Weekly Commentary 20/12/21: Unexpected returns

Bruce looks at long run UK equity index returns over the decades. Index returns have been unexciting recently, but the top performing UK companies have still achieved 35-45% CAGR in the last 10 years. You just need to know what to look for.

The FTSE 100 was 7,280 down less than -0.2% versus last week. The Bank of England raised interest rates to 0.25% which caused a rally in UK banks shares, as the increase is unlikely to be passed on to household savers, hence Net Interest Margins should increase. The Nasdaq was down almost 3% and the FTSE China 50 was down almost -5% as investors tried to anticipate how much contagion there would be from Evergrande.

This time of year people normally make predictions. I thought I’d do something different and look at historic returns. I have a couple of sources* going back more than a hundred years. In the UK the simple real annual return (with gross income reinvested) is +6.1%, so £100 invested in 1900 would be worth c £18,000 today, adjusted for inflation. The nominal amount (ie unadjusted for inflation) would be c. £2.5m according to Barclays.

There’s considerable volatility around the simple average annual +6.1% (the range is -55% to +137% in 1975.) Those two extreme years were both in the 1970s as the stock market fell -69% peak to trough in the secondary banking crisis before then recovering to grow at +24% CAGR real from 1975 onwards. The standard deviation of returns is 20%, so in any particular year investors are unlikely to experience the +6.1% return. Or put a different way: most years’ returns are unlike the “average” year. Below is a histogram of returns that I put together in Excel, using Barclays Equity Gilt data.

Negative real returns (ie below the level of inflation) do occur frequently: 39% of the years in the Barclays 120-year sample (or twice in five years). Although +6.1% is the annual average, the most frequently occurring real return is between 0 and minus 10%. I wouldn’t have expected that.

Last week I wrote that smaller UK companies (Numis Small Cap Index back to the 1950s), have done better than larger companies. The relative progress of small caps was not a consistently steady grind year over year though. In bear markets, smaller illiquid stocks are hit harder. That should be an opportunity for the well-prepared amateur investor with spare cash, because the professionals have to deal with outflows (hence forced sellers) at exactly the wrong time. So if we do have another sell off in 2022 be prepared to see it as an opportunity. Rather than try to guess what the index does (or inflation, interest rates, Q.E. tapering, unemployment or whether Russia invades the Ukraine) I think we are better off spending our time looking for companies with good financials, that offer asymmetric upside if things go well.

I think you really need to use a screening tool like SharePad to do this. AIM was up +5.6% CAGR nominal in the last 10 years, but there were plenty of companies that did spectacularly well. Somero led the pack returning +48% CAGR over 10 years, and the top 10 all returned greater than +35%.

Those top 10 (see chart on the next page) are not particularly high profile stocks, even after such strong performance. I don’t think that you could have found them by just reading the newspapers. But somehow I own three, the most recent purchases being Impax in April 2016 and SDI in July 2016. I am also a long-term holder of Games Workshop +36% and Creightons +47% CAGR, which would have been on the list except that they’re not quoted on AIM, but the LSE Main Market. NB GAW is the only one I’ve owned for the full 10Y, but all the others I’ve owned for several years.

On a 5-year view Somero hasn’t done so well (130th best performer), but I still own Impax (3rd) SDI (13th) and SLP (14th best 5Y performer) all of which I’ve owned for that full 5 year time period, though I sold a few of my Impax at 1299p earlier this year after it had generated a 30x return for me.

This was an unexpected return for me, I’m contrarian by nature and try to find unpopular, underappreciated companies. So I’ve been playing around with a binomial distribution model (which is a complicated way of saying there is a probabilistic way of calculating expected success/failure) where you know the probability (for example: owning a top 10 performing AIM stock is 10/740), and can flex the model for N number of trials (number of stocks you buy) and the model spits out the expectation of success (that is, the chance of owning zero, 1,2,3 or 4 stocks in the top 10). Intuitively it makes sense that as you buy more shares, you’re more likely to own one (or more) top 10 performer, and the model simply quantifies that. There’s obviously a trade-off though, because owning a less concentrated portfolio dilutes return (you’re better to own concentrated positions in your best performers). I’m still playing with the model, and if it reveals some unexpected insights or ways to think about this trade-off, then I intend to write it up next week.

