The goal of this research project was to answer the question: does poverty have particular psychological, economic-choice, and neurobiological consequences? For instance, in the psychological domain, does poverty lead to lower optimism, higher stress, or lower self-esteem? In economic choice, does poverty lead to more impatience, higher risk aversion, or less altruism? And neurobiologically, does poverty lead to higher levels of stress markers such as cortisol and pro-inflammatory cytokines?
To answer these questions both laboratory experiments as well as field studies have been combined. The studies take place both in developed countries (Switzerland) as well as developing countries (Kenya), with respondents who have experienced extreme poverty. State-of-the-art methods from several fields – behavioral economics, psychology, development economics, and neuroscience are used. Additionally, an outstanding group of collaborators provide expertise from immunology, anthropology, and psychiatry.
Study 1: "Poverty" in the lab: impact on psychological welfare and economic choice
In this experiment, we ask whether poverty has psychological consequences in a laboratory setting. We develop a laboratory paradigm for "poverty" and test to what extent this experimental manipulation affects psychological outcome variables, neurobiological markers of stress, and economic choice. Subjects perform an effort task, from which they can earn income. To mimic two important aspects of poverty, different groups of subjects start the experiment with different initial endowments; in addition, after a certain number of periods, subsets of participants receive exogenous positive or negative income shocks. We then ask whether these manipulations – i.e., having less money than others, and/or having less money than previously – affects stress, temporal discounting, and a number of other psychological variables. We hypothesize that being "poor" relative to others or relative to one's own income history will be associated with higher stress and more short-sighted time preferences. A total of 148 subjects have participated in this experiment, and our results show that subjects who receive the negative income shock display increased levels of present bias, suggesting that one's own income level has adverse effects on subjective wellbeing and decision-making, and that therefore poverty may causally lead to stress and short-sighted choices.
Study 2: Increasing poverty in the field: causal impact of a natural experiment on psychological welfare and economic choice
This study investigates the psychological, neurobiological and economic choice consequences of an exogenous increase in poverty. To this end, we take advantage of two unique natural experiments in Kenya. We hypothesize that larger negative income shocks lead to higher levels of stress and more short-sighted decision-making. If we obtain this finding, it would confirm the results and generalize them a) from the lab to the field, b) from a developed to a developing country.
In the first study, we use data from Elangata Wuas, a Massai region in Southern Kenya, where a severe draught in 2008/2009 led to a large negative income shock in the form of death of livestock – the main income source. The proportion of livestock lost by each household is not predicted by any observable variables – in particular, rich households could not protect themselves from the loss of livestock. This makes the loss of livestock a random negative income shock to households, and thus it can be used to identify a causal effect of an increase in poverty on cortisol levels. We studied the causal effect of shocks on levels of the stress hormone cortisol a year later, and found that baseline cortisol is significantly higher in families who lost more livestock than others. In the second study, we use rainfall variation as an exogenous source of income variation and focused on Kianyaga, a farming district populated by the Kikuyu where agriculture is the main source of income, and periods of no rain pose significant economic challenges for households. To ask if the absence of rain raises levels of stress hormones, we combine high resolution infrared satellite imagery measuring rainfall with GPS location data for each household, and salivary samples to assay cortisol. Our findings show that periods of no rain lead to significant increases in salivary cortisol, with a lag of 10 days. Thus, exogenous increases in poverty lead to increases in salivary cortisol, establishing a causal link between poverty and stress hormones.
Study 3: Reducing poverty in the field: causal impact of unconditional cash transfers on psychological welfare and economic choice
Study 4: Reducing poverty in the field: causal impact of insurance provision on psychological welfare and economic choice
These studies address the converse question: what are the psychological, neuroscientific and economic choice consequences of an exogenous decrease in poverty? We study this question using two randomized controlled trials in Kenya: we implement an exogenous poverty decrease through either randomly assigned Unconditional Cash Transfers (UCT), or randomly assigned provision of health insurance, to poor families in Kenya. We are interested in the welfare impact of these rather novel development measures, both in terms of classical socio-economic outcome variables standardly used in RCTs, as well as the novel psychological, neurobiological and economic-choice variables. In particular, we hypothesize that receiving cash transfer or insurance will lead to decreased levels of stress and more farsighted decision-making. Both the UCT and insurance studies are now ongoing in Kenya.
This project was financed for 3 consecutive years after annual reports showed that the project is progressing successfully.