Friday, April 24, 2015

Is there a ‘Bubble’ in Quality?

The entry of ‘quality’ and ‘moat’ into the lexicon of most investors has resulted in valuations of these businesses touching stratospheric levels. Consequently, investors of today are faced with some difficult decisions.

There seem to be two schools of thought emerging.

Camp 1 seems to believe that valuation has nothing to do with the sell decision. As long as a great ‘moat’ company with great management has a visible growth outlook, the best time to sell is quite literally, never.

To us in Camp 2, this seems like a case of forgotten lessons from the past. Investors who overpaid in Infosys and Hindustan Lever had to wait for a decade or more for their investments to start generating some returns in spite of continued growth of the underlying business. We are confident that for many of today’s darlings, this time is not going to be any different!!

Neither does that mean that we throw the baby out with the bathwater and exit ALL seemingly pricey positions. There is undoubtedly still great merit in holding on to some winners and one needs to guard against selling out too early from some huge long term compounders. Since trees will not grow to the sky, this post is an attempt to pen our thoughts towards building some framework identifying which trees to climb in the current frothy environment and which to jump off.

Implied ‘look out’ earnings growth

In two recent posts here & here, Prof Bakshi has provided us some excellent insights into his thinking on the valuation debate, while suggesting that valuation indeed has something to do with the sell decision.

To Quote:

I don’t think in terms of entry multiples. I do think about exit multiples though and never value a business at more than 20 times owner earnings ten years from now.

And Elsewhere

Also when I said 20x multiple ten years from now as maximum I will value the firm at, I mean it. Many of them are valued at 15x and some as low as 10x…

While most would have worked on ‘Implied’ DCF models, let us try to invert what Prof is saying and understand what this ‘implies’ as per our understanding. Few caveats before we start.

First, As Prof has mentioned, not all companies are suitable for this kind of evaluation. Second while he likes to apply a multiple on owner earnings as compared to stated earnings and has also in the past talked about pre tax earnings multiples, for the sake of simplicity we conduct this exercise with post tax stated earnings.

To illustrate numerically, let us say a company has an EPS in year 0 of Re 1. If we are able to visualise a growth of 20% in earnings, then in 10 years, the EPS should compound to Rs 6.19. Now if we apply an exit multiple of 20x to this earning, we arrive at a price of Rs 123.8. As a next step we need to discount this with OUR hurdle rate. Let us say this is 18%. On discounting with this hurdle rate for 10 years we arrive at a PV of 23.7 which translates essentially to a multiple of 23.7 times on Re 1 of current earnings.

What the above implies is the following. If you have a hurdle (expected) rate of 18% and are not prepared to pay an exit multiple of more than 20x on 10 years look out earnings, you should buy/hold this company with an earnings growth outlook of 20% only if it is available at less than 23.7x current earnings.

Now armed with clarity in understanding, let us fool around with some real numbers.

In today’s day and age, a PE of 23.7 is considered cheap for the kind of companies we are talking about J Let us be more realistic and analyse say Hawkins Cookers trading at 45 PE. What is my asking rate for growth in profits with an exit multiple of 20 and a hurdle rate of 18%? The answer is - I should buy into Hawkins only if I expect profits to compound at more than 28% over the next 10 years. [You can arrive at this by doing a simple goal seek function here]  This would translate into current profits of 35-40 crs growing to more than 400 crs in 10 years. This helps us put in perspective what we probably know i.e. Higher the entry multiple, higher the growth required to justify the purchase.

Let us look at some other examples to buttress this point, all for an exit multiple of 20 and a hurdle rate of 18%
Rs Crores
Company
Current PE (TTM)
Implied Asking rate for 10 year PAT growth
PAT (TTM)
Implied PAT 10 years forward
Jubiliant Food
80
36%
117
2444
Symphony
77
35%
127
2561
Page Industries
78
35%
185
3781
Bosch
71
34%
1050
19476
Astral Poly
70
34%
80
1456
Relaxo Footware
52
30%
82
1112
Motherson Sumi
51
30%
825
11067
Cera Sanitaryware
44
28%
65
745
Kitex Garments
36
25%
99
943
Data as of 23rd April 2015, approximate numbers ignoring exceptionals

It is thus quite evident that under these assumptions of exit multiple and hurdle rate, markets already seem to be pricing in fairly strong future earnings growth.

