Join our community of smart investors
OPINION

Long-term risks

Long-term risks
June 27, 2017
Long-term risks

History tells us this. Thanks to the Bank of England, we have data on share prices going back to 1700. During this time, the average increase in prices, adjusted for inflation, has been 1.1 per cent a year, with a standard deviation of 15.4 percentage points. If we assume that returns in one year are independent of those the next, this implies that there's around a 40 per cent chance that prices will fall in real terms even over a 10-year period. In fact, this is roughly consistent with another way of looking at the numbers. There have been 31 non-overlapping 10-year periods in the years to December 2016, and prices fell in real terms in 15 of them.

Shares, then, are by no means guaranteed to rise even over quite long periods. You might find this surprising. If so, remember three things.

First, these data exclude dividends. Over long periods, however, these account for most of the returns on equities. To put this another way, if you're spending your dividends, don't expect your capital to remain intact.

Secondly, for most of this period, the economy grew only slowly: real GDP per person rose by only 0.6 per cent per year between 1700 and 1939. That limited growth in dividends and hence share prices. The post-war boom was unusual by historic standards. And it might be that what we call secular stagnation is in fact a reversion to normal growth.

Thirdly, our attitudes are shaped not by evidence alone but by our formative years. For many of us, these years were ones in which shares did extraordinarily well - from the late 1970s to the late 1990s. This period, however, is unrepresentative of the long-run history of shares.

This poses the question. Should we therefore prepare for decades of low returns on equities, with a big risk of falls even over long periods? Several things suggest so.

One is mainstream theory. Back in 1986, Rajnish Mehra and Ed Prescott pointed out that equities shouldn't outperform cash by very much - by no more than around 1 per cent per year, even including dividends. And in fact since then, theory has been pretty much correct.

So why did the market do so well in the 20th century? One reason is simply that investors worst fears never materialised. For much of the period from 1914 to the fall of communism, share prices were depressed by the fear of catastrophe: war, prolonged depression or the collapse of capitalism. These fears weren't entirely misplaced. Will Goetzmann and Philippe Jorion have pointed out that of the 24 national stock markets that existed in 1931 10 subsequently suffered a long-term closure. "Market failure is not a remote possibility," they warn. As those fears receded, prices recovered causing longer-term returns to appear good and stable. It's for this reason that Elroy Dimson and Paul Marsh have described the 20th century as the "triumph of the optimists". Today, though, markets are not pricing in any such existential threat. And this means there's no risk premium to be reaped.

There are other reasons to fear sustained low returns.

One lies in economic policy. Previous recessions saw cuts in interest rates which stabilised output and share prices. But with rates now close to zero (and negative rates of dubious efficacy) this stabilising force will be absent.

Yes, some economists such as UCLA's Roger Farmer say that unconventional monetary policy can work. And helicopter money almost certainly would do so. But the fact is that central banks have less ammunition now than they did in the past. And this increases the risk that shares won't bounce back as much as they have in the past.

It's not just monetary policy that's a danger, however. Previous market downturns have been reversed by governments adopting pro-business policies. For example, the profit squeeze in the mid-1970s - which caused serious people to question whether capitalism could survive - led to the business-friendly policies of Thatcher and Reagan. If, however, we are entering an era of populism - as betokened by support for Trump, Brexit and far-right parties in Europe - we might get the opposite. We might instead see anti-business policies.

There are (at least) three other nasty risks:

■ Secular stagnation. Prolonged weak growth would cause persistent earnings disappointments and a downgrading of equity valuations as investors gradually cut their expectations for future dividend growth.

■ Distribution risk. Sydney Ludvigson, Martin Lettau and Daniel Greenwald have estimated that shifts in the distribution of income between wages and profits have been major causes of long-term stock market fluctuations. In the 1980s and 1990s, these caused shares to boom. But - just as in the 1970s - we might see the opposite. Tighter labour markets and the rise of populism could hurt profits even over long periods.

■ Creative destruction. It's normal and healthy for technical change to create some companies and destroy others. This can, though, be bad for stock markets if the destroyed companies are quoted while the created ones are not. This has happened before. New York University's Boyan Jovanovic says one reason why shares slumped in the 1970s was that anticipation of the IT revolution devalued incumbent companies while the beneficiaries of that revolution, such as Apple and Microsoft, had yet to list on the market.

Granted, these three risks are partially mutually exclusive: a world of secular stagnation probably won't be one of high creative destruction. But they are surely plausible risks.

Perhaps, therefore, long-term investors should not assume that equities will outperform. The notion that long-term investors should have the bulk of their money in equities might be yet another example of a legacy idea - something that made sense once but no longer is true.