Edited transcript of talk delivered earlier today at OctoberQuest 2013, Mumbai.
What a insightful Lecture. Please continue to share this information. Thank you
sir, dont have words to say its absolutely mesmerizing…………thank you so much.
sir can you do something for the those who have realized their mis fortune of not attending your classes……some online course
I really liked your thoughts on this topic. I have 2 questions
1. According to you, how does the concept of Margin of Safety fit when buying shares of exceptional businesses like ITC and Asian Paints ??
2.With the benefit of hindsight, we know that companies like Asian Paints, Bosch and ITC have made tons of money for shareholders. Would you be a buyer of such businesses at current valuations. Afterall, it was Warren Buffett who also said that great businesses need not make great investments.
Thank you very much for the article! Enjoyed reading it. And went through the pain of the lesson first hand – I bought and sold Asian Paints twice each in last 8 years, each time patting myself for the handsome gain! A P/E of 60 would be hard to gulp! Takes a bit of unlearning, not easy!
Amazing article Sir. Fully enjoyable & must read for any serious investor.
Just one question-
While doing Time Machine Valuation of Asian Paints, why are you using PBT for calculating its value??
Shouldn’t we be taking PAT or Free cash flow for calculations?? After all, taxes are a harsh reality for any business.
Thanks a lot for the article Sir.
That’s because I used pre-tax discount rate to compute present value. One can use post tax earnings but then one must also use post-tax discount rates for computing present values.
As usual a brilliant article. Thanks for enlightening us.
Is there a reason why you use pre-tax earning numbers and discount rates. Is it ideologically or conceptually different from using post-tax numbers?
Thanks Prateek. Jatin asked the same question. See earlier comments and my response.
Still difficult to understand for me, leaving the tax angle .. why PBT is used in calculation instead of dividend payout or FCF. All the retained earning is driving profit of future years so it will any way will get counted in future earning/ dividend payout, also as part of terminal value.
An example: two companies reported exact same PBT numbers for corresponding years. One paid 25% PBT as tax and second one none. Looking at these two companies, first one should be valued better then the second one. but this calculation will value them equally. It will not make much difference in first calculation on page 18 but definitely a big difference on second one.
Bhupesh, I have not left the tax angle. In my simplistic valuation model, I have merely used pre-tax earnings and pre-tax discount rates to value Asian Paints in September 2005. If you want to use post-tax earnings for valuation, then you should also use a post-tax discount rate for bringing back future cash flows to present value.
The reason I did not get into FCF (or rather owner earnings) is because I wanted to keep it simple to make a point that Asian paints was undervalued at a P/E of 25 and also because the owner earnings of Asian Paints exceed reported earnings, so my computations were not aggressive, in my view.
Excellent presentation Sir. Buffett’s idea of not valuing a business he doesn’t understand well using a higher discount rate is wonderful. It eliminates a lot of investment options and makes investing more of a focused activity. Conversely, the idea of valuing only those businesses which one understands well using the risk-free rate is very insightful.
Thank you sir for sharing this wonderful and insightful article with us. I personally believe that the difference between Graham and Buffett-Munger style of investing lies in the temperament of an individual. Some people are more fearful of losing money than making it and their focus is more on downside consequences. So, they feel safer in buying under priced securities based on current circumstances and are more conservative in their discount rate assumptions. Other sort of investors are more hopeful and optimistic in their approach and are more focused on making money instead of having the fear of losing it. They are more liberal in their discount rate assumption during valuation.
The ideal temperament according to me should be of a realist which Buffett has evolved into over a period of time. He is not scared of paying more for a quality brand like Heinz but is equally hungry about bargains and vigilant about safety of principal as his investments in Petro China and preferred stocks of various other companies shows.
For eg : If we take the case of HDFC bank whose earnings have grown at a consistent 30% for more than a decade. If we assume that the earnings grow at the same rate for the next 10 years and the stock is valued at 25 PE after that, then the market cap of the company has to move from the present 1.5 lakh crs approx to around 20 lakh crs in the next decade which looks difficult from the big picture scenario.
Moreover, hopefully seven new private banks would enter the banking business during the same decade and would make it extremely difficult for HDFC bank to have a free ride.
My question is shouldn’t the overall size and prevalence of quality competition be considered if we assume high earnings growth for a quality business ? Isn’t it difficult for an elephant to keep on compounding babies equivalent to its size ? Is there a high probability for very well managed 10-20k crs market cap companies like Bajaj Finserv and Shriram Transport Finance to grow 15-20x in the next decade as compared to an elephant like HDFC bank ?