Looking at that list above of +30% CAGR AIM stocks (Eurasia Mining being the exception) they are not speculative resources stocks, mining, oil & gas exploration, graphene, biotech or Quantum Blockchain. Instead Somero does concrete screed, YouGov market research surveys, Water Intelligence plumbing, CVS is vet surgeries, Marlowe governance/risk management and Victoria floor tiles. I quoted Buffett last week who said ten years ago he thought he could make +50% CAGR investing in smaller stocks, and I think the graph above supports that in an interesting way. These top performers do seem to be the sort of thing the Sage would buy concentrated positions in and hold for years, if the 91 year old ever decided that it was time to retire from his day job and start investing in AIM.

This week I look at Next Fifteen the digital marketing company and Midwich, the electronics distributor, plus a brief comment on the Boohoo profit warning and Creightons H1 results which they have scheduled to come out between Christmas and New Year.

Merry Christmas.

Next Fifteen Q3 trading update

This digital marketing company announced a positive trading update for Q3 Aug Oct with +24% 9M organic revenue growth year-on-year. Q3 vs Q3 last year was better than that (+26% organic, +38% including acquisitions). They say that the strong performance has continued into Q4.

Operating Segments They split the business into four divisions:

i) Customer Engagement 56% of revenue. MBooth a New York Agency that specialises in social media.

ii) Customer Delivery 22% of revenue. Account Based Marketing, which seems to be used for selling to companies and B2B marketing, rather than individuals. Their brands include Activate, Agent3, Twogetter and Shopper Media Group. This is the highest margin segment (36% adj operating profit margin vs group 23%) was showing the highest organic revenue growth (+48%) at H1.

iii) Customer Insights 11% of revenue. Market research surveys and data analytics, primarily through Savanta and MSI, which specialises in US healthcare and financial services.

iv) Business Transformation 12% of revenue. Also enjoyed a strong H1 with organic revenues up +47% seems closest to traditional management consulting, orientated towards the tech IPO market and P.E.

Balance sheet This is another people business growing by acquisition (like K3Capital, RBGP, SFOR and Alpha Financial Markets Consulting) with deferred consideration and earnouts. They completed a bank refinancing in their Q3, with a Revolving Credit Facility (RCF) for £60m, with a £40m accordion option.

I noticed several companies mentioning according options. The thinking is that when working capital is particularly uncertain, it suits both the borrower and the bank to have some flexibility to increase a line of credit incrementally (hence accordion because the musical instrument can be stretched or compressed), which explains why we are seeing a few companies borrowing in this way. Contingent consideration and other acquisition liabilities of £63m (recorded at Fair Value) are split out directly on the face of the balance sheet. The notes to the accounts do mention that there’s considerable uncertainty about this liability (if revenues are higher than expected, earn-out obligations will also be higher) but I couldn’t find a maximum amount that the company might have to pay.

NFC had £6.6m of net cash at the end of July, intangible assets were £184m versus shareholders’ equity of £120m. So tangible book value is negative £64m. That’s not a problem if cashflow remains strong (net cash from operating activities was £27m at H1). I find one really useful SharePad feature is to be able to scroll back using the “←Prev” button, to during the last financial crisis in 2007-8: Next Fifteen’s revenue grew over that time and the company remained profitable which gives me some reassurance.

Shareholders Good institutional ownership: Octopus own 14.3%, Liontrust 13.7%, Aviva 8.5%, Abrdn 6.4%, Slater Investments 6%.

Valuation The shares are trading on 22x FY Jan 2023 and 19x FY Jan 2024. While revenue has doubled in the last 4 years from £171m in FY Jan 2017, RoCE has struggled to get into double digits. So this is an example of a stock that is generating strong top line growth, but profitability can’t keep up.

