Seven stocks for cheap cash flow and good yields

The conclusion should certainly not be that it’s a bad idea to pay attention to FCF. There’s plenty of evidence that cash generation should be at the forefront of investors' minds when making investment decisions. Indeed, other screens I run in this column that focus on FCF have performed very well. And while it is undeniably a distinct possibility that I’ve simply constructed a rubbish screen, even this may be the wrong conclusion. Luck can play a major role in determining investment results over even quite lengthy periods. This is especially true when it comes to simple stock screens, which are best regarded as a first step in an investment process rather than an end in themselves. That said, I’ve decided to make a number of changes to the screen this year while trying to stay true to the overarching objective to find companies that look attractively valued compared with FCF and pay a decent dividend. 

FCF gives investors an idea of the amount of cash left over from a companies' operations that is available for equity holders (FCF to equity), or alternatively, the whole company (FCF to firm), which takes into account its operations’ other sources of finance such as bank loans. This measure of cash flow shows what cash has been generated in the year once all necessary cash outgoings are accounted for, which includes things such as tax and investment in the business, but excludes accounting charges such as amortisation and depreciation. The difference between FCF for firm and FCF for equity is that the latter includes amounts paid out in interest. 

The tangibility of FCF is one of its draws. What’s more, it’s one of the more difficult numbers for an artful finance chief to manipulate without resorting to outright fraud. These are very sound reasons for favouring FCF as a key performance measure.

Our semi-regular Further Reading column recently highlighted another great, but less often discussed, reason to focus on FCF. The column focused on a paper from O’Shaughnessy Asset Management produced in collaboration with pseudonymous blogger Jesse Livermore. It looked at the question of how the use of depreciation charges causes companies to systematically overstate earnings. The paper made the simple but powerful point that depreciation charges are based on the historical costs of investing in a business, whereas current replacement costs are usually much higher than historical costs. FCF, meanwhile, reflects investment costs in the year they are incurred. Indeed, this can be regarded as a reason why, as a rule of thumb, many investors regard a healthy level of conversion of profits after tax to FCF as being about 80 per cent. 

The fact that companies sometimes need to undertake large capital projects, such as building big new factories, means FCF can be far more erratic from year to year than reported profits, which utilises the depreciation charge to smooth out the impact of such spending over its useful life. Nevertheless, the O’Shaughnessy paper found valuations based on FCF provided a better guide to future performance than earnings-based ratios.

Great in theory, but my FCF Kings screen, which uses an enterprise value (EV)/FCF ratio as a key valuation metric, has had another poor year (see table below).

2018 FCF Kings disappointment

*Taken over

Source: Thomson Datastream

Following the most recent outing, the cumulative performance of the screen in the six years since I began running it in 2013 stands at 43.5 per cent compared with 41.7 per cent from the index. While the screens in this column are run as a source of ideas for further research rather than as off-the-shelf portfolios, if I factor in an annual cost of 1.5 per cent to represent notional dealing costs, the cumulative total return drops to 31.1 per cent. Not very inspiring.

One of the problems this screen seems to be encountering is that it is highlighting shares that turn out to be value traps rather than real value. I’m altering the screen’s criteria this year to see if I can counter this. As well as lowering the valuation tests based on EV/FCF and dividend yield (DY), I’m introducing a test to check shares either have positive share price or earnings momentum. I’m also adding a test to ensure companies have generated an above-average cash return on capital over the five years to replace a similar test that focused on a single year. Looking at a five-year average aims to get over issues associated with the lumpiness. The same logic applies to the new test for growth in cumulative five-year FCF which replaces the old FCF growth test.

One key issue faced by this screen, discussed extensively below in relation to Go-Ahead, is that while the measure of EV used in the EV/FCF ratio does take account of pension deficits, it ignores debt-like lease liabilities. Hopefully, accounting rule changes means I can rectify this by the time of next year’s screen. The full revamped screening criteria are:

■ Cumulative five-year FCF 10 per cent higher than it was three years ago

■ Operating cash conversion (cash from operations/operating profit x 100) of over 100 per cent in at least two of the past three years.

