Emerging Markets Offer Investors a Key Benefit in Diversification
Investors the world over often have a strong bias for their own home markets and fail to seek diversification through investments in other countries and markets. However, the world's financial markets seem to be substantially interlinked. The events of 2008 showed that all markets can become highly correlated.
Furthermore, popular emerging markets, such as Brazil, have shown strong correlations to the U.S. markets. I previously wrote an article on Seeking Alpha, titled "Looking For Better Brazilian Diversification", that discussed Brazilian ETFs and ADRs reviewed instruments with lower correlations to SPY. However, in concept, emerging markets still offer diversification although in times of high stress, or perhaps high euphoria, they become even more correlated to SPY.
This article will look at a fairly large list of emerging market ETFs, some with a country focus and others with a regional focus, to identify those that provide the best diversification benefits to U.S. investors as proxied by the SPDR S&P 500 Trust (NYSEARCA:SPY). This analysis is from a U.S. point of view. Investors in Canada or the UK would have slightly different correlations from their home market indices. Different markets might offer them better diversification benefits than they would to U.S. investors. The list of ETFs considered is as follows:
Emerging Market ETFs
|(NYSEARCA:EEM)||iShares MSCI Emerging Markets Index Fund||38.41||Global|
|(NYSEARCA:VWO)||Vanguard Emerging Markets ETF VWO||64.17||Global|
|(NYSEARCA:EWX)||SPDR S&P Emerging Markets Small Cap ETF||0.61||Global|
|(NYSEARCA:EZA)||iShares MSCI South Africa Index||0.60||Africa|
|(NYSEARCA:AFK)||Market Vectors Africa Index ETF||0.12||Africa|
|(NYSEARCA:EWT)||iShares MSCI Taiwan Index Fund||3.45||Asia|
|(NYSEARCA:EWY)||iShares MSCI South Korea Index||4.92||Asia|
|(NYSEARCA:EWH)||iShares MSCI Hong Kong Index||2.12||Asia|
|(NYSEARCA:EPP)||iShares MSCI Pacific ex-Japan Index Fund||3.97||Asia|
|(NYSEARCA:EWS)||iShares MSCI Singapore Index||1.87||Asia|
|(NYSEARCA:EWM)||iShares MSCI Malaysia Index Fund||1.00||Asia|
|(NYSEARCA:THD)||iShares MSCI Thailand Investable Market Index Fund||0.58||Asia|
|(NYSEARCA:VNM)||Market Vectors Vietnam ETF||0.31||Asia|
|(NYSEARCA:FXI)||iShares FTSE/Xinhua China 25 Index Fund||7.03||China|
|(NASDAQ:PGJ)||Powershares Golden Dragon Halter USX China Portfolio Fund||0.41||China|
|(NYSEARCA:GUR)||SPDR S&P Emerging Europe ETF||0.28||Emerging Europe|
|(NYSEARCA:RSX)||Market Vectors Russia ETF||3.91||Emerging Europe|
|(NYSE:CEE)||Central Europe and Russia Fund, Inc. ||0.58||Emerging Europe|
|(NYSEARCA:PLND)||Market Vectors Poland ETF||0.08||Emerging Europe|
|(NYSEARCA:ESR)||iShares MSCI Emerg Mkts Eastern Europe Index||0.04||Emerging Europe|
|(NYSE:IIF)||Morgan Stanley India Investment Fund, Inc.||0.48||India|
|(NYSEARCA:ILF)||iShares S&P Latin America 40 Index Fund||2.31||Latin America|
|(NYSEARCA:EWW)||iShares MSCI Mexico Investable Mkt Idx||1.60||Latin America|
|(NYSEARCA:GXG)||Global X/InterBolsa FTSE Colombia 20 ETF||0.20||Latin America|
|(NYSEARCA:EPU)||iShares MSCI All Peru Capped Index||0.45||Latin America|
|(NYSEARCA:ECH)||iShares MSCI Chile Investable Market Index||0.86||Latin America|
|(NYSEARCA:EWZS)||MSCI Brazil Small Cap Index Fund||0.06||Latin America|
|(NYSEARCA:TUR)||iShares MSCI Turkey Investable Market Index Fund||0.58||Middle East/Europe|
|(NYSEARCA:EGPT)||Market Vectors Egypt Index ETF||0.07||Middle East/Africa|
SPDR S&P Emerging Middle East & Africa ETF (GAF) |
The above list was assembled to provide a good mix of ETFs from different regions. The next part of the analysis is to compute correlations to SPY and determine which ones have the lowest correlation. I calculated the correlations over a couple of different time periods ranging from 19 months (due to price history) up to 36 months which is used as the standard time for some data services when calculating betas.
