Before I click 'Buy'

One of our readers, has asked me to explain few aspects of investing process. Here follows the question.
[Caveat emptor: Neither the questioner nor I, stake any claim on brevity]

Question

What I want from you is some much much needed thoughts on the following:

1. first of all, how do you get hold of the financial data of the companies and not only for the current year but for atlas a decade. Because without it I dunno how to run the calculations for the intrinsic value !!!!!!!!!!!!!! I want to understand how do you screen your companies? I mean I want to develop a sort of a library mentally about companies I want to get interested in initially my target is 50 companies. But for that I need detailed financial info!
I am constrained because when I look at company data for the last 5 yrs from sites like money control, economic times I notice that almost all the companies have undertaken huge investments compared to their previous years and most run their cashflows into negative! How am I suppose to value them????

2. secondly if you take Buffett’s advice solid companies backed by brands than I like your views that you must have noticed that all the significant FMCG cos. Indian as well as MNCs have shown significant underperformance in the last 4 yrs. My other question is how do you deal with this.

3. Lastly, I want you to guide me to develop significant idea about various companies but also their shooting range i.e. within what range the company becomes a good buy and for that I need two significant things. A) I need financial info about companies and know where to find the info to run a Discounted Cash flow for them. B) I need to have shooting ranges for the companies I am interested in to help me act decisively when the time comes

Answer
I like the question because many of the points raised here are the key points in investment strategy and decision making. However, I’ll answer them in the same order as I think about these key points.


Developing the list of companies you track:
The first part of the investment decision making is, deciding your investing horizon in terms of the companies you are actively looking at. This step is important because you can not look at thousands of companies.

One apt analogy for this problem is found in the field of oil exploration. You have got licenses to explore a certain number of exploration blocks. As you know, you can’t scratch the surface and find the oil. You’ve got to dig deep. You start drilling a few holes and gather some data. This data is analyzed and based on the outcome the explorers revise their estimated probability of striking oil. They have to decide whether to quit this well and start digging at another place. If you dig too deep into a small number of wells and fail to find oil in all but a few, your proceeds from whatever little oil you found will not be able to compensate for the cost of drilling. If you drill too many holes, your cost will again be high enough to rule out any returns. It is an optimization problem which can be solved only with experience and analytical rigor.

Many people use the screening approach for this. They do number crunching and find all the companies that fit the criteria of being an excellent business. I never succeeded in such an approach and I don’t recommend it. The financial numbers are scorecards of past performance. They may act like signboards to an excellent opportunity but then most opportunities with visible signboards are already crowded like popular tourist places. You won’t get your money’s worth in popular hill stations (you can take my word on that because I was born in one). To find the beauty of nature, you have to go to places where there are no signboards, no footprints.

You add the valuation criteria to your screening and find supposedly good companies which are undervalued. This approach typically yields many companies which are value traps. It’s hard to have the discipline required to avoid falling into these traps. When you see amazing numbers, you are already half sold to the idea and if you read about business later, you will convince yourself about the business. The net result is that you will ignore the business while focusing on the numbers.

The approach I follow to the screening exercise starts with analysis of the business. Over the past decade, I’ve built a mental library of businesses I admire due to various reasons. They may be market leaders or deeply entrenched businesses in a small niche. In all cases these are the businesses that can ride over couple of years of back to back of recessions. I read about them in papers, magazines and the Internet. I follow the business decisions they make and I analyze the results of those decisions. From time to time, when I hear about exciting things happening in other companies, I explore them and if I like what I see, I add them to my watchlist. My watchlist at any time is very small, around 40. I don’t even scratch the surface of other companies. As an individual investor, I don’t have the resources to do this. Neither does my investment strategy, perfected over the years to suit my needs and resources, require this.

Analysis of business is a very interesting subject. To get some head start on this topic you can read ‘Common Stocks and Uncommon Profits’ ^[1] and Buffett’s letters to shareholders.

Getting financial Data:

Gathering financial data is easy for the people who follow only a small number of companies. For aggregate data, there are websites. The stock exchanges, financial portals and company websites have enough data. For example in India, BSE’s site has results for the last 6-7 years. The portals like ICICIDirect have data running for 5 years. The individual company websites may have more data, like financial performance for 10 years. The best place to find accurate data is the annual reports. Most companies give their 10-year data highlighting their growth. It can be a good source of information. For example, Tata Motors’ annual report has the summary of financial data from 1946! Even if you have the 10-year data, you can see a 4- 5-year-old annual report to find the data of the years before that.

