HCEO WORKING PAPER SERIES Working Paper The University of Chicago 1126 E. 59th Street Box 107 Chicago IL 60637 www.hceconomics.org The Formation of Prosociality: Causal Evidence on the Role of Social Environment Fabian Kosse1,2, Thomas Deckers1, Pia Pinger1,2, Hannah Schildberg-H¨orisch3 and Armin Falk*,1,2 ∗ Corresponding author, [email protected] 1 University of Bonn, Institute for Applied Microeconomics 2 Institute on Behavior and Inequality (briq) 3 University of D¨usseldorf,D¨usseldorfInstitute for Competition Economics (DICE) May 2018 Abstract This study presents descriptive and causal evidence on the role of social envi- ronment for the formation of prosociality. In a first step, we show that socio- economic status (SES) as well as the intensity of mother-child interaction and mothers' prosocial attitudes are systematically related to elementary school children's prosociality. In a second step, we present evidence on a randomly- assigned variation of the social environment, providing children with a mentor for the duration of one year. Our data include a two-year follow-up and reveal a significant and persistent increase in prosociality in the treatment relative to the control group. Moreover, enriching the social environment bears the potential to close the observed gap in prosociality between low and high SES children. A mediation analysis of the observed treatment effect suggests that prosociality develops in response to stimuli in the form of prosocial role models and intense social interactions. Keywords: Formation of preferences, prosociality, social preferences, trust, social inequality. JEL-Codes: D64, C90 ∗Financial support through the Leibniz Programme and the CRC TR 224 of the German Re- search Foundation (DFG), the European Research Council (ERC Advanced Grant 340950), and the Benckiser Stiftung Zukunft is gratefully acknowledged. 1 Introduction Prosociality is a particularly important aspect of human personality and it affects a wide range of economic decisions and outcomes, e.g., the provision of public goods, contract enforcement, management of commons, governmental and judicial effi- ciency or economic growth (Knack and Keefer, 1997; La Porta et al., 1997; Fehr and G¨achter, 2002; Henrich et al., 2004; Ostrom et al., 2002; Guiso, Sapienza, and Zingales, 2009; Cooper and Kagel, 2009; Burks et al., 2016). Prosocial behavior is not only a crucial factor for the functioning of societies but also an important skill that affects health, well-being, and labor market success (Dohmen et al., 2009; Carpenter and Seki, 2011; Becker et al., 2012; Algan et al., 2016; Deming, 2017).1 The importance of prosociality in economic contexts is thus in line with findings from a vast body of empirical literature on the returns to non-cognitive skills (see, e.g., Heckman, Stixrud, and Urzua, 2006). It is also reflected in economic theory, which, pioneered by Becker's work on altruism (Becker, 1976), incorporates other- regarding preferences and beliefs as key components (e.g., Rabin, 1993; Fehr and Schmidt, 1999; Charness and Rabin, 2002; B´enabou and Tirole, 2006; Falk and Fischbacher, 2006; Fehr and Schmidt, 2006). Despite its fundamental importance, and significant advances in understanding the consequences of prosociality, little is known about how prosociality forms. This paper therefore provides evidence on the formation of prosociality in children with a particular focus on the role of the social environment. It documents that prosociality in children is malleable and how it can be enhanced. We identify prosocial attachment figures and intense social interactions as two main drivers of child prosociality.2 Prosocial attachment figures are highly prosocial individuals with a close social connection to the child. They enable children to observe and imitate a prosocial role model and thereby internalize prosocial beliefs, preferences and behaviors (Skinner, 1953; Williamson, Donohue, and Tully, 2013). Intense social interactions are joint activities that foster social reinforcement and feedback (Eisenberg, Fabes, and Spinrad, 2006; Eisenberg, Spinrad, and Knafo- Noam, 2015). In order to obtain exogenous variation in these stimuli, we randomly assigned a sample of elementary school children to an enriched social environment in form of a mentoring program. This mentoring program is a well-established non- 1Using data from the German Socio-Economic Panel (SOEP), we find for example that a one standard deviation increase in prosociality translates into 0.45 standard deviations improved sub- jective well-being, a two percentage point reduction in the probability of being unemployed, as well as higher wages, and better health (see Becker et al. (2012) and table B1). 