This paper assesses the transmission of monetary policy in a large Bayesian vector autoregression based on the approach proposed by Banbura, Giannone and Reichlin (2010). The paper analyzes the impact of monetary policy shocks in the United States and Canada not only on a range of domestic aggregates, trade flows, and exchange rates, but also foreign investment income. The analysis provides three main results. First, a surprise monetary policy action has a statistically and economically significant impact on both gross and net foreign investment income flows in both countries. Against the background of growing foreign wealth and investment income, this result provides preliminary evidence that foreign balance-sheet channels might play an increasingly important role for monetary transmission. Second, the impact of monetary policy on foreign investment income flows differs considerably across asset categories and over time, suggesting that the investment instruments and the currency denomination of a country's foreign assets and liabilities are potentially relevant for the way in which monetary policy affects the domestic economy. Finally, the results support existing evidence on the effectiveness of large vector autoregressions and the Bayesian shrinkage approach in addressing the curse of dimensionality and eliminating price and exchange rate puzzles.