This week, Scioto analysis released a report on subjective wellbeing in Ohio. One of our key findings was that age was positively correlated with happiness. In other words, older respondents of our survey were happier than younger respondents.
This result is important in and of itself, but in the context of the broader wellbeing research it is fairly surprising. The reason is because it has generally been accepted for some time now that happiness follows a U-shaped curve over the lifespan. That is, young people are relatively happy, middle aged people are relatively not as happy, and elderly people are once again relatively happy.
This is not the first time a survey has shown that the U-shaped curve might not exist. The 2022 US Happiness Report from Gross National Happiness USA found a similar trend in their survey, and the most recent World Happiness Report said “happiness among the young (aged 15-24) has fallen sharply in North America – to a point where the young are less happy than the old.”
When I said earlier that a U-shaped happiness curve has generally been accepted, I should have clarified that I meant among economists. Researchers from other disciplines, particularly psychology, have argued against the U-shaped happiness curve.
While not all economists accept the U-shaped curve and not all psychologists dismiss it, there are differences in the way acolytes of the two disciplines approach data analysis. In particular, one interesting question is whether or not to include control variables in an analysis.
Economists include control variables because they want to understand the isolated impact that age has on happiness. In other words, if we took two people who were identical except for their age, what should we expect the difference in happiness to be between them.
This is not the same as just looking at the age of our survey respondents, because there could be some other reason that young respondents were less happy. For example, young people tend to earn less money than older people.
Psychologists have questioned the use of these control variables, because it leads to a different understanding of the data. As an example, imagine if happiness was very strongly related to income, and age only had a very small impact. If this were true, then young people might report being less happy, but if we control for income then we could still see a U-shaped relationship between age and happiness.
This might lead us to say that young people are more happy than middle aged people, they just have lower incomes. Some would argue that this misses the most important point, which is that young people reported being less happy.
These two ideas can both be true at the same time. While our survey finds that young people are less happy, it does not necessarily mean that those people will become happier as they age.
What is most interesting to me is the interaction between these two conclusions. I find our evidence very compelling, especially since it is backed up by other researchers. It is extremely concerning however if there still exists a U-shaped happiness curve with respect to age for this cohort.
That would mean that at this moment, young people are experiencing extremely low levels of happiness, and we should expect them to be even less happy as they approach their forties. We might be staring down an epidemic of low wellbeing.
As is the case with almost every policy relevant research topic, we benefit from taking an interdisciplinary approach. There is no objectively right or wrong way to interpret wellbeing data, there are just different questions to ask.
Policymakers should be concerned both with how people are reporting their wellbeing today, and how factors like age will influence future wellbeing. By looking at the complete picture, we can find solutions to improve outcomes today and in the future.
Subjective wellbeing research is an exceptionally important and woefully understudied topic, particularly at the local level. More detailed information is going to be needed to fully understand exactly how wellbeing changes with age.