■ Net debt less than three times cash profits (Ebitda).

■ Five-year cash return on invested capital (CROCI) better than average

■ A pension-deficit-adjusted enterprise value (EV)/FCF ratio among the cheapest half of all stocks screened (below 19.6 times).

■ A dividend yield (DY) in the top half of all dividend-paying shares screened

■ Top third three-month share price momentum or forecast EPS growth in each of the next two years and no forecast downgrades in the last three months.

In total, seven stocks passed all the screen's tests. Details can be found in the accompanying table. I’ve also taken a closer look at the share that appears the cheapest (on the surface at least) based on EV/FCF and the share offering the highest dividend yield.

Seven FCF Kings

Source: S&P Capital IQ

Go-Ahead

At the start of this year, there was an accounting rule change that in time should prove a major aid to all the stock screens I run that look at enterprise value (EV). EV attempts to put a valuation on a company by adding net debt to a company’s market capitalisation. However, there is also a strong argument for adding in other debt-like sources of financing, such as pension deficits, which this screen factors in, and lease liabilities, which this screen does not factor in. This screen uses EV in its key valuation criteria: EV/FCF.

The previously mentioned accounting rule change requires companies to put a number on their debt-like lease liabilities. I’d argue that these liabilities can, and should, be factored into EV calculations as they represent claims on cash flows that rank above the claims of shareholders. Unfortunately, not all companies have yet released accounts based on the new rule, so the availability of data on lease liabilities is currently patchy. That means it’s not yet possible to factor reported lease-liability numbers into screens, because screens need to draw complete data sets in order to make like-for-like comparisons between all companies. 

The fact that Go-Ahead (GOG) appears to be the cheapest stock passing all the screen's tests based on EV/FCF helps illustrate why the accounting change should be so welcomed by investors. Indeed, the analysis of Go-Ahead’s EV/FCF ratio highlights a number of other potential pitfalls investors face when calculating EV.

Go-Ahead operates train and bus services in the UK as well as a fledgling international operation. This work involves leasing a lot of assets; from trains and buses to track and depots. The company says the liabilities associated with its contracted lease payments would have been valued at about £800m if included in its most recent accounts. This is expected to fall to about £400m by the end of its next financial year (27 June 2020). If last year’s lease liabilities were treated as debt, it would cause Go-Ahead’s EV to more than double, making the company’s EV/FCF valuation look much higher.

But leases are not the only factor investors need to consider when looking at Go-Ahead’s EV. The statutory net cash reported by Go-Ahead is not all that it seems. The company’s rail business is required to hold a lot of cash (£485m at the year end), which is restricted by the Department for Transport (DfT). This is needed for the functioning of the rail contract and is not the type of cash investors should be working into valuations. For that reason the company steers its shareholders towards an adjusted net debt figure to provide a truer picture of its finances. 2019 adjusted net debt was £270m. Using this figure rather than the statutory net cash number leads to a rise in EV of more than half.

All in all, if I add on the £800m of lease liabilities and use the figure of £270m net debt rather than £222m net cash, Go-Ahead’s EV increases from the figure used by the screen of £756m to £2bn. The higher number is a better reflection of the value being put on the company, and as such, my screen is arguably being duped by the way Go-Ahead is currently obliged to report its statutory numbers.

In order to recalculate the EV/FCF ratio to take account of these issues, I have to also adjust FCF to account for the 'interest payment' associated with the company’s lease liabilities. To do this I’m simply applying a 7 per cent interest rate to the £800m lease liability, which means adding £56m to the company’s £95m FCF for the year. This gives me an EV/FCF ratio of 13.2, equivalent to a 7.5 per cent FCF yield. That compares to the ratio of 8.0 used by the screen, which is equivalent to a hearty 12.5 per cent FCF yield.

