Equal-weighted Portfolio Research Articles

Lead-lag effects in the mean and variance of returns of size-sorted UK equity portfolios

We test for lead-lag effects in the mean and variance among size-sorted portfolios for the UK stock market. We construct three sets of portfolios, namely a set of size-sorted equally-weighted portfolios of different capitalization size, a set of size-sorted value-weighted portfolios of different capitalization size, and a third set of portfolios of the same capitalization size. The recently proposed Cross Correlation Function test is employed. For both sets of portfolios with different capitalization size, we find evidence of a lead effect in both the mean and the variance from large-firm portfolios to small-firm portfolios. This result does not depend on the weighting scheme used to construct portfolios, and indicates that contrarian trading strategies on large-firm portfolios are profitable. For portfolios of equal capitalization size, there is hardly any evidence of a lead-lag effect in either the mean or the variance. This suggests that the lead-lag effect is due to the difference in the capitalization size among portfolios.

Foreign exchange exposure of exporting and importing firms

The existing literature reports insignificant `total' exposure for multinational or exporting firms, where total exposure incorporates both firm-specific and macroeconomic effects. We propose a dual-effect hypothesis to explain this result which seemingly contradicts conventional wisdom. According to our proposed hypothesis, firms are affected by both the domestic economy and foreign markets. These effects are at least partially offsetting for exporters and additive for importers. The resulting predictions of insignificant total exposure for exporters and positive total exposure for importers are borne out in our tests. The literature also reports insignificant `residual' exposure for multinationals or exporting firms, where residual exposure estimates the firm-specific exposure. This result is explained by biases in the residual exposure estimates introduced by the choice of the value-weighted market index as the control portfolio. We propose an equally-weighted portfolio of purely domestic firms as an alternative portfolio to reduce such biases and report significantly negative exposure for exporters and significantly positive exposure for importers, as predicted by theory.

Foreign exchange exposure of exporting and importing firms

The existing literature reports insignificant `total' exposure for multinational or exporting firms, where total exposure incorporates both firm-specific and macroeconomic effects. We propose a dual-effect hypothesis to explain this result which seemingly contradicts conventional wisdom. According to our proposed hypothesis, firms are affected by both the domestic economy and foreign markets. These effects are at least partially offsetting for exporters and additive for importers. The resulting predictions of insignificant total exposure for exporters and positive total exposure for importers are borne out in our tests. The literature also reports insignificant `residual' exposure for multinationals or exporting firms, where residual exposure estimates the firm-specific exposure. This result is explained by biases in the residual exposure estimates introduced by the choice of the value-weighted market index as the control portfolio. We propose an equally-weighted portfolio of purely domestic firms as an alternative portfolio to reduce such biases and report significantly negative exposure for exporters and significantly positive exposure for importers, as predicted by theory.

Assessing the Size of Asset Pricing Tests under Perfect Ex Ante Efficiency

The capital asset-pricing model (CAPM) asserts that in equilibrium the market portfolio is the tangency portfolio of the efficient frontier spanned by all risky assets. Conceptually, the tangency portfolio represents an ex ante efficient portfolio, while the market portfolio represents an ex post one. In practice, these two portfolios differ in that the value-weighted market portfolio corresponds to a buy-and-hold investment strategy, whereas the tangency portfolio relates to a constant-weight rebalanced strategy. Traditional tests of (ex ante) portfolio efficiency (e.g., the famous test of Gibbons, Ross, and Shanken (GRS, 1989)), however, have mostly focused on the efficiency of the market portfolio. Hence, the size of such tests are distorted because the tests are not correctly specified for the underlying null hypothesis of ex ante efficiency. Using simulations, we find that traditional tests, including the GRS and the GMM tests, seriously overreject the efficiency of the market portfolio under various ditributional specifications. The over-rejection problem is especially severe when the efficiency of the value-weighted market portfolio is tested with respect to equal-weighted portfolios.

