Market behaviour -- No two crashes are the same

B. Venkatesh

Anup Menon

IT WOULD be great if one could sell stocks before a market crash and buy them back when prices are low. But that is not given to most among the community of retail investors.

Should one, then, sell -- at least in the immediate aftermath of a crash? As it happens, you could do worse than holding on to your nerves and riding the storm.

At least, history suggests this might well be a superior course of action when there is gloom around. Remember May 26, 1999, the day the Kargil War started? The market tanked that day, as equity values fell over 2 per cent. You might have ended up with losses on that day if you held stocks such as State Bank of India, Satyam Computer Services and ITC in your portfolio.

Satyam, for instance, lost 4 per cent that day, against the broad market loss of 2 per cent. The stock, however, rebounded three months hence and returned 18 per cent during this period.

State Bank of India and others are examples of fundamentally sound stocks with inherent resilience. An analysis of the behaviour of a broad portfolio of stocks (BSE Sensex) in the market in the wake of a steep fall in values suggests that investors may be better off holding their investments during a crisis period.

The reason? Each time the market tanks due to a crisis, it bounces back into the positive territory after some time. The time it takes for the market to bounce back is, of course, not the same for all the events observed.

Our results actually make a virtue of the typical buy-and-hold strategy. In `behavioural finance' parlance, this means that loss-aversion is a good trait to have during crashes; this theory proposes that investors do not sell when they are faced with a loss-making investment. Rather, they wait for the market to rebound and then take profits from their investment, even if it means waiting longer.

Observations

To observe the market behaviour, we chose 10 significant events that led to crashes. In doing so, we considered the probability of a similar event occurring in the future. This was done to ensure that the results could be used to structure portfolio strategies to protect losses in the likely event of a similar crash in the future.

While we agree that no two events generate similar information effects on the market, we believe that, in general, the impact will be negative. What could be different is the flow and quantum of information which, in turn, reflect the extent and time horizon of the crash.

The chosen events included both domestic and international episodes. We chose events since 1990 to ensure that our sample included crashes that occurred both in bull and bear phases. Otherwise, a crash occurring in the midst of a bull phase may not have a large negative impact on equity values as it would have if the crash had occurred during a bear phase.

Our analysis is based on daily returns on the Bombay Stock Exchange Sensitive Index (Sensex). The choice of the index poses no problems, as it is the most actively followed index and has a fairly long history. We computed log returns for the Sensex post-crash for the following periods: one week, one month, three months, six months and one year. This exercise was conducted to find out how long it took the market to bounce back from a crash.

The following are the results of our observations:

a) Immediately after the crash, the market fell sharply and the negative returns continued for over a week. This was true for all the crashes that formed part of the sample.

b) The negative returns in the market continued for a month in 60 per cent of the cases. It was also noted that such negative returns persisted for all events having a negative international implication.

c) The negative returns continued into the third month in 60 per cent of the cases

d) The losses, however, persisted only in 20 per cent of the cases when a one-year horizon was considered.

Implications

Do our results make sense? Why do stock markets crash after a negative event only to bounce back later? We probed this question and found support from the research in behavioural finance.

Equity prices are a function of the demand-supply for stocks in the market. Investors panic when a negative event occurs. This increases the supply of stocks in the market in relation to their demand, pulling down equity values.

Consider the Pokhran nuclear tests of May 1998. The blasts was perceived as incurring the wrath of the US and other developed countries, which were against nuclear tests. It was felt that these countries would restrict Indian exports into their territories.

Click here for Chart

In the event, the Indian economy would be badly hurt. Besides lower foreign exchange and higher trade deficit, the revenues of companies dependent on exports would be hit. This, in turn, would mean lower post-tax earnings -- or so was the perception. This led many investors to sell their holdings, precipitating the crash that followed the Pokhran tests.

But why does the market bounce back after a crash? Behavioural finance suggests that investors tend to consistently overreact to negative events, which leads to equity values plunging to new lows. For instance, consider Infosys Technologies during the Nasdaq crash of April 17, 2000.

The negative sentiment rubbed off on Infosys as the stock crashed Rs 664 to Rs 7857.50 on April 18. The stock continued to tumble for the next five trading days and touched a low of Rs 7296 on April 25, 2001. Ten days later, on May 5, 2001, the stock was back again to the Rs 8,500 levels.

After the market falls, there comes a point when investors indulge in bargain-hunting, perceiving value in the stocks at lower prices. It is purely based on human instinct of buying low and selling high. This is when the market rebounds and moves into positive territory. The period it takes for the market to rebound is, however, a function of the market sentiment which, in turn, rests on the perception of the implication of the negative event.

In the case of Pokhran blast, for instance, it took nearly a year for the market to move into positive territory, whereas it took just a month for the market to regain losses after the Kargil War in May 1999.

Return-enhancing strategies

Our results do not imply that the best strategy during crashes is to hold on to your portfolio. Such a strategy would help minimise your losses and is, therefore, recommended if you are risk-averse.

As for those who do not mind taking risks, it may help if you sell your high 'beta' stocks and buy them back at a lower price at a later date; 'beta' is a measure that captures the volatility relationship that the stock has with the market index -- a high beta say, 1.5 -- means that the stock moves 1.5 points for every one point move in the market index.

Institutional investors and high net-worth investors may also consider taking short positions in Index futures and single-stock futures, when it is introduced, to minimise their short-term losses during market crashes.

Our results suggest that investors are better off holding their portfolio when the market crashes. Having said that, we also need to state the following, lest our conclusion is misunderstood. No two crashes are the same. So, it is possible that the market may not behave the way it did in the past, meaning it may take a longer time to bounce back than it has in the events that we observed.

As is always said of the market, the past is not always a surrogate for the future. Perhaps, that is why crashes occur.