MEASURING FUND PERFORMANCE USING MULTI-FACTOR MODELS: EVIDENCE FROM THE INDIAN MUTUAL FUNDS AND ULIP FUNDS

Pardeep Kumar 1 , Dr. Harpreet Aneja 2 and Dr. Ashwani Kumar 3 . 1. Research Scholar, IKGPTU, Kapurthala, India. 2. Associate Professor, Gulzar Group of Institutes Khanna, India. 3. Research Guide, IKGPTU Kapurthala, India. ...................................................................................................................... Manuscript Info Abstract ......................... ........................................................................ Manuscript History


313
concluded that there is very little evidence that any individual fund was able to do significantly better than that which we expected from mere random chance [5]. E. Fama (1970) developed a methodology for evaluating investment performance of managed portfolios by suggesting that the overall performance of managed portfolios could be broken down into several components. He argued that the observed return of a fund could be due to "selectivity ability" i.e. ability of fund managers to pick up the best securities at a given level of risk and "timing ability" i.e. due to the prediction of general market price movement [6]. Thereon, several researchers provided theoretical and empirical evidences that expected returns can be explained by more than one variable i.e. by using multifactor models. Stock returns are affected by many factors influencing the change in future cash flows and a five-factor model which included expected inflation, unexpected inflation, interest rate term structure, default premium and industrial production was proposed [7]. Elton et al. [8] suggested for a three-index model, including return on large stock, small stock and bond indexes. Fama and French put forward a 3-factor model comprising, the return on the market portfolio, firm size and book-to-market ratio [9]. In addition to Fama and French"s three-factor model, Mark Cahart presented that fund managers employ momentum strategy in order to earn an abnormal return [10]. The momentum anomaly was pointed out in [11]. It was concluded that stocks which perform best over a period, tend to continue to perform well over a subsequent period, and if fund managers employ this phenomenon in order to earn abnormal returns, it should not be counted as value added. Subsequently, Carhart proposed his fourfactor model which included one extra variable to capture the momentum anomaly in addition to 3 factors from Fama and French.

Empirical Literature Review:-
Many empirical studies attempted to get a better understanding and a more accurate assessment of the fund performance. A fund that outperforms the benchmark on an after-cost basis is considered to adding value for investors. Active fund managers believe that they have the ability to better estimate the true securities" risks and returns, to spot any mispriced securities, and to time the market, thus generating excess returns for the fund and adding value to the investors. Some of the important researches in this area are as discussed below: Financial performance of five close-ended growth funds for the period February 1991 to August 1993 was evaluated and it was concluded that the performance was below average in terms of alpha values (all negative and statistically not significant) [12]. In the research [13] 191 mutual funds of New Zealand were studied from August 1991 to July 2001 by employing Fama and French and Carhart Models. It was established that New Zealand mutual funds underperformed the benchmark and performed poorly with negative Jensen"s Alfa Values. One another study over the period  found that the average net return alpha was around 0.5% below the market benchmark and there were some fund managers who significantly outperformed the market [14]. Later, in a research [15] the performance of Greek equity fund managers for the period June 2001-December 2009 was evaluated using single and multi-factor risk-adjusted performance measures and found that the managers fail to demonstrate significant stock picking abilities resulting in repeated underperformance relative to benchmarks. Managers also fail even to justify the various expenses paid by them for active portfolio management services.
In a study on emerging markets 164 mutual funds of the Polish mutual fund industry were evaluated by using Carhart Model and it was concluded that Polish mutual funds on average were not able to add value, as indicated by their negative net alphas [16]. An important research on Portuguese markets got conducted and overall performance and market timing abilities of Portuguese equity funds investing in the domestic and in the European Union market, during the period of January 2000 to December 2007 were studied [17]. It was found that while National funds were neutral performers, European Union funds underperformed the market significantly. A major study in the south Asia in China, Singapore and Thailand for twelve years (2000 to 2011) was done by using Carhart four-factor model to determine whether any equity mutual funds significantly outperformed, or beat the market. The results showed that the chance for equity mutual funds to beat the market was very slim, but the chance that equity mutual funds performed poorer than their index performance was much higher [18]. A multifactor study in India [19] used Fama-French model and the Carhart model and suggested that fund managers did not add any value to the fund returns. The return generated by the funds, were due to the risk factors already considered in these models. It raised questions on the fund houses pocketing huge fees while not adding any significant value to the return. In an article it was concluded that the active managers could add value if they timed market-turns well but it was prohibitively difficult to identify these turns. Also active funds were charging far too much just for the market discovery and it was hard to believe that why the investors were paying such prices [20].

