While my long-term income screen has yet to really distinguish itself, some encouraging signs have begun to emerge from this strategy based on a five-year buy-and-hold investment approach. Indeed, while last year's shares have underperformed the market on the total return basis, those shares picked two years ago are now outperforming (see graph).
Long-term income - class of 2012 vs FTSE 350
The overall performance reflects some major extremes on a stock-by-stock level. Defence company Chemring, which I expressed major reservations about at the time of the original screen, continues to be the laggard. Meanwhile, Restaurant Group has delivered the kind of performance one would expect from an out-and-out growth play rather than a reliable income stock. But overall things are now moving in the right direction. Dividend growth is also picking up, rising by 5.1 per cent over the 12 months compared with a disappointing 4.1 per cent the year before. The overall historic yield on the portfolio is now 4.3 per cent, compared with 4.1 per cent last year and 3.9 per cent when the screen was originally conducted.
| Name | TIDM | Total Return (29 Mar 2012 -26 Mar 2014) | LTM DY (at 29 Mar 2012 price) | DPS gr LTM | 2yr DPS CAGR |
|---|---|---|---|---|---|
| BRITISH AMERICAN TOBACCO | LSE:BATS | 11% | 4.5% | 5.6% | 6.1% |
| CAPITA | LSE:CPI | 58% | 3.6% | 13% | 11% |
| CARILLION | LSE:CLLN | 30% | 5.8% | 1.4% | 1.8% |
| CHEMRING | LSE:CHG | -33% | 1.8% | -24% | -30% |
| CENTRICA | LSE:CNA | 16% | 5.4% | 3.7% | 5.1% |
| DOMINO'S PIZZA | LSE:DOM | 27% | 3.6% | 10% | 14% |
| GLAXOSMITHKLINE | LSE:GSK | 26% | 5.5% | 5.4% | 5.6% |
| MITIE GROUP | LSE:MTO | 20% | 3.7% | 8.2% | 6.8% |
| RESTAURANT GROUP | LSE:RTN | 150% | 4.8% | 19% | 15% |
| WM MORRISON | LSE:MRW | -25% | 4.3% | 10% | 10% |
| Average | - | 28.0% | 4.3% | 5.1% | 4.6% |
| FTSE 350 | - | 25.6% | - | - | - |
Source: Thomson Datastrea & S&P Capital IQ
The fact that the screen from two years ago has started to outperform gives me more confidence in the longer-term potential of last year's screen which is currently lagging the market. It delivered a total return of 4.9 per cent over the last year compared with 8.4 per cent from the FTSE 350, which is the index from which my long-term income stocks are selected (see graph). That said, the minimal number of stocks selected last year (only five) means this portfolio is inherently much riskier (see table).
Long-term income - class of 2013 vs FTSE 350
| Name | TIDM | Total Return (2 Apr 2013 - 26 Mar 2014) | LTM DY (at 2 Apr 2013 price) | DPS gr LTM |
|---|---|---|---|---|
| CENTRICA | LSE:CNA | -5.5% | 5.3% | 3.7% |
| SMITH & NEPHEW | LSE:SN. | 23.8% | 2.6% | 5.0% |
| BRITISH AMERICAN TOBACCO | LSE:BATS | -3.8% | 4.4% | 5.6% |
| MITIE GROUP | LSE:MTO | 17.4% | 3.7% | 8.2% |
| CRODA INT | LSE:CRDA | -7.6% | 3.0% | 8.4% |
| Average | - | 4.9% | 3.8% | 6.2% |
| FTSE 350 | - | 8.4% | - | - |
Source: Datastream, S&P CapitalIQ
The premise of my long-term income screen is to use past dividend growth rates as an indicator of future growth in conjunction with a number of other tests. Broadly speaking, most stocks have failed to live up to their track records so far but most have nevertheless delivered dividend growth.
This year I've been able to augment the test with dividend forecast data pulled in from ShareScope. While my additional test based on this should help eliminate disappointments I've had to relax my requirements due to no stocks passing all of the screen's tests this year. So stocks that pass the key dividend growth test are allowed to fail one of the screen's other tests. The details of which test the nine successful stocks failed are given in the accompanying tables. This is how the screen is carried out:
| Core Long-Term Income Test |
|---|
| The key criteria for my screen is to look for stocks that are forecast to produce a top quartile "discounted" annualised yield (4.28 per cent or more) over the next five years. This approach attempts to look at both the current yield and a likely growth rate, so stocks with low initial yields but high forecast growth can make the grade. I've had to change the screening criteria this year from my previous requirement of an annualised yield of 5 per cent or more in order to generate a decent number of results. The growth rate I use to estimate a potential payout over the next five years is an average of the compound annual growth rate over 10 years, 5 years and the last year (this puts more weight on recent periods). This is then “discounted” using the risk-free return from a five-year government bond in order to find an annualised average yield over the five years on a "net present value" basis. Using history to predict the future is always going to have flaws, but combined with the other factors this screen uses, it should provide a good steer towards stocks that are likely to deliver strong dividend growth. |
| The Other Long-Term Income Tests |
|---|
■ Dividend growth forecast in the current financial year and next financial year ■ No dividend cuts in the past 10 years ■ Dividend growth of 5 per cent or more in each of the past three years ■ A compound annual dividend growth rate of 5 per cent or more over both five years and 10 years ■ Net debt to cash profits of less than two times ■ Forecast EPS growth ■ Dividend growth last year must be at least half the three-year compound average growth rate - this is to allow for stocks that may be down on their luck but not completely off track ■ My dividend forecast for next year must be at least 1.5 times covered by forecast EPS |
Nine stocks made the grade this year. I've provided a write up of the five stocks with the highest "implied" five-year yield below with the other four stocks included in the table that follows.
Nine Long-Term Income Stocks
