Join our community of smart investors

A gloomy outlook

History warns us that the medium-term outlook for shares is gloomy, despite their attractive valuations
March 28, 2018

Equity investors have a reason for hope. It’s that UK shares are cheap. The dividend yield on the All-Share index, at 3.9 per cent now, is well above its post-1985 average of 3.6 per cent.

In the short term, this is little comfort: cheap shares can always get even cheaper because markets are driven by momentum. But in the long run it is a great reason for hope because the yield has been a better predictor of returns over longer time horizons. Since 1985, for example, the yield has explained less than 2 per cent of the variation in subsequent monthly returns on the All-Share index. But it has explained almost a quarter of the variation in subsequent annual returns, over two-fifths of the variance in three-year returns, and three-fifths of the variation in five-year returns.

Insofar as history is any guide, therefore, a high dividend points to shares rising and the longer our time horizon the more confident we can be of this.

Except for one big thing. The dividend yield isn’t the only lead indicator of returns. There are others. And these send a more bearish message.

My chart summarises the story. It shows how three other lead indicators together with the dividend yield have predicted three-yearly changes in the All-Share index since 2000. (I stress that we’re talking lead indicators here, so I’m using data in (say) January 2005 to predict returns to January 2008 and not relying on contemporaneous data at all). These lead indicators gave us good notice of the tech crash and financial crisis and of the subsequent recoveries in prices. And they have worked well recently; in February 2015 they predicted that the All-Share would rise just over 6 per cent in the three years to February 2018, which it did.

The three factors we add to the dividend yield are:

◼︎ Foreign buying of US equities. Big buying leads to falling share prices. This is because it is a sign that global investors are confident enough to buy overseas shares. When sentiment is high, however, it tends to subsequently fall. And this drags down share prices.

◼︎ The ratio of the All-Share index to its 10-month average. When the index drops below its moving average, it can be a warning sign of an impending bear market. There’s momentum in share prices and it shows up even in periods as long as three years.

◼︎ The ratio of the money stock in the Organisation for Economic Co-operation and Development (OECD) countries to the MSCI world index. When this is high (relative to its trend) shares subsequently rise, and when it is low they subsequently fall. The logic here is simple. If investors have lots of cash and few shares they are likely to rebalance their portfolios towards shares, which drives prices up. Conversely, if they have lots of shares and little cash, they are likely to rebalance away from shares and towards cash, which pushes prices down.

And here’s the problem. All three of these lead indicators are bearish now.

In the past 12 months, non-Americans have bought $140bn of US shares, the most since 2011. And although the money-price ratio has risen recently thanks to the drop in share prices, it is still close to its lowest level since 2007. And, of course, prices are now below their 10-month average.

If past relationships continue to hold, these indicators outweigh the bullish message of the dividend yield. Our four indicators together point to the All-Share index falling by 20 per cent from its end-February level by February 2021 – though we’ve had over four percentage points of this already. To put it another way, it points to a less than 20 per cent chance of the market rising between now and February 2021.

Here, though, we run into a problem. These indicators also point to the market rising nicely between now and mid-2019. This is because in 2016 our indicators were sending bullish messages; in particular, foreigners were net sellers of US shares then. It is only in 2020 that our indicators predict a fall.

As an investor, I like this. As an economist, I don’t. The idea that we can predict that the market will rise and then fall seems excessively precise and depends upon the assumption that investors will ignore bearish information for months until they wake up to it in 2020. That’s a grotesque violation of efficient market theory.

There is, though, a point here even if you don’t believe the precise forecasts. It’s that we face a genuine dilemma because we have both a powerful reason to be bullish (the high dividend yield) but also good reasons to be bearish: high foreign buying of US equities and a low money-price ratio. Past relationships give us a clue as to how these will play out. Granted, that clue yields a messy implication, and one many of you might find implausible. But I’m not at all sure how to improve upon it.