2For biological and genetic determinants of prosociality, see e.g. Fehr and Fischbacher (2003), Kosfeld et al. (2005) and Cesarini et al. (2008, 2009). 1 profit program called \Balu und Du" (German for \Baloo and You"). It provides children with a volunteer mentor for the duration of one year. Conceptually, the idea of the program is to extend a child's horizon and foster the acquisition of new skills and experiences through intense social interactions between mentor and child. During the intervention, the child experiences an unrelated and highly prosocial attachment figure taking responsibility and devoting effort and time with him/her. Before and after the intervention, children and their mothers were interviewed by trained interviewers. Children participated in incentivized choice experiments and answered a short questionnaire. Mothers completed an extensive questionnaire covering socio-economic background information, interaction patterns and assess- ments of personality regarding their children and themselves. In order to yield a comprehensive measure of a child's prosociality, we collected data on three main facets: altruism, trust and other-regarding behavior in everyday life. Altruism reflects prosocial motivation, trust indicates prosocial beliefs, and other-regarding behavior in everyday life appraises prosocial behaviors. The corresponding facets have been shown to predict real-life societal and individual success (Becker et al., 2012; Falk et al., 2018). The respective measures combine information from in- centivized experiments and statements from questionnaires. Altruism summarizes children's choices in three incentivized dictator games. Trust is measured using a well-established, age-adapted and experimentally-validated three-item trust ques- tionnaire administered to the child. Moreover, we asked mothers to assess their child's other-regarding behavior in everyday life using the prosociality subscale of the widely used Strength and Difficulties Questionnaire (SDQ) (Goodman, 1997). Our measures thus provide a broad-ranging characterization of prosocial disposition using different data collection methods and statements from children and mothers. This approach reduces measurement error and potential demand effects (Hertwig and Ortmann, 2001). For the main analysis, we collapse all three facets into one joint measure of children's prosociality, but we also provide results for each facet separately. In addition, to understand the respective role of attachment figures, we elicit prosociality for mothers and mentors. As for children, this measure consists of the three facets altruism, trust and other-regarding behavior. Our sample was recruited using official registry data. It comprises families inter- ested in participating in the mentoring intervention and the interviews. We study three distinct groups that differ in terms of socio-economic and treatment status. Based on socio-demographic background information (household income, parental education and single-parent status), we classify families as either low or high SES. Among low SES families, we randomly assigned a subset of families to participate in 2 the intervention, reflecting the treatment group (Treatment Low SES). The remain- ing families with a low SES background form our intervention control group (Control Low SES). The third group comprises families with high SES background (Control High SES). Differences in social environment and child prosociality among Control Low SES and Control High SES families allow us to derive and test hypotheses regarding the two potential drivers of prosociality formation, prosocial attachment figures and intense social interaction, respectively. Comparing Control Low SES and Treatment Low SES provides causal evidence on how providing children with an enriched social environment can promote their prosociality. Overall, we collected three waves of data. After a baseline wave of interviews (wave 1), the treatment was randomly assigned and implemented. The one-year treatment period was closely followed by a post-treatment wave of interviews (wave 2) and a two-year follow-up wave (wave 3). In the main analysis, we focus on post-treatment wave 2 data, but also use wave 1 to study baseline balance and systematic attrition, as well as wave 3 data to investigate whether the observed effects are enduring. Our main findings can be summarized as follows. Using the Control High SES and Control Low SES groups, we first document a pronounced SES gap in prosocial- ity among elementary school children, amounting to 22.6% of a standard deviation. Moreover, we show that low and high SES families differ in their social environment and provide suggestive evidence that alterable stimuli in form of prosocial attach- ment figures and intense social interaction explain a substantial part of the observed SES