Paris School of Economics
Economists must get more in touch with our feelings…
There is a vast literature on the determinants of subjective wellbeing. International organisations and statistical offices are now collecting such survey data at scale. However, standard regression models explain surprisingly little of the variation in wellbeing, limiting our ability to predict it. In response, we here assess the potential of Machine Learning (ML) to help us better understand wellbeing. We analyse wellbeing data on over a million respondents from Germany, the UK, and the United States. In terms of predictive power, our ML approaches do perform better than traditional models. Although the size of the improvement is small in absolute terms, it turns out to be substantial when compared to that of key variables like health. We moreover find that drastically expanding the set of explanatory variables doubles the predictive power of both OLS and the ML approaches on unseen data. The variables identified as important by our ML algorithms – i.e. material conditions, health, and meaningful social relations – are similar to those that have already been identified in the literature. In that sense, our data-driven ML results validate the findings from conventional approaches.
Decisions about lockdown should consider people’s wellbeing as a guide
Every day, policy makers must decide whether a policy is desirable. They do so by examining its impact on a range of outcomes. But the problem is how to aggregate these disparate outcomes. For example, as covid-19 cases rise again, some lockdown measures are gradually being reintroduced across the UK. These policy choices will lead to outcomes that are good (such as fewer deaths from covid-19, less commuting, better air quality) and some that are bad (unemployment, income losses, loneliness, domestic abuse). How can policy makers aggregate these disparate effects in order to arrive at an overall assessment? To do so requires a “common currency” with which to measure all the effects. The currency we propose is the change in years of human wellbeing resulting from the policy.
At present, the most used currency is money. This is the method used in traditional cost-benefit analyses. Each of the various outcomes is valued by the amount of money that those affected would be willing to pay in order to produce that outcome. For many of the most important outcomes, however, including health and unemployment, it is difficult…
In choosing when to end the lockdown, policy-makers have to balance the impact of the decision upon incomes, unemployment, mental health, public confidence and many other factors, as well as (of course) upon the number of deaths from COVID-19. To facilitate the decision it is helpful to forecast each factor using a single metric. We use as our metric the number of Wellbeing-Years resulting from each date of ending the lockdown. This new metric makes it possible to compare the impact of each factor in a way that is relevant to all public policy decisions.