Our latest Magic Formula shares

Disinterest is in fact likely to be a real virtue for anyone using the Magic Formula. The reason for this is that the results from the screen can be volatile and it is prone to underperform the wider market in any given year. That makes it hard to maintain the disciplined approach required. But the less interest an investor pays to the performance of their portfolio, the less resolve they are likely to need to stick with their knitting. Indeed, research into real-world investment accounts has shown regular checking of portfolios tends to result in higher levels of trading, and unfortunately higher levels of trading tend to be associated with lower returns.

When Mr Greenblatt outlined the Magic Formula he took pains to highlight this point and that those who used it should not expect to beat the market every year. That has certainly proved the case in the eight years I’ve tracked a version of the screen in this column. Looked at on a start-to-finish basis, the 247 per cent total return from the screen’s top 10 stocks, or 148 per cent from the top 30, looks great compared with 71 per cent from the FTSE All-Share index. The comparison remains strong even after taking the important step of trying to factor in notional dealing costs at 1.5 per cent. This takes the total return from the top 10 to 208 per cent and the top 30 to 119 per cent.

But some of the bumps along the road (the past 12 months was a case in point) have been fairly horrible. Indeed, the largest peak-to-trough fall (maximum drawdown in finance jargon) from the top 10 shares has been a stomach-churning 33 per cent compared with 19 per cent from the All-Share. After a drop like that, even the steeliest nerved investor would be likely to experience a gnawing feeling that it was time to throw in the towel. The top 30 shares – the number of shares Mr Greenblatt recommended people bought – has a more palatable maximum drawdown of 23 per cent. The table below also shows the one-year returns of each of the portfolios picked by the screen in the years I’ve followed it.

Magic formula year by year

Source: Thomson Datastream

This screen is unusual for this column in so far as Mr Greenblatt designed it as an end in itself. By contrast, most of the other screens I run are principally viewed as a way to generate interesting ideas for further research. Given the hit-and-miss nature of the screen, and the difficulty of sticking with such a leap-of-faith discipline during tough times, it is important that the logic behind the process is simple and compelling. Fortunately, this is certainly the case with the Magic Formula.

The screen focuses on two vital considerations for fundamental analysts: the price investors are asked to pay and the quality of what they buy. The Magic Formula screen works by looking at a slightly modified form of a price/earnings (PE) ratio to assess 'value' (the price paid) and a ratio similar to return on capital employed (ROCE) to assess 'quality' – return on assets (RoA) is also considered an acceptable alternative. Details can be found below.

Value

Mr Greenblatt uses an earnings yield in his Magic Formula (this is like a PE ratio with the numerator and denominator flipped on their heads) calculated using enterprise value (EV) and earnings before interest and tax (Ebit). EV adjusts a company’s market capitalisation for cash (subtracted) and debt (added).

Quality

To measure quality, Mr Greenblatt looks at Ebit generated from 'tangible assets'. Tangible assets consist of net working capital added to net fixed assets, which reflects assets that are actually being used in a company's operations to generate profits.

All stocks are ranked for value and separately ranked for quality. The two rankings are then added together and a final ranking is produced. The screen is carried out on all non-financial, main-market and Alternative Investment Market (Aim) stocks with market capitalisations of more than £50m. Financial companies are excluded from the screen because the nature of their balance sheets means the quality measure used can give dubious results. Mr Greenblatt suggested investors target the top 20, or preferably 30, shares. I also track more concentrated portfolios which, while better performing since inception, are a lot more volatile and risky. It’s easy to understand why Mr Greenblatt recommended being more diverse.

The variable results from the screen from year to year beg the question of whether there is an optimal balance between value and quality. Intuitively, it seems to make sense that the best time to use the Magic Formula is likely to be when the market is making little distinction between the valuation it puts on high- and low-quality stocks. While the screen had good years in 2016 and 2017, it is arguably the case that the premium attached to quality shares has risen over recent years, which may be pushing results more towards riskier 'value' stocks.

Whatever the case, the picks from the screen 12 months ago looked very much like they veered towards the lower end of the quality spectrum and performed poorly.  Among the 'value' stocks in last year’s top 10 picks, the one that proved the biggest 'value trap' was outsourcer Interserve (IRV), which has recently announced a significant debt-for-equity refinancing, which has left long-term holders as bit players in the company’s future. The screen also highlighted a number of troubled retailers as top 10 picks, and although sofa retailer DFS (DFS) actually managed a slightly positive performance, the falls from Dixons Carphone (DC.) and SafeStyle (SFE) far more noteworthy. Meanwhile, last year’s top pick, newspaper group Reach (RCH), continues to follow a strategy of consolidating and cutting costs to try to offset the pain from declining circulations. Unfortunately, the market still seems unconvinced. The performance of last year's shares ordered by Magic Formula ranking are in the table below.

Tough times in 2018

Source: Thomson Datastream

*Taken over

This year there are once again a number of stocks that are possibly cheap for a good reason. The top 30, listed in rank order, along with some key fundamentals can be found in the table below.

Magic formula picks 2019

Source: S&P CapitalIQ