Join our community of smart investors

The yield signal

If history is any guide, equities are very cheap now.
April 14, 2020

Should we buy equities now? This question can be rephrased: will the historic relationship between the dividend yield on the All-Share index and subsequent returns continue to hold?

If the answer is yes, then equities are indeed a great bargain. This is because the dividend yield has been a fantastic predictor of medium-term returns. Since 1985 it alone has explained over two-fifths of the variation in subsequent three-year changes in the index. If this relation continues to hold, we can expect the index to rise more than 50 per cent in the next three years.

But will it hold? There are reasons to think not, but these are not so overwhelming as to justify selling. A 50 per cent rise is unlikely, but some rise is to be expected.

Most obvious is the likelihood that the yield will fall not because prices will rise, but because companies will cut dividends. Three things, however, make me suspect this will not prevent the market rising.

One is that even big cuts would still leave the market reasonably well valued. We’d need a cut in dividends of almost 30 per cent – as much as we saw in the 2008-09 crisis – to take the dividend yield back to its post-1985 average. And of course that average predicts rising prices. 

Secondly, we must ask: why do dividends matter? One famous piece of economic theory – the Miller-Modigliani theorem – says they don’t. If a £20bn company foregoes £1bn of dividends, it becomes a £21bn company. Investors should in principle be indifferent between that and owning a £20bn company and getting £1bn of dividends.

So why aren’t they? It’s because a dividend cut is a signal that the company’s prospects are worse than investors thought. But would cuts really be so bad a signal now? They would tell us that we’ve suffered a massive loss of output and profits because of the lockdown. But every dog on the street already knows this. And if banks cut dividends under guidance from the Bank of England while protesting that their balance sheets are strong, the adverse signal is diluted.

And thirdly, some dividend cuts must by now be priced in. Precedent tells us this. During the financial crisis shares began to rise in March 2009 even though companies continued to cut dividends for months afterwards.

Dividend cuts alone, then, are not sufficient to overturn the buy signal coming from the current yield.

But they are not the only threat. To see some others, remember that the dividend yield is (by identity) equal to a real long-term bond yield, plus a risk premium, minus expected future long-term dividend growth.

Now, we needn’t worry about bond yields rising. If they do so, it will be because investors are willing to take more risk or because their growth expectations are increasing. In such a climate, equities would do well.

Instead, a bigger danger is that the equity risk premium will stay high. One reason for this is that many companies will be genuinely riskier. They'll emerge from this crisis with less cash and more debt. Operational gearing will be higher.

Also, as Ulrike Malmendier at the University of California at Berkeley has shown, recessions have scarring effects. They teach us that the economy is riskier than we thought – a lesson that stays with us for years, even decades.

There is, however, an offsetting mechanism here. As Harvard University’s Matthew Rabin and colleagues have shown, we often project our current attitudes too much into the future and fail to anticipate that they’ll change. For example, people pay over the odds in the summer for convertible cars and houses with swimming pools because they fail to anticipate that these will lose their appeal in the winter. Similarly, we might be projecting our current pessimism and underrating the likelihood of it changing.

There is, though, another way in which the dividend yield might stay high – if expectations for dividend growth stay low.

There are several pathways via which this could happen. One would be if we get a repeat of the post-2010 experience when governments responded to the high public debt created by the recession to embark upon fiscal austerity that depressed growth. Another would be if heightened risk aversion caused by this recession causes companies to rein in capital spending plans. Or perhaps signs of inflation caused by a mismatch between the post-lockdown recovery pattern of demand and available capital and labour will cause central banks to raise interest rates. Given that the All-Share index is dominated by multinationals, these mechanisms would have to operate globally rather than just in the UK. But this is possible given that emerging markets lack fiscal capacity while the eurozone lacks fiscal intelligence.

But again, there’s mitigation here. Listed companies tend to be older, larger and more cash-rich than others. Which means they are more likely to emerge from this recession than their competitors and so could enjoy an increase in monopoly power.

Overall, then, there are good reasons to fear that the link between the dividend yield and future returns has weakened, so we should not expect the extraordinarily high returns predicted by the past relationship. But equally, it’s hard to believe that this relationship has broken down so much that today’s high yield is meaningless. There is, therefore, reason for hope.