Risk and Return in Emerging Markets

As some of you know, I spend a lot of time thinking about macro investment strategies in emerging markets (and I write frequently for the Worldview series in NYSSA’s publications The Investment Professional, and now the Finance Professional’s Post). One of my personal projects has been to develop a tactical asset allocation process for emerging market investments.

As part of this research, I took a look at the USD denominated total returns of MSCI’s emerging market equity indexes as far back as I could readily obtain data. MSCI offers data on 22 countries it classifies as emerging markets (“frontier” markets are not included here). Now, if we plot the annualized returns vs. annualized risk (measured as standard deviation), and break the periods down into 1994-2002 and 2002-2010, we get two different pictures of emerging market risk (note that the chart shows log-returns, which are slightly smaller than pure returns).



What this shows is that from 1994-2002 (red area), emerging market investors did not get paid for taking on emerging market equity risk. In fact, the more risk they ended up taking, the worse their reward. That the returns were low in this period is not a huge surprise, since 1994 opened with the Mexican crisis, moved on to include the Asian currency crisis, Russian devaluation and default, Brazilian wobble, tech-bubble, and post-September 11th recession in the US. All in all, a tough time for markets, in spite of the US run-up.

From 2002-2010, however (blue area), emerging markets show less risk and higher returns, and, moreover, a a positive return for additional risk taken. This period includes the post-2003 boom, but it also includes the financial crisis and recent returns to June 2010.

Why choose these particular time ranges? Frankly, it was determined by the data: 1994 was the starting year for which monthly MSCI total USD returns were available for all countries in the sample, and 2002 was chosen simply because it divided the time frame into two equal blocks of time. What it does show is how emerging market performance has changed over the last 16-20 years in terms of risk and return.

Another interesting observation comes from the R^2 values of the best-fit lines for each period. One should be careful not to read too much into these statistics because the time periods are not chosen for theoretical or historical reasons, but one thing that is interesting is that there is essentially zero relationship between the level of volatility and the level of realized return in the earlier period, R^2 signals that less than 1% of return variance is explained by volatility. However, this relationship is stronger (and positive) in the later period, where volatility appears to explain 9.5% of the variance in returns.

There are two relevant points here: one is that the relationship between risk and return appears to have become stronger and more consistent with financial theory in the period since 2002, perhaps as emerging market financial systems become more sophisticated and integrated. The second is that, even to the extent that reward is linked to risk in the later period, there is still nearly 90% of returns variation that is not explained by ordinary volatility-type risk, suggesting that good country selection, even at the market index level, has the potential to add substantial value to a portfolio.