Visit Interactive Investor's Churchill China discussion board to compare strategies, share knowledge and validate decisions.
Top ranked shares for the future
By Richard Beddard | Fri, 7th July 2017 - 16:17
The Decision Engine has received an upgrade. It's automatic (partially), it's systematic (mostly), and while hydramatics are a distant dream and it may never have overhead lifters or four barrel quads, I think it's greased lightning!
Decision Engine 101
To recap: The Decision Engine scores shares according to five criteria. To receive a good rating the company's business and accounting must be reasonably straightforward, it should be resilient and adaptable, it should have a profitable record, it should be managed in the long-term interests of all shareholders, and the valuation should be reasonable. I score each of the 'soft' factors requiring human judgement on a three-point scale from zero to two.
The Decision Engine calculates the current market valuation of the share and scores it on a sigmoid scale from -2 to plus +2. The Decision Engine adds the scores for each criterion to calculate the share's score, 10 is the maximum. Then it ranks the shares.
I use the Decision Engine to help me run a model portfolio, the Share Sleuth portfolio, which is published by our sister magazine, Money Observer. It helps me make quicker, more effective decisions. I don't have to do what it says, but I use it as a guide. Companies that score more than seven are in the buy zone, companies that score less than five are in the sell zone.
The intermediate holding zone is there to ensure my 'buys' are 'obviously' better investments than my 'sells'. I can sell a company in the holding zone, but I won't generally trade unless the corresponding buy scores at least two and preferably three points higher.
The Decision Engine has been a great help, but like all good tinkerers, I will never be satisfied. Occasionally my frustrations boil over, and I shove everything aside for a day of wanton spreadsheeting.
This time, I ripped the spreadsheet that powers the Decision Engine (it has over a hundred pages) in two and redesigned the pages that rank the shares. I think I've solved two problems, one presentational, and one that, I think, gets to the heart of investing and the reasons we so often make mistakes.
Showing the whole decision
Hitherto, I've only published the top ranked shares, those that I would add to a portfolio if I were starting it today. If you've enjoyed these selections, which we've called 'shares for the future', you'll be relieved to know I'm not going to stop.
Showing only the top ranked shares tells only half the story, though. The Share Sleuth portfolio started in 2009 with a notional kitty of £30,000 which I spent in its first year.
Since then the only new money for investment has come from dividends so usually when I decide to add more shares I must reduce or expel an existing shareholding to raise the money. To explain the decisions I must rank the whole portfolio along with any new shares I'd like to add.
The column headed 'score' contains the Decision Engine's verdict. I've employed Google Sheets' default colours to emphasise whether the score is sufficient to warrant a 'buy', 'hold', or 'sell'. This time, there are eleven shares in the buyzone, or shares for the future.
The column headed '%SS' shows how much of the Share Sleuth portfolio is invested in a particular share. Top-ranked Science (SAG), for example, accounts for 5% of the portfolio. Here again, the colours have meaning. I get wary if a holding exceeds 5% (highlighted in yellow), and will almost certainly reduce it if it exceeds 10% (none do, but if a holding does exceed 10%, it will be highlighted in red).
The previous version of the Decision Engine included all the companies I am following. A second's thought ought to have told me that's not satisfactory.
I know the companies in the Share Sleuth portfolio well. In many cases I have visited them, corresponded with the executives, and read annual reports from long ago. I can feel confident in my judgements and the Decision Engine's scores.
When I start following a new company my judgements are preliminary and so, therefore are the scores. It's not a fair comparison. Sometimes, the Decision Engine was telling me to add shares I didn't actually know well enough and I was forced to go off and investigate, or find a spurious reason to downgrade the interloper, or just wing it and hope my preliminary judgements were right.
Making decisions with the Engine wasn't as efficient as I'd hoped.
That's why I've ripped the spreadsheet in two. The new Decision Engine contains only 34 shares, the 28 shares in the Share Sleuth portfolio and the six shares I have pre-committed to adding to the portfolio.
They are: James Halstead (JHD), Judges Scientific (JDG), Anpario (ANP), Churchill China (CHH), Porvair (PRV) and Quartix (QTX). Since they are not in the Share Sleuth portfolio yet, the %SS column is empty.
A second spreadsheet contains all the other shares. 30 shares I cannot yet commit to.
A share jumps into the Decision Engine only when I am satisfied I know enough to be confident I'd like to own it. All six candidates for the portfolio score highly on the four company criteria but, with the marginal exception of Judges Scientific, their valuations are too high.
Since the Decision Engine is constantly updating the valuations, I'll know if that changes. And, provided the other criteria haven't changed, a decision should be straightforward.
Behavioural bias, corrected
The idea of pre-committing to trades is an old one. I probably discovered it reading James Montier's great book, Value Investing about 10 years ago, and he got the idea from one of the best known names in investing, Sir John Templeton.
In her book, Investing the Templeton Way, Sir John's niece wrote:
There are clear psychological challenges to maintaining a clear head during a sharp sell off. One way Uncle John used to handle this was to make his buy decisions well before a sell off occurred. During his years managing the Templeton Funds, he always kept a "wish list" of securities representing companies that he believed were well run but priced too high… he often had standing orders with his brokers to purchase those wish list stocks if for some reason the market sold off enough to drag their prices down to levels at which he considered them a bargain.
