When the price is increasing on back of fundamentally good results and if the company is CONSISTENTLY growing 20+% YoY and and 6-7% QoQ every quarter then I would not hurry to book profits. In fact I would buy on dips, till I am fully invested. Selling a rising stock too early and not letting it reach the full potential is a mistake that is easy to make. But to do this, I will need conviction.
If the conviction is lacking then I would take of more mathematical approach:
When the price is increasing on back of fundamentally good results then I keep an eye on the valuation band.
Take for example the simplest valuation parameter of PE. I keep track the lower and higher PE bands.
I also keep track how consistent the company is in its result. If the company is consistent then I calculate its 1 year FW PE based on its TTM growth. (NOTE: 1 Year FW calculation is a tricky bit. I do it only for consistent companies).
Now if the 1 year FW PE is GREATER THAN the HIGHER PE band then I will consider booking profits.
Most important point to note is that forward estimates should be calculated only if the CONSISTENCY is high and NOT for any company.
This is the basic framework I have. Actual decision will have more inputs and fine tuning.
Alternate investment choice having a better return on investment not available
Dont need cash now
Even if company is not growing today, or in foreseeable future, is the Total Addressable Market itself expected to grow? If yes, is it expected that this company will get its fair share ( & thus past growth), or even grow faster than the overall market by taking share away from competitors?
Decision to hold ( emotional).
If this stock were to delist for next 5 years, would I be unhappy?
Decision to Sell ( Financial)
If I convert this stock into cash and sit on it for next 3 years, I lose about 25% ( -7% cost of cash per year). I am confident I will be able to, and I will buy this stock back at at least 25% down from here in next 3 years.
@deep86 You are not alone in pondering/wondering on this aspect .
The SELL decision is certainly one of the most tricky. We keep learning with every experience. I find it also has much to do with temperament. We have seen in the Capital Allocation thread the typical sell decision for us is A. Is the business capable of growing at 20-25% CAGR for next 2-3 years? If not, the case is decided there itself (assuming there are many others who can; so far that has never been the problem in our markets). B. If the business can keep growing at 20-25% cagr for next 2-3 years or double, then can the stock price also double in the same timeframes?
So here, we are addressing mostly B - the overvaluation situation.
The first extreme over-valuation situation I faced was Astral Poly (quoting at 70x plus - or almost 3-year forward earnings) 3-4 years back. The most usual advise I got was “don’t try to fix what ain’t broken”; dont try to second guess markets; we don’t know how high Mr Market can take Astral. I coudn’t be satisfied with that as I kept asking so where would you draw the line? what if the Valuations reach 100x, will you still not book any profits?? I started booking profits in Astral in chunks as earnings failed to keep pace with the expectations built-in.
Although each individual sell decision merits its own specific industry/business cross-examination, let me share some generic comments/insights from our collective experience base - perhaps some of these may resonate with you.
The comments below should mostly be taken in the context of small to mid-size emerging businesses only (my experience base).
Markets are inherently forward looking - even in bear markets. So 1-year forward is the base rate. In bull markets like current, 2-year forwards become the optimistic norm - such as what we are experiencing now for most businesses. I have seen I don’t get too worried when valuations are within 2-year forwards. When valuations exceed 2-year forwards and start approaching 3-year forwards, I find myself uncomfortable and start selling in chunks.
For me it became important to SLOT a business in my mind (Art of Valuation) and assign a stable normalised PE for the specific business - depending on past track record, business quality, management quality, and future earnings visibility - (Industry and Competitive position). I found myself doing this in B+, A, A+, A++ slots. Astral was in A category or 20-25x earnings slot and was a sell at 70x earnings 3-4 years back.