314
From the above literature, it can be seen that there is a dearth of literature available on ULIP fund performance, whereas mutual fund evaluation studies conclude that researchers reach contradictory conclusions while evaluating fund performance. But the majority of the researchers ascertained that portfolio managers are not able to outperform the market. Through this paper, we try to apply two of the most accepted methodologies i.e. Fama & French Three Factor Model and Carhart Four Factor Model to evaluate the performance of Indian equity-oriented Mutual funds and ULIP funds.

Objectives of the Research:-
The performance of the mutual funds and the ULIP funds is evaluated through achieving the following main objectives: i. To assess whether the mutual fund managers are adding any significant value to the return generation. ii. To assess whether the ULIP fund managers are adding any significant value to the return generation.

Research Methodology:-Research Design:-
The research followed is empirical in nature. Mutual Fund and ULIP fund daily NAVs data from April-2008 to March-2016 is used. NSE Nifty 50 has been chosen as the market benchmark index and daily closing data is acquired from the NSE website. "91 Day Treasury Bills" return is taken as the risk free rate and the data is obtained from the Reserve Bank of India (RBI) website. Only those funds which have a corpus size of more than 50 crore, at least 60% equity investment in its portfolio; and which are launched before April 1, 2008 are selected using a deliberate sampling method to make the study long term and robust.

Fama and French's Three Factor Model:-
Fama and French"s three-factor unconditional model [6] was used to evaluate the performance of the sample of funds. In addition to a value-weighted market proxy, this model includes two additional risk factors, size (SMB) and book-to-market (HML). It is calculated as: r ptr ft = + βp (r mtr ft )}+ sp(SMB) + hp(HML) + ε t …..(1) Where: SMB -SMB denotes the size factor which reflects the difference between the returns on small stocks and the returns on big stocks. SMB is calculated as the average return difference of a small cap index, i.e. NIFTY Small Cap 50 and a large cap index, i.e. NIFTY 50. HML -HML denotes the book-to-market-value factor which measures the difference between the returns of stocks with high book-to-market ratios and stocks with a low book-to-market ratio. It can be calculated by the return of the top 30% of stocks ranked by book-to-market minus the bottom 30% of stocks ranked by book-to-market ratio.

Carhart Four Factor Model:-
Carhart four-factor model [10] was used to evaluate the performance of the sample of funds. It considers an additional factor to the three-factor model of Fama and French, which is the momentum factor. It is calculated as: r ptr ft = + βp (r mtr ft )}+ sp(SMB) + hp(HML) + piPR12m + ε t …..(2) where = Carhart"s alpha measure for fund p ε t, = Error Term PR12m = the momentum factor capturing Jegadeesh & Titman (1993) momentum anomaly is formed. We rank the sample securities at the end of t-1 on the basis of their 12m average returns and form two groups, i.e., P 1 & P 2 while P 1 includes top 30% of stocks on past performance, P 2 comprises of a bottom 30% of stocks. The equally-weighted daily returns are estimated for the year t and the return difference is calculated between the two portfolios. The portfolios are re-formed at the end of time"t" and the process is repeated on year to year basis.

Empirical Evidences:-Performance Evaluation of Equity Based Mutual Funds using Fama & French Model:-
The regression analysis is utilized to obtain various coefficients has given below:

Performance Evaluation of Equity Based ULIP Funds using Fama & French Model
The regression analysis is utilized to obtain various coefficients as given below in the table: Aviva Growth 0.011 0.  Table 2 shows that out of selected 60 unit linked insurance plan funds, 58 have positive alpha values. Only two funds i.e. SBI Growth Fund and SBI Growth Pension Fund show a negative alpha value. So, the fund managers are adding value to their respective fund portfolios. Moreover, it can be witnessed that out of these 58 funds only one fund, HDFC Equity Managed Fund has the significant alpha value (p values less than .05 at 5% significance level), while all the others have insignificant ones. So, it can be established that most of the ULIP funds are not making profits in excess of returns on a market index.
It can also be noticed that in the case of equity based ULIP funds excess market returns have non-significant impact on the mutual fund returns since the mean "rm-rf" beta coefficient is 0.0493 with 51 of the 60 corresponding pvalues greater than 0.05. For example Aviva Enhancer Fund has an "rm-rf" beta coefficient of -0.00219 and p=0.925 i.e. > 0.05(At 5% Significance Level). The size factor SMB with a mean value of 0.23 is significant (59 of 60 selected funds having p<0.05). Here, Aviva Enhancer Fund with SMB 0.26 (p<0.05) has significant size factor. Similarly, the value factor HML with 0.0534 mean value, have a smaller magnitude but significant impact (36 of 60 funds showing p< 0.05). But Aviva Enhancer Fund has non-significant HML factor of 0.0406 (p> 0.05). With R 2 value of 0.062 it can be inferred that just 6.2 % of returns generated were because of the market movements, size and value factors only. Therefore, the ULIP fund returns were due to many other factors which are not accounted for in the Fama and French Three Factor model.

