Across the world, the dominant narrative is that it’s hard for active managers to outperform; it’s hard for investors to identify winning managers in advance; and even harder to stick with them through inevitable cycles of underperformance. Therefore, investors should control what they can and opt for low-cost passive funds that promise only to match market performance.
This narrative runs aground in India.
It is long known, at least anecdotally, that Indian actively managed funds have outperformed their benchmarks despite their high expense ratios. (The average expense ratio for an Indian large-cap active fund is around 2.5%.)
In this article, we attempt to answer three questions:
- Do India’s active funds truly outperform?
- If they do outperform, is the outperformance explained by systematic factors?
- Does the outperformance persist?
Two caveats are in order.
First, we only considered active funds domiciled in India throughout the study; we don’t include India-focused funds that are domiciled outside India.
Second, we limited the analysis to Morningstar India’s large-cap and small/mid-cap categories and excluded the tax savings and flexicap categories, because the former categories account for the largest proportion of funds by number and assets under management. These funds also have the long track records necessary to draw meaningful conclusions.
To answer the first question, we used the same methodology as the one used in the Morningstar Active/Passive Barometer study, with some necessary adjustments.
Morningstar’s Active/Passive Barometer is a semiannual report that measures the performance of U.S. active managers against their passive peers within their respective Morningstar categories. While the U.S. study measures active managers’ success relative to the actual, net-of-fee performance of passive funds, we used the performance of appropriate Morningstar India Indexes* net of expenses across different categories, owing to dearth of passive funds across Indian fund categories.
[*Morningstar India Index targets approximately 97% of Indian equity market by market capitalization across large- (top 70%), mid- (20%), and small-cap (7%) segments.]
The below two tables show the survivorship and success rates of Indian active open-end equity funds for different holding periods by their beginning-of-period category as classified by Morningstar.
Survivorship rate indicates the percentage of active funds that started the period and survived till the end while the success rate indicates the percentage of active funds that survived as well as generated a return greater than their passive benchmarks.
The success rates of Indian active funds were generally high and even better over the long run, compared with U.S. active managers. The success rates of Indian large-cap active funds over 1- and 10-year periods were 80% and 73%, respectively, as of June 2016. The peak success rate across U.S. large-cap equity categories over the same time horizons were 29.8% (for U.S. large growth) and 20.6% (U.S. large value). The long-run success rate differential is similar in the small/mid-cap category—61.4% versus 26.9%, which indicates managers were patient for their investment calls to play out and had some confidence in investors sticking with them.
The survivorship rate of funds has remained relatively high over all time horizons.
* Flexicap was a new category introduced in 2014. At that time, many large- and small/mid-cap funds were reclassified to flexicap, hence resulting in a lower number of funds in June 2015 compared to 2012.
** The benchmark for large- and small/mid-cap funds was the Morningstar India Large Cap Index and Morningstar India Smid Cap Index, respectively. The passive vehicles based on the top 50-stock Nifty index and 30-stock Sensex index carried a fee of 1% (open-end index funds) or 40 basis points (ETFs). We subtracted 100 basis points from the benchmark performance for both categories for simplicity.
Data as of June 2016.
The most intuitive explanation for this outperformance is that Indian active managers are taking a structural bet against the market. But if the manager loads up on well-known return drivers like size, value, or momentum, it may not give an accurate representation of manager skill. A true indication of manager skill is alpha, which is a measure of an investment’s performance adjusted for these well-known drivers of return.
To ascribe this alpha, we ran a regression analysis with the following factors: market risk, size premium, value premium, and momentum. The market-risk premium--also called beta--captures the return differences between the market portfolio and the 91-day Treasury bill rate in India. The size premium captures the differences in returns between small-cap and large-cap stocks. The value premium captures the return differences between high book-value and low book-value stocks relative to price. The momentum premium reflects the subsequent return differences between the best- and worst-performing stocks in the past 12 months, excluding the latest month.
A beta higher than 1 would suggest fund managers take higher risk than the market. A positive coefficient on other factors would suggest the fund has greater exposure to that factor. The market portfolio chosen for the analysis was the Morningstar India Index. We then regressed the return differences between the fund and risk-free rates against the factors to calculate the alpha.
The above graph demonstrates the proportion of funds by the statistical significance of alpha. Alpha t-stat measures how significant the value of alpha is in a regression model output. The higher the absolute value of t-stat, the higher the probability that observed alpha is due to stock selection rather than some random factors. Additionally, the sign of alpha t-stat indicates whether the alpha is positive or negative. For example, 4.73% of large-cap funds have an alpha t-stat less than negative 2, which means their alpha is negative and statistically significant.
