The Term Structure of Growth-at-Risk

Abstract

We show that the conditional distribution of forecasted GDP growth depends on financial conditions in a panel of 11 advanced economies. Financial conditions have a larger effect on the lower fifth percentile of conditional growth—which we call growth-at-risk (GaR)—than the median. In addition, the term structure of GaR reflects that when initial financial conditions are loose, downside risks are lower in the near term but increase in later quarters. This intertemporal tradeoff for loose financial conditions is amplified when credit-to-GDP growth is rapid. Using granular instrumental variables, we also provide evidence that the relationship from loose financial conditions to future downside risks is causal. Our results suggest that models of macrofinancial linkages should incorporate the endogeneity of higher-order moments to systematically account for downside risks to growth in the medium run. (JEL E23, E27, E32, E44)