This paper replicates the estimation results of three studies on the impact of the age composition of the labor force on business cycle volatility and investigates whether they signal a meaningful long-run relationship. We show that both the volatile-age labor force share variable and the business cycle volatility measure exhibit non-stationary behavior but find no robust evidence of cointegration. Hence the estimation results reported in the literature may be spurious. This conclusion is further supported by the finding that the strong relationship (i) disappears when cross-sectional dependence is accounted for using the CCEP estimator and (ii) is highly sensitive to small changes in the composition of the sample, to data revisions, and to the exact definition of the volatile-age labor share.
