A version of this post was submitted to Seeking Alpha

Summary

• Frontier market funds have provide excellent risk-adjusted performance over their short history
• Surprisingly, frontier market funds have been less volatile than either the developed or emerging markets funds
• Frontier market funds offer excellent diversification for developed markets portfolios

In a search for performance and diversification, some investors have turned to alternative investments such as emerging market stocks.  If you are even more adventurous, you may be ready to boldly go where few investors have gone before and invest in frontier markets.  However, before you venture into this exotic asset class, you should take time to assess the risk versus rewards.  This article will review some of the Exchange Traded Funds (ETFs) and mutual funds that invest in stocks from these less developed regions of the world.

Frontier Markets

There is no universally accepted definition of frontier markets but they are generally stock markets that are in the early stages of economic development.  There are more than 115 stock markets worldwide and about 60 can be characterized as frontier markets.  Collectively, these markets have over a trillion dollars in market capitalization.  Figure 1 lists the frontier market countries as defined by a September, 2011 Standard and Poors report.

Figure 1:  List of Frontier Markets by region (click to expand)

Pros and Cons

Some of the qualitative risks typically associated with frontier markets are described below (more quantitative risks in terms of volatility will be discussed later).

• The stock markets in frontier countries are usually smaller and less liquid than their emerging market brothers.  This makes it more difficult for large investors to establish positions.
• The countries may not be politically stable and there is the possibility of widespread corruption.  For example, the Egyptian stock market fell 40% during the revolution of 2011.
• Regulatory laws may be at the whim of the party in power and may create barriers for foreign investors.  In extreme cases, nationalization of assets is a possibility.

With these risks, why would anyone want to invest in these markets? To make money, of course!  Here are some of the positive attributes of frontier markets:

• Many frontier countries have rapidly growing gross domestic products (GDP).  The International Monetary Fund (IMF) predicts that the GDP of many frontier markets will increase by more than 5% next year compared with less than 3% for developed companies.  As examples, Nigeria is expected to grow over 7% and Bangladesh by more than 6%.
• Stocks within frontier countries often have attractive evaluations in terms of price to earnings ratios.
• Frontier markets typically have a low correlation with both developed and emerging markets, thus helping to diversify your portfolio.  As an example, emerging markets were hit hard last year as tapering talks and slowing growth took its toll.  However, frontier markets had a banner year.

Frontier Market Funds

Within the boundaries of frontier markets, there are approximately 6,000 companies to choose for your portfolio.  However, it is difficult for individual investors to invest directly in these companies and I personally would stick to mutual funds and ETFs.  Unfortunately, there are not that many funds to choose from.  There is only one mutual fund that has been in existence over 5 years and most of the other frontier funds and ETFs were launched in 2010 or later.  So for this study, I will look at mutual funds and ETFs that have been around at least a year and have over $100 million in market capitalization.  Also note that I am normally not a fan of mutual funds that charge a sales load but because of the limited sample, I have included them in this analysis.  These selection criteria reduce the field to 5 mutual funds and 1 ETF, which are summarized below.

• Templeton Frontier Market Fund (TFMAX) This mutual fund is the largest and oldest frontier market fund with over $1.5 billion of assets under management (AUM).  It was launched in October, 2008 and was closed to new investors in June, 2013.  The fund has a load of 5.75%.  The fund has 94 holdings with 33% invested in Middle East countries, 22% in Africa, 16% in Europe, 13% in Asia, and 13% in Latin America.  This fund has an expense ratio of 2.1% and does not provide any yield.
• Wasatch Frontier Emerging Small Countries (WAFMX ).  This fund was launched in January 2012 and is the second largest fund with $1.1 billion AUM.  This is one of the few no-load funds focused on frontier markets.  This fund has 126 holdings distributed primarily among Africa (36%), Asia (29%), and the Middle East (24%).  As on 9 May, this fund closed to new investors that buy the fund via third party marketplaces but remains open for purchases directly from Wasatch Funds. The fund has an expense ratio of 2.3% and does not provide any yield.
• Harding Loevner Frontier Emerging Markets (HLMOX ).  This no-load fund was launched in December, 2010 and remains open to new investors.  It has about $460 million AUM.  This fund has 78 holdings distributed primarily among Africa (23%), Asia (28%), Middle East (24%), and Latin America (11%).  It has an expense ratio of 2.2% and does not provide any yield.
• Morgan Stanley Institutional Class Frontier Emerging Markets (MFMPX ).  This mutual fund was launched in September, 2012 and has a 5.25% load.  It has $511 million AUM and is still open to new investors.  The portfolio contains 62 holdings distributed primarily among Africa (59%), Asia (19%), and Europe (10%).  By policy, at least 25% of the assets must be invested in the banking industry.  The fund has an expense ratio of 2% and has a small 0.3% yield.
• HSBC Frontier Markets Fund (HSFAX) This mutual fund was launched in September, 2011 and has $152 million AUM.  It has a 5% sale load and is currently closed to new investors.   It has 67 holdings distributed among Middle East and North Africa (46%), Asia (24%), Sub-Saharan Africa (10%), and Eastern Europe (9%).  The expense ratio is 2.2% and the fund provides a small yield of 0.3%.
• iShares MSCI Frontier 100 (FM).  This is the only ETF that satisfies my selection criteria. This fund was launched in September, 2012 and currently has $803 million AUM.  This ETF tracks the MSCI Frontier Market 100 index, which is a cap weighted index of 100 stocks that meet minimum liquidity levels and have sufficient shares available for foreign investment.  Currently the fund holds 103 securities with stocks from Kuwait comprising 20% of the portfolio followed by Qatar (17%), UAE (15%), and Nigeria (15%).  However, MSCI is in the process of reassigning Qatar and UAE to the emerging market classification.  Thus, the fund is in the process of transitioning away from Qatar and UAE to other frontier markets.  The ETF has an expense ratio of 0.8% and provides a small yield of 0.9%.

