Abstract
Understanding the significance of social preferences, especially other-regarding preferences, is essential for various economic and social aspects such as market performance, collective actions, public goods provision, and pro-environmental behaviors (Fehr & Fischbacher, 2002; Charness & Rabin, 2002; Leibbrandt, 2012; Bauer et al., 2016; Iwasaki, 2023). The nature of social preferences and their measurement have a long history in both psychology and economics (De Cremer & Van Lange, 2001; Fehr & Fischbacher, 2002; Murphy & Ackermann, 2014). In psychology, social preferences refer to the personality traits and behavior to consider the welfare of others, such that individuals with prosocial preferences tend to behave more prosocially than individuals with pro-self preferences (De Cremer & Van Lange, 2001; McClintock & Allison, 1989; Murphy et al., 2011; Van Lange et al., 2011). Economics literature defines a social preference as an individual’s preference regarding the payoffs of others, such as how the individual ranks possible combinations of personal payoffs and the payoffs of others (Charness and Rabin, 2002; Levitt & List, 2007). These preferences manifest in behaviors like altruism, fairness, and cooperation. The primary traditions for measuring social preferences in economics are revealed and stated preferences. Revealed preferences are inferred from actual behavior in specific tasks, such as dictator games or public goods games, whereas stated preferences are derived from self-reported data on hypothetical or real-world behaviors (Beshears et al., 2008). Behavioral economics measures social preferences via experimental tasks where individuals make decisions that affect their own payoffs and those of others. Common examples include the ultimatum and trust games, which have been widely used to study fairness and reciprocity (Kerschbamer et al., 2017). These measures help uncover the structural properties of tasks and environments that influence social behaviors. On the other hand, psychologists rely on self-report measures that involve asking individuals to rate their willingness to engage in prosocial behaviors or their attitudes toward fairness and cooperation (Van Lange et al., 2011). These measures are used to study stable individual characteristics and are included in various large-scale surveys and panel studies (Chuang & Schechter, 2015; Kuroishi & Sawada, 2024). Self-reports offer insight into enduring traits but may lack the precision of behavioral tasks in predicting real-life actions. Additionally, psychologists utilize social value orientation (SVO) measures to gauge social preferences without incentivizing it (De Cremer & Van Lange, 2001; Murphy et al., 2011; Van Lange et al., 2011; Murphy & Ackermann, 2014). In psychology, human traits are evaluated considering the temporal stability and predictive validity of social preference measures, which are crucial factors. Self-report assessments of social preferences exhibit more excellent temporal stability than behavioral measures, suggesting they more effectively capture consistent individual traits (Bohm et al., 2021; Froehlich et al., 2021). Nonetheless, the disparity between self-reported (non-incentivization) and actual decisions (incentivization) is an ongoing debate for accurately assessing social preferences.
Although incentivized experiments have a good rationale, recent evidence suggests that incentivization may not always matter, it might have a crowding out effect (Forsythe et al., 1994; Grech et al., 2022). Also, Matousek et al., (2022), conduct meta-analysis and suggest that it does not matter systematically for the reported discount rates whether experiments use real or hypothetical rewards. Thus, several open questions exist, such as why incentivization may not always matter in experimental economics. What payment conditions can be used in laboratory experiments, and how do they affect the outcome? What are the temporal stability, convergent validity, and predictive validity in experiments with or without incentives (Svorenčík & Maas, 2016; Mata et al., 2018)?
Therefore, the proposed study aims to examine the economics of incentivization and social preferences (SVO) through classroom experiments. It is crucial to understand whether incentivizing participants affects their other-regarding preferences (SVO) authenticity. Previous research suggests that incentivization may misalign motivations and lead to biased outcomes in understanding cooperative behaviors (Bowles & Hwang, 2008; Matousek et al., 2022). Therefore, this study will explore the necessity and impact of incentivization on social preferences, contributing to the broader discourse on temporal stability, convergent validity, and predictive validity in effective experimental methodologies and the evolution of economic theories (Camerer et al., 2000).
Three aspects of the study:
Temporal Stability:
Convergent Validity:
Predictive Validity:
Experimental procedure: (Blanco et al., 2011; Bowles & Hwang, 2008; Bowles & Polania-Reyes, 2012).
1. The experiments will be conducted in a classroom setting using randomization and using a within-subject design. We begin by collecting baseline measurements of the dependent variable(s) to establish a starting point for each participant. This baseline helps us assess changes caused by the intervention, i.e., incentivization. The participants first participate in a different set of economic games without incentivization.
2. Participants play a dual role in this within-subject Design, serving as both the treatment and control groups. This unique set-up significantly enhances the statistical power by minimizing error variance associated with individual differences. The Design’s key advantage lies in its ability to compare various conditions or treatments within the same group of participants, thereby facilitating a more direct comparison of the effects of incentivization on the dependent variable(s).
3. The participants will be a diverse group comprising undergraduate and graduate students from various disciplines such as business, economics, and engineering; this set-up plays a pivotal role in our research. Their enrollment in a wide range of courses ensures a diversity in sample, with individuals exhibiting a spectrum of knowledge and decision-making skills. The experiments are designed to elicit economic decision-making behavior, with participants being assigned to first conditions as non-incentivized and later playing the same incentivized game.
Expected Outcomes:
• Contribution to the existing knowledge on incentivization in economics and social preferences.
Ex-Ante Hypothesis 1 (H₁):
Social value orientations (e.g., altruism, fairness) exhibit temporal stability over short periods (e.g., weeks/months), even when incentivization is introduced.