SIF Folio: A new strategy for selling shares (part 1)

I want to become a better seller. I’m sure many of you will agree that choosing the right time to sell a share often seems harder than knowing when to buy it. 

Indeed, I’d suggest that having a good strategy for buying but no clear process for selling is a common weakness among investors.

At a portfolio or fund level, there are two ways of making money from stocks. Firstly, you need to buy some winners. But equally importantly, you need to minimise the impact on your portfolio return from the inevitable losers. 

Over the last year, I’ve tried to address the second of these requirements by introducing two new rules to the SIF folio:

• Stop-loss rule (sell on a 25% drop)

• Sell on a profit warning

These rules have already been put to the test, with good effect. After Centamin’s gold mining plans went awry recently, I sold the shares from SIF for a -27.5% loss. At the time of writing, CEY shares are down by more than 50% from my original purchase price. 

Note: If you’re interested in understanding more about the statistical evidence supporting my profit warning rule, I’d strongly recommend Stockopedia’s The Profit Warning Survival Guide ebook. This provides excellent statistical evidence on the benefits of getting out quickly when trouble appears.

These new rules mean that my approach to selling losers has evolved considerably from my original approach of only selling after nine months, regardless of any running losses.

However, when it comes to selling profitable holdings, I haven’t really made any changes to my system since I launched SIF in April 2016.

What I do now: At present I sell shares after nine months if they no longer qualify for my SIF screen. This is the screen I use to select shares to buy. If stocks do still qualify for the SIF screen after nine months, then I keep them in the portfolio on a rolling one-month review.

This approach has certainly helped me to lock in some gains in the 4.5 years I’ve been running the portfolio. SIF is currently up by almost 30% (excluding dividends) since its inception in April 2016. 

SIF virtual fund at 16 November 2020 (Value at inception: £1m)

Including dividends, SIF is up by more than 40% since inception. 

However, using my buying criteria to govern my selling decisions has also caused me to miss out on some opportunities for big wins. Here are a few notable examples from the last year:

In the interests of transparency, you can see a full record of all completed SIF Folio trades in this Google spreadsheet.

I’ve started to believe that using buying rules to select stocks for sale isn’t ideal. I need a separate set of criteria to judge when I should drop a stock from the portfolio.

This week, I’m going to explain the broad strategy I plan to take. In a follow-up piece in the next few weeks, I’ll lay out my new selling rules in detail and explain how I’ll run the portfolio from now on.

Selling shares: possible strategies

A rule of thumb I’ve often heard mentioned for selling shares is to ask whether you’d still buy them today. This is the principle which lies behind my current selling process.

I think there’s some merit to this approach, but I can also see some downsides:

• Using the same valuation measures to buy and sell shares means that you cap gains from re-rating;

• If a stock’s quality metrics or growth rate have improved, this may justify a higher valuation without any extra downside risk;

• The trend is your friend – stocks with strong momentum scores can keep rising for longer than you might expect

In short, I think I may be letting the best stocks go too soon. That could be costly.

I’ve decided that what I need is to develop a set of criteria to use for identifying which shares to sell. I want these to help me identify stocks where the risk/reward balance is unfavourable. These are shares that I should be seeking to dispose of.

As far as I can see, there are two options which make sense.

StockRanks: I’m a big fan of Stockopedia’s StockRank system and have great respect for the results it’s achieved to date, which you can view here. Ed Croft’s NAPS portfolio is another great illustration of how well this system has worked in recent years.

However, I don’t want to end up making SIF into a pseudo-NAPS portfolio. I don’t think that would be very interesting or educational. For this reason, I’m not going to use the StockRanks to directly select which stocks to sell. What I may do is use them to order potential sale candidates, as I do in my buying screen.

A new set of rules: What I’m going to do instead is to devise a new set of rules that are designed to invert the qualities I look for when buying shares.

These new rules will cover all three of the major factors: Quality, Value and Momentum. Here are a few examples of the factors I expect to include:

• Quality: Deteriorating profitability, poor cash generation or rising leverage. Another risk may be if there’s an increased divergence between valuation and quality. I don’t want to pay BMW money for an old Dacia.

• Valuation: Pricing that’s not supported by historic or forecast earnings growth. 

• Momentum: Evidence that price momentum is weakening or that the outlook for earnings is deteriorating.

Finding the right balance between leniency and severity may not be easy. I don’t want to hold onto indifferent stocks too readily. What I want is a systematic way to identify good companies that have the potential to keep delivering. 

Should I really change the strategy?

When I launched the SIF Folio in April 2016, the remit I agreed with Ed was to try and beat the market with a rules-based system. I’ve succeeded so far, but I believe it’s possible to do better while staying with my original rules-based approach. 

SIF (blue) vs FTSE All-Share index (grey)

There’s also a second reason why I’ve decided to make a change. There are many valid approaches to investing, but I think it’s important to choose one that you’re comfortable with. Without this, my experience suggests it’s harder to make consistent decisions and to have the commitment needed to ride out difficult periods.

I’m becoming increasingly uncomfortable with the high levels of churn required by my current process. In some cases, it feels like I’m selling good stocks against my better judgement, for no reason other than a marginal statistical anomaly.

Although the main SIF folio is a virtual fund, I run a mirror portfolio in a real-money account. So I do have skin in the game and a strong motivation to do better.

I accept that in making changes to my system, I will alter the character of the portfolio slightly. Most obviously, I could end up holding onto stocks for several years, whereas previously this was very unlikely.

However, I see investing as a process of continuous learning and evolution. I think it’s time to try and take another step forward.

At some point in the next few weeks, I will finalise and share the details of my new selling system in the second part of this article.

In the meantime, I’d be very interested to hear your thoughts on the thorny topic of selling shares.

Stockopedia