Which Attitudes Will Make us Individually and Socially Happier and Healthier? A Cross-Culture and Cross-Development Analytical Model

Abstract

This paper describes a dynamic system for the interrelationships between happiness and health that considers three main attitudes to life: α, β, and γ for Aristotelian, Epicurean, and Stoic, respectively. All variables that have been shown by empirical and theoretical studies to affect individual health and happiness are included (i.e., employment, occupation, education, ethical freedom, equity in achievements). Three main approaches are considered: behavioural and statistical ex-ante, and ex-post behavioural. A model is developed to rank the three attitudes in terms of health for a given happiness level, and consequently, provide insights into which attitude should be adopted by each individual, according to their characteristics: individuals in Protestant and non-Protestant Christian societies should adopt β and γ attitudes, respectively; educated individuals should adopt a γ attitude; and poor individuals should adopt an α attitude. Based on this analysis, this paper provides insights into which attitude actually is adopted by each society by comparing predicted health and achievement levels with the observed life expectancy at birth and per capita gross domestic product levels in 107 countries, thus providing an empirical test of the analytical model. This analysis revealed a prevalence of β attitudes in Protestant Developed Countries, with larger γ shares in less income-unequal countries; a prevalence of γ attitudes in non-Protestant Christian Developed Countries, with larger β shares in more income-unequal countries; a prevalence of α attitudes in Muslim Less Developed Countries, with larger γ shares in more educated countries; and a prevalence of β attitudes in more educated atheist and Jewish countries.

Appendix

If γ = 1, under the assumptions that u = 0, fy = fh = 0, and ha* = 0 to depict the normalised situations at a country level, ∂y*/∂α = fr (em + ed−1 + 3 fr)/(α + 2 fr)², where ∂y*/∂α > 0 if fr > 0 and fr > (1−ed−em)/3 or if fr < 0 and fr < (1−ed−em)/3. These conditions are unlikely to be met, because they require the prevalence of features characterizing Protestant DCs (e.g., negative fr) combined with features characterizing LDCs (e.g., small ed and em). We can assume that y* negatively depends on α. If α = 1, under the assumptions that u = 0, fy = fh = 0, and ha* = 0 to depict the normalised situations at a country level, ∂y*/∂γ = fr [1−fr + fr (1 + em + ed + fr)]/(1 + fr + γ fr)², where ∂y*/∂γ > 0 if fr > 0 and fr > (1/2) {−ed−em + Sqrt[(em + ed)²−4]} or if fr < 0 and fr < (1/2) {−ed−em−Sqrt[(em + ed)²−4]}. These conditions are unlikely to be met, because they require the prevalence of features characterizing Protestant DCs (e.g., negative fr) combined with features characterizing LDCs (e.g., small ed and em). We can assume that y* negatively depends on γ. Note that these results are reasonable, since it is likely that, on average, y* is larger if a β attitude prevails.