Thanks and Regards,
about the “quality being systematically underpriced over the long term” argument, KO about 15 years back was 43.59 and today its 37.77. during the same interval revenue has grown from 18B to 48B. and net margins are roughly the same. and they have about 8-10% lower number of shares outstanding. it would be very difficult to argue that KO is a bad business or has lessened its moat. so, it must have been overpriced then or underpriced now. either ways, it made for a bad investment(both in absolute and relative terms). in real terms one would have lost over 30% of purchasing power. the dividends (after tax) covered some of it. in relative terms the performance was even worse, during the same period s&p 500 moved about 70% higher.
“mean reversion” is alive and well at least in the world of price to earnings ratio.
That’s a wonderful example Rishi. As you have said, KO’s momentum in fundamentals has been intact, but it’s stock price performance has been disappointing and the only reasons for that have to do with valuation (starting or current).
There will always be examples of companies that either do well in the market over long periods of time despite having poor fundamentals or doing poorly despite having good fundamentals.
However, if we stick to the base rates on fundamentals, we get a much lesser mean reversion than we get in stock market returns.
I agree with your point on mean reversion in P/E multiples. Investors in Infosys paid dearly (in terms of opportunity cost) from Jan 2000 to December 2010— a period during which Infosys continued to perform very well, but the stock did not (while Indian markets boomed).
Mean reversion also exists in stock returns when stock prices fall way behind fundamentals or vice versa. As Warren Buffett wrote in his 1986 letter:
“What could be more exhilarating than to participate in a bull market in which the rewards to owners of businesses become gloriously uncoupled from the plodding performances of the businesses themselves. Unfortunately, however, stocks can’t outperform businesses indefinitely.
Indeed, because of the heavy transaction and investment management costs they bear, stockholders as a whole and over the long term must inevitably underperform the companies they own. If American business, in aggregate, earns about 12% on equity annually, investors must end up earning significantly less. Bull markets can obscure mathematical laws, but they cannot repeal them.”
KO and Infosys are good examples of how paying a too high a price for fundamentals, even for a great business, can ensure mediocre results.
Buffett’s observation that in the long-run stock returns and underlying business returns will roughly track each other is also consistent with Graham’s observation that in the short run the market is a voting machine but in the long run its’ a weighing machine.
Mean reversion is a very powerful and useful mental construct for investors, but they must not apply it automatically to two types of business fundamentals – the great businesses and the crappy ones. Averaged out, those types of businesses, then to stay that way…
Would be it possible to know KO’s PE multiple 15 years ago?
in case of hdfc bank, one can compare it with wells fargo for reference. wells fargo is mainly in the US (300 Mn population) and has a mcap of 220 Bn. HDFC bank is 1/10th and operates in india with 5 times the population. so the opportunity is there
banking in the US is competitive due to a lot of local banks (2000+) and yet WFC has done well. so I would say we can use WFC in our thought experiment to evaluate the potential for HDFC bank
I appreciate your logic and I also thought about it but came with contrary views that even with 300 Mn population US is a 14 trillion economy whereas we are 2 trillion with a 1.2 bn population. Even if the economy grows at 8% for the next decade, we would be 4 trillion. On a 25 bn market cap HDFC bank is about 1.25% of the total economy size. If it reaches 250 billion market cap on a 4 trillion economy it would be 6.25% which is way large than 220 billion Wells Fargo, which is about 1.57% of a 14 trillion US economy. I don’t say anything is impossible but the conviction is less if things don’t add up.
I am not suggesting a linear extrapolation, but giving WFC as a reference point. HDFC bank can be 10 times current mcap, but it may take longer than 10 years.
WFC has among the best fundamentals in the large banks, is considered undervalued (buffett keeps adding to it) and also fits in the quality being underpriced theme of prof bakshi’s talk.
I could go on and on – suggest you have a look at their presentation, to get a look at their fundamentals ..it is very instructive and shows how a bank should be run.
Munger bought this stock in 2009 for DCO after holding cash for 8 years and then bought it during the crash. If there is any stock recommendation buffett has made (or come close to it), it is wells fargo. He continues to buy the stock for Berkshire even now and has held it for 15+ years.
I am aware and not denying the fact that what quality gem a bank like Wells Fargo is. Even Aditya Puri have a vision of shaping HDFC bank like Wells Fargo and I have seen him mentioning WFC in his interviews also. I think Buffett and Munger are advocates of another US Bank i.e. US Bancorp which is an equally good bank.
But my thinking experiment was whether I would be better off within a fixed time frame of 10 years by investing in HDFC Bank or other quality company like Bajaj Finserv which is relatively small and cheap at the moment. I am judging this from a neutral angle as I am a small shareholder of both HDFC bank and Bajaj Finserv. But to make concentrated bets with a small capital, you have to always consider opportunity cost also.
I did one small exercise of a few multibagger companies in the last 10 years i.e from 1.4.2003 to 31.3.2013. I concluded that people would have been better off if they would have invested in companies like Prestige, Hawkins, Nesco, Titan, Balkrishna Industries etc. instead of Nestle, ITC, HDFC bank, Castrol, Colgate and HUL.
