Time to pay up for high quality

The reason why the business of housebuilding is widely regarded as generating low-quality profits is that profits are unpredictable. Central to this is the fact that housing markets are cyclical, which causes the profitability of housebuilders to fluctuate significantly depending on whether the market is climbing towards a peak or hurtling towards a trough.

What’s more, when the market turns down, the issues of falling house prices and sale volumes are compounded for housebuilders by the huge balance sheet risks they have to take on in the normal course of business.

True, there may be a virtue to this balance sheet risk (massive amounts of capital tied up in land banks, part-built and yet-to-be-sold homes), which arguably acts as a barrier to entry and allows return on capital to get very high when markets peak (see "changing fortunes" chart). This may have been especially true over the past decade when banks have been reluctant to finance smaller players. But while high returns are a key hallmark of a quality stock, those returns need to be sustainable. In the case of housebuilders, they are not.

The downside to the balance sheet risk comes when the market turns and the value of assets owned by builders has to be written down at the same time as sales fall and operating margins collapse. In past cycles, the bust has often hit soon after housebuilders have aggressively ramped up supply, which compounds problems and has led to many historical rescue rights issues and outright failures in the sector.

There are some grounds to argue that the cycle is different this time; but not very much so. The lack of competition from smaller players may have kept a lid on some of the excesses of the past, and land price inflation (a key input cost) has been relatively benign. There is also a serious shortage of housing supply that in theory could help underpin the market. It is also of note that housebuilders have been opting to hand back substantial amounts of capital to shareholders rather than risk ploughing everything into balance sheet expansion. But there are other factors that suggest the differences this time could actually prove to be for the worse rather than better when the inevitable day of reckoning comes. Of particular note is the artificial stimulation of the housing market through government house-buying incentives such as Help to Buy.

Whatever way you cut it, as long as economic cycles exits, it is hard to regard housebuilders as 'high quality'. This is not to say housebuilders can’t make very good investments if caught at the right points in the cycle. It is also of note that housebuilders were very well represented in last year’s screen picks, which produced a not-disastrous 4.3 per cent total return over 12 months, compared with 4.2 per cent from the FTSE All-Share. However, the outperformance of the screen came despite the contribution from the housebuilders rather than because of it, and actually owes almost everything to the inclusion of Games Workshop (GAW) – a wonder stock of 2017 – among last year’s picks.

Source: Thomson Datastream

Since I began running this screen seven years ago, it has produced a cumulative total return of 251 per cent or, after injecting a small sense of reality by applying a notional 1.5 per cent annual sharedealing charge, 221 per cent (the assumption of this column is that the screens are primarily of interest to readers as a source of stock ideas for further research rather than a source of off-the-shelf portfolios). The screen’s long-term performance compares with 93 per cent from the FTSE All-Share index over the same period. 

Regardless of the good long-term performance, I can’t help but feel credulous about the screen’s output this year. The question is why has the screen highlighted so many shares that are so obviously not high quality. In my opinion, and it is something I have alluded to a number of times before in this column, the problem screens like this one face is that the slow-but-steady economic recovery and bull market experienced since the credit crunch’s nadir has allowed many cyclicals to display the characteristics of quality stocks – strong balance sheets with a prolonged record of high margins and return on capital. Indeed, in the chart below that chronicles our housebuilders’ returns on capital employed (ROCEs), taking only the past three years it seems we are looking at a sector oozing quality. It is only a look at the ROCEs over 10 years that really illustrates the cause for concern.

While the market does have a habit of pushing up the valuations of cyclical stocks until relatively shortly before the peak, the knowledge that housebuilders are very cyclical and earnings are low-quality means they nevertheless look lowly rated from an industry-blind view point. However, they are relatively highly rated compared with their own histories based on the valuation metric most commonly used to assess the sector: the price-to-book-value ratio (P/BV).

Source: S&P Capital IQ

Most of the other shares that pass this screen’s valuation test but fail one of the other nine tests are also cyclicals. Indeed, dredging though the results, it seems fair to conclude that in the current market it is very hard to find stocks that have any claim to the 'quality' billing and are also 'cheap' – it is a big ask in any market to be fair. The rude awakening from the results of this year’s screen makes me feel I need to adjust the valuation criteria. While it seems to have contributed to some strong performance results in previous years, on the basis of this year’s selection it looks overambitious or, worse still, naive.

The specific amendment I am making to the screen is to run a version that looks for stocks with a genuine value ratio (a price/earnings growth ratio amended to take account of dividends paid and debt) that is higher than average but below the top 25 per cent of shares screened. The original version of the screen looked for a lower than average genuine value (GV) valuation.

The amended screen ends up highlighting four stocks that are indeed widely regarded as 'quality', which I have provided details of in a separate table. I have also published a list of the companies passing the screen based on the old criteria, along with those passing the below-average valuation test and failing one other test. I’d really only regard three of the stocks from the stocks on this list as having claim to a reputation for quality over cyclicality: Domino’s Pizza (DOM), Sage (SGE) and British American Tobacco (BATS). Out of the three, sentiment has recently been souring towards DOM and SGE, and has been sour for some time towards BATS.

I will monitor both versions of the screen over the next 12 months. If this was blockchain, I believe the change being made to my valuation criteria would be described as 'a fork', which cryptoheads are quick to point out is a feature and not a fallibility of blockchain. For those who would prefer a more old-school explanation of my actions, in the (disputed) words of John Maynard Keynes: "When the facts change, I change my mind. What do you do, sir?" The screen’s full criteria (old and new) are:

■ NEW CRITERIA: higher than median but lower than top quartile GV ratio.

■ OLD CRITERIA: lower than median average GV ratio.

■ PE above bottom fifth and below top fifth of all stocks screened.

■ Earnings growth forecast for each of the next two years.

■ Interest cover of five times or more.

■ Positive free cash flow.

■ Market cap over £2bn.

■ Higher than median average return on equity (RoE) in each of the past three years.

■ Higher than median average operating margin in each of the past three years.

■ RoE growth over the past three years.

■ Operating margin growth over the past three years.

■ Operating profit growth over the past three years.

Quality at a price (NEW CRITERIA)

Dubious quality (OLD CRITERIA)

Cheap, but failing one other test

Source: S&P Capital IQ