Multilevel analysis of income, income inequalities and health in Spain

The aim of this study is to test the influence of personal income (absolute income hypothesis), income inequalities and welfare (relative income hypothesis) on health. A multilevel cross-sectional logit model is used with two alternative specifications of the dependent variable: self-perceived health and chronic illnesses, and six specifications of the income inequality: three positive and three normative. This study incorporates lagged values of the regional variables and interactions between the individual and the regional variables. The data is drawn from the Spanish Life Conditions Survey for 2007 and consists of 28,023 individuals over 16 from 17 Autonomous Communities.The results support the absolute income hypothesis that a higher level of personal income is correlated with a lower probability of negative health outcomes. The relative income hypothesis results are mixed with only some indicators showing a significant relationship with health. The results also indicate that being a man, being married, working and having a high level of education are related to improved health.This study emphasizes the importance both of the health variable and of the specification of income inequality, and contributes to augmenting the limited empirical evidence available in Spain on the influence of income and income inequalities on the health of the population.

► This study increases the limited evidence about the absolute and relative income hypothesis at a regional level in Spain.
► It is the first to test the relative income hypothesis with two social welfare indices.
► It is the first in Spain to incorporate the lagged values of the ecological variables.
► Two different health variables are considered to examine the sensitivity of the results to changes.