Join our community of smart investors

The end of income investing

We should stop investing in stocks for income, because income stocks comprise two very different types of share
November 3, 2020

Income investors have had a terrible time. So far this year the FTSE 350 higher yield index has lost more than 30 per cent even adding in dividend income. That’s far worse than the 9 per cent loss on the lower yield index.

This continues a long bad run. Even including dividends the higher yield index has done no better than its lower yield counterpart in the last 20 years. And it has actually underperformed it by 4.6 percentage points a year since its peak relative to lower yield stocks in June 2005.

Such underperformance overturns the conventional wisdom. Back in 1997 the Nobel laureate Eugene Fama showed that value stocks had outperformed growth around the world. In hindsight, this now looks like another example of Murphy’s law – the tendency for patterns in equity returns to disappear after strong evidence for them emerges.

Which poses the question: what hope is there for a recovery in higher-yielding stocks?

Here, we run into a problem. Ordinarily, we’d answer this by looking at the conditions in which they have outperformed in the past, and assessing how likely such conditions are to recur soon. But we can’t do this, because there seem to be no strong links between particular economic or market conditions and spells of income stocks outperforming or underperforming. The big fact about income stocks relative to growth is their steady downtrend since the mid-2000s (even before this year) rather than a tendency to do well in some circumstances but not others.

For example, there is a tendency for high yielders to underperform when equities generally do well. Since 1999 there has been a negative correlation between annual changes in the All-Share index and annual changes in the FTSE 350 higher yielding index relative to the lower yielding index. But this correlation isn’t huge and there have been exceptions to this pattern. For example, high yielders also underperformed growth when the market fell in 2007-08 and in 2015. Income stocks' underperformance during this year’s bear market isn’t typical, but it’s not freakishly unusual, either.

Similarly, there has been a tendency for income to outperform growth when shorter-dated gilt yields rise. This is because they tend (with many exceptions which we'll come to) to be more cyclical than growth ones and so do well when investors anticipate better economic times – which is when gilt yields rise.

But again, the correlation is weak. For annual changes since 1999 in income relative to growth and changes in two-year yields it is only 0.25. That’s not strong enough for us to rely much upon it.

Nor can we look to higher oil prices to help income investors. Since 1999 there has been no significant correlation between annual changes in these and in changes in income relative to growth stocks. Sure, higher oil prices help the high-yielding oil majors. But they are bad for many other income stocks such as airlines.

You might think none of this matters because high-yielding stocks are now so bombed out that they are a bargain.

Now doubt there are one or two bargains – but then there always are. But there’s a problem here too. It’s that since 1999 there has been no correlation between the dividend yield gap between income and growth stocks and subsequent returns. Sometimes, the FTSE 350 higher yield index has seemed relatively cheap – but has become even cheaper in subsequent months. This happened, for example, in late 2008, 2010-11 and earlier this year. This warns us not to buy income stocks indiscriminately even now.

All this means we cannot reliably point to conditions in which income stocks will bounce back.

But perhaps there’s a solution here. We should abandon the idea of income stocks as an asset class. Two very different types of share can have high yields.

One type are cyclicals – those that do well when investors expect an economic upturn and badly when they fear recession. These include miners, housebuilders and airlines.

The other type are defensives – dull stocks that investors believe (sometimes wrongly) have gone ex-growth. These include tobacco, utilities and telecoms.

One reason why the FTSE 350 higher yield index is so lightly linked to economic conditions is that it comprises these two very different types of share. One type holds up relatively well in bad times; the other slumps.

Sadly, however, while this distinction is sharp in theory it is less so in the real world. Many income investors were caught out this year because they thought the oil majors were defensive when they turned out not to be.

This, however, might help explain one of the few patterns in equity returns that is robust across time and place – the tendency for defensives as a class on average to outperform. One reason why they do so might be as a reward for carrying a particular type of risk – the risk that some of them will cease to be defensive when we need them to be.

The point here is that we should not invest in equities for income: you can create your own income simply by selling some stock. Think instead about cyclicals and defensives. While there might be a case for the braver investor to buy the former as a short-term punt, it is defensives that offer the better longer-term deal.