Eliminating look-ahead bias in evaluating persistence in mutual fund performance

Performance persistence studies typically suffer from ex-post conditioning biases. As stressed by Carhart [Carhart, M.M., 1997. Mutual Fund Survivorship, Working Paper, Marshall School of Business, U.S.C.] and Carpenter and Lynch [J. Financ. Econ. 54 (1999) 337.], standard methods of analysis on a survivorship free sample are subject to look-ahead biases. In this paper, we show how one can easily correct for look-ahead bias using weights based on probit regressions.First, we model how survival probabilities depend upon historical returns, fund age and aggregate economy-wide shocks, using two samples of US based ‘income’ and ‘growth’ funds. Subsequently, we employ a Monte Carlo study to analyze the size and shape of the look-ahead bias in performance persistence that arise when a survivorship free sample is used with standard techniques. In particular, we show that look-ahead bias induces a spurious U-shaped pattern in performance persistence. Finally, we demonstrate how a weighting procedure based upon probit regressions can be used to correct for this bias. In this way, we obtain look-ahead bias-corrected estimates of abnormal performance relative to a one-factor and the Carhart [J. Finan. 52 (1997) 57.] four-factor model, as well as its persistence. The results suggest that in this sample, look-ahead bias is of minor importance and does not seriously affect estimates of persistence. Our bias-corrected results closely correspond to the findings of Carhart [J. Finan. 52 (1997) 57.], implying that there is no evidence on a risk-adjusted basis for persistence in performance.     

Manager education and mutual fund performance

This paper examines the relationship between manager education and mutual fund performance, following Chevalier and Ellison [Chevalier, Judith, Ellison, Glenn, 1999. Are some mutual fund managers better than others? Cross-sectional patterns in behavior and performance. Journal of Finance 54, 3, 875–899]. We refine their analysis by investigating if the quality of the MBA program, as measured by the mean GMAT score and Business Week ranking, has any effect on performance. We find that the mean GMAT score of the MBA program is positively and significantly related to fund performance. Managers who hold MBAs from schools ranked in the top 30 of the Business Week rankings of MBA programs exhibit performance superior to the performance of both managers without MBA degrees and managers holding MBAs from unranked programs. We also find that other education variables, such as whether the manager attained a CFA designation or holds either a non-MBA masters-level graduate degree or Ph.D., are generally unrelated to mutual fund performance.     

Revisiting mutual fund performance evaluation

Mutual fund manager excess performance should be measured relative to their self-reported benchmark rather than the return of a passive portfolio with the same risk characteristics. Ignoring the self-reported benchmark results in different measurement of stock selection and timing components of excess performance. We revisit baseline empirical evidence fund performance evaluation utilizing stock selection and timing measures that incorporate the self-reported benchmark. We introduce a new factor exposure based approach for measuring the – static and dynamic – timing capabilities of mutual fund managers. We overall conclude that current studies are likely to be misstating skill because they ignore the managers’ self-reported benchmark in the performance evaluation process.     

Firm performance and mutual fund voting

We employ a new comprehensive proxy voting records database to investigate whether mutual funds consider prior firm performance when they vote on a diverse range of management- and shareholder-sponsored proposals relating to governance, compensation, and director election. We argue that prior firm performance plays a role in the monitoring effort of mutual funds as they fulfill their fiduciary duties. Results show that voting is related to prior firm performance for selected management and shareholder proposals and that it is consistent with Institutional Shareholder Services’ recommendations. Mutual funds support management (shareholder) proposals less (more) when prior firm performance has been weak. Furthermore, even when mutual funds deviate from their fund family’s voting policies, they attach importance to prior firm performance, and their voting is, to a certain degree, affected by business ties.     

Portfolio concentration and mutual fund performance

Mutual fund managers should choose to increase their portfolio concentration when their information set is valuable enough that the benefits of the expected increase in alpha more than offsets the costs of the expected increase in idiosyncratic volatility. Consistent with that idea, we find that fund performance improves after concentration increases. Because the expected costs of increased concentration vary between funds and over time, the required expected benefits before managers choose to increase concentration should also vary. Among other results, we show that the concentration-performance relation is stronger for funds with less institutional ownership and when investor sentiment is low.     

Recession managers and mutual fund performance

We find that fund managers who began their careers during recessions produce superior returns. This superior performance is not unconditional, as they exhibit better market timing than their non-recession counterparts in recessions, but do not demonstrate better stock picking in booms. Exploring managers' portfolio choices across years, we find that recession managers tilt their investments towards defensive, rather than cyclical, industries during and before recession periods. Overall, our findings support the argument that the economic conditions under which an individual initially entered the labour market exert a long-term impact on her career outcomes and decision-making.     