20% + CAGR of profits – is it possible?

While high profit growth over 10 years is not unheard of, we did a reality check to see how fast quality companies have been able to grow in the past. Out of a universe of roughly 5200 companies that we ran a screen on, 203 (3.9%) had ROCE of over 20% in at least 7 out of the past 9 years. There being no formal definition for judging quality, we used this filter as a proxy for ‘quality’ companies. Next we checked on the CAGR growth in PAT that these companies demonstrated over the nine year period.

PAT CAGR
No of companies
% of Quality
% of all listed stocks
<10%
44
21.7%
0.8%
10-20%
59
29.1%
1.1%
20-30%
44
21.7%
0.8%
30-40%
31
15.3%
0.6%
>40%
25
12.3%
0.5%
Total
203
100%
3.9%

Key observations from the back testing exercise:
  • Mere 2% of all companies were able to demonstrate both quality and high growth.
  • Of the quality companies more than 50% grew profits at less than 20% CAGR, which also implies that the balance grew at more than 20%. The narrow point being made here is that while markets are today sanguine that all the companies in the above table (and many others) will be able to compound profits at high rates for the next 10 years, past data suggests only an even probability of success.
  • The hyper growth companies (more than 30% CAGR) where the big money was made in the past is a mere 1% of the listed universe of the companies, almost akin to find a needle in a haystack.
But is it fair to use 18% hurdle rate?

To complicate matters further, many investors rightly or wrongly, have a hurdle rate well higher than 18%. Quite obviously Higher the hurdle rate, even higher the growth required to justify the purchase. You can play around with the numbers for yourself by choosing a hurdle rate to see how the nos change. If you have a hurdle rate of 30% [Seriously, we have come across investors with hurdle rates even higher] Kitex will deliver for you only if it is able to grow its profits at greater than 37% compounded for 10 years to nearly 2500 cr.

So if you think this to be a tall ask, you have only two choices. The first is to tone down your hurdle rate and continue to hold Kitex, the second to look for greener pastures.

Without opining on what an appropriate hurdle rate should be, it is important in our view for serious investors to have a realistic hurdle rate. Having too high a hurdle rate can force you out of some great businesses available at sensible valuations in favour of moderate businesses at cheap valuations, thereby exposing you to higher risk of permanent losses.

Why an exit multiple of 20x?

Finally if the required implied growth looks too daunting, investors always have the option of remaining invested in the hope that the market will provide an exit multiple higher than 20. To understand if it is a good idea, let us start with what Prof Bakshi writes on this matter.

Finally all expected return models as an exercise in discipline, even though I know that many businesses would be worth a lot more than 20x earnings a decade from now, given their profitability and growth potential even beyond 10 years.
(Emphasis mine)

Conservative principles of valuation suggest that one could be neutral between a ‘0 growth’ company and a high quality bond. Therefore seeking earnings yield of 6-7% or roughly 15x PE may not be irrational. Paying a higher multiple implies a growth expectation. Using the same principle, assuming no significant changes in general level of interest rates 10 years hence, it may not be unfair to give a 15x multiple to a no growth entity. Therefore, a 20x multiple already assumes some growth even beyond 10 years.

Put differently applying a 20x multiple means that in your judgement the business would be able to demonstrate growth not for 10 but maybe 20 years or beyond. Most would agree that the risk that we may end up rationalising the unknown is high. In this light applying 20x in our view is aggressive enough. Going even further may be nothing short of pure adventurism.

To summarize, while not all, valuations of some quality companies seem to have overshot their intrinsic growth potential by a fair margin. While buying today or holding on to these ideas may not result in permanent loss of capital given the inherent growth in many companies, those investing today should be prepared to face sub optimal returns over elongated periods. Investors would do well to be careful in their stock selection in the current environment.

Best of Manufactured Luck!!


