Asset allocation is the most important decision in your portfolio management. It is almost impossible to get what is called an ‘optimal rebalancing’. And if you have a fairly sophisticated portfolio, 2-3 items of 7-8 HAVE TO BE IN THE NEGATIVE. If that is not happening, you are not diversified enough.

Rebalancing too often increases transaction costs – more in terms of taxation – we do not have entry exit loads in India. With so many years of investing experience I am still not sure about reinvesting and asset allocation. I realize that if there is no asset allocation between a volatile asset and a non volatile asset, the returns may not be commensurate with the RISK TAKEN. So by doing asset allocation, you may not increase returns but you will surely reduce risk. If asset allocation is not done higher-risk investments that have higher long-term returns would become over-weighted by out-compounding the lower-risk lower-return positions in the portfolio. So a portfolio without reallocating could go from a 65 equity and 35 debt to 85:15 and may have also given a sub par performance.

I prefer a different re-balancing methodology is to allow portfolio allocations to go a little ahead of the ‘planned’ ratio – just let it drift slightly. I prefer triggering a trade only if a target threshold is reached.  At that time if the client has increased his own provident fund or Public provident fund in a big way, I may even postpone the trigger. Also at a slightly higher TOTAL ASSETS, I may allow the drift to be more favoring equity. If the investments grow in line and the relative weightings don’t change, no re-balancing  trade is triggered. However, if these “re-balancing tolerance bands” are breached, or the client gets jittery – when the amounts go up some people get MORE VOLATILITY AVERSE – the investment –  that crosses the line, is then bought or sold  to keep it in line.

The one important caution to the process of tolerance band re-balancing is that it requires some active monitoring of the portfolio itself, to ensure that you realize a threshold has been reached. Fortunately, though, electronic record keeping and software with triggers are available and they suggest the trade too. I prefer doing the transaction myself..with the client’s consent.

In the short term, re-balancing presents a chance to generate better returns – REMEMBER IT IS RISKY – particularly when re-balancing among investments with similar long-term return potential –as the process of selling what has become overweight and buying is underweight can TECHNICALLY help investors to sell high and buy low. The example is say shifting from mid cap to large cap because mid cap has gone up too fast. The worry is if the client says NO, and you force him too, you can look foolish after a couple of months – you may be right after 5 months, but your client would have felt that you re-balanced too early!!! So some triggers should be NON NEGOTIABLE. Tough, especially if you are a young IFA or a young (new to investing) client stuck with a stubborn IFA

The goal is to find investments that have moved to new highs (or lows) and trigger a re-balancing trade that buys/sell the investment just before it “comes back” and reverts towards its long-term average!! Equity share portfolio re-balancing is very difficult to manage. I did not sell Cholamandalm during its journey from 65 to 1014, and I look smart. If I had not sold GMR and Crest Animation when it fell I would have looked like an idiot.

So should I re-balance at the top (Equity vs Debt) or should i also re-balance groups (Murugappa group concentration is a serious worry in my wife’s portfolio), Midcap: Large cap: Micro cap…are questions which big portfolio managers grapple with.

I am doing a session for some managers on all these topics. I hope they do not kill me with such questions. Ultimately it is difficult to answer…many of these go with the flow….



Related Articles:

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

  1. Sir, can this re-balancing exercise be function of market sentiments as measured by Index valuations? For example, say when Nifty TTM P/E is about 24, indicating generally high bullish sentiments, this could be time for rebalancing and move more to fixed income, may be to go little underweight on equity. Same applies in reverse order as well. Say when Nifty TTM P/E is about 15, that could be time to go overweight equity. Your views, please.

  2. why 15 and 24? why not 16 and 26? large cap or small cap? what if large cap runs ahead of mid cap? or will u stick to a flexicap fund and let the fund manager decide the allocation between mid and large cap? Just asking 🙂

  3. Sir, nothing sacrosanct abut 15 or 24. Close to these numbers, index valuations are near to two standard deviations event, hence, adequately out of step with normal for triggering some rebalancing especially for deciding overweight or underweight of equity. Besides, this happens infrequently, every few years, hence, allows enough passive period.

    For the sake of simplicity, I am differentiating between only equity and fixed income and not within equity sub-classes because, at these levels, assumption is that all sub-classes of equity will be more or less at same level of over or under valuations.

    Of course, this is a concept. Fine tuning can be done if person has adequate time, knowledge, experience and temperament.

  4. Sir, how often we should check overlap across MF we hold? How to use that data across rebalancing? We might have used overlap check (pattu’s xls) during fund selection.

  5. in the west, an inverted yield curve (interim interest rates higher than long term rates ) typically suggests going against the flow & moving from equity to debt to safeguard capital, as a recession is surely in the making. does something like that hold good in india statistically?

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>