One big motive for preferring emerging markets equities has long been mistaken - that they would benefit from strong growth in less developed economies.
This is wrong for two reasons. One is that stock markets should discount expected growth in advance. The other is that economic growth need not benefit existing quoted companies. It might instead help workers or foreign companies that don't yet exist. For these reasons, researchers have found little correlation across countries between equity returns and GDP growth.
A better argument for emerging markets is that they are riskier than developed markets, and this should - over time - mean higher returns.
Sadly, however, this case is also questionable.
Certainly, emerging markets are risky. I'm not just thinking here of the danger that they will be hit by slower growth in China or higher interest rates in the US. I'm thinking in particular of crash risk.
On average since 1987 - when MSCI's data on emerging markets begin - emerging markets (in sterling terms) have had a beta with respect to the All-Share index of slightly more than one; when the All-Share moves, emerging markets move slightly more. However, this relationship disguises the fact that in really bad times, emerging markets do especially badly. For example, the 20 worst months for the All-Share index since 1987 have seen UK equities fall by an average of 8.7 per cent but in these months, emerging markets have fallen by an average of 11.1 per cent. In this sense, they carry downside risk.
But are investors compensated for this by higher returns? Certainly, they have been historically. But we can't guarantee that this will remain the case.
One measure of this is emerging markets' alpha - the portion of their returns which isn't due to their co-movement with UK equities. Since 1987, this has averaged 5.8 per cent a year - implying that emerging markets have given us this return even when UK equities have been flat.
Again, though, this average disguises substantial variation. Alpha was high in the early 1990s and for much of the 2000s. But in the last five years it has been zero. This is consistent with evolutionary finance, which tells us that assets might be underpriced but that such mispricings attract investors who raise prices and so bid away high returns for latercomers.
Another way to think about this is to consider the standard error of returns. The idea here is simple. It's that the returns we've actually seen are only a sample of true returns and possibly a biased sample. For example, if you'd looked only at emerging markets in 2010, you'd have seen a 20 per cent return. But it would be silly to infer that their true return was 20 per cent. Instead, you just happen to have looked at an unusually good year. So, how many years of returns must we see in order to get an idea of true returns?
Very many. We can quantify this by using the standard error. This is equal to the standard deviation of returns in our sample, divided by the square root of the number of years in our sample. Now, since 1987 emerging markets (in sterling) have had an annualised standard deviation of returns of 24.1 and we have a sample of 26.6 years. If we divide the former by the square root of the latter we get our standard error, of 4.7 percentage points. This means that while observed returns on emerging markets have been 9.8 per cent a year since 1987, there's a two-thirds chance that true returns have been in the range of 5.1 to 14.5 per cent. The lower bound is slightly below the UK's observed returns in this period, of 5.2 per cent (with a standard error of 2.8 percentage points).
This implies that we cannot be sure that emerging markets have genuinely good returns. It might instead be that they've just enjoyed more good luck than the UK.
Nevertheless, there is a case for investing in emerging markets. It's that they protect us from two long-term risks to UK shares which might have a low probability but which are very nasty.
One is distribution risk - the danger that incomes could shift away from profits, causing UK equities to underperform. This could happen because of a political backlash against increasing inequality; a shift in incomes towards the self-employed; a destruction of existing companies by technical change; or simply rising unit wage costs due to a mix of low productivity growth and rising real wages caused by falling unemployment.
A second danger is stagnation risk. A lack of productivity growth, dearth of monetisable investment opportunities, stagnation in the eurozone or balance of payments constraint might combine to cause years of slow growth. As we saw in Japan in the 1990s, this could be especially nasty. Not only might it be bad for shares - to the extent that investors are consistently disappointed - but because it means job insecurity and low wage growth, too. In this scenario, younger investors would suffer a double loss - on their human wealth as well as their financial wealth.
These dangers are long-term ones. If they happen at all, they will play out over many years. And they are probably greater for developed markets than for emerging ones. This means that long-term investors could regard emerging markets as a way of diversifying against these threats.
Paradoxically, therefore, the case for emerging markets isn't so much that they offer high returns but that - despite their high volatility and crash risk - they offer an ability to spread some risk.
