Output Growth and Its Volatility: The Gold Standard through the Great Moderation

This study examines the relationship between U.S. output growth and its volatility over the period 1875:Q1 to 2008:Q2. We examine the data for outliers and apply corrections when found. Next, we search for possible effects of structural breaks in the growth rate and its volatility. In so doing, we employ autoregressive generalized conditional heteroskedasticity and autoregressive exponential general conditional heteroskedasticity specifications of the process describing output growth rate and its volatility with and without structural breaks in the mean and volatility processes. We discover one break in the mean process – 1936:Q2 – and three breaks in the volatility process – 1916:Q4, 1950:Q3, and 1983:Q4 (or 1984:Q3). After accommodating the breaks in the mean and volatility processes, the integrated generalized autoregressive conditional heteroskedasticity effect proves spurious. Finally, our data analyses and empirical results suggest that higher output-growth volatility stimulates output growth and that higher output growth reduces its volatility. Moreover, the evidence shows that the time-varying variance falls sharply once we incorporate the three structural breaks in the unconditional variance of output.