SIF annual review: +36%, but some uncomfortable truths
What a difference a year makes. When I reviewed the portfolio’s annual performance 12 months ago, I was pleased to still be above water in the aftermath of March’s crash.
Since then, the SIF folio has had a strong year. The portfolio has gained 36%, hitting new highs and outperforming the market by 11%. This takes the portfolio’s gains over the five years since its inception in April 2016 to 56%, versus 15% for the FTSE All-Share index.
Naturally I’m pleased to have beaten the market by a material margin over half a decade. But despite this profitable performance, I’m less comfortable with the SIF strategy than I was one year ago.
In this annual review I’ll try to explain why I’m having misgivings. I’ll also look at some potential solutions.
However, I’ll start by taking a look at the portfolio’s performance over the last year. Gains have been generated by some of the biggest winners SIF has ever seen, but there have also been a few nasty losses.
SIF performance 2020/21
This may seem slightly odd timing for an annual review. The reason for this is that the SIF folio was launched on 19 April 2016. So my annual reviews tend to follow this date each year. This review covers the portfolio’s performance from 19 April 2020 to 16 April 2021.
For newer subscribers, the SIF (Stock in Focus) folio is a virtual portfolio I run here at Stockopedia. I use a set of screening rules to select stocks for the portfolio and to dictate when they should be sold.
You can find a comprehensive overview of how I run the portfolio in this piece and see the latest version of my SIF buying screen here. Finally, you can learn more about how I sell stocks here.
Moving on, here’s a summary of SIF’s performance over the last year and since inception.
% Price return | 19 April 2020 - 16 April 2021 | 19 April 2016 - 16 April 2021 | |
Cumulative | Annualised | ||
SIF folio | 36.1% | 55.5% | 9.2% |
FTSE All-Share | 25.6% | 14.9% | 2.8% |
Dividends: SIF doesn’t formally track dividends, but I maintain a duplicate record of the portfolio which does track income. Dividends did not add a meaningful amount to the portfolio’s returns in 2020/21, but they have done over the longer period since inception:
SIF 5yr total return: 71.2% / annualised: 11.4%
FTSE All-Share 5yr total return: 35.7% / annualised: 6.3% (Source: FTSE Russell/ 31 March 2021)
Comparing last year’s gain with the cumulative gain to date suggests that SIF’s performance has been somewhat lumpy. This is true. Broadly speaking, the portfolio has outperformed when the wider market has been rising. At other times, the portfolio has done little more than track the wider market.
SIF (blue) versus FTSE All-Share index (grey) since 19 April 2016
Although no strategy can be expected to outperform the market at all times, this is one area where I think there’s room for improvement. I’ll return to this later.
Finally, here’s a snapshot of the portfolio as it stands on 19 April 2021. There are 13 holdings in total, which is at the lower end of my target range:
Winners & Losers
The market rally we’ve seen since April last year has driven some of the biggest gains in SIF’s history.
Here are the top five winners from the last year - a mix of open and closed positions (as of 19 April 2021):
Volex: +145% (open position - buy write-up here)
K3 Capital: +77% (sold 30/03/21 - write-up here)
Sirius Real Estate: +71% (sold 30/03/21 - write-up here)
Record: +63% (open position - buy write-up here)
Wynnstay: +55% (open position - buy write-up here)
When the market crashed last year, I placed a temporary ban on selling shares to avoid ditching good companies at distressed prices. Events since then suggest that I should have kept the ban in place for much longer. Unfortunately, my premature return to selling in May left me forced to lock in some ugly losses.
Here are the five biggest losers from the last year (as of 19 April 2021):
Gulf Keystone Petroleum: -63% (sold 30/06/20 - write-up here)
Wilmington: -55% (sold 06/10/20 - write-up here)
Aggreko: -51% (sold 06/10/20 - write-up here. To rub salt into the wound, the temporary power provider has since attracted a takeover bid that would have left the holding in profit)
Finsbury Food: -45% (sold 28/10/20 - write-up here)
Centamin: -25% (sold 06/10/20 - write-up here)
The only one of these sales that has proved correct so far was that of Centamin, which I ditched following a profit warning. Although the gold miner’s shares now trade on quite a low rating, they’ve kept on falling and are now down by 44% on my original purchase price.
