Diversification and Asset Allocation in the Indian Market

Rohit Sharma, in his comment to my post on diversification, did not agree with my rationale.

HDFC prudence, not being an index fund, would be investing in much more risky stocks.

He goes on to add:

One must take into consideration the Risk involved in the so called 'diversification'. HDFC prudence has given better results by taking Higher Risks. I don’t think there is any thing novel in that. It is almost a Tautology.

And finally:

'Diversification', as you define in this article’, seems to be inconsistent with 'Diversification' that the investing world generally believes in i.e. picking up funds to minimize risks and not maximize profits.

Before I continue further, let me say that my intent for the previous post was to look at diversification across various instruments and not just stocks. Therefore, my choice of HDFC Prudence.

I generally agree with his rationale but have some reservations. First, HDFC Prudence is a hybrid fund. Therefore, by definition, a portion of its portfolio is invested in Debt / Bond markets which are inherently less risky than stock. Therefore, the fund could have some of the stocks which were more risky than Sensex but overall its portfolio would be less risky. This is somewhat indicated by the fact that it had very little down years.

Second, I do agree with the fact that diversification is generally done to avoid risk. However, unlike him, I believe diversification can also give rise to higher returns (for the same risk taken). This can happen when the portfolio is made up of negatively co-related instruments i.e. when one goes up the other goes down. This is what the efficient frontier in portfolio theory is all about. The following excerpt from moneychimp explains it clearly

The second important property of the efficient frontier is that it's curved, not straight. This is actually significant -- in fact, it's the key to how diversification lets you improve your reward-to-risk ratio. To see why, imagine a 50/50 allocation between just two securities. Assuming that the year-to-year performance of these two securities is not perfectly in sync -- that is, assuming that the great years and the lousy years for Security 1 don't correspond perfectly to the great years and lousy years for Security 2, but that their cycles are at least a little off -- then the standard deviation of the 50/50 allocation will be less than the average of the standard deviations of the two securities separately. Graphically, this stretches the possible allocations to the left of the straight line joining the two securities.

All said, I concur that lower risk and not higher returns is the primary reason for diversification. Also, practically, it is very difficult to find negatively co-related securities.

I agree that looking only at HDFC Prudence is not the correct way to look at the virtues of diversification. The results confirm that for higher returns you need to take higher risk. But then they also give an historical indication of how the results in India have been.

However, what I intend to do in this post (as in previous post) is try to look at diversification in different instruments and not just diversification in stocks.

I will choose a simple portfolio made up of Sensex and Bank Deposits. I will invest a fixed amount of Rs 1000 at the beginning of every year (no market timing). I will consider 5 different portfolio's of 100%, 70%, 50%, 30% and 0% stocks. For bank deposits I will consider 1 year returns. The results are:

The % age returns for each year indicate by how much the total investment grew (shrunk) during that 1 year period.

The results are not surprising but give an indication of the historical returns of different portfolios. The risk (indicated by StdDev) and the return (indicated by Average) decreases as the %age of stock decreases.

We have to take the average stock market return with a pinch of salt. StdErr is a statistical figure that indicates how close my sample Average is to the real world i.e. by how much will the average vary if I take a different sample. Higher standard error indicates that Average (i.e. returns) may not remain the same in the future.

Now let's look at the results when I balance my portfolio at the beginning of every year i.e. while instead of dividing Rs 1000 every year in stocks and deposits. I add Rs 1000 to the previous year returns and divide the whole amount into stocks and deposits. The results are:

Again, we have pretty much similar results. One important observation is the minimum return in all cases is much less than what it would be without balancing. Finally, have a look at the sheet to find the actual growth in the Rs value.

Even though the diversification might reduce your overall returns it makes getting through the bad years a little easier. This makes a difference between your continuing to invest or stopping it and thereby missing out on the best years.

Have a look at the Minimum Returns. Pick an asset allocation that matches with your risk profile. If you can withstand 20% loss in any given year (and still continue investing) then 100% diversified stocks is your ideal asset allocation. But if you cannot then you should diversify not only in stocks but in other instruments.

Comments

[Gaurav:]

This is good work. I think that you should also consider holding period in your calculation. i.e. calculate 15 year rolling returns every month for each category you have presented (if you can get data going back another 10-15 years). Then calculate the average returns and std deviation. I will bet you that 100% stocks has highest returns and the lowest std deviation. For your very long-term holdings, stocks are much better investments than any other investment, as they offer you the lowest risk - lower even than bonds - as well as the highest returns.

2007 Apr 20 10:24 AM