I find Industry tailwinds and specific business growths practically swing the needle the most. A business consistently growing at 40-50% when valued at say 40x earnings, can within a year of holding come to more reasonable valuations like 27-30x earnings. It no more appears that expensive. And curiously, when half the financial year is past us (like now) such high-growth consistent businesses may no more be expensive!! (except in businesses with seasonal variation between quarters)
The key to persistent high-valuations is high-growth sustaining in the near to medium term.Therefore, it becomes extremely important for me to be focused on the Industry Tailwinds, Competitive Position of the business remaining stable (NOT deteriorating) - which means high visibility into near-term growth NOT faltering. Sustained high-growth becomes the panacea for most of our mistakes (including staying put in over-valued territory).
It therefore became extremely important for me to learn to become ruthless (get out of love with my money-spinners, imitating Hitesh Patel, at first) in the dissection of the industry and the business - when deciding to stay put in over-valued territory. Learn to be ruthless in distinguishing HOPE vs VISIBILITY. So the first thing I ask in such a over-valuation situation is a) Is the Runway still large enough - for the leading players in the industry to keep running at the same speeds b) What’s the near-term evidence - if there have been speed-bumps, how severe were these c) in case of speed-bumps where I decide to keep faith, what makes me so confident that high-growth would return in the near-term (few quarters, a year)
I have also learnt to respect the fact that reversion-to-the-mean is a rule of nature - especially true for high-growth, high-profitability businesses. 8 out of 10 cases, there is plateauing after 3-4 years of high-growth. Very very few businesses continue to defy this rule consistently beyond 3-4-5 years. So if I make the case for sustained high-profitability growth beyond 3-4 years (Or, high-growth returning quickly, say in less than 1-2 years of what may seem like a blip/pause after 2-3-4 quarters of low growth/de-growth) I want to make sure to double-check the facts; make sure to engage with the skeptics and be able to reasonably rebut all the objections with consistent data-points based defence, not my hope-based opinion.
I find myself holding a different perspective from another popular (and probably financially correct) edict - that if you can’t buy a business at current levels, you have no business holding it, either. So yes, even when I am pretty confident about sustainable high-growth in a specific business (say, Bajaj Finance) Buy and Hold decisions for the same business are different for me in a bull market situation like current from a practical standpoint - when valuations are over-stretched. I find much higher margin of safety (higher risk-adjusted returns decision-making) if I am holding from lower levels, than if I were to buy afresh. I want to ensure I don’t lose capital, rather than bet on only one scenario playing out, however probable. However, when valuations are reasonable (near stable normalised levels for category slots) I often find myself averaging-up with conviction (If I find Industry is stable and growing/ and competitive position is demonstratively getting stronger) in consonance with the edict :). Contradictory behaviour - in higher-risk situation?
Once comfort level on Valuations are breached as above, I find it useful to sell not all in one go, but in meaty chunks. If discomfort is quite high, it’s more like a 20% trim, else 10%ish. Instead of taking a call on how high Mr Market can take the business to, its nice to spread out the selling - giving us a chance to admit that we might have been wrong, and pause - if the trajectory of upward valuation is steep. Witnessed this with Can Fin homes, and paused for a long long time, eventually easing out
In this recent bull run, I have been compellled to trim/book profits mostly on account of position in portfolio becoming too large for comfort. I have learnt not to let individual position sizes get bigger than 20-25% for any extended periods. With larger portfolios, I find the need to preserve capital has taken center-stage, at the cost of sacrificing some growth for more assured risk-adjusted returns.
Re: your specific business - Kajaria has proven to be a quality business. However, I have not invested/tracked Kajaria ever, so not the best person to comment on the specific case. But if i see that growth has hardly kept pace or faltered, I would be wary of that. And examine if that is the same fate being met by most of industry. If certain competitors are able to grow significantly even in this environment (FY 2017 and 1HFY18) I would like to understand the reasons behind. In such a situation it would be a mistake to assume the over-hyped (in my opinion) notion that post-GST the most organised players are slated to gain the most. We are probably seeing far more the case that the semi-organised are getting better organised faster. One would also examine the housing industry segments that are growing and that are not, and which businesses might be able to take advantage of the growing segment better. One would also like to take into account the fact that Tiles segment probably has relatively low share of replacement demand; that tiles is relatively a late-cycle beneficiary of housing market rebound - while making the case for the industry/business for next 2-3 years.
