In the Investors Chronicle dated 17th to 23rd May 2019 with ‘The Activist Effect’ as the main headline on the front cover, on page 32 there is an article called ‘Fund Managers are human after all – that’s the problem’, which makes a very good read although it is perhaps a bit ‘academic’. I guess that is where I come in and if I am doing my ‘job’ correctly then I hope I can convert what seems academic into something that normal people can digest.
It was written by Nilushi Karunaratne and the high level summary would be that Portfolio Managers make good Buying Decisions but make poor Selling Decisions – and the interesting bit is that some of the conclusions are perhaps worth taking onboard ourselves as Private Investors (assuming you are not a Portfolio Manager reading this !!) because, contrary to what many people think, institutional investors are often no better than we are (and many are worse). And the simple fact is that human psychological biases apply whoever you are. Later in my Conclusion bit I will address what we can learn.
The article is based around an academic study called ‘Selling Fast and Buying Slow: Heuristics and Trading Performance of Institutional Investors’ by Klakow Akepanidtaworn, Rick Di Mascio, Alex Imas and Lawrence Schmidt. The study looked at the daily holdings and trades of 783 portfolios covering the period 2000 to 2016 and analysed 2.4m Buy Trades and 2m Sell Trades made by Institutional Portfolio Managers.
The Authors analysed the Trades against a ‘Counterfactual’ Portfolio made up of random Buy and Sell decisions. When a Portfolio Manager bought an Asset, more Shares of a randomly selected Asset already in the Counterfactual Portfolio were bought and when a Portfolio Manager sold an Asset, a different Asset was sold randomly from within the Counterfactual Portfolio.
The Authors discovered that when it came to Buying, the Portfolio Managers exhibited genuine skill and on average over a year, the Portfolio Managers outperformed on their Buy Trades by 1.2% per year compared to the Random Portfolio (that may not sound much but bear in mind that Average Stockmarket Returns over long periods are something like 7% a year - so the outperformance is worth having and of course that outperformance compounds in a very beneficial way over time).
Trouble is, on the Sell Trades the Portfolio Managers let themselves down as per this quote from the Article:
‘Astonishingly, Portfolio Managers could improve their performance by 0.7% a year by simply selling their stocks at random.’
Ouch !! That’s cutting.
‘So why did this discrepancy appear? On the face of it both Buying and Selling Decisions involve analysing and assessing information and using it to forecast the future – so surely there shouldn’t be such a difference?
‘The authors contend there is an “asymmetric allocation of cognitive resources” towards buying. Put simply, Portfolio Managers do not lack the skills to sell well, they just do not pay adequate attention when they do it.’
That’s actually quite damning – it’s like they enjoy Buying and put lots of effort in but when it comes to Selling they really can’t be arsed with it and they just make snap decisions to dump a position. It’s pretty obvious that has a read-across for us Private Investors and we perhaps need to put more thought and effort into Selling Decisions rather than making snap judgements or relying on ‘Heuristics’ that might not actually be helping.
The Link just below has a definition for ‘Heuristic’ from Wikipedia and the essence is a shortcut technique that is sufficient and satisfactory rather than being an optimal approach:
The bit that really stands out for me from that Wikipedia definition is this bit:
‘Heuristics can be mental shortcuts that ease the cognitive load of making a decision.’
And I think that gets to the nub of it in the context of what we are discussing with regards to Investment and Trading. The crux is that Humans are inherently ‘lazy’ and like minimal effort and easy decision making. Heuristic methods can made the Selling task simple and no stress.
I struggled to really understand the next bit but I think it is essentially saying that when relevant News is released to the Markets, then Portfolio Managers act on this well and their Portfolios as a result outperform the Counterfactual Portfolio. An obvious example of this would be if a Company releases a Profit Warning after previously doing ok, then a Portfolio Manager who sells their Position (or at least a chunk of it) is likely to outperform a Portfolio where the ‘decisions’ are random and it is of course highly probable the Stock will continue to do badly (of course this isn’t always the case).
The problem was that on days when there was no relevant information about the Stock being released, the Portfolio Managers who took action on those days did badly:
‘However, on non-announcement days, with less information to hand, Portfolio Managers retreat to heuristics to enable fast decision-making and “substantially underperform” the random strategy.’
And it’s really interesting that when it comes to the buying side of things, there is no ‘meaningful difference’ between announcement and non-announcement days.
‘In both scenarios, the Portfolio Manager’s buying decisions outperform the counterfactual, reinforcing the notion that Portfolio Managers pay more attention to purchases.’
