May 5th 2008. Unitech Limited hits a high of 337.80 Rs on BSE. Less than 6 months later, on October 24th, Unitech hits a low of 26.5 Rs, falling 92.1%. A week later it rises by roughly the same percentage points to Rs 48.
It makes you wonder, did I miss a chance to make a quick buck? The same story is repeated in innumerable counters, in markets all across the world, and it becomes hard to optimize the risk-reward equation. Should you buy risky stocks that may rise 2-3 times or buy quality stocks even though you know they haven’t fallen significantly from their highs and aren’t expected to shoot up like a rocket?

Let me ask you a question. Suppose I give you a bet in which, if you win, you can make 20% profit. You have 75% chance of winning this bet but if you lose you will have to pay me double the returns you would have got (40%). You can calculate that this bet has a positive expected value for you. If we play this bet every year for 50 years, you will almost surely win. You must take this bet.

If you didn’t, please revise your high school arithmetic. If you did take this bet, let me, for a moment, try to wean you away from this winning bet. I give you four more options from which you can pick. In each option if you lose the bet, you lose twice the percentage returns (n% in you win, -2n% if you lose). Once you pick your bet, we will play this once every year for 50 years. These options are explained below.

 X A B C E Returns on success 30% 20% 15% 12% 9% Probability of success 66% 75% 80% 90% 99%

Going only by your instincts (remember the power of compounding!), would you keep the 20% option or change it to any other bet? If yes, which one?

For those of you who didn’t change their mind or didn’t change it to the most conservative option, it would help if you were to reassess the value of certainty or near certainty. The expected annual compounded returns are highest in bet E (8.7%), lowest in X (-12.9%) and almost negligible in A (0.9%).

Value investing is not about finding the stocks that can give you (say) 10 times returns. It is about finding the stocks which are almost certain to give above average returns. Value investors should not attempt to beat the market averages but they should focus on maximizing the certainty of good returns. The term ‘good returns’ means returns above the risk free rate of returns (from government securities, FDs, etc.). The concept of margin of safety comes from this simple mathematics.

The quest for margin of safety does not kill returns as you would expect. It does just the opposite. Your efforts to ensure the certainty of good returns limit your investment horizon to businesses which are equally certain to generate good returns due to their competitive advantage, brand value or intellectual property.

At the same time, the more certain you are, the more concentrated your portfolio will be. Diversification is not a panacea against the risk of losses. Over diversification is a bigger evil than low diversification. I’ve seen many people holding 5-10 mutual funds (and sometimes a fund of funds). Each mutual fund is diversified into 50 to 200 stocks. So you can be absolutely sure to have funded every Tom, Dick and Harry who happens to bring out an IPO. Warren Buffett was dead right when he said he would almost certainly improve your investing record if he gave you a punch card and each time you buy a new company, your card gets punched. After 20 punches, you're finished. No more stocks, ever.

Twenty stocks in a lifetime? Some day traders will say “Gee, I bought 20 last week!”

Let me give you a stock to consider (Don’t rush to buy!).
Manaksia Ltd, Price 34 Rs, EPS 15.7 Rs, Book Value 89.7 Rs, Net Current Assets per share 46 Rs, Dividend per share Rs 2, Debt to Equity 0.43, P/E 2.16, 52 week high 212 Rs, low 26 Rs.

Howsoever attractive it may look, can you be certain about a company whose name you just heard today? Can you waste one punch hole from your punch card on this? On the other hand, a leading company selling below fair value can be an investment where you can put all your investment and rest easy. In fact most of the promoters have all of their wealth concentrated in very few companies. In the last 17 years, I haven’t seen the networth of many large business houses going down in the long term but I’ve sure seen many individual investors permanently losing their money by investing in a bunch of well diversified mutual funds and booking losses when it is too late.

I’m emphasizing this point because in the current environment there is enough junk selling at bargain prices. Stay miles away from temptation. Buy the companies that are certain to remain in business 10 year hence, that are certain to earn decent returns on their capital investments, that are certain to keep their competitive advantage intact and thereby grow at least as much as the economy. All of these returns mean nothing if the management lacks integrity. Invest only in those companies who care about their shareholders and run their business in the most ethical manner. Finally, the financial figures mean nothing if the company shows more creativity in accounting than in their products. So it pays to be certain that the numbers reflect the actual state of the business.

Rationality is not as intuitive as you may think. It’s a good idea to learn from others’ mistakes. It’s not bad idea to lose few dollars on gambling if that helps you in learning to deal with probabilities and decision making under uncertain information. To learn by investing your hard earned money in stock markets is certainly suicidal.

To quote the Will Rogers “There are three kinds of men: The ones that learn by reading; The few who learn by observation; The rest of them have to pee on the electric fence and find out for themselves”

Fortunately you have the option to choose your kind.

Anonymous said...

Wow some article. After reading this and the last few articles I am really getting addicted and almost cant wait for the next one, just as i am addicted to Ajit Dayal's (equitymaster) articles.
The common thing between both of you is you speak truth and are honest. I am yet to come across articles/websites which speak the similar language. Cheers!

Anonymous said...

As per my calculations, the expected return in case X is -0.6%, in case A is 5%, in case B is 6%, in case C is 8.4% and in case E is 8.7%. What am I doing wrong? (Example: Case A-> 75%*120 + 25%*60 = 105)

The rest of the post is strong even with the above calculations but the wrong calculations can bias one's understanding more strongly.

Kamlesh Pandey said...

You are not taking into account the effect of compounding.

years N=50
returns r=0.2 (20%)
probability of success p = 0.75 (75%)

For a large number N, you will succeed N*p times and fail n*(1-p) times

EndingValue = StartingValue * power(1+r,N * p)*power(1-2r,N*(1-p))

CAGR(%) = 100 * (power(power(1+r,N * p)*power(1-2r,N*(1-p)),1/N)-1)

So expected value does not give indication of the gains. A coin toss with 50% gain on success and 50% loss on failure has expected value zero but its a losing bet if played repeatedly.

The reason is
(1+r)*(1-r) = 1 - r*r

If you lose 50% of your investment this year, you won't be able to recover it by making 50% gains next year on the reduced capital.