Factors Affecting the Implementability of Stock Market Trading Strategies

A substantial body of academic literature provides evidence of stock market trading strategies that generate appreciable abnormal returns. However, there are a number of factors that could partially or fully mitigate the ability of market participants to implement these trading strategies, such as price pressure, restrictions against short sales, and incentives to hold no more than 5% ownership in a firm. We investigate the extent to which these factors account for the evidence of abnormal returns found in seven trading strategies documented in prior research. We find that the size and return reversal trading strategies do not perform well in the presence of the various constraints, whereas the cash flow-to-price, return momentum, post-earnings announcement drift, and operating accrual strategies generally continue to generate significant positive abnormal returns. We find that the book-to-market strategy generates significant positive abnormal returns in about 50% of the scenarios we examine. We find that all of the strategies generate positive abnormal returns in the presence of the restriction against short sales, but that price impact adjustments and constraints on holding more than a 5% stake in any portfolio firm each have a large negative effect on portfolio returns. We also find that equally-weighted portfolio allocations generally perform better than value-weighted allocations and funds with a greater number of stocks and/or lower initial market capitalization also generally produce higher returns. Finally, we find that portfolios that engage in short positions perform worse than long-only portfolios due primarily to the sustained increase in stock prices during the sample period.

Are the gains from international portfolio diversification exaggerated? The influence of downside risk in bear markets

The fundamental rationale for international portfolio diversification is that it expands the opportunities for gains from portfolio diversification beyond those that are available through domestic securities. However, if international stock market correlations are higher than normal in bear markets, then international diversification will fail to yield the promised gains just when they are needed most. We evaluate the extent to which observed correlations to monthly returns in bear, calm and bull markets are captured by three popular bivariate distributions: (1) the normal, (2) the restricted GARCH(1,1) of J. P. Morgan’s RiskMetrics, and (3) the Student-t with four degrees of freedom. Observed correlations during calm and bull markets are unexceptional compared to these models. In contrast, observed correlations during bear markets are significantly higher than predicted. Higher-than-normal correlations during extreme market downturns result in monthly returns to equal-weighted portfolios of domestic and international stocks that are, on average, more than two percent lower than those predicted by the normal distribution. If the extent of non-normality during bear markets persists over time, then a US investor allocating assets into foreign markets might want to allocate more assets into foreign markets with near-normal correlation profiles and avoid markets with higher-than-normal bear market co-movements.

Portfolio Return Metric: Equal Weights Versus Value Weights

Whereas the existing literature focuses on the relation between weighting schemes and abnormal portfolio return metrics, this study extends the literature and investigates the relation between weighting schemes and raw portfolio return metrics. We show that the equal-weight portfolio return metric systematically yields higher estimates of portfolio returns for event samples than the value-weight portfolio return metric. We also demonstrate that value-weight portfolio return metric can be a biased estimator of the counterpart of the population. These results imply that the commonly used testing procedure based on the matching portfolio method and the Fama-French three factor regression can produce misleading inferences. Several remedies are proposed in this study.

An Excursion into the Statistical Properties of Hedge Fund Returns

This paper provides an overview of the most important statistical properties of individual hedge fund returns. We find that the net-of-fees monthly returns of the average individual hedge fund exhibit significant degrees of negative skewness, excess kurtosis, as well as positive first-order serial correlation. The correlations between hedge funds in the same strategy group are of the same order of magnitude as the correlations between funds in different strategy groups and relatively low. Only 10-20% of the variation in the average individual hedge fund's returns can be explained by what happens in the US equity and bond markets. Compared to individual funds, portfolios of hedge funds tend to exhibit lower skewness, higher serial correlation and higher correlation with stocks and bonds. Movements in the US equity and bond markets still only explain 20-40% of the variation in hedge fund portfolios returns though. Finally, an equally-weighted portfolio of all funds in our sample offers a 2.76% higher mean return than the average fund of funds. This strongly suggests that the timing and fund picking activities of the average fund of funds are not rewarded by a higher return.