Something that really struck me here was in the bit where it said ‘to enable fast decision-making’ – it does seem like speed and a degree of laziness or thoughtlessness is driving much of the Sell decisions. Again, that is something that we Private Investors perhaps need to think about.
The next bit looks at how the heuristic approaches actually impair the Performance of the Portfolio when it comes to Sell Decisions. Firstly they suggest that ‘limited attention’ (for this I read some degree of laziness but it is also a casual attitude and a rush to dump something so the next great Stock can be bought) means that Portfolio Managers restrict their ‘Consideration Set’ – in other words, they don’t weigh up the merits and demerits of all Stocks in the Portfolio, they just take shortcuts to reduce the Sell Candidates down to a smaller subset of the Holdings. And it seems to be Past Returns which really drive the decision – obviously this could be a mistake because a Stock that has done badly for a period is quite possibly going to be a Stock that does well in the future – in fact, simple mean reversion makes this quite likely:
‘Portfolio Managers are over 50% more likely to sell assets with extreme returns (the best and worst performing assets) than those that are just under- or overperforming. Assets with extreme returns offer investors a reason to rationalise their decisions – extreme gains may have realised their full upside potential and extreme losses may be a sign that the initial investment thesis no longer applies.’
In other words, a Portfolio Manager can easily explain why they sold a Stock that has done extremely well or extremely badly and this may need to be to their Investors in the fund via a Quarterly Update or whatever, or simply to their own internal management. And the obvious implication is that by restricting the subset of Stocks to be weighed up for selling, they are perhaps not dumping the Stocks that they actually should be selling. The Random Monkey is better at including everything in the full set of Stocks to be considered !!
The second way in which these sub-optimal selling decisions impact returns is that when it comes to selling, the Portfolio Managers tend to dump the Stocks they have less emotional commitment to. This is evidenced by them selling Stocks where they have a smaller position size and hanging on to Stocks where they have a high level of emotional commitment and have a big position that they have built up over time. This was measured by the Study’s authors by looking at a thing called ‘Active Share’ which can be found for any Fund I believe, The Active Share in effect measures the extent to which an Asset is overweighted in the Portfolio relative to the Benchmark the Portfolio uses:
‘As such, Portfolio Managers are more likely to sell low active share positions that exhibit extreme returns. However, discarding these potentially promising assets too early leads to substantial under-performance versus a random strategy.’
A crucial point here is that often a relatively small Position in the Portfolio is only small because it has been recently bought and has not yet grown by more purchases by the Portfolio Manager into something substantial. Therefore if the relatively new Asset shows extreme returns (up or down remember) then the Portfolio Manager is more likely to dump this even though it could turn out to be a big winner over time. I have to say that in my experience I have seen exactly this sort of thing happen with my own portfolio. It is certainly a strong point in favour of the idea of never taking much notice of where you bought a Stock in terms of Price paid or how much Loss or Profit you have. In reality, these things are irrelevant to whether or not you should be Selling the Stock or perhaps even buying more.
Interestingly it then goes on to say that:
‘Buying behavior correlates little with past returns…..’
So whereas Extreme Returns (or you can see this as extreme moves) are used by Portfolio Managers to inform/justify selling decisions, when it comes to buying, previous moves have little relevance.
‘The more Portfolio Managers rely on heuristic behaviour, the worse their performance. Portfolio Managers who are most prone to selling assets with extreme returns, forego almost 1.5% in returns per year.’
The next bit really chimes with me and is all about making quick snap decisions when you are stressed or worried and emotionally unbalanced – which can easily happen when Markets are tanking or whatever or if you are tired:
‘People are more likely to rely on heuristics when their attention is stretched, such as during periods of stress, and it is at these times when Portfolio Manger’s selling performance is worst.’
So that part of the text is telling us to be especially careful on those really Bad Days that we get on the Markets and it shows how we shouldn’t panic and make poorly thought through decisions and in truth it would be far better to do nothing and to go out for a walk or something !!
The next bit looks at why this happens:
‘The authors postulate it is because Portfolio Managers view selling assets as a means to an end – a “cash raising exercise for the next buying idea”. Many admit sell decisions are made in a rush, particularly when timing the next purchase.’
I really get that bit – I have put a lot of effort in recent years into being much more careful on my Selling Decisions and I continually make the point that my Number One mistake in Investing has been to sell great companies too early. This next bit is quite a damning verdict:
‘The paper suggests active funds should use Portfolio Managers to pick which assets to buy. Thereafter it might be wise, and cheaper, to hand over to the infamous blindfolded monkey…….’
