Essays on mutual fund performance in India

Guide(s)

Anshuman, V Ravi

Department

Finance and Accounting

Area

Finance and Accounting

University

Indian Institute of Management Bangalore

Place

Bangalore

Publication Date

3-31-2024

Year Awarded

March 2024

Year Completed

March 2024

Year Registered

June 2018

Abstract

Active equity mutual funds (MFs) in India charge their investors approximately $3 billion annually, and fund managers are handsomely compensated for their active management efforts. Despite this, the effectiveness of MFs in delivering superior returns to investors has been a topic of intense debate among academics and industry practitioners alike, but without any definitive conclusions. This dissertation addresses this lacuna by estimating the magnitude of performance of active funds compared to various benchmarks. In addition, the dissertation also assesses the role of the fund managers' i) market timing skills, (ii) stock selection ability, and (iii) informational advantages in driving fund performance. The first essay conducts a thorough empirical examination of the performance of active MEs in India. The existing literature is limited and faces several issues, including conflicting findings regarding the efficacy of active funds, survivorship bias, small sample size, and concerns about obsolete results of existing studies due to the exponential growth of the Indian MF industry in the past decade. Employing a survivorship-bias-free sample of 251 funds spanning from 1995 to 2021, on average, I find weak evidence of the outperformance of active funds over their assigned passive benchmark indices (0.14% monthly average 4-factor alpha). Further, the modest outperformance of active funds in the overall sample can be attributed to the period till 2005 and has practically vanished thereafter, suggesting increased market efficiency and raising questions on the usefulness of active fund management. About 20% of funds in the sample demonstrate positive and statistically significant performance estimates. While ELSS funds showed the highest average performance measures, Dividend Yield funds had the lowest average performance estimates. These findings are robust across the majority of the performance measures. including raw returns, risk-adjusted performance ratios, and factor models. Additionally. I find weak evidence of short-run persistence in performance but no significant persistence at the 1-year horizon. The persistence, too, weakens over the sample period. I find strong evidence of performance chasing among investors, but fund performance is not sensitive to new fund flows. The second essay employs the factor model and holdings-based approaches to examine the drivers of superior MF performance, i.e., the market timing (MT) and stock selection (SS) skills of fund managers. Using both approaches, on average, I do not find any statistically significant and robust evidence of MT skill among fund managers. About 10% of funds in the sample demonstrate positive and statistically significant MT skills, while a slightly higher fraction of funds (14% - 26%) demonstrates negative and statistically significant estimates of MT skills, suggesting that a considerable fraction of funds generate negative returns when attempting to time the market. Using the factor model approach, I find robust evidence of SS skill driving fund alphas. However, the SS skill estimates weaken over the sample period. Around 20% of funds display positive and statistically significant evidence of SS skill, while only a small number of funds (< 10) exhibit negative and statistically significant evidence of stock SS. The analysis also establishes that while the factor model approach clearly identifies MT skill, the identification of SS skills is less clear and is sensitive to the choice of benchmarks. Finally, the holdings-based approach, which eliminates reliance on benchmarks for MF performance evaluation, suggests that the majority of fund returns are generated by the general investment style of the funds. The approach provides further robustness to the finding of positive SS skill among fund managers. The third essay proposes a novel approach to quantify the informational advantages to fund managers by analyzing their professional peer networks. I employ the network analysis approach to examine whether fund managers who occupy more central positions within the peer network of fund managers have informational advantages and, thus, perform better. Using a novel dataset of fund-level assignments of MF managers in India from 1993 to 2021, I find that a fund manager's performance tends to improve when they have directly worked with a higher number of fund managers at the same fund. Further, the fund managers who are, on average, closer to the other fund managers in the peer network tend to perform better, underlying the importance of "speed" of information acquisition. However, fund managers who occupy "broker" positions in the network, i.e., connecting two clusters of fund managers, do not seem to exhibit superior performance. Finally, the analysis suggests that having influential or "elite" connections has a negative effect on the performance of fund managers. The results are consistent across multiple tests.

Pagination

xix, 183p.

Copyright

Indian Institute of Management Bangalore

Document Type

Dissertation

DAC Chairperson

Anshuman, V Ravi

DAC Members

Rangan, Srinivasan; Das, Debjyoti

Type of Degree

Ph.D.

Relation

DIS-IIMB-FPM-P24-12

This document is currently not available here.

Share

COinS