The results could be totally different for a different time frame. So, I conclude that no theory is perfect in investing and there is always scope for learning and refining our understanding.
Below is the link to that multibagger exercise for your reference.
One suggestion regarding your estimate of HDFC bank’s projected market cap as % of Indian GDP. The appropriate growth rate should be nominal growth rate (real + inflation)
Sir, I think the annualized returns numbers might be incorrect in a lot of places. eg: Shriram Transport Finance.
Stock returns since ’05 => 588%
Years = 13-5 = 8 years
Annualized returns = (5.88)^(1/8)-1 = 24.7% which you’ve mentioned as 71%;
Same goes for ITC, etc.
Thanks for bringing this up. I simply aggregated the total holding period return then divided it by the number of days in that period and then multiplied the result with 365. I did this for both the stock and the index returns so the results are comparable. You can see the workings from:
I have also added two new columns – Stock CAGR and Sensex CAGR.
Even though the conclusions don’t change, the method you suggested was more accurate, so thanks for pointing that out.
Updated the presentation with CAGR numbers instead of flat return numbers.
great talk prof – a ‘tour de force’
I came up with a similar conclusion with different companies such as Crisil v/s a denso or Ashok Leyland. However mine was a fuzzy conclusion, whereas you have backed the same with empirical data and multi-disciplinary explaination for the higher discounting.
what was dimly visible to me, now seems so obvious after your speech that i feel silly for not having realized it sooner …better late than never 🙂
It took Mr. Buffett 35 years to arrive at the same conclusion. 🙂
He acknowledged as much in his 1983 letter when he wrote:
“You can live a full and rewarding life without ever thinking about Goodwill and its amortization. But students of investment and management should understand the nuances of the subject. My own thinking has changed drastically from 35 years ago when I was taught to favor tangible assets and to shun businesses whose value depended largely upon economic Goodwill. This bias caused me to make many important business mistakes of omission, although relatively few of commission.
Keynes identified my problem: “The difficulty lies not in the new ideas but in escaping from the old ones.” My escape was long delayed, in part because most of what I had been taught by the same teacher had been (and continues to be) so extraordinarily valuable. Ultimately, business experience, direct and vicarious, produced my present strong preference for businesses that possess large amounts of enduring Goodwill and that utilize a minimum of tangible assets.” (Emphasis Mine)
jeremy siegel has said something similar. in ‘the future for investors’ he compares returns of various periods. He compares top end high pe companies with top end lesser PE companies and came to a conclusion that the second kind of companies give better returns.
But how would the 15 year comparision with dividends for high PE asian paints, nestle, ITC compare with lower PE higher dividend paying like Tata Investment, VST Industries, Castrol, Pirmal, GIC Housing.
I also find Godrej Industries completely missing in your blog. Is there some reason?
I have all the stocks mentioned above.
I do not have an accounts background so cant do these calculations myself. But I understand what you are saying and also agree with it. This could also be some post in future.
Thanks professor…as mentioned by Rohit you backed the reasoning with empirical data. I have a request how can we obtain the PE,PB graphs of stocks over a period of time.
[…] better and very much rewarding to go into superior businesses. Prof. Sanjay Bakshi has posted a wonderful post on the same. Yet we fail to understand the extreme negative sentiments and valuations of 2-4 PE in […]
sir i had one question on discount rates. If long term interest rates are used for discounting cash flows then the implicit assumption would be that the bond cash flows of a AAA rated company and the equity cash flows of the same company would be discounted at the same rate. Assuming in a perfect market both the bond and equity of that same company are fairly valued by the investors then the investor can expect to get same return from holding bond or equity and should be indifferent to same. While one can do a lot of work and be highly certain about the future prospects of the business, would it not be true that a priority debt would be relatively more safe given more scope to absorb errors in forecasting and hence there should be some equity risk premium over the same company’s debt. I agree it makes no sense to replace risk with higher discount rate and hence no need to have separate discount rates for various businesses but should equity as a whole not have at least a slight premium over debt in same company as otherwise we would be assuming that margin of error in forecasting is same for both debt and equity of that company. If however quality is sustainably under priced in long term then my hypothetical scenario of fair value for both debt and equity probably does not exist but still would be good to have your views on the same.
Sandeep, let me address your question by citing an example and asking you a counter question.
A company has debt of Rs 100 million and net worth of Rs 900 million. The aggregate capital employed, therefore, is Rs 1 billion. This company earned pre-tax and pre-interest earnings of Rs 350 million last year. The 35% ROIC is the consequence of an enduring moat.
