When Create Y'all Average Down?
The last post explained why I intend a amount valuation is non a necessary business office of the investment process. H5N1 decent stock notation is 15 pages on the business, 1 page on the management, 1 paragraph or fifty-fifty 1 judgement on valuation.
Valuation mightiness commonly live a laid of questions along the lines of "what produce I demand to believe" to get/not teach my coin back.
But I would prefer a elementary change to this process. This is a change nosotros receive got non done good at Bronte (at to the lowest degree formally) too nosotros should produce better. And that is the enquiry of averaging down.
--
Warren Buffett is famously fond of "averaging down". If yous liked it at $10 yous should dearest it at $6. If it goes downwards "just purchase more". And inwards the value investing canon yous volition non detect that much objection to that view.
But averaging downwards has been the destroyer of many a value investor. Indeed averaging downwards is the iconic agency inwards which value investors destroy themselves (and their clients).
After all if yous loved something at $40 and yous were wrong, yous mightiness dearest it to a greater extent than at $25 too you almost as probable to live wrong, too similar it to a greater extent than yet at $12 and could as live wrong.
And earlier yous know it yous receive got doubled downwards 3 times, turning a vii percent seat into a xviii percent loss.
Do that on a few stocks too yous tin live downwards 50 percent. And inwards a bad marketplace seat that 50 percent tin live fourscore percent.
And if yous produce non believe me this has a name: Bill Miller. Bill Miller assembled a startling tape beating the S&P ever yr for 15 at nowadays years too so blew it up.
Miller had a (false) reputation as 1 of the greatest value investors of all time: In reality he is 1 of the biggest stock marketplace seat losers of all fourth dimension too a model of how non to deport inwards markets.
How non to deport is live a imitation value investor, buying stocks on which yous are wrong, too recklessly too repeatedly average down.
--
At the other destination traders who (correctly) intend that people who average downwards die. The most famous exposition of this is a photograph of Paul Tudor Jones - amongst a slice of newspaper glued to his wall stating that "losers average losers".
And yet Warren Buffett too a few of his acolytes receive got averaged downwards many times too successfully. And frankly sometimes I receive got averaged downwards to nifty success.
At to the lowest degree sometimes - the Bill Miller slogan is correct: "lowest average cost wins". Paul Tudor Jones may live a nifty trader - but he is non a spell on Warren Buffett.
--
I would dearest it if I had an encyclopaedic cognition of every mid-cap inwards Europe too could purchase the strange startlingly practiced describe organization when tiny too cheap. But the chore is as good large. The the world is complicated too I can't encompass everything.
But when I expect at tasks that tin live achieved yesteryear a four-analyst store I receive got 1 rattling high on my listing of things we can do and should do: We should get the average downwards conclusion right to a greater extent than often.
So I receive got thought well-nigh this a lot. (The implementation leaves a footling to live desired.)
--
At a rattling large picture: averaging downwards when yous are right is rattling sweet, averaging downwards when yous are incorrect is a disaster.
At the starting fourth dimension selection the enquiry so is "when are yous wrong?", but this is a airheaded question. If yous knew yous were incorrect yous would never receive got bought the seat inwards the starting fourth dimension place.
So the enquiry becomes is non "are yous wrong". That is non going to add together anything analytically.
Instead the enquiry is "under what circumstances are yous wrong" too "how would yous tell"?
--
When yous set it that agency it becomes obvious that you must non average downwards (much) on highly levered describe organization models. And looking at Buffett he is very practiced at that. He bought one-half a billion dollars worth of Irish Gaelic Banks as they collapsed. They went merely about to zero. But he did non double down. He liked them downwards ninety percent, he did non similar them to a greater extent than downwards 95 percent.
By contrast these are the stocks that Bill Miller blew upwardly on: American International Group, Wachovia, Washington Mutual, Freddie Mac, Countrywide Financial too Citigroup. They were all levered describe organization models.
By contrast yous tin in all probability safely average downwards on Coca Cola: indeed Buffett did. It is really hard to function out a realistic circumstance inwards which Coca Cola is a zero. And if it is yet growing in that location is going to live a cost at which yous are right - so averaging downwards is going to move some agency to obtaining an average cost nigh or below that price.
Of course of didactics fifty-fifty Coca Cola is non exclusively safe. You could imagine a the world where the underlying occupation was litigation - where some undercover element is establish to live a carcinogen too where the fellowship faces an uncertain futurity of lawsuits. It is non probable - too if it happens yous are going to teach at to the lowest degree some alert that this is a circumstance on which yous could live wrong. Whatever, exterior that circumstance on which yous mightiness live warned, Coca Cola is non a leveraged describe organization model dependent acre to bankruptcy too is almost exclusively unlikely to halve 4 times inwards a row. You tin average that 1 down.
