Subjective Well-Being and Heterogeneous Contexts: A Cross-National Study Using Semi-Nonparametric Frontier Methods

This paper aims to contribute to the literature on subjective well-being by exploring the extent to which certain economic, social and institutional variables may affect the levels of well-being declared by individuals from different countries. To do this, we adopt a novel methodological approach based on frontier techniques in order to identify whether the maximum possible levels of well-being are achieved given the available resources. Specifically, the technique used to conduct the analysis is the stochastic nonparametric envelopment of data model, which combines the advantages of parametric methods and the flexibility of the nonparametric approach. This methodology has been adapted to deal with contextual variables by reformulating the original mathematical syntax of convex nonparametric least squares. Our empirical analysis is based on longitudinal information gathered from the World Values Survey for a set of 82 countries. Our results suggest that the most efficient countries in terms of well-being include mainly developing Latin American nations together with some European countries. Moreover, we find that several social indicators, such as the quality of government, the unemployment rate or different inequality indices, have a significant effect on the estimated efficiency measures.