Beyond revealed preferences: how gender, relative socioeconomic status, and social norms drive happiness and behavior

2017-07-21Can we see or measure preferences, beliefs, and feelings beyond how an individual selects among alternatives? In this thesis, I attempt to do so, in two ways. First, I look at the measurement and determinants of life satisfaction in two papers. Second, in a single paper, I look at one set of important relationships between behavior and psychophysiology. ❧ The first article, Reversing the Gender Gap in Happiness: Validating the Use of Life Satisfaction Self-Reports Worldwide, explores the common finding that women’s life satisfaction self-reports are significantly higher than men’s, on a global scale—despite substantial objective disadvantages. This seeming contradiction begs the question, are self-reports reliable? I address the conflict by observing that women and men use the response scale differently when responding to a question about life satisfaction. Once men’s and women’s response scales are normalized, women are less satisfied on average. Further, I show that other major determinants of life satisfaction do not show the same bias, validating past work in this field. ❧ Does money buy happiness? The next paper, Life Satisfaction Within and Across Countries: Societal Capital and Relative Income, contributes to the longstanding debate on the relationship between income and life satisfaction. We find that when looking across countries, life satisfaction is strongly correlated with GDP per capita—but it is also strongly correlated with measures of societal capital, such as health, education, and a healthy environment; excluding them results in omitted variable bias. Once we account for such variables, we find that absolute income (i.e. purchasing power) is not a significant predictor of life satisfaction, while relative income (i.e. income compared with one’s countrymen) is highly significant. ❧ The final chapter looks at a different side of the human mind, psychophysiology. This work finds connections between the actions individuals take in a laboratory experiment and their skin conductance responses (SCRs), which are indications of how emotionally aroused they are. We find that prosocial behavior increases in nonanonymous conditions, and that we can, to some extent, predict which individuals will change their behavior based on previous physiological responses