The people doing screening of large number of companies need data at one place and in standard format, and they can find it in databases like Reuters. When I used to do number crunching, I used to write software programs to pull data from the net. Even if you don’t know much of programming, you can use tools like Excel which allow web queries to extract data into spreadsheets, and then learn to write some macros to do that operation for multiple companies.

The important part here is to find the accurate data. The splits and bonuses make it difficult to find accurate data. The past data should be comparable to current data. There is no point in comparing when a company has gone though mergers, acquisitions , de-mergers, spin-offs or it has spawned numerous subsidiaries which now account for a significant percentage of revenues. The data you gather requires validation for correctness. Anyone doing significant amount of number crunching should establish procedures to verify the accuracy of data before the analysis. The analysis rarely happens on the data that gets reported. You do not look at sales or profits. You look these parameters in relation to other parameters like networth, average of data from past years, etc. So it can get to be a pretty daunting task for an individual investor. One word of caution, even when you gather base data, do not use the numbers that you can yourself derive from basic set of data. Don’t try to get data for P/E, return on networth because you can yourself calculate it using base figures like profits, sale and basic arithmetic encoded into formulas in a tool like Excel.

Forming the expectations about risks and returns:

This is another interesting part because it decides the asset allocation among various asset classes (fixed returns, stocks and so on). At an individual stock level, the allocation will again depend on the expected returns and risks involved.

I follow a simple approach to it. If I have the money, I see how much return (interest) I can get from the fixed income securities (FDs/Bond). Then I try to see if I have enough opportunities to beat that return without taking significant risk. The allocation to equities varies based on prevailing price levels and the amount of long term capital that you can invest and forget about. You have to look at your own finances before you look at the financials of companies. You have to avoid creating an asset-liability mismatch by investing short term funds on long term assets like equities.

If you don’t suffer from a disease called irrational optimism, it’s easy to form an opinion on risk. You have to be skeptical. Rather than saying, I’ve got a million to invest in stocks you must say: “I‘ve got a million to invest in risk free bonds. Can I make little more by investing in stocks without taking much risk?” You got to have a solid case of higher returns for every small risk you subject yourself to. Let me explain by an example.

What are the chances of the State Bank of India going bankrupt in 10 years? Would the Government of India let the bank fail when it has about one fourth of total deposits and advances of all scheduled commercial banks in India. Write down your answer in probabilities. Now consider this, SBI’s 10-year FD earns 8.5% whereas the yield of a 10-year Indian Govt. bond on Jan 16^th, 2009 was 5.61% (I don’t think it’s tax tree). About 3% difference in bond yields for a little bit of extra risk!! This can give you an idea about the extra returns you have to ask to compensate for the risk.

The Sensex went up from on Jan 16^th, 1999 to 9323 as of today. This amounts to 10.89% returns (+ dividends) from investing in India’s best 30 companies. If, in the next 10 years, we repeat this performance and the Sensex ends up at 26000, do you think the returns, mere 2% more than SBI FDs are enough to compensate for the risk?

It may well be. The FD returns are taxed whereas there is no tax on long term capital gains. Considering 1.88% dividend, yield of the Sensex, I may beat after-tax FD returns. Of course, if I carefully choose the stocks and do better than Sensex, my returns may be well above the yield from FDs.

This is how you have to think. After analyzing your finances (income, spending, pipeline of big budget purchases, kids’ education, etc.) and the returns available from fixed income securities, you have to analyze the returns that can be expected from the general market. It’s stupid to think of past peaks of markets. After the Harshad Metha scandal, it took 7 years for the Sensex to cross its last peak. After the tech bubble burst, it took 4 years for the Sensex to reach its previous highs and I haven’t yet talked about the Dow Jones in 1929!

It’s important to have a certain expectation on long term results from equities in general because most people won’t be able to do better than the leading indices. History has enough empirical evidence for that. This guesstimate goes into deciding what percentage of your networth you may allocate to the equities. Only then must you think about individual stocks.

