July 18, 2012

Buy Low, Sell High, But How?

This article was published in The Global Analyst

The greatest investment challenge for any investor is to buy low and sell high. However,. many a times we end up buying high and selling low!
While historical returns can look appealing, the exact timing of your investment can create huge divergence between your returns and historical returns. If your timing is wrong, then a Buy and Hold Strategy can backfire heavily. This is best explained through this chart that shows annual performance of Sensex since 2004. In these 8 years, we had 6 positive years and 2 negative years. If you would have invested Rs.100 at the beginning of 2005, it would be worth Rs. 234 at the end of 2011. On the other hand, if you would have invested Rs.100 at the beginning of 2008, your investment would be worth only Rs.76 now. So, a passive investment approach (what is called Buy and Hold) may yield results based on your market timing.
On the other hand, if you are an active investor trading in the market day in and day out (what is called as high frequency trading), it mostly will be a negative sum game that benefits only the brokers to earn their commissions. Also, it involves heighted level of tension on a daily basis leading to acidity and associated habits.
This article is to present a strategy that can time the market without emotions but based on carefully evaluated rules (through an intense study of the past). I believe such a strategy can generate ‘above ordinary” returns and can live up to our initial expectation of “Buy Low Sell High” ambition. However, it should be noted that there is NO investment strategy that beats a Buy and Hold strategy 100% of the time. It depends on the level of entry and further market movement.
 Before we aspire to “Buy Low, Sell High”, there are some insights from the market that are worth exploring.
1.       All markets frequently go “underwater” and stay there for longer than what you can anticipate.  For eg., if you invested at 100, and your investment value goes down to say 90, technically your investments are termed “under the water”. The problem is, both the depth and the period of time of your investments being under the water cannot be predicted. The Japanese stock market is a great example. The Nikkei Index reached its highest level of 38,915 on 29th December 1989. The current level is 8,724 (15th July 2012), after 23 years!
2.       Do we have such episodes in the Indian market? For eg., if you would have invested on 14th January 2004 when Sensex was at 6,194, you would have gone under the water immediately and you had to wait till 30th November 2004 before you could level your original investment[1]. You had to wait for 10 agonizing months during which your investments could have dropped by 24% when Sensex reached 4,708 (24th June 2004). Any investment made around November 2010 is technically still “under the water”. The probability of something going under the water is high during the peak of  bull markets. Once our investments are “under the water”, we take time to accept the fact during which time it goes even deep under the water. For eg., , if you would have invested on Jan-08 when Sensex was at 20,873, you would be under the water till 4th November 2010 to reach the same level, a staggering period of 34 months(nearly 3 years) during which time your investment value could have gone down by 60% before it recovered. Simply put, you would be staring at a loss of 60% which can be psychologically nerve wrecking. The lowest point during this period was 9th March 2009, when Sensex hit 8,160. Like how Sensex can zoom from 8,000 levels to 20,000 levels, it can also crash from 20,000 levels to 8,000 levels.
3.       When markets go under the water, we often ride it down fully rather than limit our loss by cashing out at some stage.
4.       Once our investments are “above the water”, our expectation about future profits is unlimited!
5.       Most of our buy/sell points are either determined emotionally or arbitrarily than a careful evaluation of our investment environment. Psychology plays a major role in our market timing. After we make an investment, and if we move into profit zone, we attribute that to our “skill” and not “luck”. However, when we move into loss, first we are in “denial” mode (how can I make a bad investment decision?) and then when we are deep in the red, we “resign” to our fate and do nothing. Again psychology!
The Strategy
This strategy aims to Buy low and Sell High. The strategy is explained through a series of questions:
When to Buy?
The basic idea is to invest only when market is in a “new low” as determined by a reference to a previous high. For eg.,



