies for these networks (e.g. number of close friends, political participation, membership in voluntary associations, religious involvement, doing favours, etc.). The core insight of the concept of social capital is that, like tools (physical capital) and training (human capital), social connections have value for QoL.
Social connections have value, first, to the people who are in the networks. For example, labour markets are permeated by networks, so that most people are as likely to get their jobs through whom they know as through what they know. Similarly, social connections bring benefits for health: as a risk factor for premature death, social isolation rivals smoking (Berkman and Glass, 2000). Evidence also suggests that social connections are powerful predictors of (and probably causes of) subjective well-being. Finally, the same personal activity can have different impacts on subjective well-being, depending on whether it is conducted alone or with others.
All these are “internal” effects of social networks, since they represent ways in which
social networks benefit people in those networks. Social connections, however, also have “externalities”, i.e. implications for bystanders. The literature on “social capital” has brought out clearly a number of examples of positive externalities. For example, neighbourhood networks can deter crime (Sampson 2003), and this effect also benefits residents who sit at home in front of their TV. The performance of democratic government and even the pace of economic growth may also depend on the quality of social connections within a jurisdiction. Finally, several (mainly US) studies suggest that both child welfare (infant mortality, teen pregnancy, low birth-weight babies, teen drug use, etc.) and school performance (drop-out rates, test scores) are robustly predicted by measures of community social capital. However, the “externalities” stemming from networks can also be negative. A strong sense of belonging to one group can strengthen a sense of a unique personal identity in terms of the group to which he or she belongs (Sen, 2006). This may generate fissures in national communities, and breed a climate of violence and confrontation. More generally, a longstanding tradition in economics has stressed the potential for groups to generate benefits for insiders that weigh heavily on the opportunities and QoL of outsiders. In other words, groups can foster bonds among participants but also erect walls with respect to outsiders and members of other groups. To account for these multiple effects, research distinguishes between two types of social capital, “bonding” and “bridging”, but practical implementation of this distinction in empirical research remains a challenge.
In short, a rich literature from several disciplines shows that social connections benefit people in the networks, with effects on non-participants that depend on both the nature of the group and the effects being considered. In some cases, like health, studies have demonstrated that social connections can have positive effects at both the “individual” and “aggregate” level: people with more friends live longer in part because of the biochemical effects of social isolation, and in part because public health systems are more effective in areas of higher social capital. Many of these “causal” claims are yet to be tested with an experimental or quasi-experimental design, but even on this score progress is slowly being made. A high priority for research in this field is more work on causal linkages using natural or randomized experiments.
On why social capital is important to subjective well-being, they note:
Much evidence at both the aggregate and individual level suggests that social connections are among the most robust predictors of subjective measures of life satisfaction. Social connections have a strong independent effect on subjective well-being, net of income. Moreover, the available evidence also suggests that the externalities of social capital on wellbeing are typically positive, not negative (Helliwell, 2001; Powdthavee, 2008). In other words, increasing my social capital increases both my own and my neighbours’ subjective well-being, and thus represents a coherent strategy for improving QoL for the country as a whole.
The analysis of the effects of social connections on subjective well-being is in its infancy. Much of it does not account for unmeasured individual characteristics, and most of it relies on cross-sectional data. That said, recent analyses have strengthened the case that the link between at least some forms of social connections and subjective well-being is causal. Krueger, Kahneman et al. (2008) report that, when controlling for individual fixed effects (such as personality traits), most pleasurable activities involve socializing — religious activities, eating/drinking, sports, and receiving friends. Similarly, in a recent large-scale US panel survey on religious attendance and subjective well-being, Lim and Putnam (2008) found that religious attendance at time1 (or time2) predicted subjective well-being at time2, controlling for levels of subjective well-being at time1, as well as many other covariates; the essential mechanism involved in this relation is neither theological nor psychological, but rather the strong effect of “friends at church” on well-being. Fowler and Christakis (2008) also report evidence suggesting that subjective well-being can spread in a beneficially “contagious” way from one person to another. For no other class of variables (including strictly economic variables) is the evidence for causal effects on subjective well-being probably as strong as it is for social connections.
And on measurement, the report notes:
As research on social connections is relatively new, national statistics are still rudimentary. Most researchers have relied on unofficial sources. One proxy of social connections often used is the number of associations in civil society or church to which each person belongs. However, the fragility of such measures is, by now, well recognized. A formal organization with a name and address may not correspond to any actual individual members, much less to social networks among those members. Moreover, the role of associations differs from country to country. Because of these reasons, measures of organizational density are generally not good measures of social connections, despite their frequent use for that purpose.
A related approach measures the activities assumed to be the result of social connections, such as altruistic behaviour. Thus, some research has used blood donations, membership in voluntary organizations or charitable giving as proxies. Other studies have relied on some aggregate measures of individual behaviour like voting turnout, based on the argument that, even though balloting is a private activity, participation in voting is higher in countries with a dense network of political parties or civic organizations, and that in all countries members of
these organizations have a higher probability of voting. Other studies have used proxy measures of social connections based on information on family ties, such as marriage rates, though social changes throughout the world have made these an imperfect measure of enduring interpersonal ties.
