An International Welfare Analysis of Inflation-Indexed Government Bonds
International Monetary Fund (IMF) - International Capital Markets Department
Craig W. Holden
Indiana University - Kelley School of Business - Department of Finance
May 18, 2005
We analyze the introduction of an inflation-indexed government bond in a Markowitz framework of individual agent portfolio optimization. Our theoretical metric for the welfare gain is the percent increase in wealth required to make the investor indifferent between holding the benchmark set of risky assets only vs. the expanded set of assets, which includes the inflation-indexed government bond. This metric is estimated using data on real monthly rates of return on domestic and international assets from 1993-2003. Optimal portfolios are calculated for hypothetical investors in four developed countries (Canada, France, Sweden, and the United States) and two developing countries (Brazil and Mexico) before and after introduction of an inflation-indexed government bond. We find that risk averse investors based in Brazil and Mexico benefit more than risk tolerant investors from the introduction of inflation-indexed government bonds due to the high variability of real returns on existing nominal return assets. High inflation and exchange rate risk leads to less favorable investment opportunities when indexed debt is not present. In contrast, we find that gains in the four developed economies are largest when risk tolerance is highest, but these gains are dependent on the ability to engage in short-selling which may not be allowable or feasible if inflation-indexed debt is not sufficiently liquid. Otherwise, the developed country examples indicate that demand for indexed bonds will likely be low when inflation and exchange rate risk is low, and may explain why the main holders of such debt are insurance companies and pension plans.
Number of Pages in PDF File: 34
Keywords: Inflation, indexed, exchange rate risk, bond
JEL Classification: G15
Date posted: June 17, 2005
© 2016 Social Science Electronic Publishing, Inc. All Rights Reserved.
This page was processed by apollobot1 in 1.157 seconds