[1] For the sake of simplicity, I am assuming that you invest in a ETF like structure that replicates the broad market.
On 15th April 2009, Sensex reached 11,284 a new high. You should index this as 100 and keep referencing further moves to this 100. Subsequent to this, the index offered a “new low” on 16th April at 10,947. Then offered another “new low” on 21st April when Sensex reached 10,898 and offered another “new low” on 22nd April when Sensex reached 10,817.  As per the strategy, we will invest at all such points of “new lows”.
Remember, markets can keep going down for a considerable period of time which means that it will offer you many “new lows”. You should then be prepared to invest in every such “new lows”.
How much to invest?
We should start with a defined instalment amount of say Rs.10,000 and increase the investment amount as we encounter further “new lows” tracked by the indexing explained above. The following rules can help:
As you may see, the amount to be invested is based on spotting a new low and also the level where it is spotted. Hence, it is variable and cannot be determined upfront. All you can do is to have some amount locked up in a money market fund to be utilized as and when the model screams a “buy”. As the table indicates, you may end up investing Rs.15 lakhs during the last 3 years, Rs. 24 lakhs during the last 5 year or Rs. 44 lakhs during the last 8.5 years. But there is no way you will know that in advance since you cannot predict how long the markets will fall.
What is the frequency?
By virtue of the strategy, even this cannot be determined upfront. You may have a situation where you will find the need to invest almost on a daily basis or there may be periods where you will not invest for years.
When to Sell?
You will sell when the market reaches back a high from where you started (i.e., 100).  Continuing on our earlier example, while we would have bought on 16th, 21st and 22nd April because it reached “new lows” on these days, we will sell our position on 24th April when the index reached 11,329 and surpassed earlier new high of 11,284 and reached a “new high”.
How much to sell?
We will sell all the investments that happened during this period at that day’s value.
How intensive is the strategy in terms of implementation?
The strategy involves buying and selling whenever the model screams so. However, we will not know in advance when it will scream. Going by the number of buy and sell activities, we can classify the findings as follows:
The % number of days on which the strategy will force you to act is at the peak 18.5%. In other words, we will have some trading action only on 18.5% of the time while it is 100% if you are a day trader. Hence, it is not too much of a call to act.
How did the Strategy do?
I have tested the concept for several time periods (last 3 years, 5 years, 8 years) and find that the strategy outperforms the buy and hold strategy in the 5 years and 8 years period significantly while it slightly under performs the buy and hold in the 3 years category. Hence, it is worth the time and effort. However, it should be noted that this strategy  might underperform a Buy and Hold strategy in an upward trending bull market.
When does the strategy perform the best?
When the market goes one full cycle i.e., it comes back to a point where it was before. Technically it is called “Peak-Trough-Peak” (PTP). For eg., as stated before the Sensex reached a peak on 9th Jan 2008 at 20,869 and thereafter fell continuously till 9th March 2009 to reach 8,160 after which it started climbing back slowly and reached its earlier peak of 20,869 on 4th November 2010. In other words, if you would have invested in Sensex on 9th Jan 2008 and exited on 4th November 2010 (after app 3 years) you would have just got back your money implying nil returns. However, our strategy would have screamed a buy 35 times all through the fall from Jan 2008 to March 2009. As per the investment band, you would have invested a total of Rs. 13.46 lakhs in this period through 35 buy actions with the last buy of Rs.60,000 on 9th March 2009 when the Sensex was at 8,160. Remember, as per the band, you start off modestly at Rs.10,000 investment and increase your investment value as the market braces new lows. Since after 9th March 2009, the Sensex did not encounter any new low, there was no buy call till 4th November 2010 when Sensex touched a new high and the model told you to sell. The sale value would have fetched Rs. 24.4 lakhs implying an IRR of 30.6% against a 0% return on the buy and hold.  Look how active the strategy becomes when the market falls and how inactive the strategy is when the market is rising!
So, what is the difference in terms of investment style?
Concluding Thoughts
As Warren Buffet said (which most of us do not follow), be adventurous when others are fearful and be fearful when others are adventurous. When you follow that advice, you can buy low and sell high. The idea of investing a pre-defined amount at a pre-defined time will expose your investment to market timing risk. On the other hand, the idea of investing when markets are in a free fall and cashing out when markets are in a free rise will make sure that you don’t encounter capital loss. The idea espoused above is just that. Given our limited ability to deal with losses psychologically, I feel such a strategy can be of great help. All it needs is some basic excel skills which all of us possess these days thanks to Microsoft!
Best of luck!
PS: The Author thanks Mr. Madhusoodhanan for data analysis