There is a well-established positive correlation between life-satisfaction measures and income. This paper attempts to provide some evidence on whether this correlation re‡ects causality running from money to happiness. I use industry wage di¤erentials as instruments for income. This is based on the idea that at least part of these differentials are due to rents, and part of the pattern of industry affiliations of individuals is random. To probe the valid-
ity of these assumptions, I compare estimates for life satisfaction with those for job satisfaction, present fixed effects estimates, and present estimates for married women using their husbands’ industry as the instrument. All these specifications paint a fairly uniform picture across three different data sets. IV estimates are similar to the OLS estimates suggesting that most of the association of income and well-being seems to be causal. |
