Bridging Disciplinary Boundaries (January 11 - 14, 2007)

This study draws on the premises of social capital theory (Bourdie, 1985; Coleman, 1994) to explore the relationship between informal networks of social support and saving outcomes of the working poor in an IDA saving program. The question addressed in this paper is important because IDA programs started in the United States and are now being implemented in poor developing countries where informal networks are highly prized. This study may shed light on the designing of IDA programs in this context. Indeed, existing evidence suggests that, in several developing countries, informal networks of social support have been used to connect people to asset development initiatives like microfinance and micro-credit (Grameen Bank, 2002; Milgram, 2001). Moreover, the idea of asset development through IDA mechanisms is relatively new. Accordingly, this study may help identify aspects of informal networks that may be important for performance in IDAs and similar programs. The study uses data from two primary sources: the Management Information System for Individual Development Accounts (MIS IDA) (Johnson, Hinterlong, & Sherraden, 2000), and a longitudinal experimental research conducted at an IDA site in Tulsa, Oklahoma (1998—2003). Both datasets are part of the American Dream Demonstration (ADD), a national policy demonstration promoting saving and investment among low income persons in the United States (Sherraden, et al., 2000). The sample is composed of IDA participant who completed the three waves of the survey (N=412). Univariate statistics were performed to describe the sample's demographic characteristics; a paired samples t-test to assess participation in an informal network overtime, waves 1 and III; and OLS regression to understand overall impacts of informal networks on performance in an IDA program. Results of the paired sample t-tests suggest that aspects of participating in an informal network differ significantly across the two time periods. Respondents report a significant increase in community involvement from Wave I to III (mean=3.30 vs. mean=3.62). A significant change is also reported on the measure of giving help to member/s of one's community (mean=6.72 vs. mean=6.60). Regression results indicate a significant relationship between participating in an informal network and performance in an IDA program. Specifically, a respondent's relationship with her community-giving help, is related to IDA performance (b=-56.72, t=-4.35, p<0.00). Also, the presence of another adult in the household, and the presence of children under the age of 17 in the household have a significant relationship with IDA performance. Results of this study, though modest, provide a useful overview of the impact of informal networks on IDA performance. Results suggest that while such networks may influence performance in a low resource community – where the sense of community is highly prized, e.g. the context of the Grameem bank – the effect is very modest in respect to low income persons in an advanced-market economy like the United States. These findings may have important implications for policy and program design. For example, while involvement in IDA related activities, e.g., financial education, may enhance performance in an IDA program, it may constrain involvement in informal networks of social support.