Hopefully, when James Halstead, which is close to the buyzone, or Quartix, which is further away, go green I'll not feel conflicted. I'll have received a clear signal from the Decision Engine.
Getting carried away in the heat of the moment, when shares are flying high, or suffering paralysis, like Templeton feared, when shares are cheap, are two of a number of behavioural biases I expect the Decision Engine to correct.
By separately investigating five criteria, I hope to avoid the "halo effect", being drawn into dodgy investments by one alluring factor, a glamorous chief executive say, or explosive short-term growth.
And by throwing grit into the engine, by insisting that there must be clear light, a point score of two or three, between a new addition and the share it replaces, I'm hoping to avoid the granddaddy of all behavioural biases - overconfidence.
Animalcare (ANCR) has made the most noteworthy move since I last wrote about the Decision Engine, having slumped to bottom place. This is not a reflection of its prospects, it's a reflection of my ability to determine them.
The company is effectively being taken over by Ecuphar, a Belgian rival and partner. Animalcare will be a subsidiary, and the combined entity, which was previously privately held and has grown through acquisition, is a bit of an unknown quantity.
The companies tracked by the decision engine currently are:
Brokers air charters. Air Partner (AIR) new acquisitions provide training and safety services to airline industry.
Alumasc (ALU) manufactures building products mostly for the management of energy and water: drains, gutters, roofs and solar shading for example.
Supplies generic and enhanced pet medicines. Also supplies pet identification and veterinary products.
Manufactures natural feed additives that improve animal health and thereby promote growth.
Castings (CGS) manufactures cast iron parts for heavy trucks and cars, components for engines, gearboxes and differentials for example.
Manufactures hard wearing tableware for restaurants, mainly. Also markets imported retail tableware.
Cohort (CHRT) supplies defence technology and expertise. Specialises in communications, intelligence, surveillance and electronic warfare.
Colefax (CFX) designs posh wallpaper and fabric. Also decorates big houses, trades antiques, manufactures furniture.
Dart (DTG) flies holidaymakers to European destinations and provides packaged holidays. Also transports groceries in Britain.
Dewhurst (DWHT) manufactures components for lifts, keypads and railway rolling-stock, particularly pushbuttons.
Finsbury Food (FIF) bakes cakes, bread, croissants, and pies for supermarkets and cafes.
FW Thorpe (TFW) manufactures Thorlux incandescent and LED lighting systems for factories, offices and shops. Also signage, and road and tunnel lighting.
Games Workshop (GAW) designs, manufactures and retails fantasy Warhammer miniatures for collectors and gamers.
Goodwin (GDWN) manufactures large steel components for the oil and gas, construction and defence sectors. Also processes minerals for casting jewellery and tyres.
Howden Joinery (HWDN) supplies kitchens to small builders. Manufactures the cabinets.
Manufactures vinyl flooring for offices, shops, factories, hospitals, schools, sport and leisure venues.
Acquires small businesses that manufacture scientific instruments and operates them.
MS International (MSI) manufactures naval gun systems, forklift blades and petrol station forecourt structures.
Next (NXT) retails clothes and homewares.
Portmeirion (PMP) manufactures tableware. Owns Portmeirion, Spode and Royal Worcester brands as well scented candle maker Wax Lyrical.
Manufactures filters and filtration systems for fluids and molten metals.
Designs vehicle tracking systems for business with small fleets and insurance companies.
Renishaw (RSW) manufactures machine tools and robots for the automation of factories, primarily. Also surgical robots and other whiz-bang devices.
Ricardo (RCDO) designs, engineers and manufactures advanced engines and transmissions. Operates an environmental consultancy.
Advises and does scientific research and product development for customers in medical, industrial, consumer and energy industries.
Solid State (SOLI) manufactures and distributes specialist electronic components and computer systems, often used in harsh environments.
Spruce Aegis (SPRP) designs and distributes smoke alarms and carbon monoxide detectors.
System1 (SYS1) uses proprietary market research techniques to test people's emotional response to advertisements and concepts.
Treatt (TET) sources and processes essential oils, which are ingredients used in flavours, fragrances and cosmetics. Develops flavours.
Trifast (TRI) manufactures and distributes fastenings (nuts and bolts, screws, and rivets) for automotive, domestic appliance, and electronics assemblers.
Tristel (TSTL) manufactures disinfectant systems for simple instruments and surfaces in hospitals. Also supplies veterinary surgeries, and laboratories.
Victrex (VCT) manufactures PEEK, a tough and light polymer. Increasingly manufactures components made from the material.
Vp (VP.) rents out specialist equipment and tools to construction, engineering, transport, oil and gas and events industries, and individuals.
XP Power (XPP) designs, manufactures and distributes power adapters for industrial and healthcare equipment.
This article is for information and discussion purposes only and does not form a recommendation to invest or otherwise. The value of an investment may fall. The investments referred to in this article may not be suitable for all investors, and if in doubt, an investor should seek advice from a qualified investment adviser.