Persistence in Portuguese mutual fund performance

Recent evidence suggests that future performance is predictable from past performance, that is, funds with superior (inferior) performance in the past are likely to remain good (bad) performers in the future. This research addresses the persistence of mutual fund performance in a European regional market (the Portuguese equity fund market). Some of the problems in evaluating fund persistence are identified in the context of limited sample size and using the peer group median as a benchmark for contingency table analysis of performance persistence. The criteria for assessing performance persistence based on the contingency table methodology of repeated winners and losers are presented in terms of significance statistics, adjusted for small sample bias. The adjustments are accomplished through the Yates continuity correction and Fisher's exact p-value. The appropriateness of each criteria under different circumstances is also discussed. The analysis of the returns of all Portuguese domestic equity funds, since a representative number was established, shows some performance persistence (on a quarterly basis). The persistence, however, is reduced when the returns are controlled for the various dimensions of risk. Significant risk persistence has been documented. Furthermore, for more or less frequent intervals of measurement, the industry persistence is rejected, although individual funds exhibit superior/inferior performance.     

A meta-analysis of mutual fund performance

The purpose of this article is to introduce the statistical technique of meta-analysis of regression results using as our example the Lee and Rahmann (1990) study of the performance of 93 mutual funds. Specifically, we derive and estimate the meta-analysis formulas, explicitly adjusted for correlated regression residuals, which quantify the effect of sampling error on their reported regression results. Our analysis of selectivity reveals some real variation around a mean risk-adjusted excess return of about 1% per year; while our analysis of market timing reveals some real variation around a negative mean value and confirms that the correction for heteroscedasticity does make a difference. An examination of the 80% probability interval for the mean selectivity value indicates that the best mutual funds can deliver substantial risk-adjusted excess returns.     

Explaining persistence in mutual fund performance

This study investigates the determinants of persistence in mutual fund performance. Previous research that uses factor-mimicking portfolios and characteristic benchmarks to model fund performance fails to explain all the persistence in fund returns. This study employs a model that directly relates mutual fund returns to the characteristics of the stocks held by funds. Adjusting fund returns for the size of the stocks in which funds invest and financial ratios intended to capture fund manager investment styles explains all the persistence in mutual fund returns from 1976–1985, the period in which persistence is most prevalent.     

Management team structure and mutual fund performance

We investigate the relationship between performance and portfolio management team structure of open-end mutual funds during 1997–2004. We first analyze differences in performance and risk taking between single-manager and multiple-manager mutual funds and find that the latter underperform the single-manager funds in terms of risk-adjusted returns during the 2001–2004 bear market. This underperformance is more evident among growth-oriented funds. There are no differences observed in the 1997–2000 bull market. Not all multiple-manager funds, however, are managed by pure teams. When we compare the performance of single-manager and pure-team funds we do not find any differences in performance. The underperformance of multiple-manager funds documented in previous studies comes from multiple-manager funds that employ many investment advisors and, therefore, their exact management structure is unknown. We also document differences in management structure reporting between Morningstar and CRSP.     

Timing and selectivity in Portuguese mutual fund performance

We extend the international evidence on timing and selectivity skills of fund managers by applying the Henriksson and Merton [Henriksson, R., Merton, R., 1981. On market timing and investment performance. II. Statistical procedures for evaluating forecasting skills. J. Bus. 54, 513–533] model to Portuguese based mutual funds investing in local, European and International equity.

The results show that managers do not exhibit selectivity and timing abilities, and there is even some evidence of negative timing. Furthermore, we observe a distance effect on stock selection performance, since fund managers that invest locally seem to perform better that those who invest in foreign markets. However, this effect is reverted with respect to market timing skills of fund managers, suggesting that International fund managers are more focused in market timing strategies.     

The persistence of European mutual fund performance

This paper examines the performance and persistence in performance of style-consistent European equity mutual funds between 1988 and 2010. Using a large survivorship bias-free sample for six European countries, we document strong evidence of persistence in benchmark-adjusted returns over 1-year time periods as well as over longer periods. We find statistically and economically significant performance persistence for time horizons of up to 36 months, although persistence is much more pronounced for the top and bottom performers. Thus, past performance of European mutual funds have explanatory power for future performance and investors can obtain useful evidence from past performance data.     

Climate change news sensitivity and mutual fund performance

In the presence of rising concern about climate change that potentially affects risk and return of investors' portfolio companies, active investors might have dispersed climate change risk exposures. We compute mutual fund covariance with market-wide climate change news index and find that high (positive) climate news beta funds outperform low (negative) climate news beta funds by 0.24% per month on a risk-adjusted basis. High climate news beta funds tilt their holdings toward stocks with high potential to hedge against climate change news risk. In the cross section, such stocks yield higher excess returns, which are driven by greater pricing pressure and superior financial performance over our sample period.     

