Many intertemporal open economy macro models imply a theory of consumption smoothing channels; thus we build an empirical model to analyze the intertemporal smoothing role of saving components (fixed investments, inventories and trade balance) through the use of VAR impulse responses to different types of shocks. We find that for the OECD countries the bulk of intertemporal smoothing has been carried out domestically, via gross fixed investments and inventories, but the trade balance has also played a relevant – albeit volatile – smoothing role. We also characterize the dynamic behavior of each component: the trade balance and inventories are mostly used as short-run smoothing tools while fixed investment provides more and more smoothing over time. We can also address some empirical puzzles, such as the “excess sensitivity of investment” anomaly (Glick, R., Rogoff, K., 1995. Global versus country-specific productivity shocks and the current account. Journal of Monetary Economics, 35, 159–192) and the “saving-investment correlation puzzle” (Feldstein, M., Horioka, C., 1980. Domestic saving and international capital flows. Economic Journal, 90, 314–329).

By fully exploiting the statistical properties of panel data, this paper improves upon existing methodologies to estimate consumption smoothing at least in three respects. First, we model explicitly incomplete risksharing as well as incomplete intertemporal smoothing, and couch the two mechanisms in a unified framework. Second, we fully exploit simple panel data analysis in order to measure degrees of both risksharing and intertemporal smoothing taking place in a given set of economic regions. In particular, we are able to measure not only the smoothing of idiosyncratic shocks, but also the dependence on aggregate (non-diversifiable) shocks. Third, we distinguish neatly between the effects of temporary vs. permanent shocks. This can be done by taking advantage of the complementarity between the “within” estimator and the “between” estimator in a panel regression. We apply the above methodology to a panel of 23 OECD countries in the period from 1955 to 2005. The main finding is consistent with the puzzle of negligible international risksharing, as domestic consumption growth – once properly analyzed – does not depend on aggregate income growth. Our analysis shows that industrial countries have tended to absorb output shocks mostly through intertemporal smoothing. About 25% of all temporary shocks are smoothed this way, while a comparable fraction of permanent shocks determine consumption growth.