The above Mr Donald is pure gold. I have much to learn. Only the second post in recent times that I think I, and anyone, would need to read 10 times over to understand. Thank you.
Valuation based selling is one approach, like you have mentioned is what I have been using mostly. Another approach I also use is a trailing stop approach. The trailing stop approach works well for me in those companies where valuations are way above my comfort level, but are market favourites. Like Bajaj Finance you mentioned. I don’t want to out-guess the market. So, my sell point would be say a 20-25% below it’s previous high. Most of the time I have seen it allows me to ride the upturn quite nicely without selling out too early.
The 20-25% stop point will vary based on my conviction about the company and the level of under/overvaluation.
Thanks for your pointers. Trailing stop approach is novel for me.
Can you explain more. How is this useful in the context of an emerging business that slowly gets discovered and is still in the early growth phase. Did you mention applying trailing stop loss for Bajaj Finance?? In what context?
I am unable to appreciate how to use a “previous high” level for BFL - as the valuation curve has been a one way journey up and up - from about say 9x-13x earnings levels during 2013 and mid 2014, continuing to be valued higher and higher progressively each year to current 40-45x+ levels, with more consistent performances - leaving competition far behind, increasing the distance with peers every year. Current 52w high of ~2000, and a 20-25% stop loss would mean selling a chunk at 1500-1600 levels?? What am I getting wrong
Wouldn’t that be akin to second-guessing Mr Market? I find that any trailing stop loss anywhere in this early growth stage - would have generated false signals and been at odds with the growth trajectory of the business, and prove to be sub-optimal?? Same would likely be the case with most quality emerging businesses during the phase Mr Market reaches consensus on consistency of performance and superior competitive positioning/strategy, isn’t it.
Ok… so let me try to explain a little bit. What I have seen is as long as the growth keeps coming, the stock price does not correct much. We saw this in many many cases, Astral, Mayur, Poly Medicine, Symphony, Cera etc… At most the correction is 10-15% due to extraneous factors. Say I bought Astral and after a few years it’s up 5x. And valuations have moved from 12 PE when I bought to 50 PE. In my opinion it is overvalued. But the business growth is intact. So, the idea then is to not second guess the market. I don’t know if 50PE is high or low. The market may decide to give it 200PE, like it did to Infosys at one time. So, my approach is because I am 5x up, I will put a trailing stop say 30% below the high price I have seen. If my thesis is correct, it will not correct that much from the recent top, but reverse and continue it’s upward journey. And I will keep revisiting my stop upwards to that 30% mark.
The only practical problem with this approach that I have come across is what to do when the stock just flat lines and does nothing. Like PI in the last 2 years or Symphony. Then the SELL decision is based on my understanding of their next 2-3 year growth prospect (your first point).
The major advantage of this approach is I have been able to eliminate in most cases selling too early and capture the momentum in price if there is any.
Coming to the specific example of say Bajaj Finance. The reason I have a trailing stop is I feel it is overvalued at the current level. And the reason I am holding is exactly as you mentioned, because of the quality of the business, ability of the management to garner growth etc etc. But there is also the possibility that it may time correct at this level for 2-3-4 years till the earnings catch up. Or RBI comes up with some new guideline that is unfavorable to the company.
As I mentioned before, a good company will not price correct significantly, except in demon like situations,when it becomes a call one needs to take whether to keep or sell. So a 30% stop will make sure I lock in the gains. I can always get back later if the business keeps doing well.
Ah! got it. You mean - in the consistent performance phase, usually the 25-30% trigger doesn’t get breached, so you are good.
Actually in the growth phase I would be greedy - that the 25-30% trigger gets breached - allowing a Buy at 30% cheaper levels . Like we could actually buy BFL at bulk ~800 levels in Nov/Dec/Jan - think the previous high was 1180 then.
Employing stop loss trigger would have sent me conflicting signals. Wouldn’t have been able to resolve, I think.
