Asset Pricing
Applied Econometrics
International Finance
Monetary Economics

Empirical Finance
International Finance
Investment Analysis
Macroeconomics

May 1999 (Oral) Econometrics, Finance
May 1998 (Written) Macroeconomics and Microeconomics

Asset Co-Movement and Diversification Benefits

Professor William N. Goetzmann (Chair)
Professor K. Geert Rouwenhorst
Professor Robert J. Shiller
Professor George J. Hall

May 2003

Ph.D., Economics, Yale University, May 2003, expected
M. Phil., Economics, Yale University, December 2000
M.A., Economics, University of Miami, May 1997, with Distinction
B.A., Quantitative Economics, Nanjing University, July 1995, with Distinction

Yale University
   Summer Research Fellowship, 2002
   Dissertation Fellowship, 2001
   Pre-Dissertation Research Fellowship, 1999, Yale Center for International and Area Studies
   Yale University Fellowship, 1997-2000
University of Miami
   Dean's Scholarship, School of Business, 1995-1997
Nanjing University
   Best Paper Award of the Student Research Contest, 1995
   The Creativity Fellowship for Outstanding Undergraduate Research, 1993-1995
   Ranked Second among the Entering Class of Science Majors, 1991

Doctoral Courses
   Econometric Theory II, for Professor John Rust, Spring 2001
   Macroeconomic Theory I, for Professor Robert Shiller, Fall 2000
M.B.A. Course
   Investment Analysis, for Professor Roger Ibbotson, Yale School of Management, Fall 2002
Undergraduate Courses
   Financial Markets, TBA, Spring 2003
   Introductory Macroeconomics, for Professors William Nordhaus and George Hall, Fall 199

Research Assistant for Professor William D. Nordhaus, Department of Economics, 2001
Research Assistant for Professor William N. Goetzmann, Yale School of Management, 1998

Deutsche Bank Global Markets, Hong Kong, Associate, Summer 1999, Winter 2000
Valuengine Inc., Stamford CT, Consultant, 2000
Centennial Olympic Games of Atlanta, Volunteer and Language Agent, Summer 1996

"Macroeconomic Factors and The Correlation of Stock and Bond Returns," 2002 [Job Market Paper]

"Long-Term Global Market Correlations," NBER Working Paper 8612, 2001, with William N. Goetzmann and K. Geert Rouwenhorst, under review

"Conditioning Bias and Testing the Constancy of An Unconditional Correlation Matrix," 2001

"Macroeconomic News and Stock Returns," Working paper, Department of Economics, 2000

"The Dilemma of China's Infrastructure Industries: An Application of the Harrod-Domar Model of Investment," Chinese Economic Issue 1994-1 (in Chinese)

"Statistical and Economic Measures of Diversification Benefits"
"The Term Structure of Inflation Expectations"

American Finance Association Annual Conference, Washington D.C., 2003, [Paper Accepted]
International Monetary Fund Conference on Global Linkage, 2003, [Paper Accepted]
Western Finance Association Annual Conference, Park City, Utah, June, 2002, presenter
Inter-University Graduate Student Conference, Princeton University, October, 2002, presenter

Diversification is said to be the only free lunch in finance. However, the amount of risk that can be reduced by forming a diversified portfolio depends on the correlation structure of the assets in the portfolio. When diversification benefits vary over time, their implications for domestic and international investors also change. My dissertation analyzes correlation structures and diversification benefits in both the domestic and international contexts. The goal is to investigate the co-movement properties of the most common asset classes, develop tools to examine asset co-movements, and improve the measures for evaluating diversification benefits. I also attempt to understand asset co-movement properties by establishing a link between asset return correlations and fundamental economic variables.

