This was the first official presentation of this project. It is hard to think of a better audience than the one at Midwest Macro - always full of both young and accomplished macroeconomists, eager to point out any flaws, and offer suggestions for improvements. Following a two-year hiatus the meeting took place in the scenic foothills of the Rocky Mountains in Logan, UT. The presentation itself was met with a very positive reception and a few ideas for fine-tuning.
 

In the paper, we first show that the inflows and outflows of Foreign Direct Investment are positively correlated - when German companies build new factories in France, French companies also tend to build new factories in Germany. In the standard international business cycle model only net flows matter, and inflows and outflows are negatively correlated. We extend the standard model to allow for domestic and foreign ownership of physical capital to be less than perfect substitutes. The counterfactual, negative correlation of inflows and outflows is reversed when the elasticity of substitution between domestic and foreign ownership of capital is sufficiently low. We then use our model to re-evaluate the welfare gains from financial integration. Foreign ownership of capital has a positive impact on measured total factor productivity, because the effective capital stock is larger than the sum of its domestic and foreign components, leading to larger output and consumption. We find that these welfare gains can be very large - many orders of magnitude larger than the gains stemming from insurance against country-specific risk alone.