Of those pre-interest earnings of Rs 350 million, Rs 10 million go to the lenders as the debt carries an interest rate of 10%.The balance 340 million become pre-tax earnings to be shared between the taxman and the stockholders. The pre-tax ROE comes to about 38% (340/900). Assume further that the earnings of the company will continue to grow for a long time, without any equity dilution. The current market cap, however, is only Rs 1,800 million i.e. the stock is selling at just two times book value offering you an starting earnings yield of about 19% (340/1,800)
Inflation is 12% p.a.
Which is the risker security in the capital structure? Equity or Debt? Use Buffett’s definition of risk for thinking about the answer to this question.
Sir in this case clearly it is the debt which is riskier as return on debt does not even cover inflation and pound notes of future will not even allow me to maintain my current purchasing power. The equity on the other hand has current earnings yield of 18% and with improving earnings without further dilution this will improve further going forward on the opening valuation. even if earnings were to remain static equity would still be better if most of those earnings convert to cash flow and management distributes those earnings as dividends.
However, if we take the asian paints example in the note and for discussion sake assume that opening PE was indeed 60 and not 25 and thus fairly valued, based on what we reasonably assumed at that time then which one would be more risky, equity or debt as both can be reasonably expected to generate c.12% return per annum based on our simplistic assumptions. For equity one would have assumed high growth rates in following 5 years and this could entirely be based on very strong convictions based on commercial knowledge and understanding but still prone to error while for debt one would need to make only very modest assumptions. Equity may have potential to both outperform or underperform relative to your forecasts given the opening valuation based on same discount rate as debt. If we assume inflation of 10% then at that price would the debt not be a better risk to assume than equity in the same company even though both are fairly valued using the same discount rates. Of course both could be bad options compared to much better other options but just comparing the two, which would be better.
Apologies if i am being a bother but ten years before when i studied from you at MDI i did not have the maturity to ask these questions and hopefully you do not mind clarifying these concepts now:-)
Sandeep, so we agree that even within the same capital structure, equity could be less risky than debt. This is VERY counter-intuitive to most investors, ESPECIALLY those who got their “knowledge” by reading up standard academic corporate finance text books and not reading Graham or Buffett instead.
The market price of any security (stock or bond) in relation to its underlying value will always enter into the risk picture. Other things remaining the same, the higher the price/value ratio, the higher the risk of permanent capital loss, and lower the subsequent expected return.
Now, I will try to answer the question you originally asked. You asked that if the stock and bonds are correctly priced, within the same capital structure, should they offer the same return? Let’s see if that happened with Asian Paints. Let’s assume that in September 2005, the stock DID sell at P/E of 60 instead of 25. So, the price would have been Rs 118 instead of Rs 49. So, imagine that you bought it at Rs 118, held it for the next 7.5 years (Sept 2005 to March 2013) and sold it at the then prevailing market price of Rs 491. Under those conditions, the stock price return of this investment operations would have been 21% p.a. See https://db.tt/g7cgS11h for computations. That’s a return which exceeds the returns on our hypothetical bonds in Asian Paints.
So, by now, we have established two things.
First, being a bond investor, even in a great company, will be a very risky proposition as compared to being an equity investor in that same company at a bargain price. That’s because inflation will destroy long-term purchasing power resulting in permanent capital loss (and that’s what risk is in Buffett’s worldview) while growth in earnings (partly caused by inflation itself) would benefit the equity investor.
Second, even if you pay UP for quality and pay a FAIR price for the equity of a great business, the returns on its equity would exceed the returns on its bonds. Isn’t this logical from a businessman’s viewpoint? When you go into a business, that business requires A LOT OF INTELLIGENT EFFORT. You can always invest in a debt security which requires LITTLE EFFORT. So, to be compensated for taking that extra effort, you seek an additional return. So, the fair price that you will be willing to pay is one that will deliver you that extra return. In the case of Asian Paints, we see that even an investment at fair price in Sept 2005, at the P/E of 60, would have delivered you 21% p.a. over the next 7.5 years. Also notice, this 21% return would have occurred DESPITE P/E multiple contraction from 60 in Sept 2005 (we paid fair value remember) to 42 in March 2013.
Graham had it all figured out. In his book, “The Intelligent Investor,” he wrote:
“It has been an old and sound principle that those who cannot afford to take risks should be content with a relatively low return on their invested funds. From this there has developed the general notion that the rate of return which the investor should aim for is more or less proportionate to the degree of risk he is ready to run. Our view is different. The rate of return sought should be dependent, rather, on the amount of intelligent effort the investor is willing and able to bring to bear on his task.” (Emphasis mine)
1. People’s notion of risk is all screwed up thanks to the standard academic finance textbooks they read instead of reading Graham and Buffett. Indeed, as we just saw in this example, what is perceived as low risk by most people (bonds of a great company) actually turns out to be more risky than the stock of that great business, EVEN AT A FAIR PRICE;
2. Low risk (when Buffett’s risk definition is used) translates into high return and high risk into low returns;
3. Paying up for quality (but not overpaying for it), not only carries low risk, it also offers a high return. Buying great businesses at fair prices is better than buying mediocre businesses at bargain prices as Buffett says— something I tried to substantiate in my OctoberQuest talk by providing anecdotal, as well as, empirical evidence.
