Measuring inequality of subjective well-being: A Bayesian approach

This article proposes a new measure for inequality of subjective well-being and shows the empirical results using a Bayesian ordered probit model. We introduce a new concept called “regret” as a measure for inequality of subjective well-being. Regret is the probability with which a respondent who chooses an option in a multiple-choice question pertaining to subjective well-being does not choose any other option indicative of better well-being. Regret is estimated in connection with demographic factors using the Markov chain Monte Carlo (MCMC) method and data of the World Values Survey. Furthermore, the relationships between regret and GDP per capita and the changes therein are shown to investigate those between inequality of subjective well-being and economic conditions.

► We propose a new measure for subjective well-being called “regret”.
► Regret is computed by using a Bayesian ordered probit model.
► Regret enables time-series and cross-national comparisons of subjective well-being.
► Relationship between regret and per capita GDP is negative across countries.
► Inequality of subjective well-being differs from one group of countries to another.