Benefits from international portfolio diversification have been widely studied and well documented. During recent years, however, the landscape for international investments has changed appreciably. Liberalization and deregulation of financial markets, particularly emerging markets, and increases in privatization worldwide have brought expanded investment opportunities for global investors. The growing influence of foreign equity investments attracts the attention of policymakers attempting to stabilize their financial markets and to achieve their domestic economic goals. Similarly, investors make their geographic portfolio allocation decisions and establish trading strategies based on an increasing number of choices.
This study addresses these issues by studying the behavior of U.S. investors in the 18 largest emerging country equity markets and in the 18 largest developed country equity markets around the globe. Specifically, the focus of the study is to investigate: (1) momentum strategies of buying past winners and selling past losers, (2) measures of investment performance for U.S. investors, and (3) effects of adjusting for risk and for momentum effects in evaluating investment performance.
The analyses of trading behavior and investment performance of U.S. investors are based on their monthly purchasing and selling activities in foreign markets over the 1992–2003 period. The investigation is motivated by previous studies which indicate that investors often earn significant profits by buying past winners and selling past losers in the U.S. market. Rather than investigating only the domestic market, this study investigates momentum strategies globally and attempts to identify whether superior performance, if it exists, is the result of the simple trading strategy or is the result of risk taking and/or momentum influences. While most past studies focus on performance measures from simulating momentum strategies, this study uses actual trading data. Because momentum trading can have major effects on capital flows and market stability and investor performance can have major effects on portfolio allocations, information gleaned from this study should prove useful to both policymakers and investors.
The investment performance of U.S. investors in global markets is evaluated using three measures of returns: raw returns, risk-, and momentum-adjusted returns. Comparisons of strategies and performance are made between developed markets and emerging markets and over different holding periods.
The paper proceeds as follows: Section 2 reviews literature, Section 3 presents data and methodology, Section 4 reports empirical results, and Section 5 concludes.
This study investigates trading behaviors and investment performance of U.S. investors in the world's 36 largest international equity markets from December 1992 to June 2003. The evidence shows that U.S. investors consistently tend to employ winners-momentum trading strategies by buying past winners in emerging markets (at all time horizons), developed markets (at 11 of 15 time horizons), and global markets (at 12 of 15 time horizons). Conversely, they employ losers-contrarian trading strategies by buying past losers in all three market segments and at all time horizons. Considering both winners-momentum and losers-contrarian behaviors, in general, they purchase more past losers than past winners. These U.S. investors are winners-momentum traders and losers-contrarian traders in emerging markets for all time horizons and tend to buy more past winners than past losers in the long-term. In developed markets, U.S. investors tend to be winners-momentum traders in the short-term and midterm while they exhibit losers-contrarian behavior over all time horizons. Then tend to buy more past losers than past winners over all but the shortest time horizons. When U.S. investors consider emerging and developed markets jointly, their behaviors parallel those of developed markets only.
Thus, there is strong evidence of winners-momentum trading in all market segments, a behavior possibly tending to cause market instability as prices diverge from fundamental values. Interestingly, however, there is even stronger evidence of losers-contrarian behavior. Thus, U.S. investors tend to buy past winners but they buy even more past losers, a stabilizing behavior for equity markets and especially important during periods of financial crises.
Investment performance is evaluated by the use of three measures of returns—raw returns, risk-, and momentum-adjusted returns—over time horizons varying from 1 month to 30 months. In emerging markets, U.S. investors reveal strongest performance when risk-adjusted returns are measured, followed closely by raw returns measures. However, once returns are momentum-adjusted, U.S. investors performance drops drastically and becomes negative for 10 of the 15 time horizons. Thus, superior returns appear to result from momentum trading rather than risk taking.
When developed markets only are considered, U.S. investors’ performance is poor by all three measures of returns, with the performance being almost identical for raw returns, risk-, and momentum-adjusted returns.
When viewed from a global perspective, including emerging and developed markets, performance results are slightly better than for emerging markets alone and appreciably better than for developed markets alone. Again there are indications that any superior returns result more from momentum trading effects than from risk taking effects.
One interesting performance finding is that long-term performance is, in general, superior to short-term and medium-term performance. This is especially true for emerging and combined markets. This implies that U.S. investors may have an information advantage over local investors in that they are able to predict long-term price movements. This is particularly true in emerging markets while the advantage does not appear as strong in developed markets. This finding is similar to that of Chan et al. (2000) who find that momentum profits fall when they exclude their six emerging markets from their 23 country sample. It also supports earlier studies that identify momentum-trading strategies in emerging markets, for example, Richards (2002), Grinblatt et al. (1995), and Froot et al. (2001). However, Hameed and Kusnadi (2002), who base their momentum trading results on individual stocks in six Asian markets, find little evidence of momentum profits. Further, the indications of minimum momentum trading in developed markets are not in keeping with earlier results for other developed markets, such as those of Jegadeesh and Titman (1993) for U.S. markets, Grinblatt and Keloharju (2000) for Finland, and Karolyi (2002) for Japan. The most unique results are the significant differences between emerging and developed market trading behaviors. These differences follow through to similar differences in performance measures. Overall and in general, U.S. investors perform best in global markets, less well in emerging markets, and poorly in developed markets.
The study's findings have important implications for both investors and policymakers. Investors can profit from recognizing U.S. investors’ winners-momentum strategies, activities which appear to cause past winners to rise in the short run. The short-run superior performance, driven primarily by winners-momentum rather than by losers-momentum and risk taking, suggests that U.S. investors’ behavior reflects new information to the market. These investors perform even better in the long term, suggesting that investors have a good chance to profit in global markets over both short- and long-term horizons. Thus, U.S. investors may be superior price discoverers who are able to analyze markets’ fundamentals and long-term trends better than local investors.
The implications for policymakers in emerging markets differ from those in developed markets. For emerging markets, winners-momentum trading is consistently strong but losers-contrarian trading is stronger. Thus, rather than selling losers, U.S. investors continue to buy and these purchases of past losers are greater than the purchases of past winners. This should be a comforting situation in that capital outflows during a period of financial turmoil would be less than with losers-momentum trading. This interpretation, however, must be evaluated with care because the study period encompasses the Mexican, Asian, and Russian financial market crises.
For the developed markets, winners-momentum trading remains strong and again losers-contrarian trading overrides winners-momentum trading. Risk-taking and momentum effects on investment performance are weak and the short-term superior performance is mainly driven by a simple trading strategy.
Overall, the markets reflect winners-momentum trading and losers-contrarian trading. Momentum effects on investment performance are strong but risk-taking effects are weak.
