People with low incomes exhibit higher temporal discounting compared to richer
people, i.e., they are more likely to prefer smaller and sooner over larger and later
payments. Here we test whether this relationship reflects a causal effect of income on
discounting. In doing so, we distinguish whether changes in discounting reflect the
experience of a negative shock, or the resulting lower income levels. Participants in a
laboratory experiment randomly received different starting endowments, creating “rich”
and “poor” groups. All participants then performed a real effort task to earn money,
following which subgroups of participants received positive and negative income shocks.
Importantly, the magnitude of the shocks and the initial endowments were designed
such that they resulted in exactly identical income levels between those groups that did
and those that did not receive a shock. This design allows us to identify the effect of
income shocks on discounting, while exactly controlling for income levels.