Investor monitoring and differences in mutual fund performance

A number of mutual funds cater exclusively to institutional investors. Although institutional funds might be a natural place to look for “smart money”, agency costs associated with delegated monitoring may lead to less monitoring and worse overall performance. We split institutional funds based on proxies for the degree of investor oversight, and we find that institutional funds with low initial investment requirements and funds with retail mates perform significantly worse than other institutional funds both before and after adjusting for risk and expenses. Tracking error is especially important in the flow-performance relationship of institutional funds with high minimum investment requirements.     

Cross-sectional learning and short-run persistence in mutual fund performance

Using monthly return data of more than 6400 US equity mutual funds we investigate short-run performance persistence over the period 1984–2003. We sort funds into rank portfolios based on past performance, and evaluate the portfolios’ out-of-sample performance. To cope with short ranking periods, we employ an empirical Bayes approach to measure past performance more efficiently. Our main finding is that when funds are sorted into decile portfolios based on 12-month ranking periods, the top decile of funds earns a statistically significant, abnormal return of 0.26 percent per month. This effect persists beyond load fees, and is mainly concentrated in relatively young, small cap/growth funds.     

Measuring bond mutual fund performance with portfolio characteristics

This paper studies the performance of U.S. bond mutual funds using measures constructed from a novel data set of portfolio weights. Active fund managers exhibit outperformance before costs and fees generating, on average, gross returns of 1% per annum over the benchmark portfolio constructed using past holdings (approximately the same magnitude as expenses and transaction costs combined). This suggests that fund managers are able to earn back their fees and costs. There is evidence of neutral ability to time different portfolio allocations (sector, credit quality, and portfolio maturity allocations) and only a subgroup of bond funds exhibit successful timing ability. One performance measure based on portfolio holdings predicts future fund performance and provides information not contained in the standard measures. These results provide the first evidence of the value of active management in bond mutual funds.     

A statistically robust decomposition of mutual fund performance

Previous decompositions of risk-adjusted mutual fund performance might deliver biased results. In this paper, we provide new reliable insights on the drivers of mutual fund performance by decomposing risk-adjusted performance of U.S. equity mutual funds using the Generalized Calendar Time regression model. According to our results, out of all previously considered fund characteristics, only the negative effect of lagged fund size and the positive effects of lagged performance and lagged family size remain highly significant. Our analysis further suggests that much of the variation in previous empirical results can be attributed to methodological issues.     

Managerial multi-tasking,Team diversity,and mutual fund performance

This study examines the impact of multi-tasking teams on fund performance. We find that while managerial multi-tasking has a negative impact on fund performance, teamwork can mitigate the adverse effect associated with managerial multi-tasking, which is indicative of superior performance of funds managed by multi-tasking teams. More importantly, it is the characteristics of the multitasking team that contribute to these superior results, which can be attributed to network cognitive diversity, suggesting that extended networks, facilitated by indirectly-connected managers via local teammates, can largely enhance the scale of cognitive diversity, thus generating significant gains through information pooling and integration. In assessing possible mechanisms for the observed superior performance, we find evidence of improved decision-making induced by network cognitive diversity through both transmission and sharing of value-relevant information, and speedy information diffusion.     

International evidence on ethical mutual fund performance and investment style

Using an international database containing 103 German, UK and US ethical mutual funds we review and extend previous research on ethical mutual fund performance. By applying a Carhart multi-factor model [Carhart, Journal of Finance 57 (1997) 57] we overcome the benchmark problem most prior ethical studies suffered from. After controlling for investment style, we find no evidence of significant differences in risk-adjusted returns between ethical and conventional funds for the 1990–2001 period. Our results also suggest that ethical mutual funds underwent a catching up phase, before delivering financial returns similar to those of conventional mutual funds. Finally, our performance estimates are robust to the inclusion of ethical indexes, which, surprisingly, are not incrementally capable of explaining ethical mutual fund return variation.     

The value of coskewness in mutual fund performance evaluation

Recent asset pricing studies demonstrate the relevance of incorporating coskewness in asset pricing models, and illustrate how this component helps to explain the time variation of ex-ante market risk premiums. This paper analyzes the role of coskewness in mutual fund performance evaluation and finds evidence that adding a coskewness factor is economically and statistically significant. It documents that coskewness is sometimes managed and shows persistence of the coskewness policy over time. One of the most striking results is that many negative (positive) alpha funds, measured relative to the CAPM risk adjustments, would be reclassified as positive (negative) alpha funds using a model with coskewness. Therefore, performance ranking based on risk-adjusted returns without considering coskewness could generate an erroneous classification. Moreover, some fund characteristics, such as turnover ratio or category, are related to the likelihood of managing coskewness.