The four other positions listed above have all delivered share price recoveries and (mostly) positive operating updates since being sold. SIF would be a more valuable portfolio if I’d held onto them all.
Lessons learned after five years
So what have I learned after five years and one major stock market crash? Here are the positives:
After five interesting years in the markets, I’m happy to have beaten the index conclusively. It seems that the SIF system does have some merit.
On the whole, I’m also happy with the kind of companies I’m buying and the prices paid for them.
I enjoy investing using a structured, quantitative approach.
However, there are some negatives, too. I’ve grown increasingly unhappy with the portfolio’s trading style, which is based around a default holding period of nine months - you can see previous analysis on holding periods here).
There are three main reasons for my growing discomfort::
Philosophically, I want to be a long-term investor. In my non-SIF personal holdings, I expect to hold shares for many years, ideally indefinitely.
Mathematically, I believe I’m missing out on the opportunity to enjoy the compound gains that good companies deliver by reinvesting their earnings at high rates of return.
Financially, I dislike the relatively high trading costs I’m incurring. As I mirror the SIF holdings in a portion of my own portfolio, trading regularly has a real cash cost to me. This isn’t just an abstract consideration.
What should I change?
I think that the best way to introduce this section is to present some alternative performance statistics.
What would have happened if I had never sold any of the stocks I'd bought for SIF? As it happens, I've kept a record of this.
SIF average actual holding period gain: 7%
Average buy-and-hold gain: 28%
I should caution that for various reasons, these numbers are not exactly correct. But I'm confident they are accurate enough to be a reliable indicator. I've shared this data through a Google Sheet here for anyone who would like to explore SIF's past holdings.
What’s interesting is that for much of the portfolio’s lifespan, my holding period gains have matched or exceeded the buy-and-hold gain. But over the last couple of years, this difference has eroded and gradually reversed.
This seems to confirm my view that by selling good stocks too soon, I’m missing out on the transformative gains good companies can provide. Scalping more modest profits on a wider range of stocks is not compensating for these missed opportunities.
Of course, it's possible to argue that the current gains across the market aren't sustainable and that they may reverse. Perhaps they will. But when and by how much? On balance, I find it hard to see the gap closing sufficiently to justify my high turnover of stocks.
The new selling rules I introduced last year were an attempt to mitigate this problem. Essentially, I adapted my screening rules to allow higher valuations and slower growth on existing holdings. This was done in an effort to prevent me giving away too much upside by selling winning stocks too soon.
I think this change was a move in the right direction, but I'm not sure that it's enough.
I reckon it's time to step up the challenge. I want to gradually evolve SIF into a buy-and-hold portfolio, where I spend more time analysing potential buys and less time trading.
Rest assured, I am not going to abandon my screen-based approach for selecting stocks. On balance, I think it’s been successful and continues to have potential. However, I may tweak some buying criteria and will make more significant changes to my policy on selling stocks.
I accept that this strategy shift leaves me open to accusations of style drift. More prosaically, given the portfolio’s outperformance versus the market, I may be trying to fix something that isn’t broken.
However, I've always made clear that the SIF portfolio is an open-ended experiment whose progress I document here each week. In that context, I think it makes sense to try and improve its performance.
I’ll lay out my plans here before I make any changes to the way I run the portfolio. Watch this space.
In the meantime, I’d be very interested in your views on the issues I’ve raised or any other aspects of SIF you’d like to discuss.
Finally -- as always, thanks for reading and commenting so generously over the last year. It’s been hair-raising at times, but the quality of contributions from the Stockopedia community is always excellent and much appreciated.
Disclosure: At the time of publication, Roland owned all the shares listed in SIF.
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.
25 comments
Enjoyed reading your thought process Roland. I agree that the rules you use for selling are as important, and arguably more important, than buying.
