This paper empirically investigates the relationship between long-run economic growth and output volatility for the time series experience of 25 OECD countries between the years 1960 and 2013. Given the low number of observations, we reject, based on Monte Carlo simulations, the obvious choice of Garch estimation, and instead propose a pooled OLS estimator between a filtered GDP series that eliminates the cyclicality and the fluctuations around this trend. We find strong empirical evidence for a positive relationship between output variability and economic growth. This relationship seems to confirm theoretical literature which proposes such a positive relation.