Disclosure: Examples used in this post is for educational purpose only and is not intended as a buy or sell decision on any stock. Neither us nor our family members have invested in any of the mentioned examples over the past 1 year

25 comments:

  1. Great write-up and a much needed one...good use of the theory of 'Inverting'. What interests me is Prof's usage of the exit multiple. Although not sanguine on 20x, it more importantly transcends the traditional multiples based on the industry type. Guess quality of earnings, growth, management and sustainability has a lot of role to play irrespective of other factors. Never imagined a textile company commanding rich current valuations. Compounding at high internal rates of return and longevity of the same i guess are the crucial factors. Thanks for the great post guys

    ReplyDelete
    Replies
    1. Thanks Abhishek,

      We agree completely. Absolute valuation principles should have little to do with industry. You have given an example of textile. Similar examples exist in auto ancs, real estate etc. where some companies due to great management, moat, high ROE and growth are able to command premium valuations. People regularly bring in industry multiples to justify a purchase quite often on a relative basis. But relative valuation has little fundamental basis to begin with. Thanks for sharing your thoughts

      Delete
  2. Interesting post. I have not come across any other article trying to address this riddle. However, in order to do fair comparison, I think you need to do similar study of 20% CAGR and asking growth rate for a set of companies with moderate quality (say ROCE of 10% o 20%). I suspect that results will be eye opener and would show that universe of moderate quality companies is heavily laden with land mines.

    Now, let us focus on what are the options available to equity investor? First is to stick with high quality and accepting lower hurdle rate. Second is keep hurdle rate fixed and go down quality curve in trying to achieve this hurdle rate. Assuming that future is unknowable and investor is not attempting to time the market by going in to cash, what is the most optimum way for an investor? I guess sticking with quality will be the one as going down the quality would increase the risk of getting seriously injured while trying to achieve high returns.

    Would be glad to know your views.

    ReplyDelete
    Replies
    1. Hi Ankur,

      Thanks for your very relevant comments. Firstly, without running a fresh screen on moderate companies, let me reiterate that the morale of this post is not to suggest a move from quality to moderate businesses but simply attempting to build a framework on when to sell quality on the basis of valuations.

      Going down the quality curve is infact a poor choice as already mentioned in the post. The optimum thing to do would be to accept a lower hurdle rate, but not lower than say 10% because at that point fixed income alternatives start appealing. We agree that that the future is unknowable and therefore like to think in probabilistic outcomes, something like a base case, realistic case and bull case scenario. To protect against jumping off too early from quality, we would like to apply the exit multiple on the bull case and then use the low appropriate hurdle rate. So after giving an exit multiple of 20 to your bull case scenario and discounting at 10% hurdle rate, you arrive at a price of say 100. What should you do if the price today crosses 100?

      In our view you should sell. There is no case to hold if the market is making a certain payment today for your most optimistic uncertain outlook far out into the future. This we believe is not timing but prudence in decision making.

      Delete
    2. Well, I agree that valuation is a big riddle and there is probably no right or wrong answer. However, what I find intriguing in your post above is underlying assumption that exit multiple of 20 is somewhat sacrosanct figure. I know Prof Bakshi has written about this and, given the league he is in, I do understand that it is probably prudent to follow his clues. But, if you do historical back checks on P/E multiples, you will find that many of companies continue to quote at high multiples even if growth slows. Nesltes and Levers of the world fall in this category. Some continue to maintain respectable growth and keep commanding high valuation. In fact, your study above shows that roughly 25% companies in your quality samples grew over 30% pa. This is very commanding growth over a long period of time.

      So, when I sell out and move to cash, I am, in fact, trying to time the market, waiting for corrections to take position once again. Frankly, it is big judgmental call to time a secular growth story.

      There are three choices. First is continue holding quality stocks with some x% probability of getting lower than 10% returns (which implies 100-x% probability of getting higher than 10% returns). Second is move to cash and ensure 100% probability of 7% after tax returns. Third is go down quality curve. Having dismissed third option, I still think that first option is better in a sense that it gives some probabilities (100-x%) of getting better than 10% returns. Based on your study above, I would put that as roughly 25% which is not bad I would say.

      Very intelligent folks like you guys (I seriously mean it. What I am reading on your blog is pretty high class stuff) should not be prisoner of one single criteria for your sell decision. Probably, it would be worthwhile to focus on what additional filters you might apply to narrow down your search for high quality with high sustainability of growth (just like those 25% high grower businesses in your study). May be then probabilities of making more than 10% returns even at holding on high valuations would become better than 25% currently I am putting in and may be it would give more confidence to hold on even at high valuation.