As Donald mentioned, most of us have faced this dilemma in the past and probably will continue face the same in future. Even though it may sound too overarching…indeed there is no one size fits all approach possible. In my opinion this has to be case specific both in terms of the business that one is holding and one’s investment style and/or return expectations.
For me what has worked is, typically, overvaluation is not a good reason to sell a high quality business where I see strong entry barriers and growing strength of business model. However, any change in my underlying assumptions about the quality of business/business model/industry dynamics is DEFINITELY a trigger point for sell decision.
One of the reasons I feel that overvaluation is not a good reason to sell,especially for high quality business, is…if we look at past history of all high quality business franchises, even if they have not traded cheap in the past, they have generated an alpha by a wide margin without taking commensurate risk (of generating alpha by participating in stories where business model is still evolving or competitive advantage is yet to be established/confirmed). I did a small exercise where I looked at some high quality businesses (Asian Paints, Nestle, Pidilite, Dabur, Emami) going back to Circa 2006, I did a 10 year DCF analysis based on 2006 numbers, with some very optimistic assumptions on topline growth, margin expansion and profit to cash flow conversion, 3% terminal growth rate and 12% discounting. In most of the cases,in 2006 market cap was much above/near the DCF number even under very optimistic assumptions. Hence, if one would assume that, odds would be against one to invest in those companies at such high valuation and make high risk adjusted returns. However, contrary to that, in ensuing 10 years, those companies would have generated 20-33% CAGR + dividend.
As I mentioned, most of these five companies in 2006, market was factoring in very optimistic assumptions and hence were richly valued, however, inspite of that, one could have generated significant alpha over market by remaining invested (or even buying!!) at that point of time.
Now the cache here is to figure out…the quality of the business and ability for the business to grow for long period of time. So, if one is confident about high quality of the business, it may make sense to stay put.
Having said that, I feel, the decision above is also a function of what is the hurdle rate for investment that one have in mind. So, for a mircocap investor, who has hurdle rate of 35% CAGR, this approach may not make sense while someone with more modest expectation of 20-25% CAGR over longer period of time, it make make lot of sense. One tool that has been immensely helpful to me, and weeds out this dilemma, is the expected return framework as suggested by Prof.Bakshi. I have been clued into that as it does not look at valuation in isolation. It combines,expected growth rate, entry multiple and exit multiple, to provide a picture that is wholesome. If according to one’s assessment, the expected return is lower than hurdle rate, one can sell otherwise, one may hold.
Actually you see in this particular case the 30% stip loss did not actually get breached
My take is currently in BFL, I don’t bring any new perspective to the table. When I first bought it was not a consensus buy. Stock was available at 2x book, people were concerned about possible NPAs in consumer lending and not much institutional coverage. Today it is the darling of institutions and extremely well researched. In this type of a stock, I would also take market view as an input. If price is severely correcting (above 30%), then I want to be open to the idea that the bigger players know more than me.
So, net net, it needs to be taken on a case By case basis. But a trailing stop seems to work well in most situations for me.
My BUY/SELL decision is mostly based on a reverse DCF approach.
I typically buy when I see a big difference between the growth rate implied by the current price and what I think the base growth rate for the business can be. If the implied growth rate is 6% and I think the business can do 10-12% I BUY. Selling framework is the inverse of the above. I also use some base heuristics like Operating Cash Flow yield vs the 10 year G-Sec and FCF yield (where applicable)
Needless to say this calls for some excel modelling and discipline to make a distinction between the base case, optimistic case and pessimistic case. A pure PE approach does not work for me, nor does the funda that a stock needs to grow earnings at 20% to justify a 25 PE which is a very lazy approach that is too easy to provide insights.
Practical example - When Cera peaked out at 2950 in May 2015, my reverse DCF told me that the company had to do 20% earnings growth over the next 10 years and get an exit multiple of 35 at the end of the period to justify the price. Low probability event and I did sell some though my execution wasn’t ideal, I do keep a tail in such scenarios if I believe the business has a long enough runway ahead.