I had a look at the spreadsheet, but I think that the analysis is flawed (specifically, comparing buy and hold average gains to holding period gains). The 7% figure and 28% figure are measuring very different things.
at least, I think they are. You don't appear to be allowing for time in the calculation, as the 7% is a return over the average holding period for the stocks you sold (i.e. purchase date to sell date), and the 28% is the average of purchase date to today. Its highly likely that returns will be higher if you hold the stock for longer. But will they be higher than the gains you made on the other stocks you bought instead?
The comparison I think you need is between the annualised return you made from your current SIF sell strategy (holding period >9mths, apply sell criteria, reapply monthly) vs. a never sell strategy (never re-apply buy criteria, hold indefinitely).
I had a bit of a ruminate on the same topic here. https://www.stockopedia.com/co.... In doing so it popped up some odd insights about how and when Stockranks work, and when they appear to work in reverse.
The test of an effective sell strategy is (I think) the return on your alternative use of capital net of frictional costs exceeds the return you would have made had you done nothing. So you have to compare over a consistent time period the return you made using your buy strategy vs. the return you would have made if you simply did nothing. These also need to allow for the frictional costs associated with dealing, stamp duty, bid/offer spread, capital gains tax etc.
Roland, I just want to add my own thanks to you for your articles. Over the years, I have come to realise that my own investing philosophy, the thing that allows me to sleep easily, even in troubling market times, has been very much informed by yours. Other writers, I read with much scepticism. You are very sound, very grounded. I would see "style drift" as calm and reasonable "style evolution" and, as such, along with your insight of it, an added benefit for your readers.
Thanks for this very helpful article. I agree with the review and examination of the underlying strategy. I have also overtraded on some shares which I liked in principle but traded on the trend only to regret selling later on, so I understand where you are coming from and encourage you to complete the process.
Giving full disclosure, I often look at the SIF and take on some of the holdings based on alignment with my objectives, the sound rationale and analysis, so I have more than a passing interest.
Hi Roland, I've given up on commenting on stockopedia but on this occasion I just wanted to say thank you for the interesting articles and evaluations which are so helpful for amateurs like me. I think that you are right to examine your strategy and there is nothing wrong with that even though it doesn't appear to need 'fixing'. I don't think you should restrict yourself to a holding period. You should hold on to a share for as long as it fulfils your criteria and performs only selling when it doesn't. I have held onto shares for years through dips and drops because I believed in them to be solid companies. I'm very glad that I did. I have analysed my own portfolio and have found that if I had used stop losses I would have been worse off in the longer term although I do think that they are a good idea. Thank you and all the best to you.
I’ve never applied strict stop losses; I evaluate the share constantly and either do nothing, buy or sell (sometime a portion) depending on what’s going on with the company, it’s valuation and the wider markets.
Hi Roland,
Really enjoyed your reports over the years and your very respectable performance. The compounding of dividends is one of the areas that are hardly ever explained. I read somewhere that 90% of returns are in the form of dividends.
Look forward to reading how your investment process evolves.
BSV
Hi BSV,
Thanks for your comment. You make an interesting point about dividends. I think what you're referring to is the extra return obtained by reinvesting dividends, rather than withdrawing the dividend cash.
Reinvesting dividends manually (or automatically through accumulation units in a fund) means that your investment capital increases each year by the value of the dividends. This extra capital will then itself generate capital gains and dividend income in future years (hopefully).
This is the compounding effect - generating returns from previous years' income. Over time, the gains from compounding snowball. These figures assume a fixed capital sum, with 5% income reinvested annually:
Of course, one would hope that share price gains over these periods would add to the total return - these numbers are gains generated by reinvesting income only.
However, in the interests of balance, I should point out that dividends are not the only way to benefit from compounding. Possibly, they aren't even the best way.
As fund manager Terry Smith is fond of pointing out, good companies can reinvest excess profits at higher levels of return than most investors.
For example, if we use our dividend cash to buy shares in a business trading at 2x book value, then each £1 of dividends only buys 50p of equity (book value).
If the company concerned generates a 15% return on equity, that means our dividend cash will effectively earn a 7.5% return.