To compare these funds with developed and emerging markets, I have also included the following two ETFs for reference.

SPRD S&P 500 (SPY).   This ETF tracks the S&P 500 index and is a good proxy for the developed markets.  The fund has an ultra-low expense ratio of 0.09% and yields 1.8%.

iShares MSCI Emerging Markets (EEM).   This ETF provides broad cap-weighted exposure to emerging market equities.  It tracks the price of 800 securities in 21 emerging markets and has an expense ratio of 0.7%.  It yields 2%.

Risk versus reward 5 years

To determine the relative performance of the frontier market funds, I plotted the annualized rate of return in excess of the risk free rate (called Excess Mu in the charts) versus the volatility for the past 5 years. The Smartfolio 3 program (www.smartfolio.com ) was used to generate the plot shown in Figure 2.

Figure 2. Risk vs. reward over past 5 years (click to expand)

As we have noted, only TFMAX had a 5 year history.  It should be noted that the return associated with this mutual fund did not account for the load that had to be paid initially to purchase the fund.  Figure 1 is a somewhat surprising plot because the frontier markets were less volatile than the S&P 500 and significantly less volatile than emerging markets.  To be honest, I did not expect this result but it can likely be attributed to the fact that individual frontier countries had a low correlation to each other.  This is an illustration of modern portfolio theory (MPT) that states that if the constituents of a portfolio are not highly correlated, the combined portfolio will have a volatility that is less than the constituent volatilities. MPT was an amazing discovery made by an economist named Markowitz in 1950 and his work was so revolutionary that he was awarded the Nobel Prize.

Another explanation for the low volatility may be the relatively small amount of foreign ownership in frontier markets.  When a large group of investors crowd into an asset there is a tendency for traders to follow the hot money and increase volatility.  The fact that frontier markets are not liquid enough for the large players may have decreased the volatility.  If hot money does discover the frontier markets, prices and volatility could explode upward.

As Figure 2 illustrates, frontier markets generated a higher return than emerging markets and accomplished this with lower volatility.  The comparison to the S&P 500 is not as straightforward.  The S&P generated higher returns but also had larger volatility.  Was the increased return worth the increased volatility?  To answer this question, I calculated the Sharpe Ratio.

Sharpe Ratio

The Sharpe Ratio is a metric, developed by Nobel laureate William Sharpe that measures risk-adjusted performance. It is calculated as the ratio of the excess return over the volatility. This reward-to-risk ratio (assuming that risk is measured by volatility) is a good way to compare peers to assess if higher returns are due to superior investment performance or from taking additional risk. In Figure 2, I plotted a red line that represents the Sharpe Ratio associated with EEM. If an asset is above the line, it has a higher Sharpe Ratio than EMB. Conversely, if an asset is below the line, the reward-to-risk is worse than EMB.  Similarly, the blue line represents the Sharpe Ratio associated with SPY.

Using the Sharpe Ratio, it is apparent that TFMAX has a much better risk-adjusted performance than EEM and about the same risk-adjusted performance as SPY.  This is very impressive since the S&P 500 has been in a rip-roaring bull market over the past 5 years.

In order to include additional funds in the analysis, I reduced the look-back period to 1 October, 2012. The results are shown in Figure 3.

Figure 3. Risk vs. reward since 1 October, 2012 (click to expand)

Some interesting observations are apparent from the figure.  All the frontier market funds have been on a tear lately, handily beating both SPY and EEM on an absolute basis and a risk-adjusted basis.  The best risk-adjusted return was booked by MFMPX but as we have indicated, the return data does not take into account the required load payment.  The two no-load funds (HLMOX and WAFMX) lagged on a risk-adjusted basis but their no-load nature makes them serious contenders.  The lone ETF (FM) had good absolute returns but had a relatively high volatility (similar to SPY).  Even though FMs risk-adjusted performance was less than that of the mutual funds, it still had better risk-adjusted performance than either SPY or EEM.

Diversification

Next I wanted to see if the diversification promised by these frontier market funds lived up to expectation.  To be diversified, you want to choose assets such that when some assets are down, others are up.  In mathematical terms, you want to select assets that are uncorrelated (or at least not highly correlated) with each other.  I calculated the pair-wise correlations associated with the selected funds.  The results are provided in the 19 month correlation matrix shown in Figure 4.  As is evident from the matrix, these funds provided excellent diversification.  FM had the highest correlation with SPY and EEM but the correlation was still small at 50%.  The mutual funds were even less correlated with the reference ETFs, scoring about 30% to 40%.  The funds were only moderately correlated among themselves, with correlations in the 50% to 70% range.  So my conclusion was that these funds had lived up to expectation and provided an excellent amount of portfolio diversification.

Figure 4. Correlation over past 19 months (click to expand)

Bottom Line

The loaded funds have done well and for very long term investors, the load is not as significant.  However, for shorter time horizons, I would only purchase no-load funds or the ETF.  With the stock market now in its fifth year, I would keep a close eye on these funds but as long as the bull is healthy, frontier markets should provide a