Stock Valuation and Investment Strategies

This article studies the relative investment performance of several stock-valuation measures. The first is mispricing based on the valuation model developed by Bakshe and Chen (1998)and extended by Dong (1998) (hereafter, the BCD model). The BCD model relates, in closed form, a stock's fair value to (i) the firm's net earnings per share (EPS). (ii) the expected future EPS growth and (iii) long-term rate. The second is a value/ price (V/P) ratio based on the Lee-Myers-Swaminathan (1999) residual-income model. The other measures are all indirect valuation indicators, including book/market (B/M), earnings/price (E/P), size, and past return momentum. These measures are shown to possess distinct properties. For example, the B/M, E/P and V/P ratios are highly persistent over time: high (low) B/M stocks tend to maintain high (low) B/M ratios. But, the BCD model mispricing is highly mean-reverting: an overpriced group will eventually become underpriced (in about 1.5 years on average), and vice versa. More importantly, the BCD model mispricing, momentum, size V/P and B/M are, in decreasing order, significant ex ante predictors of future returns. The best investment strategy is to combine the BCD model mispricing with momentum rankings. Indeed, if one would hold an equally-weighted portfolio of stocks that are the most underpriced and that have top momentum, the average monthly return from 1979 to 1996 would have been 3.18 percent, with a monthly Jensen's alpha of about 1.5 percent.

Political Cycles and the Stock Market

Contrary to the widespread opinion that Republicans are good for business, we find that the average excess returns in the stock market are higher under Democratic presidents. The empirical difference between Republican and Democratic mandates, which is about 9% annually for the value-weighted portfolio and 16% for the equally-weighted portfolio, is economically and statistically significant, and is also robust in subsamples. Moreover, there is a remarkable monotonicity in the difference of returns for size-decile portfolios, from \% for large firms to about 20% for small firms. We test three plausible explanations of the above findings. First, we test the hypothesis that the observed correlation is entirely due to political variables proxying for business-cycle factors. Second, we explore whether the observed relation can be explained by shocks, i.e whether the effect is concentrated around the election dates, when new information is revealed. Lastly, we investigate the possibility that varying volatity and price of risk during the two presidential regimes might account for our findings. We reject all three hypotheses. As it stands, the difference in excess returns during Republican and Democratic presidencies is a puzzling feature of the data that cannot easily be accommodated by asset pricing models.

Are the Gains from International Portfolio Diversification Exaggerated? The Influence of Downside Risk in Bear Markets

The fundamental rationale for international portfolio diversification is that it expands the opportunities for gains from portfolio diversification beyond those that are available through domestic securities. However, if international stock market correlations are higher than normal in bear markets, then international diversification will fail to yie ld the promised gains just when they are needed most. We evaluate the extent to which observed correlations to monthly returns in bear, calm and bull markets are captured by three popular bivariate distributions: (1) the normal, (2) the restricted GARCH(1,1) of J. P. Morgan's RiskMetrics, and (3) the Student-t with four degrees of freedom. Observed correlations during calm and bull markets are unexceptional compared to these models. In contrast, observed correlations during bear markets are significantly higher than predicted. Higher-than-normal correlations during extreme market downturns result in monthly returns to equal-weighted portfolios of domestic and international stocks that are, on average, more than two percent lower than those predicted by the normal distribution. If the extent of non-normality during bear markets persists over time, then a U.S. investor allocating assets into foreign markets might want to allocate more assets into foreign markets with near-normal correlation profiles and avoid markets with higher-than-normal bear market co-movements.

A profitable trading rule using mean reverting investment trusts discounts

This paper makes use of the finding that discounts of individual investment trust shares are mean reverting to their sector mean. This suggests limits to the degree to which individual shares in a sector can vary in price behaviour before competitive pressures bring individual shares back into line. Hence a buy low-sell high strategy can be constructed and its performance compared to an market index based on an equal weight portfolio of all shares in a sector. Such a stratagy outperforms the market index for the two main, conventional investment trust sectors for the sample period 1976–1995.

Testing for mean reversion in heteroskedastic data based on Gibbs-sampling-augmented randomization

Previous work reported that heteroskedasticity did not affect the sampling distribution of the variance ratio, or had assumed that the investigator knew a priori the pattern of heteroskedasticity. This paper uses the Gibbs sampling approach in the context of a three state Markov-switching model to show how heteroskedasticity affects inference and suggest two strategies for valid inference. We find that test procedures which ignore the pattern of heteroskedasticity are biased, rejecting the null hypothesis of no mean reversion too often. We present a resampling strategy that standardizes historical returns, using the Gibbs sampling approach to allow for uncertainty in parameters and states while conditioning on the information in the data. A second strategy is to estimate the VR from standardized data. Again, Gibbs sampling makes appropriate use of the information in the data. For CRSP stock returns 1926–86 we find that evidence of mean reversion is substantially weakened. Gibbs sampling estimates of VRs are closer to unity, and the pattern shifts to shorter lags. There is weak evidence of mean reversion in returns on equal-weighted portfolios and essentially none for the value-weighted. Thus, when returns from high variance periods are appropriately standardized, the sample evidence for mean reversion changes in pattern and significance.