In contrast, if the company itself is able to reinvest surplus cash at 15%, then it will do so at 1x book value, generating a 15% return on every £1 reinvested. All else equal, this will be reflected in superior share price performance to a company that distributes the surplus cash to shareholders.
Depending on circumstances, shareholders may also pay tax on dividends before reinvesting them, eroding future returns still further.
I hope this makes some kind of sense. Personally, I prefer companies which pay some dividends. This is partly because I like to have the option of income. But it's also because I like management to be disciplined in its use of surplus cash -- money paid out as dividends can't be reinvested in the business. So my hope is that a company that maintains a reliable dividend will have to be more discerning about its spending plans.Cheers,
Roland
Hi Roland,
Yes reinvested dividends compounds even better. There's a nice article link below.
https://www.thebalance.com/the...
I am agnostic about dividends generally but am positive when it is being offered.
I find Terry Smiths' thoughts on dividends confusing. As he should view his returns as an investor better than most companies, surely he should be in favour of dividends as he should be better able to allocate the capital.
I think that Mr Buffet & co might say we like getting dividends from companies we own but have no plan to give them at Berkshire, as we are better at allocating capital.
BSV
Terry Smith invests a lot in US markets where generally dividends are lower than the U.K. even some large cos there don’t pay any dividends. Dividends in the FTSE 100 are about the only growth there is, maybe that’s where the 90% returns comes from, I wonder if that is why our top 100 companies don’t grow. Elephants don’t gallop is a saying which applies to the U.K. but certainly not in the US.
I think you're right to subject your system to continuous development. Tweak an element when you see it has limitations or doesn't suit your personal style as well as you first thought. However, take care to avoid being influenced by hindsight particularly on disposals - sellers remorse is a powerful emotion. Just because the market rose after your disposal does not mean you were 'wrong' to have sold. The alternative; when your holdings tank and you don't have a stop loss, will bring similar resentment.
Be comfortable with your selling rules and accept they have to play out over bull and bear markets - bringing different results in each case, but since you don't know what tomorrow will bring, the best you can do is be happy with your rules.
Of course, long term buy and hold means there is (virtually) no circumstance when a sale is warranted so selling rules are less important. The rub is, of course, that you are going to see significant loss of capital on some holdings and that those picks of big gainers will need some heavy mupltiples to bring you a return you'll be happy enough with not to want to tweak the system again to minimise your losers and pick more winners.
Good luck in the quest! Reality TV for stock watchers!
Re: Buy and Hold
Many years ago i alighted on GB Group as a really good company. I regularly traded IT between 20p and 30p, made a bit of money and didn't i think i was clever.
Price today? 864p!
DUH!!!
C
Roland,
Thank you for posting your regular SIF results and your on-going thoughts on the strategy. It's fascinating to have a real life and evolving process which is constantly being tested in the market.
For anyone who's been investing for a good length of time, they will recognize your gradual shift towards a more "buy and hold" approach. That seems to be where we all end up eventually. However, if you are leaning towards this strategy, do you not feel that you should re-visit your criteria for share selection. If you're planning to reap the benefit of the compounding affect of some big winners, then shouldn't you be screening to identify those shares that are most likely to deliver these compounding returns over the long term. This may involve giving a higher weighting to the quality metrics over value and accepting that shares that seem "expensive" now may be your best bet for the long term. I'd be interested to know if you have plans to re-look at the selection criteria.
All the best.
Hi Roland
I would say thank you for all your articles, they are always appreciated, and well thought out and it is also a value that you are prepared to put your own money into the system and not just run as a virtual portfolio.
I think that all of us say that we try to learn more each day, to learn from each decision we make, selling low, buying high, holding too long, selling too early. So this is not a fault by yourself but the sign of a wise investor.
After all if we were all so clever, none of us would be on here we would have sold up to live of our £Squillions as my daughters would say.
Thank you
Tommy
Of the list today I hold Volex (LON:VLX) I sold Wynnstay (LON:WYN) on the last dip to book the gain, to watch it bounce though it may with a few more days prove wise in the rear view mirror! Or not...