Operationally levered describe organization models
Not every describe organization model is as as rubber as Coca Cola. Indeed almost every describe organization model is to a greater extent than unsafe than Coca Cola. H5N1 non financially levered mining stock can halve 5 or half dozen times. If yous receive got a mining fellowship that mines coal at $40 per tonne, has no debt too the cost is $60 a tonne it is going to be really profitable. But prices below $40 (highly possible) volition receive got profits negative. Add inwards some environmental construct clean upwardly too some closing costs too it is exclusively possible that a stock loses 95 percent of its value. Averaging downwards when downwards 40%, some to a greater extent than when it halves, too so halves 1 time to a greater extent than too it volition yet lose 2 thirds of its value. The divergence betwixt averaging materials similar that downwards too doing what Bill Miller did is only 1 of degree.
It is yet a disaster. And yous volition receive got proven Paul Tudor Jones adage: losers average losers.
Obsolescence
There is some other iconic agency that value investors lose coin - too that is technical obsolescence. Kodak was made obsolescent too was a value stock all the agency downwards to bankruptcy. The circumstances on which yous mightiness live incorrect (digital photography going to 95 percent of the market) could receive got been stated pretty clearly inwards 1999.
You mightiness affair it was worth owning Kodak as a "cigar butt stock" - enough of cash time period too bargain amongst the futurity later. There was a reasonable purchase instance for Kodak the whole agency down. But technical obsolescence is ever a agency yous should live wrong. When the threat is obsolescence you are non allowed to average down.
Bill Miller averaged Kodak down. Ugh.
---
If I could improve our formal stock notes inwards whatever agency I would similar an ex-ante description of what circumstances nosotros are allowed to average downwards a item stock, too how much.
We receive got a default at Bronte - too the default at Bronte is that nosotros receive got a maximum percent for a stock (typically say nine percent but ofttimes as depression as 3 percent depending on how nosotros assess the jeopardy of the stock) too as the fund managing director I am allowed to pass that whenever I want but I am non allowed to overspend it. If nosotros receive got a 6 percent seat amongst a nine percent loss trammel too it halves I am allowed to add together 3 percent points to a greater extent than to the exposure. But that is it. Simon, beingness the jeopardy manager, isn't specially fussed if add together the extra when the stock is downwards thirty percent of 50 percent, but I can't add together it twice. If it is a seat on which nosotros concord nosotros are allowed to jeopardy nine percent so I am allowed to jeopardy nine percent.
We volition non autumn for the value investor trap of losing xviii percent on a vii percent position.
We receive got made a change of this over time. And that is every half dozen to nine months I teach some other percent indicate to add. That is at Simon's discretion - but the thought is that the easiest agency to detect out whether yous are incorrect is to wait. After a yr or 2 the underlying occupation volition usually move public. If fourth dimension has non revealed novel data so nosotros are allowed to jeopardy more.
But nosotros tin too should produce ameliorate amongst ex-ante descriptions nether the circumstances inwards which nosotros are prepared to add together too circumstances where nosotros are not. The occupation is that yous tin current of air upwardly inwards a mindset where yous ever where yous wish to add, where yous intend the the world is against yous too yous are right too yous volition merely live proven to live right.
Clear ex-ante descriptions of the effect (which require competent describe organization analysis) mightiness assistance amongst that problem.
--
The bad instance of averaging down
The iconic bad province of affairs to average downwards is a levered describe organization model involving fraud. It is surprisingly mutual because people who run highly levered describe organization models receive got rattling rigid incentives to prevarication or to encompass it upwardly when things plough to custard. I tin intend of 2 recent examples: Valeant too Sunday Edison.
Much to my shame I added to my (small) seat inwards Sunday Edison as it fell. Ugh. But also this was a highly levered describe organization model too thence yesteryear Definition the form of house where losers average losers. I should non receive got done it - too I won't inwards future.
But the highly levered describe organization models apply fairly generally. When Bill Ackman rang Michael Pearson too asked if there was whatever fraud at Valeant he already had the incorrect mindset. Then he added to a large belongings inwards a fellowship amongst over thirty billion dollars inwards junk-rated debt. Losers average losers.
Incidentally our half dozen calendar month dominion (before yous were allowed to add) would receive got saved Mr Ackman a lot of extra losses. Time has revealed enough well-nigh Valeant. And it would receive got saved me at Sunday Edison too.
--
Whilst I intend that soul bespeak me (as per the concluding spider web log post) for a valuation on every stock is absurd, I intend it is exclusively reasonable for them to inquire "under what circumstances would yous average down". If yous can't respond that yous in all probability should non ain the stock. I should insist on it amongst every long investment.
