To answer this, we must know what it is that drives emerging markets returns.
In fact, just eight variables can explain three-quarters of the substantial variation in annual returns on MSCI's emerging markets index (in sterling terms) since 1995. Four of these are especially important.
One is commodity prices. A 10 per cent fall in the S&P/GSCI is associated, other things equal, with a 3.6 per cent fall in emerging markets on average. The recent fall in both commodity prices and emerging markets is therefore not unusual.
But the causality here is mixed. In some cases, weaker commodity prices cause emerging markets to fall; lower oil prices, for example, have clobbered Russian stocks. In other cases, though, it is the things that cause lower commodity prices that hurt emerging markets, such as weaker growth in commodity-using economies, such as China.
A second factor is the Japanese stock market. A 10 per cent fall in this (in sterling terms) is associated with a 6.8 per cent drop in emerging markets in sterling terms. What's good for Japan is good for emerging markets. If we control for Japanese equities, however, there is no significant correlation between US equities and emerging markets.
Third, there is the dollar-sterling exchange rate. A stronger pound is associated with emerging markets falling in sterling terms. The effect here, though, isn't as strong as you might expect: a 10 per cent rise in the pound is associated, on average, with only a 2.1 per cent fall in emerging markets in sterling terms.
Fourth, there are US corporate bond yields, as measured by the average yield on Baa-rated bonds. A one percentage point rise in these is associated with an 11.3 per cent drop in emerging markets. Tighter US financial conditions or increased credit risk (caused by, say, weaker US growth) are bad for emerging markets.
On top of these factors, there are some things that predict emerging markets returns. One of these is a momentum effect: emerging markets tend to rise in the 12 months after they have been high relative to US stocks. Sadly for emerging markets investors, this force is against them, as emerging markets are low relative to the US. Offsetting this, though, is a valuation effect: emerging markets tend to fall after they have been high relative to Japanese stocks. With this ratio only slightly above average, it isn't a factor either way.
A more hopeful indicator for investors is US Treasury bond yields. Historically, emerging markets have tended to do well after these have been low - which is the case now.
We can use these factors to get an idea of where emerging markets are heading. If past relationships continue to hold and if commodity prices, Japanese equities, sterling and US corporate yields don't change, then MSCI's emerging markets index will rise 12 per cent in sterling terms in the next 12 months, with only a one-in-six chance of it falling.
There are a lot of 'ifs' in that statement. But it has its uses.
For one thing, it alerts us to the risks. I would highlight here US corporate bond yields. If rises in the fed funds rate cause a general sell-off in these, emerging markets would suffer. Another risk, of course, is lower commodity prices. Here, though, we have some luck. These are partly predictable by the Chinese money stock, and there are signs that this has picked up in the last two months - which might point to an end to the drop in commodities soon.
Secondly, all this tells us that if you are holding emerging markets as well as commodity stocks or commodity ETFs and Japanese stocks, then you are not as diversified as you might think, because all these assets tend to move in the same direction.
Maybe we can't tell where emerging markets are going - the future is unknowable. But it might help to know what the risks are.
The statistics
This table shows the formal model I'm using, It does a good job of explaining the drop in emerging markets in the 12 months ending July, suggesting that there's no sign of any breakdown in it.
Instead, the biggest failures of the model came in 2007, when it under-predicted returns, and in 2013 when emerging markets did worse than predicted.
The intercept looks high. But it implies that if all variables were at their post-1994 averages, then the MSCI emerging markets index would rise by 1 per cent per year - which isn't odd at all.
| ||||||||||||||||||||||||||
I confess to being a little unhappy with the combination of a negative coefficient on the ratio of emerging markets to Japan and positive coefficient on the ratio to US prices. But the facts aren't always consistent with a nice tidy story.