Ultimately, however, all these indicators are inadequate proxies of social connections, and reliable indicators can only be constructed through survey data. Only personal reports allow measuring the many and evolving forms of social connectedness. In recent years a number of statistical offices (in the United Kingdom, Australia, Canada, Ireland, the Netherlands, and most recently, the United States) have begun to gather and report survey-based measures of various forms of social connections. As an example of these endeavours, Appendix 2.2 presents the list of the questions included (since early 2008) in an annual Supplement to the
November US Current Population Survey, which has traditionally probed respondents about voting in national elections.56 These questions have been selected after extensive vetting by the Census Bureau and the Bureau of Labor Statistics for reliability, intelligibility, and inoffensiveness; they cover several manifestations of civic and political engagement, as well as other forms of social connections (such as number of friends, or frequency of contacts and
favours done for neighbours).
Measuring social connections, however, goes beyond measuring these particular aspects. Also needed are suitably tested questions assessing people’s experience in a range of other domains. Some of the most important domains that could be explored through dedicated surveys include the following:
• Social trust. Despite only moderate test-retest reliability, the canonical social trust question has been asked thousands of times in many countries: its behaviour is well understood, and it allows many comparisons across time and space.59 Moreover, at the aggregate level (e.g. states/nations), responses are highly stable over time (even when individual-level stability is low), suggesting that this question measures a very predictive characteristic of communities. Data on social trust are also significant determinants of subjective well-being. Compared to this canonical question, questions about “lost wallets” are potentially more reliable, since they are more specific and quasi-behavioural. Opportunities for comparison across time and space are, however, scant, and we still lack studies of its variability across time for the same person.
• Social isolation. Lack of contacts with other people in normal daily living is both a symptom and a cause of social distress, and it can lead to a downward spiral affecting morale and reducing social and economic opportunities. Social isolation can be measured through questions asking people about the frequency of their contacts with others or about how often they spend their time socializing with family members, friends and work colleagues or with other people in sports, religious and cultural associations. Social connections are also a function of living arrangements (i.e. living alone) and employment status (e.g. having a job). Research has highlighted strong associations between the degree of social isolation of each person and measures of their well-being, self-assurance, ability and power of action, and activity (Ringen, 2008)
• Informal support. Questions about the availability of social support in case of need have been used in many countries. The Gallup World Poll includes a yes/no question about friends or relatives “you can count on”; answers to this question are highly predictive of subjective well-being but have little discriminatory power (about 90% of respondents answer this question affirmatively). This suggests that alternative formulations or more nuanced responses may be needed (e.g., “if yes, how many?”, or specifying the type of help expected in various specific contingencies).
• Workplace engagement. For many people in many countries, a large fraction of all their social connections are with workmates, either in the workplace or outside of it. Helliwell and Huang (2005) found that trust in workmates is a robust predictor of life satisfaction. Questions about trust in workmates have been tested in some countries (e.g. North America).
• Religious engagement. While there is general agreement that religion is an important form of social connections, the introduction of specific survey questions (even when extensively vetted) raises issues of political sensitivity. Identifying suitable formulations of these questions is important as, in most countries, religious engagement is a robust predictor of subjective well-being and (in many countries outside Europe) it is a large fraction of all social networks.
• Bridging social capital (i.e. friendships across lines of race, religion, class, etc.) is the most important under-measured form of social connections for many outcomes. The informal advisory committee to the US Current Population Survey recommended that a suitable measure of bridging social capital could take the form of follow-on probes of the form, “Of these close friends, how many are… (White, Black, Asian, Hispanic, Catholic, Jewish, college-educated, etc.)”. This bridging question is least susceptible to response bias and political correctness.
The Economist reports on the commission in Measuring What Matters:
” Man does not live by GDP alone. A new report urges statisticians to capture what people do live by…HOW well off are Americans? Frenchmen? Indians? Ghanaians? An economist’s simplest answer is the gross domestic product, or GDP, per person of each country. To help you compare the figures, he will convert them into dollars, either at market exchange rates or (better) at purchasing-power-parity rates, which allow for the cheapness of, say, haircuts and taxi rides in poorer parts of the world.
“To be sure, this will give you a fair guide to material standards of living: the Americans and the French, on average, are much richer than Indians and Ghanaians. But you may suspect, and the economist should know, that this is not the whole truth. America’s GDP per head is higher than France’s, but the French spend less time at work, so are they really worse off? An Indian may be desperately poor and yet say he is happy; an American may be well fed yet fed up. GDP was designed to measure only the value of goods and services produced in a country, and it does not even do that precisely. How well off people feel also depends on things GDP does not capture, such as their health or whether they have a job. Environmentalists have long complained that GDP treats the despoliation of the planet as a plus (via the resulting economic output) rather than a minus (forests destroyed).”
See earlier Social Capital blog post on “Advances in Social Capital Measurement“.
See full 292 page “Report by the Commission on the Measurement of Economic Performance and Social Progress” here.
See Economist’s article, “Measuring What Matters” (9/17/09)