4. Investors should listen to Graham and think about the relationship between INTELLIGENT EFFORT and RETURN, instead of risk (which is equated with uncertainty and/or volatility by most people) and return.
Many thanks sir. that was a very useful discussion which helped me think through the whole point. I think one reason why Asian Paints achieved 21% return in our hypothetical example was due to the issue of systematic under pricing by the market, which you mentioned earlier. You would note that in our fair value calculation, we had assumed a 2% terminal growth rate, which in effect means the business will be valued at 1/(12%-2%) i.e. 10x multiple at the end of the forecast period. The competitive advantage period for companies like these, especially in a market like India, is very long. Even if you assume a terminal growth rate of 7% after 7.5 years, you are still implicitly assuming fair value PE of only 20x at the end of forecast period. As against these values of 10x or 20x as used by the market in fair value calculations in 2005, the actual PE in March 2013 was 42, which while lower than initial PE of 60 is still much higher than the implicit 10x or 20x used in initial calculations. Thus the market underestimated the length of competitive advantage period and thus assigned much lower terminal value at the beginning and hence undervalued the stock. do let me know if my thinking has been flawed.
You have hit the nail on the head Sandeep except that I will be VERY wary of using terminal growth rates of 7%. People can go a bit crazy doing that. Instead of projecting terminal growth rates, why not figure out implied growth rate implicit in current market value of the firm and then ask yourself if the CAP is mispriced or not? I think that approach will be better because the inversion trick de-biases the analyst from the “goal seek” function embedded in her brain.
Completely agree sir. I used 7% only to make the specific point. in my opinion terminal growth rates should be no more than 4-5% which can be conservatively expected to be the minimum long term sustainable growth rate of Indian economy. If it were US I would probably take only 1-2%. In reality companies with this kind of moat would probably grow faster than overall economy but as a conservative investor you have to keep your margin of safety. I would be keenly waiting for your valuation lecture on KKCL to see how you evaluate whether or not CAP is mispriced in a stock. many thanks again.
Great Discussion. But I seriously think there is a big mistake. If I buy something at full valuation, my future should mirror the discount rate by definition!
As Sandeep pointed out, either the terminal rate was very low, or we should have explicit period of longer period (say 20 years), and then have low terminal growth rate.
So assuming the market is correctly valuing Asian Paints now (not overvaluing or undervaluing it), maybe the correct multiple would have been 80-100 and not 60. Because at correct PE, future returns would have been equal to discount rate.
But we cant do backward investing. Paying 60 PE to get 21% CAGR COULD be a fool hardy game, which would not have been a fat pitch/ no brainer/ 1 step hurdle.
I would really request if you can do some example where you say how much return would you expect by investing at 50 PE TODAY in Nestle or 80 PE in Justdial, say over next 7 years.
Ash, the idea that “if you buy something at full valuation, my future should mirror the discount rate by definition” makes sense if we KNOW for SURE what FULL VALUATION is. And to know what the full value of anything, one needs to know the future cash flows from now till eternity. An impossibility!
The reason why investors, who in Sept 2005, bought Asian Paints at P/E of 60 representing “full value” at the time, and still did better than the discount rate over the next 7.5 years is that they were not really buying at “FULL VALUE” because that calculation assumed a 2% growth rate from 2013 onwards but Asian Paint’s stock price in March 2013 is not being priced by the market with the assumption that from March 2013 onwards, the growth rate in earnings will be only 2%, given that earnings from FY11 to FY13 grew at the rate of 15%. So, in Sept 2005, even if you had perfect foresight till March 2013, and were very conservative about earnings growth beyond March 2013, and consequently arrived at a “full value” at P/E of 60, you’d have been wrong as the value was more than that…
The key point to remember here is that for businesses that will grow their earnings even at a moderate pace, but for a long long time, seemingly high P/E multiples which deter many value investors, are actually too low. This also means that if you buy a great business at what appears as a “fair price” then you will still end up compounding your money at far more than the discount rate you use to arrive at that “fair price.” The longer the competitive advantaged period (CAP), the more likely its worth a lot more than what most value investors think.
And so, the focus should shift from P/E multiples to the duration of CAP. And P/E multiples should be the RESULT of the valuation exercise and not the CAUSE of it.
Incidentally, Asian Paints was just one example illustrating these points. I am sure most investors who follow Indian markets know of many more examples of companies with long duration CAPs, which have, in the past, delivered exceptional long-term shareholders returns even if entry prices were “seemingly high”.
