Why stop losses might take the sting out of selling losing shares
There’s a maxim in investing that you should cut your losers and run your winners. But the problem is that individual investors tend to do the opposite. Research shows that profitable stocks are often sold too soon, while losing stocks are held too long (in the hope they’ll recover) - and that damages returns. It’s a behavioural bias so well known that it has its own name - the disposition effect. The good news is that the humble ‘stop loss’ could be one way of avoiding this particular emotional trap.
What we know about selling losers
Joel Greenblatt tackles the issue of selling shares in his book, You Can Be a Stock Market Genius. In typical forthright style, he says: “The bad news is that selling actually makes buying look easy… however, here’s a tip that has worked well for me: Trade the bad ones, invest in the good ones.”
Greenblatt stresses that this advice isn’t quite as useless as it might seem! His point is to know the kind of company you’re buying and hold it for as long as it performs as you expect. That might mean cashing out early from shares that suddenly face an uncertain future. But other positions might be held for many years. The logical conclusion is that you should sell badly performing positions as soon as possible.
Greenblatt’s comments echo the advice of many successful investors. For instance, it’s been noticeable in all my recent interviews with highly successful UK investors, that they all take a similar line on selling losers. In simple terms: get out early and you’ll feel a lot better. Fund manager Mark Slater has a general principle of selling on bad news, Robbie Burns advocates strict stop-losses, while Lord Lee of Trafford also advises having a fixed plan on when a holding should be sold. I know that the subject of my next interview, Giles Hargreave, takes a similar approach.
In academic literature, the statistics on the trend for investors to hold losers speak for themselves. When I’ve written about the disposition effect in the past, I’ve mentioned work by Terrance Odean. His classic study of 10,000 accounts at an American broker between 1987 and 1993 found clear evidence of investors selling winning positions over losing positions. And they paid the price - on average the losing stock that was held went on to fall by 1.0% against the market, while the winning stock that was sold actually gained 2.4%.
Can stop losses stop this problem?
In The Little Book of Behavioral Investing, James Montier refers to Odean’s findings as a reason why investors should consider using stop losses. Put simply, a stop loss is an instruction to a broker to sell a position at a pre-arranged price. If the share price falls unexpectedly, the stop loss will usually act as a safety net by automatically selling the holding. Importantly, it takes away the stress of making a sell decision in the heat of the moment. There are a few moving parts with stop losses, which we’ve covered here. But arguably, as Montier explains: “Stop losses can act as triggers to prevent you sliding down the slippery slope of the disposition effect.”
Some of the most recent research on this subject has confirmed Montier’s view. Last year UK researchers (studying the UK market) picked over the trading records of individual investors at a brokerage between 2006 and 2009. They went looking for evidence of the disposition effect in their sample and, if it did occur, they wanted to know whether stop losses had any impact on it.
Their conclusion was that “stop losses used as part of investment decisions are an effective tool for inoculating against the disposition effect.” More specifically, they found that the investors who used stop losses tended to be less experienced - and that when they weren’t using stop losses, it was this category that often shied away from selling losing positions. By comparison, they found that more sophisticated investors who had more experience in investing were simply less prone to selling winners and holding losers.
What does it all mean?
The most recent research confirms that not everyone is susceptible to selling winners and holding losers - sophistication and experience in the market has an influence. It suggests that we all need a degree of self-awareness, knowing whether we’re prone to making rash and potentially costly decisions. If it is a risk, then using stop losses might be worth thinking about.
On a final point, there are very few subjects that provoke more debate than how, when and where to set stop losses. It’s most certainly an individual decision based on risk appetite, among other things. We’ve looked at it here, but there are some great articles taking varying views on the subject by Grindertrader here and John Rosier here.
Disclaimer
This is not financial advice. Our content is intended to be used and must be used for information and education purposes only. Please read our disclaimer and terms and conditions to understand our obligations.
26 comments
The 3 golden rules of investing
1. Cut your losses
2. CUT YOUR LOSSES
3. Run your profits
I have a very strict stop loss system.
I make running stop loss between 10% and 20%
When I buy a share I look it up on sharescope to bring the 1 year chart up. I then run an 10% stop loss line under the share price. If it does not fit I try up to 20% and to get the best fit.
In my spreadsheet make a column that check if the share price moves up, if it does then I move the stop up too.
This works for me, and that's all the agony out of shall I sell or not.
It a bit more complicated than that, but that the basics.