Interesting thoughts as ever Roland and as you know when to hold and when to sell is something I've been focussing on over the last year or so.
I have had a bit of a play with your numbers in the spreadsheet and it certainly does look like the stocks you sold have mostly gone up. And by somewhere in the region of 20% on average following your sale (although the median, that minimises the impact of outliers, is c. 11%.)
However, I don't think that in isolation is such a useful view. If I have worked the right magic on your figures then of the stocks you sold more than 1 year ago the CAGR since selling averages only 2.6% (median = 2.4%).
That seems to suggest that your rules mostly made the right call to sell.
(I filtered out the stocks you sold less than a year ago as compounding CAGRs forwards came create complete nonsense - including the more recent sells shows an average CAGR of 36.5% but a median of only 4.6% !!)
I agree that a fixed term holding period is a bit arbitrary, but overall I would give your sell decisions a B+ based on this quick analysis.
I suspect that it would be pretty difficult to back-test the recent change in selling rules which gives you more scope to hold, prior to November 2020; but it might be insightful to review the stocks you have held over the last 5 months that you would have previously sold?
I'm not sure if you are edging to a hold for infinity style? Or relaxing the hold criteria further?
The problem I can see with the former, would be that you run out of funds to invest unless you have a super healthy dividend stream or are constantly feeding new funding into the portfolio. So it might not IMHO be a particularly realistic proposition.
Anyway, just food for thought, hope it is of some value.
Why not run two portfolios? SIF Buy and Hold and SIF Trading.
The difference in the SIF Trading would be editorial consideration of Price and Volume, Support and Resistance, Moving averages of various lengths, maybe 21 day, 90 day and 200 day, (shorter spans would be too demanding) and perhaps Relative strength. The tools are available and it would highlight another feature of Stockopedia.
The higher transaction costs of SIFT would be compensated by avoiding stressful draw-downs and holding higher cash balances at strategic moments.
SIFT articles could alternate with SIF B&H where there will often be nothing to do except watch the paint dry between reporting dates.
Really interesting; thank you.
Thank you Roland. There has been a lot of work put in by you in doing the research and presenting the results in a well thought out and clear argument as to why you are or are not to purchase a particular share. I and I am sure that other readers have been able to learn a lot.
But all of us change the way we invest and try to analyse what the market is doing to our investments over time. It would be good for the SIF portfolio to develop and change when reason and the opportunity present themselves.
I tend to be a buy and hold investor and I am far from the only one reading your articles. We would be interested in at least a tilt towards this angle when the shares under discussion offer this possibility.
Hi Roland, thank you for this update and your individual analytical write-ups on companies of interest; you go beyond the call of duty in running and sharing this virtual portfolio even if the research work is part and parcel of your own portfolio.
My lateral thought idea is to ask why not park SIF on existing rules and start over with a SIF2, a clean break to introduce whatever course your evolved thinking takes?
I look forward to seeing how your SIF strategy and performance develop.
Regards,
Andrew
Thanks for sharing Roland. I remain a fan of a Quantitative style of investing. I agree with you on holding for the longer term. The changes you will implement moving forward are a good thing in my opinion. Many 'super investors' adapted their style over the years so it's good thing. I would be concerned if one had to make drastic or regular changes because that would not enable one to see if the style has traction.
Shame about that K3C sell
Indeed. Although I do worry about the cyclical and acquisition risks with K3 Capital (LON:K3C), it is most likely a stock I'd have chosen to continue holding at this time.
Roland
It would be interesting to see the allocation page for the folio.
Hi Indie,
I didn't include this due to space constraints, so here it is now (as of 20/04/21). Note that SIF is currently around 30% cash, which isn't reflected below.
Here's how this breaks down in terms of styles. I'm pleased to see there are no red ones:
Hope this helps,
Roland
Hi Roland - Great update and like me you have been agonising over selling in the last year, Oh and we are both about 30% cash. Not sure how that makes me comfortable though!
Thanks, Roland. Reading your thoughts is very reassuring as the questions and dilemmas are very familiar. Thanks for being so open.