Financial analysis involves comparisons of multiple variables. For example, an analyst may look at an automobile-maker and decide the valuation is cheap compared to its peers. This involves balance-sheet analysis across various parameters.
He should also examine scenarios of vehicle price change, and changes in the costs of key inputs like metals, plastics and electronics, and perhaps, in the tax structure. He may also need to examine currency. The stock may be a good buy for domestic investors but not for those with forex risk.
Once this process of variable analysis starts, it's difficult to cut it off. If the industry is labour intensive, labour relations and laws may be critical. There could be management changes with key personnel retiring or moving on.
If the business is located in a water-sensitive or power deficit region, there may be issues. There may be social or environmental problems that lead to activist protests. There could be other policy changes that reduce or enhance demand.
There aren't hard and fast rules. Some variables change quickly, others slowly. They interact with each other in ways difficult to model. By the time you finish an exhaustive analysis, the price itself may have changed significantly.
Good analysts will take an overview and focus on the variables they consider important. This is very subjective since different people with the same information assign different weights to different pieces.
I'm lazy enough to adopt a minimalist approach. If you buy a stock at random, you have roughly 50 per cent chance of getting it wrong. If you do a cursory analysis before buying, you still have 50 per cent chance of getting it wrong. If you invest in cold towels and coffee and spend several nights programming multi-variable computer analysis, you still have a 50 per cent chance of being wrong.
The greatest investors seem to be able to improve that ratio slightly but not by a huge amount. It is conservative to assume that you will generally have a 50:50 ratio of winning investments to losing ones. So I assume that there is a 50 per cent chance that any investment decision I take will be wrong and consider the implications.
The key question is not how I make an investment decision. It is this: How much money can I afford to lose on every wrong decision if one out every two investment decisions is wrong (assuming equal investments)? Presumably less money than made on the correct decision.
That understanding can be further refined.
If 50 per cent of your investment decisions are wrong, you can only afford to lose substantially less on the wrong decisions than you make on the correct ones.
The overall equity returns – the sum of gains and losses - must exceed what you can make from alternative, less risky investments such as debt.
I can, without trouble, make around seven to eight per cent return from debt at present. Hence, I need to make somewhat more – let's say 10 per cent overall, from equity. If half of any money I deploy in equity is making losses, the remaining half needs to generate 20 per cent, plus whatever losses are being incurred, to make equity investments worthwhile as a whole.
Let's say I set a hard and fast rule of exiting any equity position that loses 10 per cent. In that case, assuming equal investments in every equity position, I need to make at least 25 per cent on the equity investments that do gain.
This thought process actually helps substantially in guiding investment decisions. I am only interested in equity investment that appear to have the potential to generate at least 25 per cent upside at current prices.
The time variable also comes into the picture. If you're a long-term investor, you are hoping for big returns in two-three years, rather than a profit in the next six weeks.
In that case, your entire return - loss matrix has to be scaled up to account for compounding effects. You must seek say, 25 per cent compounded over three years – that is, roughly 95 per cent. On the downside, you should be prepared to lose 10 per cent compounded for three years – that is about 33 per cent.
The few times I've discussed this philosophy with investors, I've usually received responses to the tune of “This is too negative an approach”. However, it seems to work for me. Before I make any trade, I work through the scenarios arising from potential loss. Try it. It may work for you.