Latest stories

Arithmetic return is what you see, geometric return is what you get

A

I am reading a fairly enjoyable book ‘A Mathematician plays the stock market’ (see section ‘currently reading’ under sidebar). Found the discussion on geometric mean v/s arithmetic mean (for returns) fairly relevant for an investor.

Let me explain

Arithmetic mean (or returns) is the simple average of returns over the time period being considered.

For ex: consider a stock X that has the following returns for the year
Week 1 : + 80 %
Week 2 : -60 %

The average return for a two week period is +10%. The ‘average’ return for the year would be 1.1^52 = 142 times the original investment. If average return is what an investor gets, then anyone can be fabulously rich in no time.

Geometric mean of the same stock will be derieved by the following formulae
Total return = (1+0.8)*(1-0.6)*(1+0.8)…….( for 52 weeks) = 0.000195

The scenario in the above formulae is that the investor makes a positive return the first week, followed by negative the next and then positive and so on. So the real return he/she actually gets is less than 1 % of capital invested

Another example
An investor is on the lookout for a hot mutual fund. He looks at the mutual fund rankings and sees a fund, which has returned 100 % last year. He invests his money in the fund. The hot fund promptly proceeds to lose 50% next year ( reversion to mean or maybe bad luck ).

The return the fund publicizes is 25 % (average for 2 years ). An investor who was invested for 2 years is lucky to get his money back. The performance chasing investor looses half his money. The fund manager and his company get their asset management fee and are able to show great performance at the same time.

Reminds me of a famous title of a book – ‘where are customer’s yatch?’

There is another interesting discussion happening on the BRK board on MSN (registration required ) on the same topic.

Concept of variant perception

C

I have been reading a book ‘No bull’ by Micheal steinhardt. He was hedge fund manager and was able to deliver around 30% returns for almost 30+ years.

I found his concept of variant perception useful. According to micheal, every investment idea should be explainable in 2 minutes and four points

  • The idea
  • The consensus view around the idea
  • The variant perception of the analyst
  • The trigger event which would unlock the value

For example, if there is a solid growth company which is expected by the market to grow at 20%, and as an investor my expectations are close to the same number, then I am not going to make more than the cost of equity if the actual numbers meet the expectations (for more detailed understanding of how to evaluate market expectation read the book Expectations investing)

I have used this concept of variant perception in some form although not exactly in the same manner as explained by micheal. Let me give an example

Marico in the year 2003-2004 was selling at around 10-12 time trailing earnings. A simple DCF would easily show that the market was discounting 2-3 years of competitive advantage period -CAP (for detailed understanding of CAP, please read this article) with growth in low teens. Now if one looks at the brands, the history of their New products and their distribution network and management,it is easy to see that this company could grow in low teens for considerably more than the market implied CAP of 2-3 years.

So basically my variant perception was not centred around the growth (which is the the usual variant perception generally given by most of the analysts) but around the CAP of the company.

Marico now sells for a much higher PE and the growth was also much higher than implied by the market (around 15% +).

To a certain extent, one can see the same kind variant perception being exploited by warren buffett, except that he is a genius at recognising such businesses with CAP much higher than implied by the market price (ex: coke, GIECO etc)

My problem with stock screens

M

Most of us know the problem with simple stock screens such as one’s based on low P/E ratio, low P/B ratio etc. A lot of stocks which get filtered through the screens are typically companies with poor economics. I have tried to overcome this problem by building a screen which has the following additional screening criteria

  • An ROCE/ROE of atleast 13% or more
  • No loss during the past 5 years
  • Above average growth over 5 years in NP
  • D/E < 2

Adding the above stock screens has filtered out companies in the following industries (partial list below)

  • auto components
  • bank
  • cement
  • Chemical
  • Shipping
  • Fertiliser
  • Shipping
  • Paper
  • Textile

I have started analysing one company at a time under each of the classification. Unfortunately the reason these companies have filtered out is either due to a cyclical uptick in the industry (cement, shipping, paper etc) or it is tier II company in the industry with high operating leverage and has seen a reduction in interest cost. Due these reasons , the recent PE, ROE etc of these companies is good, debt is down and these stocks look good.
My concern is how these companies would fare once the cycle turns downwards. Let me explain using the example of shipping industry which I am analysing currently.
The main companies in the shipping industry which have filtered out through the screen are

  • mercator lines : High asset addition recently through debt which has resulted in high earnings and high ROE. The risk to the business is high if the business cycle reverses as the company may be unable to service its debt
  • Varun shipping / Shreyas shipping: high operating leverage, high debt and high growth in earnings in recent times due to high shipping rates. Earnings risk is high due to operating leverage
  • Essar shipping : High earnings due high shipping rates. Also ROE is high to asset revaluations. This stock looks interesting and worth investigating further.

I guess the stock screen is throwing up a lot of companies which may be statistically cheap but not really a value stock. So essentially I am not be able to come up with a list of companies which are great value. I guess it is to a certain extent an indication of the kind valuation levels existing in the current market (The same filter in 2003 gave much better companies). So I guess I will have go through the entire list and maybe at the end (the list has 100+ companies) come up with a few good stocks. It defintely not a waste of time because it helps me to understand more companies/ industries which could be helpful in the future

any suggestions on improving the above screens ?

Comparing performance when invested capital is low

C

Good article (free registration required) on mckinsey quarterly on how to evaluate performance, when the invested capital is low in a business (like IT services, FMCG, consulting services etc)

http://mckinseyquarterly.com/article_abstract.aspx?ar=1678&L2=5&L3=5

Some excerpts

A more useful way to measure performance is to divide annual economic profit by revenue.2 Grounded in the same logic as conventional ROIC and growth measures,3 this metric gives executives a clearer picture of absolute and relative value creation among companies, irrespective of a particular company’s or business unit’s absolute level of invested capital, which can distort more traditional metrics if it is very low or negative. As a result, executives are better able to evaluate the relative financial performance of businesses with different capital-investment strategies and to make sound judgments about where and how to spend investment dollars.

In application, this approach will vary from business to business, depending on what is defined as volume and margin. In a people business, such as accounting, the margin would likely best be broken down into the number of accountants multiplied by the economic profit per accountant. In a software business, however, it would be better calculated as the number of copies of software sold times the economic profit per copy of software; in this case, deriving the margin from the number of employees wouldn’t make sense. But in all cases, this approach can provide a more nuanced understanding of performance across businesses or companies with divergent levels of capital intensity.

Equally important, economic profit divided by revenue avoids the pitfalls of ROICs that are extremely high or meaningless as a result of very low or negative invested capital. Economic profit, in contrast, is positive for companies with negative invested capital and positive posttax operating margins, so it creates a meaningful measure. It is also less sensitive to changes in invested capital. If the services business mentioned previously doubled its capital to $20 million, its ROIC would be halved. But its economic profit would change only slightly and economic profit divided by revenue hardly at all (to 6.8 percent, from 6.9 percent), thus more accurately reflecting how small an effect this shift in capital would have on the value of the business.4

my thoughts : The above metric is not sufficient to evaluate. I would still consider a low capital intensive business superior compared to a capital intensive one , even if the above ratio is low , as a low capital intensive business could have higher free cash flow (provided both have similar competitive advantages ) and hence could be worth more.
The above metric is good to look at, but i would not base my decision on it (or any other single metric)

Subscription

Enter your email address if you would like to be notified when a new post is posted:

I agree to be emailed to confirm my subscription to this list

Recent Posts

Select category to filter posts

Archives