Allan A. Schmid (2002) discussed distinguishing motives from the outputs of social capital in a recent issue of the Journal of Economic Issues. Most social capital researchers view social capital as social connectedness that, at the individual level, is embodied in interpersonal networks (Dasgupta 2002; Durlauf and Fafchamps 2004). While a popular measure of social capital--the so-called "Putnam's Instrument"--identifies social capital with participation in voluntary organizations, (1) Schmid characterized these associational activities as the outputs or consequences of social capital, not social capital itself. Schmid viewed motives as social capital. For him the crucial question was: What motivates people to be involved in community and associational activities? The suggestion was that the definition of social capital should focus on its sources rather than its consequences (Schmid 2002, 751). Viewing social capital from this perspective focuses on an individual's or group's sense of obligation or sympathy toward another individual or group. In this paper we critically examine this view of social capital.
One of the reasons social capital has generated widespread interest is that researchers have been consistently documenting that social capital has real and significant consequences in all walks of life. The basic argument is that social capital--at the individual as well as the aggregate levels--promotes cooperation, collaboration and coordination, and thereby has a variety of micro and macro level outcomes. Among macro level outcomes are political participation and good governance (Putnam 1995, 2000; DiPasquale and Glaeser 1999), increased judicial efficiency and decreased government corruption (LaPorta et al. 1997), economic performance and regional variations in growth (Knack and Keefer 1997; Serageldin and Grootaert 1999), and rise of high school in the Midwest in the early part of the century (Goldin and Katz 1999). At the micro level, it has been argued that social capital influences cooperative movements (Paldam and Svendsen 2000), income (Narayan and Pritchett 1999), children's school drop out rates (Coleman 1988), labor market outcomes (Putnam 1995; 2000), and the likelihood of positive individual outcomes, especially in the case of disadvantaged youths (Furstenberg and Hughes 1995). It is therefore imperative, that we begin the study of Schmid's view of social capital by studying its consequences.
Schmid's view of social capital is embodied in motives. Schmid classified three motives: sympathy, norm, and reward. He presented a "lost wallet" vignette experiment to study these motives. Schmid, however, did not test the hypothesis that he proposed, namely, the impact that this motive structure has on outcomes such as associational activities. Therefore, to establish this link is a natural follow-up to Schmid. In this paper, we present the results of a survey, The Georgia Social Capital Survey, that conducted the same vignette experiment used by Schmid. (2) Our survey, however, gathered additional information that allowed us to explore links between Schmid's notion of social capital (motives) and its outcomes (various associational activities).
The outcome itself, namely associational activities, is the subject of widespread interest among social capital researchers. Robert Putnam's research is well-known for linking associational activities with a variety of social attributes. In John Helliwell and Robert Putnam (2000), an index of associations, among other measures, is associated with higher growth. Edward Miguel, Paul Gertler, and David I. Levine (2001) show that industrialization is associated with the rising density of organizations. Dora Costa and Matthew Kahn (2003) show that the decline in volunteering is strongly related to higher female labor force participation. Abdul Munasib and Jeffrey Jordan (2005) found that associational activities have positive effects on the environmental awareness of farmers. Sjoerd Beugelsdijk and Ton Van Schaik (2001) found that in European regions, group participation helps explain growth. Deepa Narayan and Lant Pritchett (1999), using household data from rural Tanzania, show that indices based on memberships in groups, characteristics of the groups, and household values and attitudes, affect per capita household expenditures. Michael Carter and John Maluccio (2003) found that the number of associations in communities and the interaction of family income with community income help ameliorate effects of individual specific economic shocks. In Christiaan Grootaert (2000), an index based on the number of memberships in associations, diversity of memberships, number of meetings of associations, index of participation in decision-making, measure of cash contributions to associations, measure of time contributions to associations, and the measure of orientation toward the community are statistically significant in explaining per capita household expenditures.
Our preliminary analysis showed that there are some correlations among associational involvements and motives, particularly sympathy and norm. However, when we employed regression analysis and included additional controls to explain variations in the measures of associational involvements, we found that the motive structure matters less. We also studied the determinants of motives and concluded that they may not be adequately explained by observed variables.
Additional questions in our vignette experiment provided additional insights. We reinterpreted what Schmid conceptualized as "motivations for other" as the respondent's level of "generalized trust." Our findings showed that generalized trust has some effect on associational activities, which is consistent with an extensive body of literature that argues that generalized trust promotes cooperation (Coleman 1990; Fukuyama 1995; Paldam 2000).
The following section briefly discusses Schmid's view of motives as social capital and the vignette experiment that identifies these motives. The third section explains the Georgia Social Capital Survey, which is a unique survey that focuses on social capital at the individual level. The results and analysis are presented in the fourth section and in the fifth section; we present our analysis of Schmid's view of social capital, followed by the conclusions.
Motives as Social Capital
In Schmid (2002), social capital is "a person's or group's sympathy toward another person that may produce a potential benefit, advantage, and preferential treatment for another person or group beyond that expected in an exchange" (750). Schmid used the so-called "lost wallet" vignette experiment where the respondent was asked whether she would return a lost wallet. If she answered in the affirmative, in the follow-up question she was asked to allocate 100 points among three reasons for returning the wallet: first, sympathy for the person who lost the wallet (sympathy); second, returning is the ethical action (norm); and third, self-serving reward or praise (reward). The way the respondent allocated points--to sympathy, norm, and reward--revealed her motive structure, thereby measuring her level of social capital.
If a person would return the wallet due to a feeling of ethical obligation--that it is the right thing to do--they are consistent with a code or societal norm. There is no requirement to care for the other person; the action is a result of norms of learned behavior. Of course, such norms or rules can change, altering the result. Returning the wallet out of a desire for reward or praise also requires no caring.
On the other hand, returning the wallet out of a sense of sympathy or obligation for another person does indeed require affinity or caring. Defining social capital as trust, reciprocity, and mutual obligation means those who are motivated by affinity or caring may exhibit higher levels of social capital--or higher levels of the manifestation of social capital--in terms of community behavior.
Georgia Social Capital Survey
The analysis in this study was based on a telephone survey of Georgia residents using a random dial approach. Random digit dialing (RDD) probability sampling was used to ensure all residents of Georgia a near-equal probability of selection. The survey was conducted by the University of Georgia Survey Research Center between June 13 and July 1, 2003. The study design called for conducting a total of 500 telephone interviews; to achieve the 500 completed interviews, 1,238 phone contacts were made, representing a 40.4 percent response rate.
The survey covered only residential households. Non-response numbers included businesses, unavailable respondents, non-working numbers, answering machines, no answer/busy, or strange noises. The 500 responses represent a statistically valid sample of the population of Georgia at the 95 percent confidence interval (with a sampling error of +/-4.3 percent). The survey was pretested by administering the instrument to 60 people outside the Athens, Georgia, local area. Additional pretesting was conducted statewide following revisions. The pretesting resulted in 61 survey questions, including demographic information.
The first survey question asked if respondents would return a lost wallet with $1,000 to an owner who was unknown to them. In the follow-up, respondents were asked to allocate 100 points among three reasons for returning the wallet: sympathy, norm, and reward. Since mixed motives exist, the respondents were not asked to choose one motive but to allocate importance using 100 points.
Additionally, the respondents were asked whether they believed others would return the wallet if the respondent had lost it. Again, as a follow-up question, they were asked to allocate 100 points for the motivation of others in returning the wallet.
After asking the lost wallet question the survey asked a number of questions about associational activities. The questions were selected from...