Happy for How Long? How Social Capital and GDP relate to Happiness over Time

What does predict the evolution over time of subjective well-being? We answer this question correlating cross country time series of subjective well-being with the time series of social capital and/or GDP. First, we adopt a bivariate methodology similar to the one used used by Stevenson and Wolfers (2008), Sacks et al. (2010), Easterlin and Angelescu (2009), Easterlin et al. (2010). We find that in the long (at least 15 years) and medium run (6 years) social capital is a powerful predictor of the evolution of subjective well-being. In the short-term (2 years) this relationship weakens. Indeed, short run changes in social capital predict a much smaller portion of the changes in subjective well-being, compared to longer periods. GDP follows a reverse path: in the short run it is more positively correlated to the changes in well-being than in the medium-term, while in the long run the correlation vanishes. Secondly, we run trivariate regressions of time series of subjective well-being on time series of both social capital and GDP, which confirm the results from bivariate analysis.