I am not trying to prove that paying what appears to be high P/E is a good idea. Most of the time, it isn’t. Instead, I am trying to disprove, by showing substantial disconfirming evidence, that ignoring what appears to be an “expensive” stock is a costly mistake made by value investors— a mistake, which is underweighted by them because it does not show up in their P&L (as it’s an opportunity loss), but hugely affects their long-term net worth…
We should focus on the disconfirming evidence.
What kinds of companies deliver exceptional returns over long periods of time despite seemingly high entry prices? What was common between those companies? Are there companies having similar attributes today? And what is likely to happen to investors who buy them today and hold them for a couple of decades?
In calculating terminal growth rates of a company especially for companies with ability to raise prices, assumption of upto 4-5% means that the growth rate will not keep up with the long term inflation rate which is in the range of 8%-12% p.a. is it how it should be?
Pricing power in a business is not about beating the rate of inflation (which is an approximation of price rise for people in general) announced by the government. Instead, pricing power for a business is about ability to raise prices unilaterally (because it can), or in response to the price inflation with respect to the inputs used by that business, without fear of loss of business volume AND market share.
In 1976, Nestle India consumed 24 million kgs of milk at an average cost of about Rs 2 per kg. 34 years later, in 2010 (the last year for which quantitative info about raw materials is available), the company consumed 406 million kgs of milk at a cost of about Rs 21 per kg. Over the same period, prices paid for Skimmed Milk Powder rose from Rs 13 per kg to Rs 140 per kg and for sugar from Rs 3 per kg to Rs 34 per kg. So, prices for its three key inputs rose by about 10 times in 34 years.
Raw materials consumed as a percentage of revenues was 7.2% in 1976 while in 2010 it stood at 4.5%, so it’s margins were not adversely impacted because of input inflation. Rather, they rose DESPITE the company experiencing significant input inflation. But did the company beat the rate of inflation out there in the economy at large? Let’s see.
The rise in raw material prices of 10 times in 34 years translates into a rise of about 7.01% a year. But the consumer price index, over the same period rose by a factor of 12.6 times, implying an inflation rate of 7.7%. So, technically speaking, one of the greatest long-term value creators, failed to beat the rate of inflation…
When evaluating businesses, investors should focus on the input price inflation for that business instead of the inflation rate in the economy. Likewise, when thinking about investment returns, investors should focus on their own costs incurred to maintain or enhance their living standards, instead of worrying about the inflation rate in the economy generally.
Earlier reply – I think looking at CAP again is more of a hindsight thing rather than foresight. We had examples like airtel, kodak, xerox where CAP suddenly goes. Yes, there is discomforting evidence where 60 PE has made money, there is also discomforting evidence where Infosys, HLL, Coke did not make money from high PE even though profits increased.
Last reply – Sorry, could not understand this – shouldn’t the sales realization achieved and not the raw material price increase be compared with general inflation to say the company created value over rate of inflation or not?
You’re missing my point Ash. I am not arguing in favor of a P/E of 60. Nor an I am arguing against it. I am arguing in favor of reducing the role of current P/E ratio while evaluating investments in growing businesses having moats which are likely to deliver them long duration CAPs.
Excellent post – such stock specific examples are often missing in most value investing discussions.
I wanted to flag a point here – Mean Reversion in Graham’s concepts is in the context of Mr. Market – that an individual stock price will revert to its own intrinsic value over time, regardless of the vagaries of the market. It is in this context of market price that the book’s opening quote should be interpreted as well.
The mean reversion defined in the post as “under-performing businesses would improve and outstanding businesses would deteriorate” – seems to suggest that there exists some sort of averaging out in underlying business fundamentals (and not in price) – this may be a concept that exists elsewhere but is not the core of Graham’s use of mean reversion.
Thanks Rajiv. In my view, the core of Graham revolves around mean reversion – in business and in the stock market. Here are a few examples of Graham’s strategies which depend on mean reversion in business fundamentals (which presumably will also result in market’s weighing machine function playing out)
1. The Relatively Unpopular Large Company Theme (from The Intelligent Investor). Here is what Graham wrote:
“If we assume that it is the habit of the market to overvalue com- mon stocks which have been showing excellent growth or are glamorous for some other reason, it is logical to expect that it will undervalue—relatively, at least—companies that are out of favor because of unsatisfactory developments of a temporary nature.”
2. Low-Priced Common Stock Theme (from Security Analysis) – already discussed in the OctoberQuest transcript. Here both factors played out – reversion in fundamental performance, and a shift from a bear market to a bull one.