The type of business also is an important input to the reverse DCF process, if the business serves segments like FMCG where the range of outcomes is not very wide, it is not meaningful to sell out for minor incidents of overvaluation since the Terminal Value is where the juice is, not the earnings visibility for the next 2-3 years.
However if it is a cyclical business, one needs to be more conservative even at the risk of selling out too early. Classic example being Sanghvi Movers which went from 70 to 400 in 1.5 years as the company became profitable again. However the stock price is now back to 150 range since the company did a massive capex and the wind energy segment saw a fall in demand. Such businesses where the end demand is cyclical and can suddenly change has a range of outcomes that are far wider, in such cases I am happy selling out prematurely. It is another story that I could not execute very well here as well but I saved myself a whole lot of trouble by being conservative.
I always ask myself - is my valuation thought process for a business consistent with my evaluation of the quality of the business? If not mistakes will eventually be made that will cost me. If I believe something is a high quality business (irrespective whether the market prices it that way) I will give a longer rope to the business than to a business which may not rank very high in my evaluation.
This thought process is my conclusion from -
reading most of Michael Mauboussin’s papers
The investment checklist I have developed that is optimized for my own thought process and style of investing
Borrowing ideas from Morningstar framework on classifying businesses into wide/narrow/no moat
Unless all of these tie in it is very tough for me to take a conclusive call on selling. Buying I admit is way easier and something that I am more confident about
If one has to summarize, @Donald approach to selling is based on Analysis Edge, or…behaving as an owner of the business @basumallick approach is based on market edge, assuming that the information and analysis edges are already taken care of, and thus acknowledging the psychology of the market /buyer in addition to, or over-riding the behavior of owning the business @zygo23554 approach is more classical reverse DCF. @desaidhwanil on the contrary provides the dis-confirming evidence about selling on overvaluation!
When I check my notes on Cera in March 2015, I find the note that t ~2500, the stock is a screaming sell, with a upside-downside of 80%. The valuation model I used was for buying, i.e. what is the margin of safety IF I buy at 2500. So a downside of 80% is clearly very high ( obviously, i used very strict constraints!), but then the stock did go down to 1470 in Feb 2016, i.e. about 43%. But now in November 2017, its roared back to 3500. From 2500 to 3500, I still have a 40% return over 2.5 years - not great, but I will take it. Now if I had another business to buy in 2015 which would have exceeded this return as per the three edges, I would have allocated capital to that business , at the cost of Cera, assuming no additional capital allocation. Using @basumallick’s 30% trailing approach, I would have sold out at 0.7*2950 = ~2000…and thereafter either bought back Cera itself at 1470 a year later, or another business. So the unknown factor here is, would I have allocated the capital to Cera again in Feb 2016, or another stock anytime during March 2015 to Feb 2016? Given that its not easy ( for me !) to identify good businesses and good managements and the industry tailwinds being supportive, would I have still stuck to Cera ?
Some more thinking certainly on the cards for me here - thanks to the very insightful posts !
VP is full of insights. Problem for me is that I discovered it late. Used to read only printed underlined materials. Is there any way to get some of these immortal threads in Word format or PDF. Tried to copy paste but found it difficult. Some of the people here are very knowledgeable & I request them to write a Book, alone or in collaboration of various members of VP. Thanks
I tried. It shows variable pages from 10 to 98 but does not get downloaded. Once it showed first 15 pages and I was able to download it. Later on I tried to copy only Summary (37 pages) it was downloaded easily (but without replies, even if you keep open all the replies manually). Thx
Interesting approach on the stop loss.How would you take (for example Canfin which has corrected almost 30% and now bounced back sharply). Simply selling out again is like timing the market.In a bull market correction of 20-30% is a given depending upon situations. It clearly boils down to individual styles and comfort and also what has worked in the past. I am wondering how do people keep stocks for 5-10 years without selling to capture compounding benefits if we keep selling.
.
. Like we could actually buy BFL at bulk ~800 levels in Nov/Dec/Jan - think the previous high was 1180 then.