MEL SILVERSTEIN AND XUAN CHEN University of Southern California KENNETH HELLER Indiana University* We propose that, although moderate amounts of intergenerational support are beneficial to older adults, excessive support received from adult children and provided to children may be harmful by virtue of eroding competence and imposing excessive demands, respectively. We tested our hypotheses using a sample of 539 older participants in the University of Southern California Longitudinal Study of Generations. Lagged regression models are estimated to predict nonlinear change in positive and negative mood over 3 years. Results reveal that among the unmarried and those with low expectations for support, a greater volume of support from children initially elevates positive mood but after the improvement reaches an asymptote, greater support begins to depress positive mood. Providing support to children reduces depression associated with being unmarried in later life but does not appear to increase distress at high levels. Key Words: aging, depression, intergenerational support, well-being. Research findings are equivocal about whether aging parents derive psychological benefits from exchanging social support with their adult children. The inconsistency of findings may be due to the assumption, implicit in the use of linear models, that more support is better. We suggest that excessive support received from family members may increase distress by inducing dependence and eroding the autonomy of the older recipient and that excessive support provided to family members may increase distress by being burdensome to the older provider. Echoing a question first raised by Wheaton (1985), we ask, Can there be too much of a good thing? In this analysis we use nonlinear models to test both the positive and negative consequences of intergenerational social support exchanges in later life. The central hypothesis guiding this study is that intergenerational social support, although beneficial for the psychological well-being of older parents at moderate levels, may be harmful at high levels. RECEIVING SOCIAL SUPPORT Research on the psychological consequences of receiving social support in later life has produced a mixed set of findings. Some studies show that social support improves the psychological wellbeing of older people or buffers the impact of stressful events (Krause, 1986; Norris & Murrell, 1984; Russell & Cutrona, 1991; Silverstein & Bengtson, 1994; Thompson & Heller, 1990), but other studies find that social support has little bearing on well-being (Dean, Kolodny, & Wood, 1990; Krause, Liang, & Keith, 1990; Lee & Ellithorpe, 1982; Lee & Shehan, 1989; Umberson, 1992). However, several studies find that support received from others increases distress among older people (Arling, 1987; Dunham, 1995; Greene & Feld, 1989; Lee, Netzer, & Coward, 1995; Penning & Strain, 1994). These latter findings are generally interpreted in terms of the loss of autonomy and control associated with relying on others for the satisfaction of basic needs. Several perspectives on social support in the aging family suggest that over involvement by relatives-especially adult children-may cause distress in older persons. Overly vigorous support from children violates a deeply seated desire for independence on the part of many older parents (Blieszner & Mancini, 1987; Cohler, 1983; Pyke & Bengtson, 1996; Townsend & Poulshock, 1986). Research shows that parents generally prefer to remain functionally autonomous for as long as possible before relying on adult children for support and generally expect less support than their children are willing to provide (Lawton, Silverstein, & Bengtson, 1994). Indeed, at high levels of intensity, intergenerational social support may cause a painful reevaluation of the relationship around the difficult issue of role reversal (see Chappell, 1991). …