Opinion Technology is changing marketing, because companies are more willing to spend money when they can see a “Return on Investment” on their advertising budgets and use data analytics to support their decisions. NFC seems similar to SFOR, which I wrote about last week, but with significantly more attractive valuation metrics. That might be because Sir Martin Sorrell is good at presenting his investment case or he may have been more shrewd with acquisitions than Next Fifteen.

I’m wary of this narrative around data, analytics, marketing. For years people have spoken about how great the data insights from Dunn Humby, the data analytics firm that Sir Martin Sorrell talked about on Mello last week (now owned by Tesco) were. But this doesn’t show in Tesco’s profit margin or share price performance.

Not every problem can be solved with data When I was a management consultant, in our head office in West London Dunn Humby was on the floor below us. The project managers that I used to work with ignored the designated parking spaces, and frequently parked their BMWs in Edwina Dunn’s parking space. She used to complain, but consultants have a thick skin and don’t care about upsetting people. So both Tesco’s profitability and Edwina’s occupied parking space show that Dunn Humby’s recipe of data and analysis can’t solve every class of business problem.

Midwich Trading Update to 31 Dec

This video equipment distributor with a December year end said that trading was stronger than expected. The Board expects that adj FY PBT will be not less than £30 million (more than double £14m adj PBT in FY 2020, around the 2019 adj PBT £31.2m). The shares were up +10% on the day of the RNS.

History Midwich was founded in 1979, with a particular focus on disk drives for the BBC and Acorn Electron home computers. That puts it in illustrious company of semiconductor designer ARM, which also started out with Acorn computers (the original A in ARM stood for “Acorn”, but they later changed this to “Advanced”). In the mid-1990s Midwich shifted focus to Audio Visual (AV) distribution to trade buyers. Revenues peaked at £200m in 2002, but it ran into difficulties and the computer and peripheral distribution was closed down. In the mid-2000s, it was back on the front foot and started growing by acquisition and expanding internationally (revenues grew from £134m in 2006 to £314 in 2015).

Listed on AIM in May 2016, at 208p raising £74m two thirds of which went to the selling shareholders (Stephen Fenby and Anthony Bailey) and one third issuing new shares and raising money for the company. This valued the group at £165m on admission. Since then it’s continued to grow by acquisitions, most recently eLink in Germany and NMK in UAE and Qatar. I find it very helpful that they’ve set out this history clearly on their website. They now have 20K of trade customers (professional AV integrators and IT resellers) and their products are used by commercial and educational institutions rather than household consumers.

Though you might expect a seller of AV equipment to benefit from Covid, they need trade shows to take place in order to showcase their wares and make sales. These events have been delayed and cancelled so they have not been a “covid winner”. The shares fell by half from 630p 2 years ago to trough just above 310p in October last year before recovering with the “vaccine rally”.

Balance Sheet As you’d expect from a business that has grown by acquisitions, the balance sheet is top heavy with intangible assets and goodwill, which at £76m exceed shareholders equity, so the tangible book value is negative. There is some deferred consideration too, but less than £10m so not large in the context of the group. Net debt ex. leases was £56m at the end of June, and they say that they have access to facilities of £180m.

At the H1 there were some very large negative movements in working capital, so that cash profits (PBT with depreciation, amortisation and share based payments added back) was £16m, but cash from operating activities was minus £8.7m. The commentary says that H1 is normally more working capital intensive, plus due to chip related supply chain difficulties they decided to increase inventory to ensure continuity of supply.

Trade and other payables due within 12 months were £137.5m at H1, which is larger than inventories of £118m, and receivables of £127m. For context H1 sales were £390m.

Ownership Stephen Fenby, the Group Managing Director, who joined as Finance Director in 2004, and has been MD since 2010 still owns 19.5%. Abrdn 11.7% and Octopus 11.0%, Granular Capital, Fidelity and Liontrust all own greater than 5%.

