History is definitely important when it comes to understanding the risks to dividends. But it's not the only thing to consider. For example, the record shows that real estate, general retail, and banks are more reliable than most areas when it comes to avoiding dividend cuts. However, most investors would understandably balk at buying into these sectors in order to secure a reliable income-stream right now.
It may be possible to use accounting data to identify which sectors and individual shares are most in danger of having to slash payouts to investors. A 2001 study by Andrew Benito and Garry Young of the Bank of England found that companies that cut or omit their dividends often have similar characteristics. Specifically, they discovered that low levels of cash flow and high levels of income gearing were associated with dividend cuts. Also, smaller companies and those with greater investment opportunities are more likely to skip payments.
Using these findings as inspiration, I went through the 550 members of the Datastream UK Total Market index. I ranked every company for which data was available based on its forecast dividend cover for the next two years, its financial gearing, cash flow generation, and interest cover. The database did not contain a complete set of figures for every company. While I left such companies in the final list, I have put greater weight on the ones that I was able to rank on every single test.
As expected, the companies with the very highest dividend yields mostly come out as those with worse-than-average dividend-cut risk scores. Heading the list is Taylor Wimpey – with a trailing dividend yield is 36 per cent – and a risk score of 9 out 10. That said, there are firms that come out as less dicey than average. Topps Tiles – with a yield of 21.5 per cent – has a score of just 4.4, somewhat less than the market average of 5.5.
Above-average yield, high risk
| Company | DS Sector | Price (p) | Yield (%) | Score (10= worst) |
|---|---|---|---|---|
| BARRATT DEVELOPMENTS | Home Construction | 105.5 | 34.71 | 4.4 |
| TOPPS TILES | DIY retailers | 46.25 | 21.51 | 4.4 |
| HAYS | Business Training | 77.5 | 6.77 | 3.8 |
| MORGAN SINDALL | Heavy Construction | 569 | 6.68 | 4.2 |
| AGA RANGEMASTER | Household Products | 170 | 6.6 | 4.6 |
| COMPUTACENTER | Computer Services | 121.75 | 6.57 | 4.8 |
| RESTAURANT GROUP | Restaurants & Bars | 115 | 6.3 | 4.8 |
| KIER GROUP | Heavy Construction | 1001 | 5.83 | 3.4 |
| NEXT | Apparel Retailers | 994 | 5.53 | 3.6 |
| LOW & BONAR | Construction & Materials | 99 | 5.08 | 4.8 |
| MISYS | Software | 157.25 | 5.03 | 4.6 |
| GLAXOSMITHKLINE | Pharmaceuticals | 1150.5 | 4.78 | 4.8 |
| VITEC GROUP | Industrial Engineering | 375 | 4.75 | 4.6 |
| BP | Integrated Oil & Gas | 519.5 | 4.66 | 3.8 |
| VP | Support services | 228 | 4.61 | 3.4 |
| IMI | Industrial Machinery | 449 | 4.5 | 4.6 |
[Source: Datastream. The sectors are those used by Datastream and aren't the same as ICB sectors commonly used elsewhere. The links point to more articles, comment and data on each company]
One approach to picking safer high-yielders would be to combine our insight into historic trends with the risk scores. So, we might pick out high-yielding companies from sectors that have fared well in previous dividend recessions, and that also have low dividend-cut risk scores.
The real estate sector has had one of the least risky dividend-cutting records over time. Nevertheless, a whole cluster of real estate companies figure prominently among the riskiest shares, although they are not included them in the table here as data was not available for every risk-scoring measure.
Above-average yield, moderate risk
| Company | DS Sector | Price (p) | Yield (%) | Score (10= worst) |
|---|---|---|---|---|
| PENDRAGON | Specialty Retailers | 11 | 36.36 | 8.4 |
| ITV | Broadcasting | 44.6 | 7.06 | 8.4 |
| MARSTON'S | Restaurants & Bars | 183.75 | 6.98 | 8.4 |
| UNITED UTILITIES GROUP | Multiutilities | 699.5 | 6.59 | 8.4 |
| KCOM GROUP | Fixed Line Telecoms | 38.5 | 7.32 | 8.2 |
| MITCHELLS & BUTLERS | Restaurants & Bars | 239.5 | 6.07 | 8.2 |
| SEVERN TRENT | Water | 1344 | 4.88 | 8.2 |
| FINDEL | Specialty Retailers | 167.75 | 13.23 | 7.8 |
| YULE CATTO | Specialty Chemicals | 123.5 | 7.77 | 7.8 |
| PERSIMMON | Home Construction | 313 | 16.36 | 7.6 |
| LOGICA | Computer Services | 105 | 5.52 | 7.6 |
| SMITH (DS) | Containers & Packaging | 101 | 8.71 | 7.4 |
| PREMIER FOODS | Food Products | 85.25 | 7.62 | 7.4 |
| TATE & LYLE | Food Products | 389.75 | 5.8 | 7.4 |
| REXAM | Containers & Packaging | 352 | 5.68 | 7.4 |
| NORTHERN FOODS | Food Products | 56.25 | 10.76 | 7.2 |
[Source: Datastream. The sectors are those used by Datastream and aren't the same as ICB sectors commonly used elsewhere. The links point to more articles, comment and data on each company]
In spite of the message of my risk-scoring system and of history, there are certain sectors that I would not rely on for dividends at the moment. Housebuilders, banks, real-estate and retailers are all in crisis and this overrides my findings.
Of course, no system of screening can be infallible. It's worth asking where the risk-scoring approach here could fall down. The cash-flow and interest cover figures I have used are all based on trailing data. But profits are close to a cyclical high point, so we would expect ratios that include profits and cash to worsen. The use of forecast dividend cover data partly gets around this problem, but also introduces another. Stockbrokers' analysts are notorious for excessive optimism in their estimates of tomorrow’s earnings. As a result, my scoring system could end up being too rosy as well.
