Most active investors I speak to have no real strategy for selling shares.
And if they do it’s usually a simple one, such as selling on dividend cuts or selling if a company’s share price falls by 20%.
I think this is a huge mistake.
Choosing what shares to sell and when to sell them is incredibly important. In fact, your selling strategy should be just as well-thought-through as your share-buying strategy, and just as rigorously applied.
For me, selling is all about improving the portfolio by weeding out weak and/or overpriced companies. As with the rest of my investment strategy, I follow a number of rules when selling.
Rule #1: Have a written selling strategy
The first part of your selling strategy should be to write your strategy down.
This may seem obvious, but it’s often overlooked as investors tend to focus on what to buy, because that’s a more exciting question.
What do I mean by a written strategy?
I mean it should be written down. It should be quite detailed. It should tell you when to sell. It should tell you what to sell. It should limit your ability to make ad-hoc, knee-jerk, off-the-cuff decisions.
This sort of thing isn’t to everyone’s taste, but I’m a big believer in documenting, applying and continuously improving best practice, and that applies to selling stocks as much as it does to building skyscrapers.
Rule #2: Give yourself time to think before you sell
When one of your investments declines by 25% in a single day, it’s entirely natural to want to sell immediately. In my experience though, the knee-jerk response is usually a bad response.
Of course, it depends on the specific situation, but if you’re investing in quality companies with long track records of dividend payments and growth, such rapid declines are often driven by changes to investor sentiment, rather than by changes to the company’s long-term prospects.
For me, having time to think about which shares to sell means only making one sell decision every other month.
That doesn’t mean I sit around for two months thinking about what to sell, but it does mean:
- Knowing in advance when my next sell decision will be made (e.g. at the start of September, November, etc.)
- Having plenty of time to make that decision (hours or days rather than minutes)
- Being able to ignore share prices most of the time, because if I only make one trade on one day of the month, why should I care what share prices are doing the rest of the time?
To see what this sort of bi-monthly selling schedule looks like, here are my last three sell decisions:
Rule #3: Sell the opposite of what you buy
Whatever system you use for comparing companies in order to decide which ones to buy, use the same system for selling, but in reverse.
In my case I’m looking for companies that have (among other things):
- a long and unbroken record of dividend payments
- good long-term growth
- good growth quality
- good long-term profitability
- a strong balance sheet
- a low share price relative to earnings and dividends
So when it comes to selling, I’m primarily looking to get rid of any holdings that have:
- suspended dividends for more than a year
- negative long-term growth
- inconsistent, low-quality growth
- weak long-term profitability
- lots of debt and pension liabilities
- a share price which is high relative to past earnings and dividends
Why might a portfolio which is supposed to be full of high-quality, attractively valued stocks have one or more low-quality, overvalued stocks?
The answer is that things change; companies can go from good to bad and shares can go from cheap to expensive.
In practical terms, I use the UKVI stock screen as the primary driver of both buy and sell decisions because it favours stocks with the features I’m looking for.
I analyse and then potentially buy companies that are as close to the top of the screen as I can get.
When it comes to selling (every other month, remember) I do the opposite. In other words, I look at my current holdings and look to sell the holding with the lowest stock screen rank. There’s a bit more to it than that, but that’s the basic idea.
For example, here’s a screen shot showing the three lowest-ranked holdings for the Model Portfolio:
These three companies all have attractive valuations at first glance (green in the Yield, PE10 and PD10 columns) but very weak fundamentals (red in the growth rate, growth quality, profitability and debt columns).
It is the combination of valuation and fundamental factors that matters, and for these companies, their combination is not especially attractive (although they’re not especially bad either; out of 204 stocks on the UKVI stock screen Vodafone still manages to come in 133rd place).
Because of their low stock screen rank (due to their weak growth and profitability), it’s likely I’ll sell these companies within the next year or so unless they can turn things around or I decide to give them a stay of execution (see Rule #4 below).
If you don’t use a stock screen then perhaps you use “buy rules”, where all new purchases must have (for example):
- a yield of more than 4%
- ten-year dividend growth of more than 5% per year
- ten-year average return on capital employed of 10%
- debts of less than three times earnings
Over time it would be reasonable to expect some of your holdings to end up with some combination of:
- low yield (perhaps the share price took off like a rocket)
- low growth (perhaps the company has begun to struggle)
- weak profitability (ditto)
- excessive debts (perhaps management has borrowed money to upgrade the company’s equipment)
An easy way to apply this “sell the opposite of what you buy” rule is to sell holdings that meet the fewest of your criteria.
For example, if you have holdings that fail every one of your tests then they’re going to be obvious sell candidates. But you might have holdings that only pass one or two of your buy rules, so those holdings might also be on your hit list.
This is a very different approach to the sort of knee-jerk “sell on bad news” approach that many investors use, but it’s well suited to the slow-and-steady mindset of the true long-term investor (especially when coupled with my rule on regular but infrequent selling).
Rule 4: Give yourself some wiggle room, but not a lot
Rules are great, but you have to know when to break them (or at least bend them a little).
For example, a super-simple selling system would be to occasionally sell the holding with the weakest combination of “purchase criteria”.
In my case, that would mean looking at the stock screen rank of each holding every other month and then selling the company with the lowest rank.
But that isn’t what I do.
I’m well aware that no stock screen is perfect and that sometimes a bad company can rank well on a screen, just as a good company can rank poorly.
So instead of just selling the lowest-ranked holding, most of the time I’ll look at the five lowest-ranked holding and choose which company to sell from that relatively small group.
To help me decide which company to sell, I look at additional factors such as capex intensity, acquisition history, cyclicality or geographic diversity, and even “soft” factors such as my thoughts on their future prospects.
The aim of all this is to remove the stock which appears to have the least good fit with the rest of the portfolio and its high yield, high return and low-risk goals.
This approach gives enough flexibility so that I feel like I’m actively involved in the decision-making process, rather than mindlessly following a set of mechanical rules. But it also stops me from making up the rules as I go along.
I might, for example, decide to sell the holding with the 27th lowest rank out of 30, perhaps because I’m unsure about the long-term future of its industry, or perhaps because I really like the lower-ranked holdings, despite their lower rank.
That’s a personal judgment rather than something which can be calculated from the company’s accounts, but it would still be entirely reasonable to sell the 27th lowest-ranked holding rather than the one with the absolute lowest rank.
In making that sort of decision I’m being guided by the numbers but not completely controlled by them.
And that’s fine because I think intuition and gut feel still have value, as long as they’re based on a sound, evidenced-based framework.