International Finance and Macroeconomics

October 9, 2009
Charles Engel and Linda Tesar, Organizers

Andrew K. Rose, UC, Berkeley and NBER; and Mark Spiegel, Federal Reserve Bank of San Francisco
Cross-Country Causes and Consequences of the 2008 Crisis: Early Warning

Rose and Spiegel model the causes of the 2008 financial crisis and its manifestations, using a Multiple Indicator Multiple Cause (MIMIC) model. Their analysis is conducted on a cross-section of 85 countries; they focus on international linkages that may have allowed the crisis to spread across countries. Their model of the cross-country incidence of the crisis combines 2008 changes in real GDP, the stock market, country credit ratings, and the exchange rate. They explore the linkages between these manifestations of the crisis and a number of its possible causes from 2006 and earlier. The causes they consider are both national (such as equity market run-ups that preceded the crisis) and, critically, international financial and real linkages between countries and the epicenter of the crisis. They consider the United States to be the most natural origin of the 2008 crisis, although they also consider six alternative sources of the crisis. A country holding American securities that deteriorate in value is exposed to an American crisis through a financial channel. Similarly, a country that exports to the United States is exposed to an American downturn through a real channel. Despite the fact that they use a wide number of possible causes in a flexible statistical framework, they are unable to find strong evidence that international linkages can be clearly associated with the incidence of the crisis. In particular, countries heavily exposed to either American assets or trade seem to behave little differently than other countries; if anything, countries seem to have benefited slightly from American exposure.

Laura Alfaro, Harvard University and NBER; and Faisal Ahmed, University of Chicago
The Price of Capital: Evidence from Trade Data

Alfaro and Ahmed use highly disaggregated data on trade in capital goods to study differences in the price of capital across countries. This strategy is motivated by the fact that most countries import the bulk of machinery equipment from a small number of industrialized countries. The researchers find the price of imported capital goods to be negatively and significantly correlated with the income of the importing country. Their results, which differ from findings using Penn World Tables data, caution against discounting a role for the higher price of capital goods in explaining the higher relative price of capital to consumption goods observed in poor countries.

Olivier Jeanne, Johns Hopkins University and NBER; Damiano Sandri, IMF Research Department; and Eduardo Borensztein
Macro-Hedging for Commodity Exporters

Borensztein and his co-authors use a dynamic optimization model to estimate the welfare gains of hedging against commodity price risk for commodity-exporting countries.They show that the introduction of hedging instruments, such as futures and options, enhances domestic welfare through two channels: by reducing export income volatility and allowing for a smoother consumption path, and by reducing the country's need to hold foreign assets as precautionary savings (or by improving the country's ability to borrow against future export income).Under plausibly calibrated parameters, the second channel may lead to much larger welfare gains, amounting to several percentage points of annual consumption.

Nicolas Coeurdacier, London Business School; Pierre-Olivier Gourinchas, UC, Berkeley and NBER
When Bonds Matter: Home Bias in Goods and Assets

Recent models of international equity portfolios exhibit two potential weaknesses. First, the structure of equilibrium equity portfolios is determined by the correlation of equity returns with real exchange rates and non-financial income; yet empirically, domestic equities don't appear to be a good hedge against either risk factor. Second, equity portfolios are highly sensitive to preference parameters. Coeurdacier and Gourinchas address both issues. They show that in more general and realistic environments, 1) the hedging of real exchange rate risks occurs through international bond holdings, since relative bond returns are strongly correlated with real exchange rate fluctuations; and 2) domestic equities can provide a good hedge against non-financial income risk, conditional on bond returns. The model delivers equilibrium portfolios that are well-behaved as a function of the underlying preference parameters. Empirically, the authors find reasonable support for the theory for G-7 countries. They are able to explain short positions in domestic currency bonds for all G-7 countries, as well as significant levels of home equity bias for the United States, Japan, and Canada.

Oscar Jorda, UC, Davis; and Alan M. Taylor, UC, Davis and NBER
The Carry Trade and Fundamentals: Nothing to Fear but FEER Itself

The carry trade is the investment strategy of going long in high-yield target currencies and short in low-yield funding currencies. Recently, this type of trade has seen very high returns for long periods, followed by large losses after large depreciations of the target currencies. Based on low Sharpe ratios and negative skew, these trades could appear unattractive, even when diversified across many currencies. But more sophisticated conditional trading strategies exhibit more favorable payoffs. Jorda and Taylor apply novel (within economics) binary-outcome classification tests to show that their directional trading forecasts are informative, and use out-of-sample loss-function analysis to examine trading performance. The critical conditioning variable, they argue, is the fundamental equilibrium exchange rate (FEER). Expected returns are lower, all else equal, when the target currency is overvalued. Like traders, researchers should incorporate this information when evaluating trading strategies. When the authors do so, some questions are resolved: negative skewness is purged, and market volatility (VIX) is uncorrelated with returns. Other puzzles remain: the more sophisticated strategy has a very high Sharpe ratio, suggesting market ineffciency.

NBER Videos

National Bureau of Economic Research, 1050 Massachusetts Ave., Cambridge, MA 02138; 617-868-3900; email:

Contact Us