What long-term returns might we reasonably expect on UK equities? The answer is: we cannot tell, which makes accurate long-term financial planning impossible.
The problem is that we face both risk and uncertainty. My chart below shows the risk. It shows annualised returns on the All-Share index since the dates on the horizontal axis. So, for example, it tells us that since December 1988 the index has returned 4.7 per cent per year after inflation if we had reinvested dividends but just 1 per cent if we had spent them: the figures are before tax.
This tells us two big things. One is that dividends matter a lot. Without them, returns have been low and often negative. In fact, since 1987 the index has risen by less than retail prices. History tells us, therefore, that it is only by reinvesting dividends that we have a hope of decent long-term returns.
But only a hope. My chart also shows that you would have made poor returns even with dividends if you had invested at the wrong time. Since 2000 real returns have been less than 2 per cent a year. And they’ve been negative over the past five years. In this sense, even longer-term returns are risky.
The idea that it doesn’t matter when you buy equities because they will come good over the long run is therefore wrong. Timing matters.
Which brings us to some good news. History tells us that we can tell when to buy by using the dividend yield. A high yield predicts high returns. Taking all months between 1986 and 2015 together, the dividend yield alone can explain over half the variation in subsequent annualised returns.
With the yield now well above its long-term average, it is now pointing to good longer-term returns. If post-1986 relationships continue to hold, the current yield predicts annualised real returns of 4.7 per cent including dividends and 0.9 per cent excluding them.
This is not as comforting as it seems, however. For one thing, even past relationships point to a one-in-six chance of share prices falling over the long run. That's considerable risk.
And for another, we face uncertainty – in the sense that we cannot be confident that past relationships will hold.
One reason why a high yield has predicted good returns in the past is that it has not stayed high, and prices have risen as the yield has fallen.
There’s a danger, though, that the yield might stay high. One way in which this could happen is simply that this recession might have a scarring effect. In reminding us of the fundamental unpredictability of economic activity, it might permanently raise risk aversion.
Investors might require a permanently high yield for another reason, however – that they expect sustained slow growth.
I stress 'sustained'. We might well see a brief post-lockdown boom in profits thanks to the release of pent-up demand, high government borrowing and increased monopoly power among surviving companies as their competitors go bust.
But this bonanza might be short-lived. BP’s recent cut in its forecast for longer-term oil prices highlights a double threat for many listed companies.
One is that we’ll see weak aggregate growth. Even before the pandemic, western economies were stuck in secular stagnation. While this recession might relieve this problem if a wave of bankruptcies raises the profits of the survivors and hence the motive to invest, this is only one possibility. There’s also a danger that post-crisis efforts by individuals and (less forgivably) governments to reduce debt will depress demand. And boardrooms might also be scarred by this recession into being less willing to authorise expansion and capital spending.
The other problem is that BP might be right to fear that we’ll get a different type of growth, perhaps less carbon-intensive. This is a problem because you can’t teach old dogs new tricks. Peter Rousseau and Boyan Jovanovic have shown that companies often embody specific vintages of organisational capital – they are good at doing what they were founded to do, but cannot easily adapt. New types of growth – such as the IT revolution – are thus often led by new companies, which means the devaluation of older companies.
All this poses a massive headache for all of us. If we cannot know future long-term returns we’ve no hope of saving the right amount or of knowing how big a pension pot we’ll need in retirement.
We can do some things to mitigate this. Younger people should save regularly. If you invest the same amount each month you’ll automatically buy more stocks when prices are low and (hopefully) expected returns are higher. We should also diversify internationally, because longer-term returns can differ a lot from country to country. We should hold private equity as well as quoted stocks, as future growth might come from the latter. And we should also hold non-equity assets such as cash, gold and bonds despite their negative real expected returns as these protect us from nasty surprises about longer-term growth.
Whether all this is sufficient, however, is doubtful. The fact is that those of us saving for our old age are to a large extent in the dark. Which is why some of us favour a higher state pension – because there are a few jobs that governments can do better than individuals.