3. High Cost Producers and Leveraged Companies Theme (from Security Analysis). Graham writes:
“When a rise in the price of the commodity occurs, there will ordinarily be a larger advance, percentagewise, in the shares of high-cost producers than in the shares of low-cost producers. The foregoing table indicates that a rise in the price of copper from 10 to 13 cents would increase the value of Company A shares by 100% and the value of Company B and C shares by 300%. Contrary to the general impression in Wall Street, the stocks of high-cost producers are more logical commitments than those of the low-cost producers when the buyer is convinced that a rise in the price of the product is imminent and he wishes to exploit this conviction to the utmost. Exactly the same advantage attaches to the purchase of speculatively capitalized common stocks when a pronounced improvement in sales and profits is confidently anticipated.”
He is essentially saying that at the bottom of a commodity cycle investors should buy the most inefficient companies and the most leveraged ones because they will will bounce back the most when the industry turnaround happens as their profits will rise the most in percentage terms…
4. Bargain Issues – here Graham focuses on “average past earning power” and compares it with current market value and recommends stocks which have high earnings yield (i.e. low P/E) ratios based on average plus a strong balance sheet. He knows current earnings are low and the market is inefficient by focussing on near term earnings instead of the average past earnings (implying the current earnings will eventually revert back to the normal average earnings)…
Dear Prof. Bakshi,
First of all, please accept my sincere thanks for putting up the transcripts, notes, presentations and case studies prepared for BFBV class. It has been of tremendous help to many people like me in moving up the learning curve. I would like to take this opportunity to share my thought process and observations on the idea of “paying up” for values. I am struggling to still get full hang of this idea and look for your guidance on the same.
First of all, I have observed, that many a times, some people take “paying up” too far and take that as an excuse to justify any valuation saying that one has to pay up for a great business. They quote, Warren Buffett’s example of paying up for See’s candy/Coca Col/ Gillette. However, from whatever preliminary research that I have done, Buffett did not pay 40-50 times earning in any of these great businesses. He paid 12-13 times for See’s candy, 17 times for coca cola (which he considered as one of the greatest business in the world). So my first doubt is “how much is too much?”. Again, I am not alluding that one should get anchored at some P/E multiple of P/B multiple. However, beyond certain valuation, risk-reward tend not to be in favour of investor. odds of winning starts to be against the investor (though, it does not preclude the chances of winning altogether!). And I understand it correctly, the whole idea in Buffett-Munger philosophy is to invest in “mispriced bets” where odds of winning are predominantly in investor’s favour.
Secondly, my understanding is that irrespective of “investing approach” followed by any of the great value investors, the common thread that binds them is margin of safety. Now, if we decide to pay for quality, how do we ensure that there is enough MoS in the price that we are paying for? In order to determine margin of safety, one has to calculate IV of the business with foresight (and not with hindsight). In many cases where people are willing to pay 30-40 times earnings, MoS can only exist if one assumes that business continues to grow @ 20+% rate for next 15-20 years at constant and/or increasing margins with same asset deployment efficiency. Now what are the odds of this scenario playing out considering that business environment can be very dynamic and many of the factors not present today can crop up in next 5-10 years and change the game completely. Sometimes, people apply these kind of assumptions even to businesses where they clearly see that competitive intensity is rising consistently. So, I am struggling, that even after I conclude that the business is “great” business and have reasonable certainty of cash flows and ROE, what is the reasonable “Ceiling” on growth rate one should assume to arrive at IV and hence MoS?
The last question is about making capital allocation decision. I very well understand, that some of the great businesses like Asian Paints, Nestle, ITC etc have been trading at premium valuations all the while and have still been great wealth creators for investors. However, while, one is looking at “opportunity set” available, where numbers of “good” businesses ( growing at good rate, having some pricing power, operating in duopoly/oligopoly, have reasonably good moat of brand and distribution reach) are available at very reasonable price where MoS is high even after making very conservative assumptions, how should one take “capital allocation decision” on paying up high value for “great businesses”?. In other words, on the continuum of low quality/cheap business – extremely high quality/high valuation; there exist a set of businesses which are decent quality and available at very attractive price. How should one decide capital allocation approach along the continuum of businesses?
Your observations/inputs will be of great help.
Thanks & Regards,
You are welcome Dhwanil.
I agree with your first point.
Now let’s come to your second point where you’ve focused on margin of safety. One of the things that value investors have to struggle with when they are contemplating a transition from pure Graham & Dodd approach to value investing to Buffett/Munger/Fisher style is to understand that margin of safety can come from multiple sources. Graham focused primarily on only two such sources: a low price and wide diversification.
For a Buffett/Munger/Fisher style investor, margin of safety could come from many other sources as well:
First, margin of safety could come in the form of a “superior business model”— one which has a sustainable moat which makes life difficult for competitors and which enables the business to earn a high return on invested capital over a long time period (competitive advantaged period). The businesses I talked about in the OctoberQuest conference have solid business models which have proven to be extremely resilient and have enjoyed long CAPs.