Theoretically, in a reasonably efficient market, applying a stop-loss to the stocks in your portfolio will give the same mean return as shorting the index the stocks are drawn from, but with higher volatility.
Nick could you expand on this or link to somewhere which describes this further, sounds interesting! Is this assuming mean reversion?
It's a consequence of MPT (Modern Portfolio Theory) and it all hinges on what "efficient" means. Basically if you pick a portfolio as a subset of an index (or by some rule, such as SR>90) then the portfolio will have a mean return the same as the index but with higher volatility. If you then apply a stop-loss rule the rule will sell stocks "at random" from your portfolio. The subset of stocks it sells will also have the same property (mean return same as index) but since you are selling it is equivalent to shorting the index.
If you argue that the stop-loss rule is not selling "at random" but in reaction to a price move which is indicative of the stock's future performance then in mathematical terms you are arguing that the market is not "efficient". But then we would have to debate whether stop-losses are in fact able to exploit an inefficient market. However before you get that far, check whether your stopped-out stocks as a group track the underlying index after being sold. If they do, you are selling at random.
A good easy-to-follow explanation of MPT is here:
http://www.moneychimp.com/articles/risk/riskintro....
(The Wikipedia explanation is way too technical to be useful as an introductory text.)
So having tried stop losses and having been badly burnt (selling £RDSB and Rio Tinto (LON:RIO) right at the bottom, since when they have made a very good recovery) on the referendum result I decided to top up a number of my shares. Stop losses would have had me sell a large part of my portfolio. As you may have seen in other posts about SCS (LON:SCS) and SafeCharge International (LON:SCH), the result has been remarkably pleasing. I'm not saying stop losses do not work (clearly in some cases they do) but it seems pretty pot luck when this is. I certainly will never leave an automatic stop loss with my broker again after one stock apparently dropped 10% for just long enough for my trade before recovering to its previous price. One piece of advice I read here I think (which I like) is to only use a stop loss on company news and not on general market movements. This is why I didn't sell on the referendum result and it has worked out well so far.
When I was a regular small-time amateur investor/trader twenty years ago I determinedly held on to a falling share over several weeks, buying more all the way to the bottom until cashing out at 2p/share (peak had been around £1 I think it was). There was no internet back then so only insiders knew that the company's assets (golf-courses) had been ludicrously over-valued and repeatedly signed off by a useless (big-name, well-paid) auditor. So I always admired people who used a stop-loss, and after reading this article was inspired to give it a go... but now I'm not so sure.
Within the space of six or seven weeks I've learned that determining to sell once a share has shed a certain percentage has worked against me on four out of four occasions (each time after a ten per cent fall). Soon after bailing out I've watched prices turn around and tick back up a half or even a couple of per cent every day or so with hardly a pause. I would have made good gains on all of them, but on only one of those four occasions did I buy back in - luckily, as it turned out. It's hard buying back in at a higher price than I had sold at a week or two before. They way I look at it now was that the value had been there all along, my timing was slightly out and I lost my nerve.
I dont think the markets have been particularly volatile since the beginning of August, but if I use a stop-loss again it will be set at 20%, if not 25%, and applied only if I can see a reason for a sliding share price.
I think other Stockopedia articles on over-trading and loss aversion are more helpful.
I use a 10% trailing stop loss in a googledocs spreadsheet which is formula driven. All I have to do is increase the stop price if the SP is rising and if "SELL" flashes red I review the situation. Unless there is no good reason for the drop, I'll sell. It is not perfect but it gives me a kick in the pants to do something.
I find when to sell a far harder decision to make than when to buy.
The key word was "guaranteed" but you also forgot the tax implications too after I am so dead, I don't care who else benefits!
Incidentally, thinking of investment bankers driving expensive cars, I justify mine by the numbers of hours I spend researching and I presume so do they. As a matter of interest, has anyone costed the time they spend managing their portfolio and deducted it from their overall investment earnings? Or do we think of it as a hobby; the "golf that pays for itself."
but that's just the thing. They don't even need to guarantee 4% to get people to part with their money.
We give them 16%+ and half of our capital and we're grateful. lol.
When we could earn it ourselves, pay 40% tax towards supporting society, whilst giving our 60% to the ones we love or causes that mean a great deal to us. In the meantime before our inevitable death, we enjoy the benefit of the 16%+ returns.
It's a no brainer for me but each to their own.
for me a div isn't to important, I focus on growth within the share price. div can always be cut if the company isn't producing the required results, but yes its nice to have a div while im waiting for the share to trend upwards.
