The Hindu Business Line : Diversification or Di-worse-ification?

Investors load their portfolios with large number of stocks or funds, believing this means better diversification. But this is not so. Also, Markowitz portfolio diversification is not optimal for retail investors as they are typically “returns maximisers” without defined investment horizon. The article suggests style-diversification and risk management rules as ways to minimise downside risk.

B. Venkatesh

Portfolio diversification is a much-touted virtue that is often misunderstood. Many investors hold 50-100 stocks in their diversified portfolio. The number ranges between 15 and 25 in mutual fund portfolios.

This article examines if diversification is meaningful for retail investors, given the paradigm shift in their investment behaviour since Harry Markowitz developed the mean-variance model. It appears that Markowitz diversification may not be optimal f or retail investors, for two reasons.

One, retail investors are “returns maximisers” unlike institutional investors who construct portfolios to match a target return or liability structure within a defined investment horizon. And two, a gradual contraction in investment horizon makes it difficult to construct Markowitz-efficient portfolio.

Given this, retail investors should be more concerned about risk management and style-diversification.

This means that investors need not stack their portfolios with 25 mutual funds or 100 stocks. Perhaps, five different styles of mutual funds or 10-15 stocks could be enough so long as appropriate risk management rules are in place.

Markowitz’ APPROACH

In a seminal paper that won him a Nobel Prize, Harry Markowitz suggested that investors could reduce their risk by constructing a portfolio of assets with low cross-correlations.

This would be possible if stocks in the portfolio were not largely affected by the same factors that could bring down their prices.

Such an optimal portfolio is constructed using factors such as expected returns, variability of returns and correlation among assets in the portfolio.

Markowitz’s Mean-Variance approach has since been modified to include regime switches or changes in the market structure. Studies have shown that assets become highly correlated during market crashes. A tail-adjusted mean-variance model accounts for the high correlation during such events.

Though institutional investors use a Markowitz-based model, it seems to have little relevance for retail investors. Institutional investors mostly construct portfolios to match their liability structure or target returns within a defined investment horizon. Often, retail investors are “returns maximisers” with no specific investment horizon.

It is difficult to construct a meaningful Markowitz-efficient portfolio in such returns-maximising world. The reason is that correlations and asset price volatility are time-varying.

That is, 5-year price volatility of an asset will different from its 10-year volatility as will its relationship with other assets. So, carving a diversification programme based on a 10-year relationship among assets may not be optimal if the retail investor later decides to fold the portfolio in just three years.

Besides, there has been a sharp contraction in the retail investors’ investment horizon. This has come off the need to continually trade on the portfolio, triggered by the volatile price movements in recent years.

Now, correlations among assets may not hold in the short term because of high short-term volatility. A diversified portfolio constructed on unstable relationships will not be optimal.

Style Diversification

Diversification is important because the future is uncertain. If an investor “knows” how an asset will move, she will place all money in that asset.

The problem is that retail investors do not engage in diversification as styled in the Markowitz model. Take mass-affluent investors whose asset size is Rs 50 lakh or less. They have 15-25 funds in their portfolio and engage in mutual fund diversification, not portfolio diversification.

These investors predominantly load their portfolios with growth funds. All growth funds follow similar investment styles- the portfolios are a mix of large-cap and mid-cap stocks. Having 10 growth funds would mean that the portfolio is over-exposed to the same universe of stocks.

This would subject the portfolio to high risk if the market tanks, for these growth funds will typically move in the same direction.

An optimal strategy would be to buy funds with varying investment styles. An investor should buy a large-cap fund and add a mid-cap fund as a diversification strategy. Sector funds also help in diversification, if that sector has low correlation with the rest of the portfolio. Investors could consider spreading their exposure among two or three funds within a style universe to satisfy their need for mutual fund diversification.

Retail investors who directly buy stocks suffer from the same problem as their mutual-fund counterparts. Having 50 per cent portfolio exposure in 25 different large-caps does not strictly diversify risks.

It would be instead optimal to buy 3 to 5 large-cap stocks and apply strict risk management rules to protect downside risk. An investor can, thus, have not more than 15-20 stocks in the portfolio, depending on the asset size.

It is important to construct a portfolio with profit-maximising objective and then to place risk management rules around it to minimise downside risk.

Investors should not delude themselves into believing that diversification by itself will prevent downside risk. It will not and January 21, 2008 is a case in point.

(The author is an investment strategist. He can be reached at enhancek@gmail.com)

More Stories on : Stock Markets | Investments | Micromotives

Article E-Mail :: Comment :: Syndication :: Printer Friendly Page

Stories in this Section
Madras Cements: Buy

Does the IIP offer clues to corporate growth?

Templeton India Equity Income Fund: Invest

DBS Chola Opportunities Fund: Invest

Graphite India: Buy

Amtek Auto: Buy

Index Outlook

Diversification or Di-worse-ification?