The response of competitors to announcements of bankruptcy: An empirical examination of contagion and competitive effects

We find, like [Lang, L.H.P., Stulz, R.M., 1992. Contagion and competitive intra-industry effects of bankruptcy announcements: An empirical analysis, Journal of Financial Economics, 32(1), 45–60], that large firm bankruptcies generate a dominant contagion effect. A value-weighted portfolio of competitors' stocks experiences a significant loss of 0.56% in the three days centered around the Chapter 11 announcement. This represents an average loss of $3.32 for all the competitors combined for every dollar lost by the bankrupt firm. In addition, we find that small firm bankruptcies also generate a dominant contagion effect among smaller sized competitors; an equally-weighted portfolio of all competitors has a significant 0.12% drop. In a new approach to separate the contagion and competitive effects, we compare the stock price reactions of competitors who themselves subsequently file for bankruptcy in the next three years (candidates for contagion effect) with those who do not do so (candidates for competitive effect). As expected, candidates for contagion effect experience a significant, negative three-day stock price reaction of −4.68%. However, contrary to expectations, candidates for competitive effect also have a significant, negative return (−0.49%), suggesting that the competitive effect is weak at best since it is dominated by the contagion effect even in this sample. Other procedures to identify candidates for competitive effect generally yield similar findings. Finally, we analyze competitors' stock price reactions based on selected characteristics (e.g., industry concentration, and leverage), with similar results as before. One explanation for the failure to detect a competitive effect is that the impact may already have been incorporated in stock prices prior to the filing for Chapter 11. Consistent with this explanation, we find significant positive stock price reactions by competitor stocks for the hundred days prior to the bankruptcy announcement.

Using Stock Return Data to Measure the Wealth Effects of Regulation: Additional Evidence from California's Proposition 103

In contrast to literature that has used the event study method to examine the stock market reaction to California's Proposition 103, we estimate the abnormal performance of both equally-weighted and value-weighted portfolios of securities. We also estimate a cross-sectional regression in value terms; that is, regressing the abnormal change in dollar value of the firm's equity against direct premiums written in automobile and other insurance lines in estimating relative exposure to Proposition 103. Results from the valueweighted portfolios indicate that Proposition 103 had little impact on the aggregate dollar value of the insurance industry. Cross-sectional regressions indicate that exposure, income, and size measures are not related to changes in firm value.

Hypothesis testing with the sharpe and Treynor portfolio: Performance measures given non-synchronous trading

Using the Scholes and Williams (1977) model, consistent estimates of the variances and covariances of security returns under conditions of non-synchronous trading are derived. The effects of using such corrected estimates on the measurement of portfolio performance are illustrated by assessing the performance of a value- and equally-weighted portfolio of all the stocks issued under the Quebec Stock Savings Plan (QSSP).

The Investment Performance of the Common Stock Portfolios of Property-Liability Insurance Companies

This paper has examined the investment performance of the common stock portfolios of 20 property-liability insurance companies over the period 1958–1967. The performance of these portfolios was compared with the performance of both equal-weighted and value-weighted random portfolios of common stocks. The evidence indicated that the returns earned by the insurance companies were significantly lower than the returns earned by random portfolios of equivalent risk. While this study is not precisely comparable to the previously published studies of mutual funds in terms of either the methodology employed or the time period considered, the results are quite similar. An additional group of institutional investors can be seen not to have outperformed the “market.” The presumption of inferior performance is even stronger in the case of the property-liability companies because all the comparisons in this study were made on a gross basis. Most of the studies of mutual fund performance have compared fund returns net of investment expenses to the gross returns from random investments. In those instances where comparisons were made on a gross basis, the performance of fund portfolios was found to be very similar to the performance of random investments.