The final bit of this superb article looks at what Private Investors can do to improve their performance and to avoid falling into the traps that Heuristics drive Portfolio Managers into:
Put simply, pay attention when making your selling decisions. The primary finding of this study is that poorly informed decisions inevitably lead to disappointing returns. Institutional Investors create value through educated buying and proceed to cannibalise it via ignorant selling. Particularly when portfolio rebalancing, avoid selling positions with extreme returns without re-evaluating the individual investment case and whether the underlying rationale for including it in our portfolio has changed.
That is self-explanatory I think and really it is saying take time over selling decisions and make sure you carefully consider the Investment Case and you should be doing this across your whole portfolio, not just for Stocks that have done brilliantly or awfully.
I see the sense of this next bit although I have never done it. That is something Readers can weigh up and consider if it might be a useful exercise for them to undertake:
Additionally, re-evaluate your past selling decisions and track your selling performance. Professional investors direct a large amount of resources at informing and analysing the performance of buying decisions. By contrast similar feedback is virtually non-existent on the sell side.
Alarmingly, one Portfolio Manager informed the study that once they sell an asset, they “delete the name of the position from the entire research universe”. Rather than a form of self-flagellation, revisiting your past sells can provide valuable insight into your selling behaviour and reveal the (bad) habits you have picked up.
The Article that inspired my Blog is very much concerned with Portfolio Managers and how they behave but on the basis that we are all Humans (well, I am assuming most of us are, although I do near the Monkey House at Whipsnade Zoo has many WD Readers !!) then I am sure the lessons from this topic are very relevant to how we do things as Private Investors (or Traders for that matter).
To me the simple, basic, message that can be drawn from this is that Portfolio Managers tend to put immense effort and thought and analysis and suchlike into their Buying Decisions but when it comes to Selling they take many shortcuts and employ mental Heuristics that lead to sub-standard and sub-optimal Selling Decisions. I have no doubt this applies to us Private Individuals with our activities regarding the Markets and the obvious action we must take is to put more effort into Selling Decisions in general and that covers not just deciding when to Sell a particular Stock but also thinking holistically over our Portfolios and not limiting our ‘Consideration Set’ to just a few outlying Stocks that have done extremely well or done extremely badly.
Another aspect which is thrown up by this Article is how Portfolio Managers tend to dump Stocks that they have the least Emotional Commitment to. This is something we need to be very aware of and we need to be very careful to avoid falling into a similar trap where Great Stocks that we have only recently bought are dumped in favour of something we might have held for a long time and built a big Position in.
The danger is that the Stock we are hugely in love with has delivered most of its upside and the Stock we have little commitment to might be on the verge of a big run upwards – the lesson here is that we must consider carefully all the Stocks we hold and not simply be selling them because we are bored or because we are not passionately in love with them. If we are Selling just to bring a new Stock Idea into the Portfolio then we could be making a huge mistake in thinking this new Idea is much better than Stocks we already hold (and know much better as a result). It often strikes me that people buy Stocks out of boredom – that is a very bad habit.
Another consideration is the way the Portfolio itself is constructed and we need to think hard about what any New Stock would bring to the Portfolio in terms of Diversification and Risk (both potential Downside Risk and Upside Risk) and when deciding what Stock to Sell we should also be considering this overall Portfolio viewpoint and we should perhaps not be selling something that is unique and brings useful diversification to the overall Portfolio. A Portfolio should be viewed as a Holistic thing and it is worth thinking about how certain scenarios would affect the Portfolio – for example, is your Portfolio resilient for a Recession to hit or do you really have a ‘Fair Weather Portfolio’?
I continually repeat the mantra that my biggest single error that I have made over and over for many years is to Sell Great Stocks far too early. My Investing History is littered with examples where I might have taken a very nice Profit out of a Stock but simply by having held on I would have made many multiples more on my Capital gains and I would also have picked up rising Dividends and have had lower Trading Costs and lower costs from having to hunt for New Stocks etc. I am eager to do anything I can to reduce my repetition of this error and such things as TopSlicing go a long way towards reducing the tendency I have had in the past to sell Great stocks too early.