The graph also suggests that even after adjusting for all factors there is still some outperformance of active funds that cannot be attributed to known factors. Also, this outperformance is more prominent in the small/mid-cap fund universe where fund managers have a wider pool of companies to choose from with little or no analyst coverage. Overall, this confirms that stock-selection skill has been a more determining factor for the outperformance than systematic factors.
That active funds in India have consistently generated sizeable alpha in the face of their foreign counterparts’ struggles is truly impressive. Considering only past returns, however, is a retrospective assessment of what-could-have-been. From a prospective investor’s viewpoint, it is also equally important, if not more, to analyze whether the performance of mutual funds has been persistent over time. Investors want to be able to gauge the sustainability of the factors driving the outperformance. In essence, they want to unravel the what-is-to-come.
In order to explore the persistence question, we replicated the methodology used by the Morningstar analysts Alex Bryan and James Li in their 2016 study “Performance Persistence of US Mutual Funds.” We then divided the funds in each category into quartiles based on their returns in the past period and computed the percentage of funds that posted above category-median returns in the subsequent investment period and called this the quartile’s success rate. (We used quartiles instead of quintiles used in the original study owing to insufficient funds in initial periods.)
For example, if there are 10 funds in a category quartile at the onset of the sample period and seven went on to post above-median returns, the success rate for that quartile would be 70%.
Finally, we analyzed the differences in the success rates of the first and fourth quartiles in both categories and reported the average of success rate differentials across rolling one-year periods to eliminate any time-period bias. We performed our analyses for one-, three-, and five-year holding periods (with one-, three-, and five-year lookback periods, respectively, for quartile division) to test for persistence across investment horizons and adjusted our findings for survivorship bias. The data we’ve used to calculate and compile our findings spans a 14-year time frame, from December 2001 to December 2015.
The results of the study, as illustrated in the table below, indicate that top-quartile funds manage to maintain a significant success rate differential over bottom-quartile funds across categories and across the holdings periods considered in the study. The one exception is small/mid-cap funds over a five-year horizon, which produced an average success rate differential of negative 14.8%.
Another compelling discovery in our study was the widening of the difference between top- and bottom-quartile success rates as we moved from shorter to longer holding periods within the large-cap fund universe. The success rates had widened from 13% to 19% when we moved from a one-year to a five-year time horizon. The opposite occurred in the U.S. study. There, the success rate difference for large caps was 6% to 7% over one year. That result shrunk to negative 1% to negative 3% over the five-year horizon.
The findings of our persistence study, in many ways, corroborate the conclusions presented by our analysis of factor-adjusted alphas. Both analyses indicate the presence of manager skill and acknowledge its contribution toward fund performance.
The largely positive success-rate differentials across categories and investment horizons, for example, are evidence that managers in the top quartile, by virtue of the skill advantage they possess, perform better than the relatively unskilled bottom-quartile managers and that they persist in doing so.
Interesting questions remain
A widely debated topic is whether active investing makes more sense for investors to attempt in less-efficient markets such as India. It would be interesting to see how active funds in other emerging markets fare using the same framework we used here.
Regardless of the fact whether markets are efficient or inefficient, outperformance is a zero-sum game, gross of fees. If Indian equity managers have outperformed their benchmarks, there must be investors who have posted below-benchmark returns. According to data published in the 2016 midyear report of Morningstar’s Active/Passive Barometer, the success rates of U.S.-based diversified emerging-markets stock funds have remained less than 50% over a 10-year holding period. This steers us to the idea that perhaps successful active investing is a local phenomenon, where domestic fund managers stand in a better position to exploit inefficiencies of the market. In addition, domestic fund managers have the flexibility to hold stocks that may not be accessible to a foreign equity investor. One example of this in India is HDFC Bank. The stock is in most Indian large-cap funds, but it cannot be held directly by a foreign investor or fund manager because of foreign ownership restrictions. (This stock is available in the United States as an ADR.)
It would be interesting to compare how local managers fare against their offshore peers in India and across emerging markets. While we looked at style factors, another natural starting point would be a holdings-based attribution analysis through a sector lens: What is the relative importance of sector selection vis-a-vis stock selection when it comes to explaining performance? These are some of the areas for future research.