Abstract

Easterlin Paradox states that there is a linear-logarithmic relationship between income and life satisfaction. Accordingly, the increase in income level increases the life satisfaction level of individuals at the beginning, but after a point this effect becomes stable. In this study, the Easterlin Paradox is discussed at the macro level. For this purpose, a heterogeneous data set consisting of a large group of countries was used. In order to determine the most appropriate functional form for the relationship in the paradox, a unique method was used, unlike the studies in the literature. Expressed as fractional polynomial models, this method uses non-integer powers of variables to determine the structure of functional relationships. This method has the advantage of determining the best fit from a large set of models. In the study, fractional polynomial models were used by integrating panel data models. According to the results, there is a sigmoid, cubic-logarithmic relationship between life satisfaction and economic growth. Accordingly, in countries with low economic growth, while the effect of income on life satisfaction increases exponentially, in countries with high growth, this effect first increases similar to the logarithmic structure, then this increase loses its effect but does not become stable. So although the Easterlin Paradox is more likely to prevail in high-income countries, it cannot be said to be exactly valid. In contrast to the more general findings obtained by classical methods, the use of fractional polynomial models enabled more detailed results regarding the validity of the paradox.