Here is a new and very important paper by Victor Stango and Jonathan Zinman, here are some of the main results, noting that each and every paragraph is important:
Our first finding is that biases are more rule than exception. The median consumer exhibits 10 of 17 potential biases. No one exhibits all 17, but almost everyone exhibits multiple biases; e.g., the 5th percentile is 6.
Our second finding is that cross-consumer heterogeneity in biases is substantial. The standard deviation of the number of biases exhibited is about 20% of its mean, and several results below suggest that this variance is economically meaningful and not substantially inflated by measurement error.
Our third finding is that cross-consumer heterogeneity in biases is poorly explained by even a “kitchen sink” of other consumer characteristics, including classical decision inputs, demographics, and measures of survey effort. Most strikingly, we find more bias variance within classical sub-groups widely thought to proxy for behavioral biases than across them. E.g., we find more bias variation with the highest-education group than across the highest- and lowest-education groups.
Our fourth finding is that our 17 biases are positively correlated with each other within-consumer, especially after accounting for measurement error following Gillen et al. (2019).1Across all biases, the average pairwise correlation is 0.13, and 18% have p-values < 0.001. Within six theoretically-related groups of biases (present-biased discounting, inconsistent and/or dominated choices, risk biases, overconfidence, math biases, and limited attention/memory), the average pairwise correlation is 0.25 and 29% have p < 0.001.
Our fifth finding is that there are also some important correlations between biases and classical inputs. Classical inputs and demographics may not explain much of the variance in biases (per finding #3), but some of them are correlated with biases in patterns that align with prior work. Most notably, the average pairwise correlation between cognitive skills and biases is -0.25. Cognitive skills are strongly negatively correlated with most biases, but positively correlated with loss aversion and ambiguity aversion. Other classical inputs are relatively weakly correlated with biases, except for a few expected links between patience and present bias, risk aversion and aversion to uncertainty and losses, and risk aversion and math biases that can lead to undervaluation of returns to risk-taking.
Overall not encouraging! But perhaps some of that is also what makes life more meaningful, at a high cost admittedly.