Working Paper: NBER ID: w26888
Authors: David G. Blanchflower; Carol L. Graham
Abstract: A number of studies – including our own – find a mid-life dip in well-being. We review a psychology literature that claims that the evidence of a U-shape is "overblown" and if there is such a decline it is "trivial". We find remarkably strong and consistent evidence across countries and US states that statistically significant U-shapes exist with and without socio-economic controls. The US is somewhat of an outlier with evidence of an early uptick in the raw data with some variables – but not in others – that disappears when controls are included. We show that two of the studies cited by psychologists suggesting there are no U-shapes are in error; we use their data and find the opposite. The effects of the mid-life dip are comparable to major life events like losing a spouse, losing a job or getting cancer. They are clearly not inconsequential.
Keywords: wellbeing; midlife dip; happiness; U-shape; psychology; economics
JEL Codes: I31
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| age (J14) | wellbeing (I31) |
| midlife dip (J26) | wellbeing (I31) |
| age (J14) | U-shaped curve in wellbeing (I31) |
| socioeconomic factors (P23) | wellbeing (I31) |
| U-shaped curve in wellbeing (I31) | life satisfaction (I31) |
| midlife dip (J26) | comparable to major life events (J17) |