Evaluating the stocks

With roughly the same thinking, as described above, you can analyze the allocation of your funds to one stock vs. another. It’s not possible to figure out the target levels to where the stock can reach but it’s possible to find levels where the risks from the investments are outweighed by the possibilities of returns.

It’s not necessary to do a full blown discounted cashflow analysis. You need to understand the business dynamics and form a rough idea of the profitability levels and expected growth. Too many people fall into the trap of focusing too much on numbers and flashy models. Intelligence, qualification in finance, and number of years of experience with investment companies positively correlate to your likelihood of falling into this trap. I can teach investing to a small town illiterate businessman and I can learn a lot about business from him but it’s very difficult for me to talk to ‘educated’ people because they are not willing to give up their models which don’t work. To me, they offer the same amusement that a person, who builds and sells abacus^[2], will get when he sees the adults who refuse to outgrow the abacus (I earn my living by building analytical modeling and decisioning software applications).

John Maynard Keynes has said: “It’s better to be roughly accurate than precisely wrong.” In all complex systems which are beyond the grasp of precise modeling and numerical analysis, you’ve got to remember this.

When you know your calculations are, at best, intelligent guesses, you would leave a margin of safety. If the price falls below your estimated value, your margin of safety increases. After a certain point you’ve got to say, “Well, I can come out a winner on this one”, and invest. You would analyze the situation over time as more information becomes available and the margin of safety changes. You can make additional investments if the need be or even exit at a loss if you come across something you had not considered while making the decision. Over the years, this process will get hardwired and you will not do the calculations, just the way while driving, once you are good at it, you don’t calculate the speed at which you can take a sharp turn. You just do it. Once you’ve gained enough success, confidence, skills and experience, there will be times where the odds are such that you can go all-in^[3].

Thanks for a long question because it reduces the guilt of giving a long explanation that I’m so used to.

On the specific query on the underperformance of FMCG companies – the 4-year period is a short one. The rise in commodity prices has worked against them while the producers of both agri and other commodities have gained, which makes the show look disappointing. However, these are stable businesses which won’t go bust in recession unlike many other commodity players. They won’t crumble under loans like indebted companies. These won’t report huge losses due to defaults like financial companies. These businesses earn hard cash, the most precious commodity at today’s date. The problem is that the metrics about returns are available on demands but there is no metric available which describes the risk. The beta^[4] of the stock is the biggest bullshit of finance. So, I think you shouldn’t focus too much on underperformance in the last 4 years. Having said that, I haven’t found too many attractive FMCG companies in India and I would say that you need to imbibe the essence of what Buffett has said rather than trying to achieve the same results as his spectacular FMCG bets. He himself has done much more than that.

That ends my answer..

Regards

Kamlesh

References

1. Common Stocks and Uncommon Profits: A book by Phillip Fisher.

Buffett once said that his investing is 85% Graham and 15% Fisher. When you read the book you will see a clear impression of Fisher’s ideas in Buffett’s investing style.

My review of the book

http://www.unfairvalue.com/2006/04/book-review-common-stocks-and-uncommon.html

2. Abacus: An abacus, made of a bamboo frame with beads sliding on wire, is a calculating tool used primarily in parts of Asia for performing arithmetic processes. You can do various arithmetic operations including square roots, cube roots, etc.

http://www.ee.ryerson.ca/~elf/abacus/

3. All-in: An all-in move in poker is where you bet all your chips in one single move. If an opponent calls and wins, you are out. All-in should be done only when you are dead sure about your win (due to strong cards or your belief that everyone else will fold). It’s an expert move but very popular in the depiction of the game in movies. That’s why those who are novice at poker foolishly go all-in at the drop of a hat.

In investing too, the inexperienced investors play huge gambles turning a blind eye to the odds stacked against them.

4. Beta: A mathematical measure of the sensitivity of rates of return on a portfolio or a given stock compared with rates of return on the market as a whole. A beta of 1.0 indicates that an asset closely follows the market; a beta greater than 1.0 indicates greater volatility than the market. In effect, if the perception of investors about the value of stock is always wrong, it will have lower beta than a stock where the market is sometimes right and the valuation models will assign lower risk to such stock. That explains the ‘bullshit’ part