John
Valuation mightiness commonly live a laid of questions along the lines of "what produce I demand to believe" to get/not teach my coin back.
But I would prefer a elementary change to this process. This is a change nosotros receive got non done good at Bronte (at to the lowest degree formally) too nosotros should produce better. And that is the enquiry of averaging down.
--
Warren Buffett is famously fond of "averaging down". If yous liked it at $10 yous should dearest it at $6. If it goes downwards "just purchase more". And inwards the value investing canon yous volition non detect that much objection to that view.
But averaging downwards has been the destroyer of many a value investor. Indeed averaging downwards is the iconic agency inwards which value investors destroy themselves (and their clients).
After all if yous loved something at $40 and yous were wrong, yous mightiness dearest it to a greater extent than at $25 too you almost as probable to live wrong, too similar it to a greater extent than yet at $12 and could as live wrong.
And earlier yous know it yous receive got doubled downwards 3 times, turning a vii percent seat into a xviii percent loss.
Do that on a few stocks too yous tin live downwards 50 percent. And inwards a bad marketplace seat that 50 percent tin live fourscore percent.
And if yous produce non believe me this has a name: Bill Miller. Bill Miller assembled a startling tape beating the S&P ever yr for 15 at nowadays years too so blew it up.
Miller had a (false) reputation as 1 of the greatest value investors of all time: In reality he is 1 of the biggest stock marketplace seat losers of all fourth dimension too a model of how non to deport inwards markets.
How non to deport is live a imitation value investor, buying stocks on which yous are wrong, too recklessly too repeatedly average down.
--
At the other destination traders who (correctly) intend that people who average downwards die. The most famous exposition of this is a photograph of Paul Tudor Jones - amongst a slice of newspaper glued to his wall stating that "losers average losers".
And yet Warren Buffett too a few of his acolytes receive got averaged downwards many times too successfully. And frankly sometimes I receive got averaged downwards to nifty success.
At to the lowest degree sometimes - the Bill Miller slogan is correct: "lowest average cost wins". Paul Tudor Jones may live a nifty trader - but he is non a spell on Warren Buffett.
--
I would dearest it if I had an encyclopaedic cognition of every mid-cap inwards Europe too could purchase the strange startlingly practiced describe organization when tiny too cheap. But the chore is as good large. The the world is complicated too I can't encompass everything.
But when I expect at tasks that tin live achieved yesteryear a four-analyst store I receive got 1 rattling high on my listing of things we can do and should do: We should get the average downwards conclusion right to a greater extent than often.
So I receive got thought well-nigh this a lot. (The implementation leaves a footling to live desired.)
--
At a rattling large picture: averaging downwards when yous are right is rattling sweet, averaging downwards when yous are incorrect is a disaster.
At the starting fourth dimension selection the enquiry so is "when are yous wrong?", but this is a airheaded question. If yous knew yous were incorrect yous would never receive got bought the seat inwards the starting fourth dimension place.
So the enquiry becomes is non "are yous wrong". That is non going to add together anything analytically.
Instead the enquiry is "under what circumstances are yous wrong" too "how would yous tell"?
--
When yous set it that agency it becomes obvious that you must non average downwards (much) on highly levered describe organization models. And looking at Buffett he is very practiced at that. He bought one-half a billion dollars worth of Irish Gaelic Banks as they collapsed. They went merely about to zero. But he did non double down. He liked them downwards ninety percent, he did non similar them to a greater extent than downwards 95 percent.
By contrast these are the stocks that Bill Miller blew upwardly on: American International Group, Wachovia, Washington Mutual, Freddie Mac, Countrywide Financial too Citigroup. They were all levered describe organization models.
By contrast yous tin in all probability safely average downwards on Coca Cola: indeed Buffett did. It is really hard to function out a realistic circumstance inwards which Coca Cola is a zero. And if it is yet growing in that location is going to live a cost at which yous are right - so averaging downwards is going to move some agency to obtaining an average cost nigh or below that price.
Of course of didactics fifty-fifty Coca Cola is non exclusively safe. You could imagine a the world where the underlying occupation was litigation - where some undercover element is establish to live a carcinogen too where the fellowship faces an uncertain futurity of lawsuits. It is non probable - too if it happens yous are going to teach at to the lowest degree some alert that this is a circumstance on which yous could live wrong. Whatever, exterior that circumstance on which yous mightiness live warned, Coca Cola is non a leveraged describe organization model dependent acre to bankruptcy too is almost exclusively unlikely to halve 4 times inwards a row. You tin average that 1 down.