A superior business model makes it a “low risk business” to be in the first place (leaving the question of price aside – let’s ignore that for now as I will address it later.) Agreed till now? Good.
Second, margin of safety comes from a bullet proof capital structure. Notice, the companies I referred to favorably in my talk all had debt-free balance sheets (except Shriram Transport Finance which has a leveraged balance sheet as its a finance company but nevertheless it’s hgot one of the strongest balance sheets in that business) with plenty of liquidity on the asset side of those balance sheets. How can this not be a source of margin of safety?
Third, margin of safety comes from high quality management— one with very good operating skills, capital allocation skills, and integrity.
And finally, margin of safety comes from a low price.
For a pure Graham & Dodd Style investor, who primarily focuses on low price, its important to not only get the margin of safety from the large difference between intrinsic value and market price. It’s also important to practice wide diversification- as an additional source of margin of safety. A Graham & Dodd investor has only two friends – a low price, and wide diversification.
In contrast, a Buffett/Munger/Fisher style investor has, as I have pointed above, have many friends. So, they can afford to pay up a bit (and maybe quite a bit), but so long as they don’t overpay, they should do quite well.
Your last question is tough. It requires one to decide if you should compromise or not. There are arguments in favor of never compromising with quality and waiting for the right opportunity to buy on a few occasions only and being in cash rest of the time. Then, there are arguments in favor of compromising a bit and not letting go of very good (but not great) opportunities because there are many many more of those than the really great ones and waiting is boring carries opportunity costs etc etc…
Different people have different approaches and I would not like to impose you with my approach, which, by the way, has also evolved over the years…
Thanks for writing.
Thanks for elaborate reply addressing each of the points. I quite appreciate it. You brought a whole new perspective that MoS for great businesses may stem from many sources other than price. I think a very valid point. However, what still remains unclear to me whether such MoS arising out of competitive advantage/strong capital structure and great management can be quantified? I do understand these factors address the issue of “certainty” of cash flow going forward and protection of “margins” even in adverse business environment. On the other hand, how will the impact of these factors be gauged in terms of “quantum” of future cash flows? Because, when one pays up 25+ PE, a investor is paying a large chunk for “presumed growth” component. Don’t you think, here is one place where making “highly optimistic assumptions” (such as growth rate of 20+% for more than a decade) may lead to situations where one is prone to make erring on aggressive side?
Again, as you very rightly said, moving from Graham style investing to Munger/Fisher style investing will require recalibration of number of concepts. Though, I am currently experimenting with “Graham style” and “GARP” but yet to taste the waters on paying up quite a bit for great businesses. But, for the sake of having “first hand” experience, would surely like to give it a try.
I will share with you my approach. maybe it will be of some help.
if I am sure of the greatness of a business, then I usually link the p/e I am willing to pay with the combination of ROE and growth rate of the company. I don’t see any point in paying 40 p/e for a company with 10% growth and 40% RoE. however, I am willing to pay 30 forward p/e for company growing at 40% and 50% ROE. of course you have to look at balance sheet, FCF, capital needs etc.
other test is – If I pay 40 p/e today, will that become 20 p/e on today’s price (preferably with say 3% yield on then dividend and current price) in say 3 years because of growth, under reasonable scenario? i think that is what will create real MoS. though action after 3 years should be on then p/e and dividend yield.
so, look at page industries – its been a 12 bagger in last 4 years. now that is a combination of 3 times p/e expansion (from 15 to 45) and 4 times eps growth (25 to 100). had it stayed at rs. 350, it would have become a 3.5 p/e and 14% yield stock – a highly unlikely scenario – giving the real MoS.
in the next 4 years, will the p/e expand 3 times (45 to 135) – 90% chance not. will eps grow by 25-30% – likely. will p/e fall/ rise/ remain same – might remain same.
so i can then decide whether to buy/ hold/ sell accordingly, % allocation, based on my opportunity set/ comfort/ return expectations.
there was a good writeup in one of motilal’s wealth creation studies. companies which have payback period of <1 are huge wealth creators, where they define payback period as current market cap divided by next 5 years profits. if you say this could be a great business, you should be able to have some idea of profits of next 5 years. note this method indirectly looks at growth and RoE.
Another point to note is p/e is only a shortcut to DCF. so if you look at P/FCF, lot of companies which look cheap on p/e basis will look expensive on P/FCF basis – which is the real metric. So we always need to look at FCF component of eps.
so the current p/e of kajaria or cera might be 15-20, but P/FCF will be 35-40, while the p/e and P/FCF of Page/ CRISIL/ old Titan are much more aligned. one can argue that these are in growth phase, so FCF will be low, but on should be cognisant of optical lower p/e.
Comments are closed.