25 Cornerstone Value shares

The share picks from last year’s O’Shaughnessy Cornerstone Value stock screen were forged in fire. Markets were plummeting as investors digested the far-reaching implications of the global Covid-19 pandemic. Worse still, it was the type of stocks this screen seeks to identify – value stocks – that were at the front of the line for a pummelling. Indeed, at the time of publishing the screen results last year, the 25 stocks it highlighted could easily have been regarded as some kind of stock-market suicide note.

Fast forward 12 months and things feel slightly different. Not least because this screen has massively outperformed the FTSE 350 index, which I benchmark it against.

It’s particularly pleasing that this screen has had a belter over the last 12 months because it marks a fine finish to my first decade of monitoring it. The funny thing is, though, the screen is not meant to register this kind of standout performance. 

That said, my screen is not entirely true to the name on the box. Due to the diminutive nature of the UK market compared with the US, I’ve had to tinker with the classic Cornerstone Value screen that American quant-legend Jim O'Shaughnessy set out in his original edition of What Works on Wall Street. Specifically, my screen looks for shares in companies that are much smaller than those that O’Shaughnessy was interested in. What’s more, in order to get meaningful results the screen focuses on only 25 shares rather than the 50 recommended by O’Shaughnessy. Something else to point out is that in later editions of his classic book, O’Shaughnessy updated his valuation criteria to a composite measure of value rather than just dividend yield.

Nevertheless, my Anglicised version of the classic Cornerstone value screen has done well over the last decade, producing a cumulative total return of 124 per cent compared with 69 percent from the FTSE 350. Adding in a 1 per cent charge each year to account for dealing costs the total return drops to 102 per cent.

It's a good result and the kind of outperformance the screen was designed by O’Shaughnessy to achieve; decent, but not shoot-the-lights-out. However, over recent years it has failed at one of its other objectives – the tinkering I’ve had to do could be a reason. This objective is to outperform while staying closely aligned with the market. 

In the early years of this screen up until mid 2017, it seems to be doing a startlingly well in achieving this goal. However, since then, as the performance of 'growth' stocks decoupled from 'value' stocks, the screen underperformed. Meanwhile, its recovery in late 2020 massively outdid the index. Indeed, while the strong returns from the last 12 months look great from the perspective of outperformance, it can be seen as a failure from the perspective of trying to incrementally eke out higher returns.

This result could be totally random. However, if a reason needs to be given for performance going off-kilter since mid-2017, my guess would be that it reflects the divergence of returns between growth and value stocks. This kind of thing is not unprecedented, but the extent to which fortunes have diverged is unusual and extreme. 

Indeed, as recently mentioned in my Ideas Farm column, O’Shaughnessy Asset Management has itself suggested what has happened in recent years is unusual but typical of a technological revolutionary cycle. Prior to the information technology revolution, it suggests the last of such revolutions ran from 1926-41 and involved the discovery of cheap fossil fuels, production lines and mass-adoption of the automobile.

Big picture stuff aside, the nuts and bolts of the Cornerstone Value screen are reassuringly straightforward. The screen looks for the highest-yielding shares of what it calls “market leaders”. The original identification of market leaders is based on rigorous backtesting by O’Shaughnessy and requires companies to display:

• Market capitalisation of at least £500m (my Anglicisation).
• Greater-than-average number of shares outstanding.
• Higher-than-average cash flow per share.
• Turnover of more than 1.5 times average.

There is one alteration I’ve made to this year’s screen to take into account the exceptional year we’ve had. 

A lot of companies have cut dividends in response to tough trading caused by the pandemic. As the screen focuses on historic yield, I’ve added a check to make sure the historical dividend is the real thing. All shares must have a forecast dividend yield that is at least equal to the lowest qualifying historical yield identified by the screen – ie, the 2.3 percent historical payout from Direct Line Insurance.

This year’s 25 stocks can be found in the accompanying table along with some fundamentals. As usual, a more detailed excel table is available to download.

Source: FactSet

*FX converted to £