I study the implications of international risk sharing for emerging economies. On the empirical side, I document two new facts. First, the Backus-Smith puzzle is more severe in emerging than in industrial economies. Second, international investment correlations in emerging economies are lower than those in industrial economies. I also confirm two previously documented findings: 1) international consumption correlations in emerging economies are low and often negative, whereas they are positive in industrial economies, and 2) there is excess volatility of consumption relative to income in emerging but not in industrial economies. Altogether, these facts imply lower levels of risk sharing in emerging vis-à-vis industrial economies. On the theoretical side, I develop a multi-country general equilibrium model with asymmetric trade that can explain these facts using two key transmission mechanisms: 1) inelastic trade and 2) stochastic shocks to trend growth. On the one hand, due to inelastic trade, a negative productivity shock at home can generate a depreciation of the terms of trade that further reduces domestic wealth. A strong enough negative wealth effect results in a drop of total demand of the domestic good; thus explaining the perceived lack of international risk sharing. On the other hand, due to stochastic shocks to the rate of productivity growth and a low share of world trade, emerging economies have a lower scope for risk sharing which further decouples emerging economies' international comovements of prices and quantities. I calibrate the model to match within-country and across-countries' business cycle statistics for Mexico vis-à-vis US.