The anatomy of subjective well-being

Abstract

This paper contributes to the literature on subjective well-being (SWB) by taking into account different aspects of life, called domains, such as health, financial situation, job, leisure, housing, and environment. We postulate a two-layer model where individual total SWB depends on the different subjective domain satisfactions. A distinction is made between long-term and short-term effects. The individual domain satisfactions depend on objectively measurable variables, such as income. The model is estimated using a large German panel data set.

Introduction

The recent issue of this journal, devoted to the theme of ‘Subjective well-being and economic analysis’, may be seen as a significant step towards the lifting of the virtual ban on measuring utility that has dominated economics since Robbins (1932). To be honest, it should be noted that various prominent economists, such as Frisch (1932) and Tinbergen (1991) always refused to take such a stand. Van Praag (1968), Easterlin (2001), and Holländer (2001) a.o. make a strong case that this anathema has actually caused a stagnation in the development of economic analysis.

In the last decade, but prior to the work published in JEBO, scattered economists have started to study subjective well-being (SWB)¹ as a serious subject. See, for example, Clark and Oswald (1994), Di Tella et al. (2001), Frey and Stutzer (2000), McBride (2001), Oswald (1997), Pradhan and Ravallion (2000), and Van Praag and Frijters (1999). Earlier studies include Easterlin (1974), Van Praag (1971), and Van Praag and Kapteyn (1973).

This paper extends this line of research by making a first attempt to develop a joint model based on satisfaction with life as a whole and on domain satisfactions. Domain satisfactions relate to individual satisfaction with different domains of life, such as health, financial situation, and job. Satisfaction with life as a whole can be seen as an aggregate concept, which can be unfolded into its domain components.

Most studies in this literature have the following structure. Individuals are asked how satisfied they are with their life as a whole or with a specific domain of it. They are invited to cast their response in terms of a small number of verbal response categories, such as ‘dissatisfied’, and ‘very satisfied’. Alternatively, the categories are numbered from 0 or 1 to 5, 7 or 10, where ‘most dissatisfied’ corresponds to level 0 or 1 and ‘most satisfied’ with the highest level. The responses are explained by ordered probit or logit models, using objective variables, such as age, income, gender, and education. When two respondents give the same answer, they are assumed to enjoy similar satisfaction levels, implying that ordinal comparability is permitted. In other words, ordinal interpersonal comparability is a basic assumption in these models. Next, the effect of the explanatory variables on individual well-being can be assessed. Additionally, one can also consider the substitution ratio between explanatory variables.² This paper aims at a somewhat more sophisticated model in which we will assume that satisfaction with life is an aggregate of various domain satisfactions.

This paper is structured as follows. Section 2 presents the model and the estimation procedure. Section 3 describes briefly the data, introduces the satisfaction questions used in the empirical analysis, and highlights the main underlying assumptions. Section 4 shows and discusses the estimation results. Section 5 concludes.