But the primary purpose of your house is as a home not an investment. If you own a portfolio of houses as investments it may well make sense to sell if prices are falling.
The primary purpose of my portfolio is dividend yield. I'd only move house if the benefits of living in it reduced beyond a certain level. Perhaps a more accurate scenario would be a second home I let out?
So to be clear you are saying that you wou;d invest £1m with me and you will be happy for me to only give you back £500k if I guarantee a 4% yield?
Given that Inheritance tax will take 40% once I'm dead ( which will provoke the inevitable sell-off), we're not far away. Guarantee? I like that a lot! Just think what I can do with your £40K annual blessing! You're far more generous than some equity release deals. Are you sure you want to do this?
Since switching over to Stockopediamy annualised return rate over the last three years is approx 20%.
Why would I want to give 16% away plus half of my capital?
There's a reason why investment bankers drive expensive cars.
Edit: to correct the approx symbol
The trouble with any inflexible stop loss system is surely that the entire market is fluid and prone to being messed with at any moment by global events, loose mouthing from bankers and politicians and the actions of speculators manipulating the market. I often see sudden falls in value of certain companies in my portfolio and search in vain for any logical reason. I tend, therefore, to use the criterion of yield as my trigger to sell. Because my over-riding strategy is buy and hold for dividends for reinvestment within an ISA, as long as the reason for the drop is not a significant profit warning that reduces the dividend yield below 3.6 percent of the money originally invested, my criteria are often still satisfied. i compare this situation to the one where the value of the house I live in fluctuates. While I enjoy living in the property, why would I sell it because someone else values it less?
Thanks Ben!! I'm keeping quiet LOL
NWF probably wont bounce back till the milk dairy prices go up. Its whats bringing it down. The time to buy back in is when milk goes back up
vik2001, Yes had NWF (LON:NWF) too and that went the way of Cambria Automobiles (LON:CAMB) though at least hasn't bounced back as yet...
NWF is also another example of a stock that was part of the NAPS and has kept falling. in this case fair play to stop loss being set at 20%
I am not entirely convinced about stop losses given that they can lead to selling a share which after dipping below a stop loss then recovers all the lost ground and some.... In the last year or so I have sold Staffline (LON:STAF), Avon Rubber (LON:AVON), and the latest one Cambria Automobiles (LON:CAMB). The latter share had fallen by 20% since the start of 2016 and Stockopedia's momentum indicator had fallen from 89 to 67. All this despite record UK car sales which suggested they might be losing out to competitors (?). Then, today they announced bumper sales and profits to send the share price up 10%+. Very strange and frustrating for the stop loss approach - inevitably if break the rules and hang on to share it never comes back!! No doubt someone will argue by the 'law of averages' in the long run an investor should come out ahead so maybe I am just unlucky!!
I have or had Cambria Automobiles (LON:CAMB) and Staffline (LON:STAF). Staffline (LON:STAF) I first bought at 600p, they fell to 480p, I bought some more within a few weeks they were back up to 600p and then upwards, I sold out too early last year. Cambria Automobiles (LON:CAMB) I still have but yes I was considering selling as falling but best to wait for the results and then decide. I think a lot depends on the market, in a bull market think stop losses work better, in this sort of market its difficult. Probably best to get out early and then review when its stopped falling.
I have just finished reading a book by Nicolas Darvas - Wall Street the Other Las Vagas, which details his Momentum approach and the stop loss box system he employed. His trading philosophy is summarized as; "I buy stocks when they are dear and sell as soon as they show signs of becoming cheap". No disposition effect here and a clear cut strategy to move on.
My stop loss is not quite the same and is a trigger of -15% loss from the purchase price, or if the company story and/or if the financials significantly deteriorate. Above a -15% loss would niggle at me each time I looked at my portfolio and detract me from concentrating on the positive outcome of where to invest next. For me too it's best to take the hit and move on. Ian
P.S. The Darvas book although written back in 1964 is still a worthwhile weekend read.
There seems to be very little research into this subject, but here are three papers testing slightly different stop loss approaches.
One of the problems with stop losses is the way the market makers can widen the spreads and sometimes push prices to levels they think stop losses orders might be placed at.
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=968338
http://www.lunduniversity.lu.se/lup/publication/1474565
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2407199
My interpretation of the Greenblatt quote doesn't imply stop losses. Unless your expectations are about momentum.
But does imply selling when the reasons for the original purchase prove to be mistaken.
nice article as always, always enjoy your guru screening articles to.
thanks