Causality in the link between income and satisfaction: IV estimation with internal instruments

Usually, it is expected that income increases life satisfaction. In recent years tough, research emerged that shows how subjective well-being, including satisfaction, influences objective measures, as for example income. This would then require explicit identification strategies for estimating effects of income on life satisfaction. I address this issue using German SOEP data and Lewbel's (2012) method, which generates instruments from heteroscedasticity. This allows identification of two separate causal effects in the link between income and life satisfaction: (1) income affecting satisfaction and (2) satisfaction affecting income. This analysis focuses on life satisfaction and equivalized income, because this is the income measure most welfare analyses use to assess utility of income. Results show no significant effects of income on life satisfaction, but effects of satisfaction on income. This suggest that the effect of income on life satisfaction may be overstated in standard approaches that do not account for this reverse causality - possibly due to reverse causality, which is likely rooted in response behavior, rather than income generation.