Correlations can change over time and also due to length of time used (e.g. 30 vs. 36 months). It should be noted that the 19 month time frame was only included to accommodate some ETFs with short price histories and is probably not sufficiently long, but should provide a general perspective.
Correlations range from -100% to 100%. A correlation of negative 100% offers the best diversification benefits in terms of reducing portfolio volatility. A correlation value of 100% offers no diversification benefit. Very few securities offer negative correlations. U.S. treasuries, gold, and emerging market bonds offer very low correlations. The following table shows the correlations for the ETF list:
Correlations to SPY
|Ticker||19 Month||30 Month||36 Month|
The first observation is that a general broadbased emerging market ETF like EEM, VWO or ILF are highly correlated already to SPY. So to find better diversification an investor would need to place more concentrated investments on ETFs with much lower correlations such as ECH (Chile), IIF (India) or EWM (Malaysia). It appears that both India and Malaysia offer pretty low correlations.
Some of the newer ETFs might like VNM (Vietnam), EGPT (Egypt) and GXG (Colombia) also seem like they might have low correlations; however, there is still little data. ESR and PLND seem pretty correlated to the SPY. The second observation is that for specific country ETFs there is some variation due to time frame, while the broader ETFs seem to show less variation.
The wrinkle for reducing portfolio volatilty is that something with a reasonable correlation but a high volatility actually will not lower the portfolio volatility. This is the case with many emerging markets, while you might be diversifying your portfolio you are still increasing the volatility of your investment portfolio. The following table shows the monthly return volatility of these ETFs:
|Ticker||36 Month Volatility||36 Month Normed to SPY||15 Month Volatility||15 Month Normed to SPY|
Source: Calculated from returns from monthly split and dividend adjusted prices from Yahoo Finance.
With the exception of FXI and EWM, all emerging market ETFs have higher volatilty than SPY. However, depending on your perspectives, this might not be bad thing. While volatility is commonly used as a measure of risk, volatility, like correlations, are historical measures.
In contrast, risk is a forward looking perspective and so the notion is, there are more things that can happen than will happen. However, I do believe both correlations and volatility provide good starting points for thinking about the future. The above table also shows that volatility is impacted by the timeframe for the calculation.
In looking at building a diversified portfolio, I think finding reasonable opportunities with low correlations is key. What this article did not get into was any assessment of whether a specific country ETF is a good investment nor does it address the liquidity issues of some of the smaller ETFs, some of the smaller less common ETFs might have liquidity issues which can erode the actual realization of any investment returns. This analysis should be used in addition to your own investment research.
It can be applied in two ways:
- Start with ETFs that have low correlations to the SPY and identify the most promising investments there.
- Use it as a check, if you have two good opportunities perhaps than choose the one with the lower correlation.
As noted earlier, India (PIN and IIF), Malaysia (EWM), Chile (ECH and CH) Vietnam (VNM), and Colombia (GXG) might be good prospects. For example, (THD) had the best returns over both the 12 month and 36 month timeframe. The chart below shows ETFs that provide good returns:
|Ticker||12 Month Return||36 Month return|
Source: Calculated from monthly split and dividend adjusted prices from Yahoo Finance.