Chapter I, Macroeconomic Factors and the Correlation of Stock and Bond Returns, examines the correlation between stock and bond returns. It first documents that this correlation in G7 countries increased steadily from zero in the early 1960s to 0.5 by the mid-1990s, and then reverted back to zero by the turn of the century. The asset pricing model employed in this paper shows that the correlation of stock and bond returns can be explained by their common exposure to three macroeconomic factors: unexpected inflation, expected inflation and the real interest rate. The link between the stock-bond correlation and macroeconomic factors is examined using three successively more realistic formulations of asset returns dynamics. Empirical results indicate that major trends in the stock-bond correlation are determined primarily by the uncertainty about expected inflation. Unexpected inflation and the real interest rate are significant to a lesser degree. Forecasting this stock-bond correlation using macroeconomic factors also helps improve investors' asset allocation decisions. One implication of this link between trends in stock-bond correlation and inflation risk is the Murphy's Law of Diversification: diversification opportunities are least available when they are most needed.

Chapter II, Statistical and Economic Measures of Diversification Benefits, develops a utility based economic measure for diversification benefits, calculated as the maximum premium that an investor is willing to pay for holding a more diversified portfolio. This measure is contrasted with the conventional mean-variance spanning tests based on Hubermann and Kandel (1987) and the GMM tests based on Hansen and Jagannathan (1991). The utility based economic measure allow one to evaluate the expansion of the investment opportunity set by combining both the risk and return properties. It also offers a flexible framework to examine how investors with different tolerances for risk may respond to the expansion of the investment opportunity set. This paper empirically investigates whether, over the last fifty years, the greater accessibility to G7 stock markets, the major fixed income markets, and the emerging markets has significantly expanded the mean-variance efficient frontier that a U.S. based investor faces. It shows that, although the U.S. stock market has become increasingly closer to the mean-variance efficient frontier of G7 stock markets, the successive introduction of the major fixed income markets and the emerging markets still help reduce the overall risk and increase the expected return. The diversification benefits are substantial and should not be ignored.

Chapter III, Conditioning Bias and Testing the Constancy of An Unconditional Correlation Matrix, discusses the type I and type II errors of conditioning bias (Boyer, Gibson and Loretan 1999, and Forbes and Rigobon 2002), and develops the tools to test shifts in unconditional correlations. It shows that, although conditioning bias remains an important caveat in testing the constancy of unconditional correlation, its magnitude can be sharply reduced in a more general setting of factor models. This paper examines the conventional constancy test of univariate correlation coefficient under various factor model assumptions, with which the conditioning bias can be directly adjusted. It also extends the test to multivariate setting to examine the equality of two correlation matrices. This multivariate test is based on the asymptotic distribution of the correlation matrix derived by Browne and Shapiro (1986) and Neudecker and Wesselman (1990). The test statistic is shown to have a Chi-square asymptotic distribution. Small sample properties and the power of the test are discussed. Bootstrap studies of the robustness of the test show that it works well as a basis for distinguishing among periods of differing asset correlation structures. Applying this test to historical data of major stock markets shows that we can convincingly reject the constancy hypothesis of the global correlation structure between various periods in international financial history.

Chapter IV, Long Term Global Market Correlations (With William N. Goetzmann, and K. Geert Rouwenhorst), is a study of global equity markets in the long-term historical context. In this paper we examine the correlation structure of the major world equity markets over the past 150 years. We find that correlations vary considerably through time and are highest during periods of economic and financial integration such as the turn of 19th to 20th century, the Great Depression period and the current period. Our analysis suggests that the diversification benefits to global investing are not constant, and that they are currently low compared to the rest of capital market history. We decompose the diversification benefits into two parts: a component that is due to variation in the average correlation across markets, and a component that is due to the variation in the investment opportunity set. There are periods, like the last two decades, in which the opportunity set expands dramatically, and the benefits to diversification are driven primarily by the existence of marginal markets. For other periods, such as the two decades following World War II, risk reduction is due to low correlations among the major national markets. From this, we infer that periods of globalization have both benefits and drawbacks for international investors. They expand the opportunity set, but diversification relies increasingly on investment in emerging markets.