      Your views as always are welcome.

      Delete
    3. Hi Ankur,

      Thanks for your thoughtful comments. Glad you liked some of our thoughts on the blog.

      We have tried to elaborate on the rationale of the exit multiple in the post itself. As discussed there, 20x is infact not sacrosanct. It is fair to pay higher, but only in the event that you are successfully able to envision high growth well beyond 10 years, maybe 20-25 years. We prefer to be conservative when asked to pay for growth that far out into the future. We like to base our decisions on absolute merits, we tend to disregard relative multiples or for that matter historical PE bands which you mention about.

      You point out three choices. There is infact a fourth, which is to move from one quality to another quality idea with better prospects but still below your hurdle rate. Most investors find this difficult to do because you need to conquer multiples emotional biases.

      Also selling decision is a process. Cash is an outcome. The right decision has to be taken. Worrying about cash creation is focusing on the outcome while ignoring the process. This is not about timing in our view.

      In most if not all things in life, it is wrong to overdo anything however good it may be. There is no reason that the same should not apply to stocks. A framework therefore needs to be developed to help identify that point of transition. This post is that attempt.

      Lastly, we wish to focus on preserving our wicket like Dravid rather than worry about how to hit the next ball for a six like Sehwag. A lot of our thinking and writing is what it is, because of this mindset.

      Delete
  3. Excellent stuff!! I have read it multiple times to memorize and ingrain it in day to day investment philosophy.

    ReplyDelete
  4. Excellent post. Enjoyed reading it, the comments, and also the conversation I just had with both the authors.

    ReplyDelete
    Replies
    1. Thank you sir. It was indeed an honour and great privilege. Look forward to your teachings in this incredible journey of learning.

      Delete
  5. The important question that follows from the expected owner earnings say 10 years down the line is - what conditions are necessary to achieve that objective - does the company need to improve margins, increase market share, reduce costs and how much the industry conditions are expected to hold . It seems with slower changing industries with limited competition will be easier to visualize and one could potentially be comfortable with higher multiples. Also good points about the hurdle rate, personally I think investors should be willing to accept a lower one for a great business over a higher one for a mediocre one because of better risk-adjusted returns and much lower chance of permanent loss of capital (which is in line with Buffett's rule no 1 of never losing money)

    Your blog is quite intriguing and insightful btw, I must be thankful to Professor Bakshi for sharing the link on Twitter. Keep up the good work guys :)

    ReplyDelete
    Replies
    1. Thank you Indraneal for sharing your thoughts. Glad you like the blog..

      Delete
  6. This is definitely not a criticism, only some thoughts:

    1) Assuming that you belong to the value investing camp, which provides that the lower the risk, the higher the return, how come you have used a higher hurdle rate to start with? I have also seen the work of other (value) investors who use a discount rate very different (read higher) from the risk-free rate (say, govt bond).

    2) Coming to high quality stocks bought at reasonable prices in the past, if we find now that current price is factoring unsustainable growth rates, your advice is to sell. However, the bigger problem is what to do with that cash. What if you do not find good enough stocks for a long period of time? Wouldn't it be more sensible to continue to hold on (with reasonably higher dividend yields because of your low cost base, and moderate growth prospects)? If we are talking about dot.com prices of technology stocks at their peak, I can understand that it was prudent to sell and make a killing because there was an absolute certainty that prices would fall. In the case of our high quality stocks (and I am not including Kitex et al) such as those in the (relatively) non-discretionary products line, high valuations are sort of given, unless of course we see a crisis like the one in 2006. It is one thing not to buy these stocks at high prices, and completely another to continue to hold on at high current prices (when your purchase price is way lower). Sell decisions are taken when we realize that our assumptions were incorrect; valid assumptions during purchase rendered invalid due to varied reasons; there is a better alternative now; or you badly need cash (this is unfortunate). Any other occasion to sell does not make sense. Of these the 3rd reason is pretty solid, but should be available.

    I am not suggesting that the current market prices are good. On the contrary, it is one of the most expensive ones and it looks like it is fast approaching to seek the second position (the first being Jan 2000).

    ReplyDelete
  7. When it comes to indexing, which one do you think is better - index mutual funds or exchange traded funds?