This Article and the Blog I have created as a result of it have reinforced my obsession with addressing this issue of selling Great Stocks too early. The big stand out for me is that when it comes to Buying Decisions I put in lots of effort and consideration but when it comes to Selling, like most people, I probably put only a fraction of the time and effort in that I do for the initial buying etc. This is something that I really need to address although to be fair to myself I am getting better at letting my Great Stocks run.
All the time my thinking is evolving and I am at the stage now where my whole approach is going towards being extremely fussy about what I allow to enter into my Portfolio and I want to find Stocks that are likely to be enduring and which I can hold for perhaps 5 to 10 years if need be. I want to be in a situation where I buy a Great Stock and over time I get to know more and more about it and the last thing I want to do is to sell it.
It strikes me as very muddled thinking to take the view that you can put in lots of time on Research and Analysis about a Stock and then you can buy it and then sell it very soon afterwards so you end up buying yet another Stock. To my mind you get to know a Stock properly once you actually own it and you have held it for several years and your knowledge about the Company and how the Shares move and suchlike just builds and builds over time. To me, the idea that you can Sell a Great Stock that you know really well and then move into something else which obviously you could never know as well in such a short time, is flawed.
Obviously this brings up the knotty subject of Hedging and if anything it reinforces my view that Hedging is the way to go and that selling Great Stocks when I am cautious about the overall markets is not the way to go. But as my repeated and recent experiences show it is vital to Hedge using Stoplosses to avoid having to battle against a rising Market which is not really much fun. It is ok, if you do it in small size, but when you have a big Hedge it is painful.
And we cannot avoid the even more knotty subject of Stoplosses. Having produced this Blog and thought very hard about this I am even more convinced that Long Term Investing and use of Stoplosses are not all that compatible. An essential element of Long Term Investing is that you need immense patience and you need to put up with long periods of time when not much will be happening on a particular Stock you own but you need to give your Investments time for the thesis to play out. It is different if you are very much a Trader whereby you cannot afford to have ‘Dead’ money and it is vital that you are recycling Cash out of Stocks that are failing to work quickly for you and you are shoving that Cash into other Stocks that you think are much more likely to perform in a shorter timescale.
When you look at that definition of a Heuristic technique from Wikipedia it is obvious that a Stoploss is such a method and it is designed to make a very quick Selling decision for you and to simplify things in a big way. That of course has positives and that in itself could perhaps be a good reason to use Stoplosses, but when you consider the evidence from what Portfolio Managers have been doing over many years, it seems like the speed and snap nature of Stoplosses might not actually be the best way. Yet again it really comes down to personal preference and what you are comfortable with.
I do however see a lot of sense in the idea of ‘Sell on the first Profit Warning’ and that is a form of Stoploss in my view but is triggered not by an arbitrary and perhaps irrelevant Price Level, but rather it is triggered by a failure of some sort within the Business and often this selling at such a point can prevent further problems down the line – although of course as with everything with Investing and Trading this is not always the case.
Another possible approach regarding Stoplosses is to use them so that if a given Stoploss Level is hit then you sell half your Position. This would have the benefit of reducing your Exposure which will help a lot if the Stock continues to fall but also you will still be in the game and if and when things start to improve, you could buy a lot more and ‘Average Down’ which ultimately could work out very well. It also means that your Research and understanding of the Stock is also not wasted although of course this is arguably a ‘Sunk Cost’ and there will be times when you have got your Investment Thesis wrong and you really should be dumping the Stock. The point is that even then a Selling Decision should never be rushed and you need to put the effort in to make sure you are making the best possible Decision you can.
Anyway, however we consider various techniques that make Selling Decisions for us, it is probably worth putting some brain power into thinking carefully about this and perhaps there is some way to track performance and in effect ‘back-test’ an approach. For myself, I am very clear that Stoplosses are not the way I want to go on my Long Term Investments (but I am most definitely going to use them on Index Trades) and what this also leads to is that I need to put considerable thought and effort into any Sell Decisions I make.
I hope you found that a useful and thought-provoking bit of scribbling; I have most definitely found it a useful Blog to write.
This one discusses Averaging Down:
The Psychology of the TopChop (there are links at the bottom to the first 2 parts):
The Hedging one – had a lot of outings lately !!
This old blog discusses Stoplosses and the idea of a ‘half’ Stoploss:
The Appeal of Long Term Investing:
This Blog covers the fascinating and important concept of Compounding over time:
Welcome to my Educational Blog Page - I have another 'Stocks & Markets' Blog Page which you can access via a Button on the top of the Homepage.
Please see the Full Range of Book Ideas in Wheelie's Bookshop.