How to do equity research
I have a brilliant friend who has worked for many big and important media houses. He told me once – when he was heading a magazine – “I have made up my mind, please do not confuse me with the facts!”.
This was when he was doing a story on mutual funds and unit linked plans.
In life I have found many such stories. A friend used to work for Hindustan Unilever (once upon a time when it was actually called Hindustan Lever). A new entrant had launched a product called ‘Nirma’. As a research guy he had to prove how Surf was better than Nirma. Every day he would have to do some experiment. Once he got tired and said ‘let me PRINT the conclusion that “Surf is superior to Nirma” – it really does not matter what I write above that, does it?”
Ability to read a lot, get contrarian views, speaking to the clerks, workers, and old frustrated employees IS A MUST WHEN YOU are researching the details of an old company. In one such company we found out that the MD and Chairman had created a partnership firm into which lots of commissions were diverted. This was told to us by a CA who worked with them during an Income Tax raid. This was a shock to us because of the reputation of the company. It is a top listed company – the raid had happened in a different company – not the main company.
Similarly there are some questions for which I can design calculators which will tell you EXACTLY what i want it to tell you. For example recently I found (in a very reputed personal finance magazine) a retirement calculator which assumed CPI (consumer price index) of 3.1% and had a ‘buy vs. rent’ calculator which assumed that over 20 years you will pay 12% interest on your home loan, and your investments will grow at 7% per annum! Now this is a funny assumption! In a country where interest rates are 12% if equity returns are 7% (for a period of 20 years) what happens to risk premium? Why are stock exchanges in existence at all?
The sub-prime problem in the USA has happened only because of this! ASSUMPTIONS. There is an old saying ASSUME – makes an ASS out of U and ME. If you take any number and extrapolate it over long periods of time, you can prove anything. But just as the trees do not grow to the heavens, mathematical projections should be used with some brains. I asked my daughter “If a cow has 4 legs at age 1, how many legs will it have at age 4?” In her keenness to show off her ‘table’ skills she said 16. However when this happend she was 7 years of age and not the head of risk in a BOND rating company.
The NHS (national health service) in UK is notorious for its delay. Once upon a time it used to take 4 hours waiting to get an abortion done. Then the time increased to 5 days. Surely an over enthusiastic excel freak can prove at in another 58 years it will take about 10 months to get an abortion done.
Similarly the population of Mumbai grows by a %age far greater than the population of India. In an excel sheet I will be prove that at some point in time the population of Mumbai will be greater than the population of India. Mathematically it should not be too difficult.
Your luck again. I do not head the risk function of any bank. I know some of them – they are the biggest risk the banks run! When you are doing equity research you come across a lot of data – you need to develop the skills to know how much of that research is worth keeping and how much worth junking. Basics.
LTC
“I do not head the risk function of any bank. I know some of them – they are the biggest risk the banks run!” – EPIC
Shankar
Hi Subra,
Looking for a book from you, covering A to Z on equity research for starters.
Aditya modi
Great caveats subra
How to get rid of hindsight bias
Almost every literature in equity research is laced with this bias..
Anshuk Jain
“”Similarly the population of Mumbai grows by a %age far greater than the population of India. In an excel sheet I will be prove that at some point in time the population of Mumbai will be greater than the population of India. Mathematically it should not be too difficult.””
No you can’t.