A broad array of domestic institutional factors-including problems with the originate-todistribute model for mortgage loans, deteriorating lending standards, deficiencies in risk management, conflicting incentives for the GSEs, and shortcomings of supervision and regulation-were the primary sources of the U.S. housing boom and bust and the associated financial crisis. In addition, the extended rise in U.S. house prices was likely also supported by long-term interest rates (including mortgage rates) that were surprisingly low, given the level of short-term rates and other macro fundamentals-a development that Greenspan (2005) dubbed a "conundrum." The "global saving glut" (GSG) hypothesis (Bernanke, 2005 and 2007) argues that increased capital inflows to the United States from countries in which desired saving greatly exceeded desired investment-including Asian emerging markets and commodity exporterswere an important reason that U.S. longer-term interest rates during this period were lower than expected. This essay investigates further the effects of capital inflows to the United States on U.S. longer-term interest rates; however, we look beyond the overall size of the inflows emphasized by the GSG hypothesis to examine the implications for U.S. yields of the portfolio preferences of foreign creditors. We present evidence that, in the spirit of Caballero and Krishnamurthy (2009), foreign investors during this period tended to prefer U.S. assets perceived to be safe. In particular, foreign investors-especially the GSG countries-acquired a substantial share of the new issues of U.S. Treasuries, Agency debt, and Agency-sponsored mortgage-backed securities. The downward pressure on yields exerted by inflows from the GSG countries was reinforced by the portfolio preferences of other foreign investors. We focus particularly on the case of Europe: Although Europe did not run a large current account surplus as did the GSG countries, we show that it leveraged up its international balance sheet, issuing external liabilities to finance substantial purchases of apparently safe U.S. "private-label" mortgage-backed securities and other fixed-income products. The strong demand for apparently safe assets by both domestic and foreign investors not only served to reduce yields on these assets but also provided additional incentives for the U.S. financial services industry to develop structured investment products that "transformed" risky loans into highly-rated securities. Our findings do not challenge the view that domestic factors, including those listed above, were the primary sources of the housing boom and bust in the United States. However, examining how changes in the pattern of international capital flows affected yields on U.S. assets helps provide a deeper understanding of the origins and dynamics of the crisis.

No abstract

, and conference participants at the Federal Reserve, Bank of Canada, and NBER for helpful comments. James Coonan and Corinne Land provided superb research assistance. NBER working papers are circulated for discussion and comment purposes. They have not been peerreviewed or been subject to the review by the NBER Board of Directors that accompanies official NBER publications.