But, arguably, it’s not only the top half of Go-Ahead’s EV/FCF that’s contributing to a shonky number. The lumpy nature of cash flows may be giving a misleading impression of Go-Ahead’s sustainable FCF. Indeed, in the most recent financial year capital spending on the bus business plummeted from £100m to £50m as the company spent significantly less on new London buses. However, the division’s capex is expected to bounce back again next year to £90m, making the 2019 FCF number rather flattering. If I make a £30m adjustment to FCF to smooth out this erratic spending, we end up with an EV/FCF ratio of 16.5, equivalent to a FCF yield of 6 per cent. On that basis, the value on offer is much less attractive than the screen suggests.

While Go-Ahead may not be the FCF deep-value play the screen took it for, based on my adjusted EV/FCF ratio it would still make the cut for this screen, which following this year’s criteria changes requires companies to be cheaper than the median average of 19.6. What’s more, while the dividend is not expected to grow, management believes it can be maintained based only on the company’s bus business. Broker Liberum puts the dividend cover based on the bus business alone at 1.5 times compared with 1.7 times for the group as a whole. The support to income offered by the bus business is good news because the company’s rail division has been struggling badly. Indeed, despite generating around three-quarters of the group’s revenue, rail generates only about a fifth of profit.

The real thorn in rail’s side is its Govia ThamesLink joint venture which achieved notoriety with commuters through a chronically botched timetable change and the ultimate intervention of the DfT. Following a settlement, the group hopes to make a mini-margin from the franchise of 0.75 to 1 per cent over the rest of its life. Meanwhile, the group has just received a short extension to its South East franchise, but on worse terms. 

Buses are performing okay, but margins slipped slightly last year from 9.8 to 9.5 per cent as bus yields failed to keep up with cost increases. Management has guided to a flat year ahead. The group should start to benefit from contributions from overseas contracts in this year, though. All in all,, for a 6 per cent FCF yield, there does not seem that much to get excited about. Had the 12.5 per cent figure stood up, the stock would have looked a lot more interesting. 

Babcock

For anyone that is familiar with the travails of support services group Babcock (BAB), it may seem odd for its shares to be highlighted by a screen focused on cash flow. One of the major criticisms that has been levelled at the group over the past five years has been its inability to convert a decent wedge of its profits into cash. However, this screen is chiefly interested in the how much cash a company generates rather than how it relates to reported profits (the exception being the operating cash conversion test). On this basis, Babcock’s FCF is high enough and its share price is depressed enough to make it of interest.

As with Go-Ahead, Babcock’s EV/FCF valuation is flattered by the fact this screen currently ignores lease liabilities. But even if I factor in the company’s own estimate of lease liabilities at £606m and an associated £25m interest charge, the EV/FCF ratio is still a relatively lowly 12.7 (a 7.9 per cent FCF yield) compared with the 11.9 (8.4 per cent) used by the screen. Importantly, the screen does account for Babcock’s hefty pension deficit which is likely to require annual top-up payments of around £70m to help sort things out. 

While there have been false dawns before, Babcock may be at an interesting juncture in attempts to turn around performance. This year Babcock’s management has bitten the bullet and seriously reset earnings expectations lower. 

The accompanying downgrade graph shows the trajectory of EPS forecasts for three historical financial years as well as the next two year. Over a five-year period, Babcock's shareholders have faced a near constant trickle of downgrades which have had an insidious effect on sentiment, as illustrated by the graph charting the deterioration in two key valuation ratios (EV/sales and forward 12-month PE ratio). However, a sharp downgrade in June marks the company’s reset, and the hope is that these lower forecasts will finally provide a level from which progress can be made. 

The company has set out plans to boost FCF and hopes to achieve annual FCF of £280m over the next five years, although a below-average £250m is earmarked for 2020. Importantly, if the FCF target can be achieved, the shares high yield should be well supported. If successful, a number of large contract bids this year could also help sentiment.

The company has come under attack from short sellers for aggressive accounting. On this front, some reassurance can be taken from recent bid interest from Serco. Meanwhile, broker Liberum has attempted to model some harsh adjustments to accounting policies and puts the maximum damage at about a fifth of its forecast earnings. 

So there are some reasons to hope Babcock may have reached the bottom. That said, the accounts remain hard to fathom and, following a litany of disappointments over recent years, buying the shares represents a bold and contrarian move.