Rare economic disasters, such as the Great Depression and World War II, play a central role in assessing the uncertain macroeconomic future and in understanding the pricing of assets. The PI's previous cross-country research back to 1870 used existing and new macroeconomic data to quantify the role of rare disasters in explaining major asset-pricing puzzles, including the high equity premium and low risk-free rate of return. The current project plans a substantial expansion of the long-term national-accounts data for up to 40 countries. These data are integral to the current project but will also be of great value for studies carried out by many economists. The current application will use the new data to study further the interplay between rare macroeconomic disasters and asset markets.

One part of the research considers asset pricing at the super-national level, such as the OECD or various regional aggregates. This application requires the construction of measures of per capita consumer expenditure and GDP at super-national levels. Another part fits the distribution of disaster sizes to a power-law density function, thereby allowing for potential disasters that exceed any observed within the large cross-country sample back to 1870. This extension makes it possible to explain the observed equity premium with a lower degree of risk aversion. Further work will use the full cross-country time series of per capita consumer spending and GDP to estimate the probability of disaster, the evolution of macroeconomic aggregates during disasters, the probability of returning from disaster to normalcy, and the extent to which disasters have only temporary effects on GDP and consumption (because of strong post-disaster recoveries). This analysis will show how asset pricing depends on the likely duration of disasters and on the partly temporary nature of disaster effects. Further work will use stock-index options prices and quality spreads in interest rates to derive measures of time-varying disaster probabilities. This analysis can explain the volatility of asset prices and also allow for time-varying equity premia. A final extension will study in detail the Great Influenza Epidemic of 1917-20 as a shock that led to a widespread macroeconomic contraction with troughs typically in 1920-21.

Broader Impacts. The recent international turmoil in housing, financial, and commodity markets and the strong responses of the Federal Reserve and other policymakers reflect fears of a new Great Depression. Further study using long-term, cross-country national-accounts data is needed to understand the frequencies, sizes, and macroeconomic role of these kinds of rare economic disasters. This study seeks to clarify the significance of these events for asset markets, economic fluctuations, welfare, and optimal government policies. The first research need, critical for understanding rare events, is the construction of more long-term data. This need has been addressed in research to date and in planned future efforts for several countries, including the United States. These data are central for the present study but will also have numerous uses for other macroeconomists. The second research need, a central part of the proposal, involves further progress in the underlying theoretical and empirical analysis of rare disasters, economic fluctuations, and asset pricing.

National Science Foundation (NSF)
Division of Social and Economic Sciences (SES)
Application #
Program Officer
Georgia Kosmopoulou
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
National Bureau of Economic Research Inc
United States
Zip Code