The International RBC Model Finally Works!

We show that incorporating uninsurable countercyclical income risk into a standard international RBC model can qualitatively and quantitatively account for the quantity puzzles in open-economy macro, namely (i) the Backus-Smith puzzle, (ii) the Backus-Kehoe-Kydland puzzle, and (iii) the weak correlation between the trade balance and the exchange rate. We also show that our model can simultaneously account for the Fama puzzle and the evidence that high interest rate countries have stronger currencies—which representative-agents models that rely only on financial or demand shocks cannot jointly account for. Crucially, our model resolves all these puzzles while relying solely on productivity shocks and thus generates the observed domestic and cross-country macroeconomic comovement.