Every body and his uncle, aunt and cousin likes to talk about Fund manager evaluation. It is extremely difficult to even assess past performance – some managers get lucky very often! More than being difficult, it is pretty useless.

Useless, yes you heard it right – because you cannot use the so called ANALYSIS to predict the future performance. Read everything about mutual funds with the following in mind:

PAST PERFORMANCE IS NOT AN INDICATOR OF FUTURE PERFORMANCE.

  • Ranking fund managers by 1, 3 and even 5 year’s performance is of very little use to predict future performance

Each fund manager has a different style. Look at Naren and look at Samir Arora. Chalk and Cheese did you say? Sunil Singhania and Balasubramaniam of Birla. Are they not fundamentally different? Even Madhu Kela vs Sunil Singhania – do we know how their stock picking was different? A manager who focuses on valuation (buys deep value) versus a manager who focuses on momentum (buys companies with growth ) will have different performance over similar periods of  time frames, but that is no indication that one manager is better than the other. It simply means that in that time frame one manager’s approach was ‘luckier’ or more liked during that period when the fund managers were being evaluated. YOU should have had the luck (skill?) to pick the ‘luckier’ fund managers. IN retrospect of course you will know “how Value would have worked” during that period of the market in the country in which you are operating.

Hence it is important to consider annualized, calendar year, rolling returns and of course CAGR in any performance evaluation. Doing this over long periods of time also tells you whether there is consistency  of approach. You can ONLY worry about the consistency of the process / approach, and really not about the result on a peer comparison basis. For example I am happy holding Franklin India Blue Chip, Franklin Prima  Plus, Hdfc Equity, and looking at their long term performance. Surely at different points in time DIFFERENT STYLES have been appreciated by the markets.

Annualized performance can be skewed by one significant return experience during the time period considered, whereas calendar year  or financial year performance may make it easier to see anomalies in the return flow and, help identify what drove those returns.  We can use that information to understand that manager versus the market conditions. This in theory should help us choose a manager WE BELIEVE will be in favor in the upcoming market conditions. Having said that, what if the fund manager’s mandate tells you NOTHING about his style. For e.g. Hdfc Equity has no ‘value’ or ‘growth’ style. Icici Pru Discovery has a value style. Franklin India Blue chip has ‘Growth at a Reasonable Price’ as its philosophy.

Now, determining what manager style could be in favor is not easy for sure. And the fact that we do not know whether the fund manager will be honest to his mandate / style. However, in a fund allocation task considered vital by the IFA,  HE  must accept that this is a much better approach than not considering it at all! A simpler approach is of course is to capture multiple styles – so holding Franklin Bluechip and Icici Prudential Value Discovery over say multiple years maybe a sensible approach. At least it has worked for me.

 

Fund manager performance vs. peers

The most common evaluation is to look at how a manager or fund did against his “peer” group over various periods of time. The argument around manager/fund style is just as common (more common did you say?) versus peers as it is against benchmarks. Annualized and financial or calendar year performance versus peers provides a more complete picture.

  1. Inaccurate peer group – Most of the times in publicly available databases the fund or manager is not classified properly. They use holdings-based screens to determine where a fund belongs, but they may miss a broader investment approach the manager employs. An example of this is a multi cap fund manager – but is largely exposed to mid-cap securities at the moment. HOWEVER what if the returns ACTUALLY came when he was more in LARGE CAP and the portfolio is different at the time of evaluation. What happens if he changes his portfolio 2 weeks after the evaluation? When the fund manager does not state his philosophy with clarity, it is difficult to put him in the correct peer group. Cannot blame the independent vendors – what is the basis on which will they classify? However remember the limitations.
  2. Broader peer group classifications – If  a peer group is extremely broad in nature that it puts funds/managers that have very different styles and throws them into one general bucket, does it really help? How will you compare Dynamic Bond fund managers? Past performance could be the WORST indicator of future performance!  When looking at the peer group, one mandate may look more favorable than the other, but it is difficult how the fund manager will recompose his portfolio. Like for example if I were a fund manager holding shares of Jindal Steel, I would have behaved differently as a fund manager of an Endowment policy (where you cannot see my portfolio) or a fund manager of a mutual fund (where every T, D and H can comment) I may have behaved differently. HOW WILL YOU as an analyst know how I will behave? For example FT took the loss on its chin, but I Pru waited and collected the whole amount. Who knws which is a sensible thing to do?
  3. Differences on how the product is being managed – This one is almost impossible to pin point. It is driven by how a manager/fund approaches the whole business, and how much the CIO influences the final decisions. Will a Ramdeo Agarwal or a Sameer Aurora blossom under a Naren? Take bond funds – a high-yield bond and small companies debt are different and specific asset classes, but can have a large variety of return differentiation. Is it fair to compare the managers? Also ‘high-yield’ may require some cross boarder deals. Will we be able to distinguish fund manager performance? What about Currency fluctuations? What if the portfolio had actually done well, but the currency fluctuation ruined the fund performance results?
  4. Getting carried away by statistics and star ratings: The human mind likes to ‘simplify’ everything to a single number hence we love ratings. A 5* rating for a fund does not mean it is a good buy – remember PAST PERFORMANCE…… !! When you see Alpha, Beta, Gamma, there is a tendency to say ‘wow..this must be right’. If you cannot understand the commentary regarding style, composition of portfolio, market conditions, etc. just the Greeks do not help. Also when you read all this along with standard deviation, reversion to mean, purity of style, etc. you really do not know whether to act or worse what act to do!!

So, next time you are evaluating a fund/manager, or reading a fund manager review,  don’t believe everything you think. Do not believe everything you hear or read. Some times, there are underlying aspects of asset classes, philosophies, fund manager changes, categorization and other influences. It is like equity markets themselves – there are just too many influences at play. Understanding these obstacles is the first step in overcoming them and choosing more appropriate options for ourselves. Now add to all this the noise and genuine feelings and need of  clients – fund picking is easier than fund retention and fund retention justification.

Phew. Sorry, long read, and still incomplete…

http://seicblogs.com/advisor-practice-management/investments/advisor-goals-based-investing/dont-believe-everything-you-think-common-pitfalls-in-fundmanager-evaluation/

 

Related Articles:

Post Footer automatically generated by Add Post Footer Plugin for wordpress.

  1. by the time “subra” sir completed this article
    somewhere some “fund manager” already “damaged” the fund and jumped to other fund!leaving investors to losses!
    so much “sucking” potential these “suckers” have!

    99% of people work hard to make rich people “richer”!
    that is why “gov ts” stopped printing “NOTES”!

    surendra

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes:

<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>