How risky is consumption in the long-run? Benchmark estimates from a robust estimator

The long-run standard deviation of consumption growth is a key moment in determining risk premia when agents have Epstein--Zin preferences. This paper studies a new estimator of the long-run standard deviation shown to provide a superior bias/variance trade-off and better confidence interval coverage than previous methods. In the postwar period the long-run standard deviation of consumption growth is estimated to be 2.5 percent per year with an upper bound to the 95-percent confidence interval of 4.9 percent. The analogous values in the longest available sample are 4 and 5.6 percent. These values can be taken as benchmarks for future calibrations.