Authors: Simona Baldi (University of Genoa), Luciano Canova (Eni Corporate University) and Gianluca Lentini (University of Milan)

This paper analyses whether individuals are influenced by the weather when reporting subjective well-being. To our knowledge, this is the first attempt to use meteorological variables trying to exploit temporal dimension of the day of the interviews: we basically follow a paper by Akay and Martinsson (2009) which tests the effect of the day of the week in which the interview takes place on subjective well-being. Climate constitutes by definition an exogenous variable with respect to individual answers of respondents to the question about happiness, so that coefficients on the regression may be reasonably interpreted as a causal relationship. 

First, we combine climate data provided by the Department of Physics of the University of Milan with Bank of Italy’s Survey on Household Income and Wealth (SHIW) 2006, in order to create an original dataset containing both information on climate and on subjective well-being. Climate data, at daily resolution, refer to: precipitation amount, relative humidity, sunlight (in terms of relative eliophany) and temperature. We exploit the temporal dimension (day of the interview) to try and detect the impact of the weather on the well-being of a sub-sample of respondents, contained in SHIW, who answer a question related to subjective well-being. To make it clear, we answer both questions: "What is the effect of the weather, today, on happiness, today?" and "What’s the effect of the change in weather today, with respect to previous days, on happiness, today?". By controlling a set of observed individual characteristics, we run an ordered probit using as dependent variable the answer to the question of subjective well-being and we find a positive effect of sunlight and a negative effect of increasing temperature.