Operationally levered describe organization models
Not every describe organization model is as as rubber as Coca Cola. Indeed almost every describe organization model is to a greater extent than unsafe than Coca Cola. H5N1 non financially levered mining stock can halve 5 or half dozen times. If yous receive got a mining fellowship that mines coal at $40 per tonne, has no debt too the cost is $60 a tonne it is going to be really profitable. But prices below $40 (highly possible) volition receive got profits negative. Add inwards some environmental construct clean upwardly too some closing costs too it is exclusively possible that a stock loses 95 percent of its value. Averaging downwards when downwards 40%, some to a greater extent than when it halves, too so halves 1 time to a greater extent than too it volition yet lose 2 thirds of its value. The divergence betwixt averaging materials similar that downwards too doing what Bill Miller did is only 1 of degree.
It is yet a disaster. And yous volition receive got proven Paul Tudor Jones adage: losers average losers.
Obsolescence
There is some other iconic agency that value investors lose coin - too that is technical obsolescence. Kodak was made obsolescent too was a value stock all the agency downwards to bankruptcy. The circumstances on which yous mightiness live incorrect (digital photography going to 95 percent of the market) could receive got been stated pretty clearly inwards 1999.
You mightiness affair it was worth owning Kodak as a "cigar butt stock" - enough of cash time period too bargain amongst the futurity later. There was a reasonable purchase instance for Kodak the whole agency down. But technical obsolescence is ever a agency yous should live wrong. When the threat is obsolescence you are non allowed to average down.
Bill Miller averaged Kodak down. Ugh.
---
If I could improve our formal stock notes inwards whatever agency I would similar an ex-ante description of what circumstances nosotros are allowed to average downwards a item stock, too how much.
We receive got a default at Bronte - too the default at Bronte is that nosotros receive got a maximum percent for a stock (typically say nine percent but ofttimes as depression as 3 percent depending on how nosotros assess the jeopardy of the stock) too as the fund managing director I am allowed to pass that whenever I want but I am non allowed to overspend it. If nosotros receive got a 6 percent seat amongst a nine percent loss trammel too it halves I am allowed to add together 3 percent points to a greater extent than to the exposure. But that is it. Simon, beingness the jeopardy manager, isn't specially fussed if add together the extra when the stock is downwards thirty percent of 50 percent, but I can't add together it twice. If it is a seat on which nosotros concord nosotros are allowed to jeopardy nine percent so I am allowed to jeopardy nine percent.
We volition non autumn for the value investor trap of losing xviii percent on a vii percent position.
We receive got made a change of this over time. And that is every half dozen to nine months I teach some other percent indicate to add. That is at Simon's discretion - but the thought is that the easiest agency to detect out whether yous are incorrect is to wait. After a yr or 2 the underlying occupation volition usually move public. If fourth dimension has non revealed novel data so nosotros are allowed to jeopardy more.
But nosotros tin too should produce ameliorate amongst ex-ante descriptions nether the circumstances inwards which nosotros are prepared to add together too circumstances where nosotros are not. The occupation is that yous tin current of air upwardly inwards a mindset where yous ever where yous wish to add, where yous intend the the world is against yous too yous are right too yous volition merely live proven to live right.
Clear ex-ante descriptions of the effect (which require competent describe organization analysis) mightiness assistance amongst that problem.
--
The bad instance of averaging down
The iconic bad province of affairs to average downwards is a levered describe organization model involving fraud. It is surprisingly mutual because people who run highly levered describe organization models receive got rattling rigid incentives to prevarication or to encompass it upwardly when things plough to custard. I tin intend of 2 recent examples: Valeant too Sunday Edison.
Much to my shame I added to my (small) seat inwards Sunday Edison as it fell. Ugh. But also this was a highly levered describe organization model too thence yesteryear Definition the form of house where losers average losers. I should non receive got done it - too I won't inwards future.
But the highly levered describe organization models apply fairly generally. When Bill Ackman rang Michael Pearson too asked if there was whatever fraud at Valeant he already had the incorrect mindset. Then he added to a large belongings inwards a fellowship amongst over thirty billion dollars inwards junk-rated debt. Losers average losers.
Incidentally our half dozen calendar month dominion (before yous were allowed to add) would receive got saved Mr Ackman a lot of extra losses. Time has revealed enough well-nigh Valeant. And it would receive got saved me at Sunday Edison too.
--
Whilst I intend that soul bespeak me (as per the concluding spider web log post) for a valuation on every stock is absurd, I intend it is exclusively reasonable for them to inquire "under what circumstances would yous average down". If yous can't respond that yous in all probability should non ain the stock. I should insist on it amongst every long investment.
John
No comments