This seems odd. In 2009 equities were obviously risky and so should have yielded much more than safe assets. Today, though, equity volatility is lower than it was then and there seems little immediate risk of a deep recession. So why should shares be as lowly priced relative to bonds as they were then?
To see the answer, remember a simple mathematical fact - that the gap between the dividend yield and the yield on safe assets is equal to the equity risk premium minus expected long-term dividend growth. In 2009, this gap was high because the equity risk premium was high because of the short-term dangers facing the market. Today, though, the problem is one of long-term risk - the danger that long-term dividend growth will be low. As the University of Chicago's John Cochrane points out, the risks to long-run equity returns come from risks to dividend growth; this is because, in the long run, share prices rise at roughly the same rate as dividends.
There are two different risks to long-run growth.
One is the danger of prolonged low growth in real GDP, and hence in profits and dividends.
The mere fact that long-term real bond yields have been trending downwards for years warns us of this possibility; many economists believe that, on average, real interest rates should be roughly equal to trend GDP growth rates.
Of course, low real rates are due in part to a global savings glut. But they also reflect a lack of profitable investment opportunities due to a combination of slower technical progress, the difficulty companies have in turning innovations into profits, and the fact that capital cannot easily substitute for labour. And low investment means slow growth.
There is, though, a second danger: that incomes will shift from profits to wages. If this happens, we could get low or no dividend growth even if the overall economy grows well.
There's little doubt that distribution risk has worked to the benefit of equities in recent decades: New York University's Sydney Ludvigson and colleagues estimate that all of the big rise in US share prices since the 1980s has come because of a shift in incomes from wages to profits. Toby Nangle at Columbia Threadneedle says equities have done well because weaker long-term growth has been offset by labour's weak bargaining power.
But this could go into reverse. If unemployment in the UK and US continues to fall, labour's bargaining position could increase. And even if workers' strength in the labour market doesn't increase, it might do so in politics, in the form of a backlash against the wealth and power of the 1 per cent.
There's also the danger that the profits of existing companies could fall because of creative destruction; new techniques and new companies make old ones obsolete. Karl Marx warned that cheaper production methods force profits down because existing companies can't compete against newer ones. And subsequent research has proved him right; Yale University's William Nordhaus has shown that companies capture only a "minuscule fraction" of the total returns to innovation; the rest flow to customers or workers.
You might think the threat of creative destruction is inconsistent with the prospect of slow aggregate growth; the former is a story of rapid technical change, but the latter is a story of the lack thereof. But there might be no such contradiction. We could have weak overall growth while what technical change there is harms existing companies. This is what happened in the 1970s. New York University's Boyan Jovanovic has shown that the value of listed companies then fell sharply in large part because the IT revolution made some of them obsolete while the winners from that process were not yet quoted on the stock market: Microsoft and Oracle, the US's two largest software firms, were formed in the mid-1970s but only listed on the stock market in 1986.
It's possible that existing quoted companies will be too sclerotic to respond to technical change (such as robotisation) or to secular stagnation. If so, investors could see little, if any, dividend growth. It's such a risk that helps to justify a high gap between dividend and bond yields.
So, what can investors do about these dangers? Here are three possibilities.
First, consider a small position in venture capital or private equity funds. These give us exposure to growth outside listed firms. Such exposure, though, carries a price; the performance of such funds has been massively variable, not least because picking winners among new companies is so difficult as to be pretty much impossible.
Secondly, be wary of long-term positions in particular stocks. We can't predict the long-term winners and losers from creative destruction. For a long-term investor, tracker funds are easier.
Thirdly, hold some cash. Doing so protects us from opposite risks. One is that the risks to long-term growth recede. If this happens, shares should rise, but there'll be a big sell-off in bonds. In this case, cash saves us from losses on bonds. The opposite danger is that the risks to dividend growth actually increase, in which case cash will protect us from falling share prices.