    ReplyDelete
  8. look at "investors are likely overstating the importance of elevated P/E levels as it relates to potential market performance in the coming months," http://www.businessinsider.in/The-stock-markets-current-valuation-tells-us-zero-about-whatll-happen-next/articleshow/47074044.cms

    ReplyDelete
  9. Great article. Sankaran Naren once said in a panel discussion that market throws different opportunities at different times. There was a time when all hyped stocks were very expensive and quality businesses with great moats were fairly cheap. It was a good time to buy them. Now quality businesses have run up very high and a fair number of good companies are available for fairly cheap. Very astute philosophy.

    ReplyDelete
  10. Great article guys. Just wanted to bring up an old article of yours, 'Licensed to speculate'. Guess it is finally time to buy the sugar stocks if we are to subscribe to the cyclical nature of sugar theory. What say?

    ReplyDelete
    Replies
    1. Thanks Ankit. Our thinking over the past couple of years has changed quite a bit and we now tend to look for quality over gruesome businesses like sugar. Lets simply say that the markets have taught us a lesson and we now wish we hadn't written that article :)

      Delete
  11. If you watch the Purinju NDTV interview, it is clear that Best quality doesn't give best return. Warren Buffet himself said, great stocks dont have to be great businesses.

    ReplyDelete
  12. Excellent work guys. Keep up the good work.

    ReplyDelete
  13. Very insightful. Your blog has quite high quality articles. Wish you guys wrote more often.
    I disagree with the view that one should be neutral b/w a high-quality bond & a no-growth company (both at yields of 6-7%). Wouldn't one prefer a high-quality bond over such a company? Shouldn't we demand higher earnings yield from a no-growth company over a AAA bond yield? A AAA bond contractually gives us the calculated yield while in equity we are still bearing some risk (de-growth).
    I think yield demanded from a no-growth company should be an inverse of our hurdle rate of investing into equities (18% in this case) which translates into a PE multiple of 5.5x.
    I may be wrong here. Would love to hear your views. Thanks.
    - Prashant

    ReplyDelete
    Replies
    1. Thanks Prashant for writing in with your very relevant insights. Thought this thread was long dead :)

      There can be two kinds of no growth companies. The high quality one (High moat able to sustain current business while not grow it, High ROCE, Great management) and the other of poor quality (possessing opposing characteristics).

      I would agree with your assessment for a poor company but not necessarily the former.

      Since the management of such companies is assumed to be of high quality, they would not be expected to burn cash without accompanying growth and much rather return most of it to shareholders. If they do not do so, in any case return ratios will trend lower over a period.

      In theory, if i were to get such a company at a 5-6 PE, i would be happy to jump into it because it implies a dividend yield of more than 15% under the assumptions above.

      Delete
  14. Would this model work for Banking sector, if the equity is not diluted for growth

    ReplyDelete
  15. In your analysis, return on capital has not been taken into account as you have considered only the high quality businesses.. and therefore little to choose between lets say 35% and 40% ... but in case lets say if we are comparing two businesses with returns 20% and 40%, available at 20x and 40x forward earnings respectively.. both the businesses are expected to be grow at similar rates in future... how would you choose between two ?

    ReplyDelete
  16. Dear Manufactured Luck Editorial Team,

    Most of the so called cases of Indian Value Stocks given in the your example-list like Symphony etc... are nothing but "intense, collusive manipulations of account books & market price" by Operators & Company management. No Indian stock can "sustainably go UP without manipulation". There is a Group of Few Big Market Operators who are doing all these manipulations since last more than 30+ years...through the help of their cronies & associates & implement their schemes through large numbers of Shell Pvt. Ltd or LLP companies.... So Only thing we need to learn is to timely ride the sectoral cycles of the markets e.g. Cement, Sugar Paper etc...MUST ALWAYS make your investment in multiple lots & sell part of it when you get 50 or 100 % profits to make your rest of the investments FREE OF COST... You can HOLD whatever is FREE ... No worries about Margin-of-safety... Repeat the Cycle ... you always get great opportunities every few years to profit from such Cyclical Up moves.... Wish you all great profit & peace of mind .... warm regards to all hard working investors... AJAY

    ReplyDelete