Valuation The shares are trading on 20x PER FY Dec 2023F, given the balance sheet I can’t see investors paying a higher multiple for this stock, because of the nature of where they’re situated in the value chain. So I think any further share price progress will need to be driven by continued improvement in fundamental accounting numbers, rather than a multiple re-rating. That does seem possible though.

Opinion MIDW has a low gross margin of 15%, as the nature of the business means that value is easier to capture by brands which Midwich is distributing. However SharePad shows an encouraging average RoCE of 20% since IPO, if we exclude 2020 as being an exceptional year when you’d expect Midwich to be loss-making (statutory PBT FY 2020 was minus £1m). I think this has potential to grow as AV sales pick up as organisations realise that they may have to invest in AV equipment, but I’m not very familiar with this type of company, so not a high conviction idea for me.

Boohoo Trading Update to end of November

The online fast fashion retailer released a disappointing trading update for the 3 months to the end of November, due to customers unexpectedly sending back their online purchases. These are the sort of unexpected returns you want to avoid as an investor. They were not helped by continued disruption to the international Boohoo business, and significant ongoing pandemic-related cost inflation. Possibly we can add this to the list of BOTB, G4M, FDEV and WINE as Covid winner turned re-opening losers. But Boohoo did not have a particularly good 2020, and has not taken part in the vaccine rally either. I think problems are less driven by the pandemic, and more to structural forces on profitability.

Sustainable Returns? The last 3 financial years RoCE has averaged an impressive 30% at Boohoo (versus negative for Ocado and ASOS 12% avg last 3 years.) It strikes me that online retail is a difficult business to maintain a +30% RoCE, as the long run history of ASOS (graph below) shows.

I will not try to call the bottom at Boohoo here.

Creightons H1 to September

Regular readers will know one of my favourite subjects is companies “voluntary disclosure”, ie not what management are obliged to disclose like statutory profits, but information that they think helps with their investment case. Very often when information is missing this is also a strong signal.

“I studied silence to learn the music” as Finnish operatic metal band Nightwish sang.

Creightons, the manufacturer of toiletries and fragrance brands H1 is to the end of September. The company is listed on the LSE main market rather than AIM, so the FCA listing rules apply: The half yearly report must be published as soon as possible and in any event within 90 days of the end of the period to which it relates. In exceptional circumstances, the UK Listing Authority may grant an extension of this time limit.

Leo picked up on the fact that Creighton’s results hadn’t been released yet, and ran a poll on Twitter.

Creighton’s then released a statement saying that their H1 results would be on the 30th December, just before the end of the mandatory reporting time.

Add to this that there was Director selling in October and November, most recently Martin Stevens, Deputy MD, sold at £1 per share (RNS 25 Nov). There is supposed to be a “close period” ahead of results being released of at least 30 days (previously 2 months according to the old FSA handbook).

It seems odd that the two rules are not explicitly linked, and that company insiders are allowed to sell shares between the company’s H1 results date and those results being publicly released to the market. I hold the shares, but I’m unimpressed by this behaviour. I’m not suggesting that anything illegal has gone on, but this looks to be far from best practice.

Insiders selling in October and November then the company waiting 3 months before publishing results between Christmas and New Year. Most people will be on holiday and not paying much attention. This voluntary disclosure choice does not help the company with their investment case, in my view. I will await those Creighton’s results with considerable interest. There won’t be many other companies reporting I intend to read them very closely.


The author owns shares in Creightons, Somero, Impax, Games Workshop and SDI

* A book by Antti Ilmanen called Expected Returns which looks at historic returns across geography and different asset classes. I also have an old copy of the Barclays Equity Gilt Study 2016, which has equity returns going back 121 years to 1900. Every year the bank used to give this away for free, but now it’s only available to clients. So I’ve used their data up to 2015, then used SharePad to calculate returns for the FTSE All Share up to the end of this year.

Bruce Packard

Leave a Reply

Your email address will not be published. Required fields are marked *