Performance Evaluation of Equity Based Mutual Funds using Carhart Model
The regression analysis is utilized to obtain various coefficients as given in the forthcoming table:  It can be detected from the table 3 that in the case of equity based mutual funds excess market returns have very significant impact on the mutual fund returns since the mean "rm-rf" beta coefficient is 0.826 with all corresponding p-values less than 0.05. For example Baroda Pioneer Growth Fund has an "rm-rf" beta coefficient of 0.909 and p=0.0 i.e. < 0.05(At 5% Significance Level). The size factor SMB with a mean value of 00.142 is having very small magnitude but significant (82 of 84 selected funds having p<0.05). Here, Baroda Pioneer Growth Fund with SMB 0.084 (p<0.05) has significant size factor. Similarly, the value factor HML with -0.0269 mean value, have a smaller magnitude but significant impact (68 of 84 funds showing p< 0.05). But Baroda Pioneer Growth Fund has nonsignificant HML factor of -0.02597 (p< 0.05). The fourth, newly introduced factor PRIM has again negative but very small mean value i.e. -0.00818 which is non-significant (1of 84 have p<0.05). Baroda Pioneer Growth Fund with PRIM 0.01020 (p>0.05) has non-significant momentum factor. With R2 value of .81 to 0.89 it can be inferred that 81% to 89% of returns generated were because of the market movements, size, value and momentum factors only. Therefore, only approximately 11% of the mutual fund returns were due to many other factors which are not accounted for in the Carhart Four Factor model

Performance Evaluation of Equity Based ULIP Funds using Carhart Model
We use the regression analysis to obtain various coefficients as given in the next table:   Table 4 shows that out of selected 60 unit linked insurance plan funds, 58 have positive alpha values. Only two funds i.e. SBI Growth Fund and SBI Growth Pension Fund show a negative alpha value. So, the fund managers are adding value to their respective fund portfolios. Moreover, it can be witnessed that out of these 58 funds not even a single fund has the significant alpha value. So, it can be established that most of the ULIP funds are not making profits in excess of returns on a market index.
It can be discovered from the table 4 that in the case of equity based ULIP funds excess market returns have very non-significant and small magnitude impact on the ULIP fund returns since the mean "rm-rf" beta coefficient is 0.0374 with only 9 corresponding p-values less than 0.05. For example Aviva Enhancer Fund has an "rm-rf" beta coefficient of -0.00185 and p=0.94 i.e. > 0.05(At 5% Significance Level). The size factor SMB with a mean value of 0.231 is having very small magnitude but significant (all selected funds having p<0.05). Here, Aviva Enhancer Fund with SMB 0.084 (p<0.05) has significant size factor. Similarly, the value factor HML with -0.0269 mean value, have a smaller magnitude but significant impact (68 of 84 funds showing p< 0.05). But Aviva Enhancer Fund has non-significant HML factor of -0.02597 (p< 0.05). The fourth, newly introduced factor PRIM has again negative 326 but very small mean value i.e. -0.00818 which is non-significant (1of 84 have p<0.05). Aviva Enhancer Fund with PRIM 0.00457 (p>0.05) has non-significant momentum factor. With R2 value of 0 to 0.1 %, it can be concluded that only maximum of 10% of returns generated were because of the market movements, size, value and momentum factors. Therefore, about 90% of the ULIP fund returns were unexplained by the Carhart Four Factor model.

Conclusions:-
The study has evaluated the performance of selected 84 equity-oriented mutual funds and 60 equity-oriented ULIP funds in India over the period April 2008 to March 2016 by employing multifactor models Fama & French Three Factor Model and Carhart Four Factor Model. It can be concluded that the mutual funds are adding value to their respective fund portfolios, whereas the ULIP funds are not adding any significant value to their respective fund portfolios. So, the large amount of expenses charged by the ULIP fund managers is not justified. The further research studies can be conducted in future by focusing on different other multi-factor models.