Emerging market bonds

Emerging market debt has a long history in the international debt markets, going back to the early nineteenth century. Beginning in 1993, contemporary U.S investors have had access to the emerging market bond sector through the medium of mutual funds. Historically, most emerging market debt was in the form of external sovereign debt, denominated in an external currency (dollar, pound, euro). Today a majority (3/4) of emerging market debt is issued in local currency. Thus investors have access to funds holding bonds in U.S dollars; funds holding bonds in local currencies; and funds that hold both. Emerging market bond mutual funds are actively managed. Recently the ETF marketplace has offered investors indexed portfolios of emerging market bonds.



Returns and risks
Prior to the Mexican default in the early 1980's, most sovereign emerging market debt was a product of bank financing. The establishment of so-called "Brady bonds" during the ensuing Mexican debt restructuring transferred emerging market debt to the public bond markets. The data history of emerging market bond investment is therefore brief. Available data begins in 1991 with the creation of the JP Morgan EMBI index, an index of U.S. dollar emerging market bonds.

As Erb, Harvey, and Visconti noted in 1997, emerging market bond returns from 1991-1997 outperformed the performance of the S&P 500 index. They also noted that this short initial return sequence also showed high volatility and negative skewness, suggesting that downside risk was high, although not realized in the short term returns data. This insight was proved correct, as the "Asian Contagion" financial crisis in 1998 produced negative returns in 1998. Returns from 2000-2010 have been somewhat less volatile (as measured by standard deviation, see table below), but the asset class has retained its quality of negative skewness. Erb, Harvey, and Visconti state:

"One difficulty with measuring skewness is that it likely changes through time. Therefore looking at past data may give no indication of future expected skewness. This is the so called “peso problem” in economic theory. Looking at past currency movements, you may see little variation in rates during a managed float regime. However, there is a probability of a devaluation that you cannot detect from looking at past data. This is just the definition of negative skewness."

Emerging market bonds are subject to three major risks:
 * Interest rate risk. Most emerging market bond portfolios hold intermediate to long term bonds. Thus the portfolios have high durations and high interest rate risk: bond principal will fluctuate with changes in interest rates.
 * Currency risk. The marketplace is now dominated by local currency bond issues. These bonds bear currency risk, the risk that changes in exchange rates can result in gains or losses.
 * Sovereign risk. The risk that a country will default on its external non-local currency debt. Closely associated with this risk is inflation risk affecting local currency bonds, where a bond's principal value can effectively be destroyed by high inflation.

'''Table 1. Emerging Market bond index returns'''

Correlation
The Vanguard Institutional white paper, Global fixed income: Considerations for U.S. investors, provides correlation data between emerging market bonds and major asset classes for the period between 2002-2010. Over this period, dollar based emerging market debt had higher correlations (between 0.55 and 0.65) with US stocks, developed market stocks, and emerging market stocks, as well as with US bonds; and lower correlations (between 0.30 and 0.45) with hedged and unhedged international bonds. Local currency bonds had lower correlations (0.30) with US stocks, developed market stocks, and emerging market stocks; and higher correlations (between 0.45 and 0.60) with US bonds, hedged international bonds, and unhedged international bonds.

Erb, Harvey and Visconti note that emerging market bond correlations with other asset classes are subject to shifts, depending on the state of the economy and markets. Thus during the 1998 "Asian Contagion" currency crisis, emerging market bonds where highly correlated with US stocks, emerging market stocks, and high yield bonds, all of which fell in price, and where negatively correlated with US treasuries, which rose in price.

Default risk
Bond defaults and restructurings have a long history. Prior to the nineteenth century, defaults were common among European nation states. With national independence movements beginning in the early nineteenth century and continuing to this day, bond defaults and restructurings became prevalent among these new "emerging markets."

Defaults and restructurings tend to occur in clusters. Historical clusters include:
 * 1824-1834
 * 1867-1882
 * 1890-1900
 * 1911-1921
 * 1931-1940
 * 1976-1989
 * 1998-2004

The following table, based on the Reinhart, Rogoff, and Sevastano paper, Debt intolerance, provides historical default and restructuring data (1824-2001). The table also shows the amount of time countries have spent under default or restructuring, which has often been considerable. Reinhart, Rogoff, and Sevastano also note that defaulting nations have also frequently been high inflation nations, having experienced inflation at or above 40% per month.

The following table, drawn from the Vanguard Institutional paper, What's Next for the Eurozone and based on data supplied by Moody's, lists emerging market defaults and principal losses over the 1998 - 2006 period.

Arguments for holding emerging market bond funds
In the Vanguard Institutional white paper, Global fixed income: Considerations for U.S. investors, conditional recommendation for the emerging market bond asset class is enlisted in two main instances:
 * 1) Investors seeking total market international bond exposure. The Barclays Aggregate Global Bond Indexes do not include emerging market bonds.
 * 2) Potential for added diversification. The return properties of emerging market bonds: higher expected returns, higher volatility, and expected low correlations to other asset classes, may appeal to certain investors who wish to include it as a portion of a broadly diversified international bond investment.

In the first edition (2006) of author Rick Ferri's book, All About Asset Allocation, the author included emerging market bonds as a potential portion of a widely diversified taxable bond allocation. The author suggested the following allocation:

In the second edition of All About Asset Allocation (2010), Ferri no longer includes emerging market bonds in his fixed income allocations, citing the high costs (0.50% to over 1.00%) incurred in implementing emerging market bond portfolios.

Arguments against holding emerging market bond funds
Larry Swedroe et al., in their book, The Only Guide You'll Ever Need for the Right Financial Plan: Managing Your Wealth, Risk, and Investments argue against including emerging market bonds in investment portfolios. The authors cite the higher default risk and extreme volatility associated with emerging market bonds as negative features of this asset. They make the following clarifying arguments about the negative features:
 * They are illiquid.
 * There is the potential for large losses. They exhibit negative skewness and high kurtosis.
 * While having low correlation to other asset classes, the correlation of risk to equities tends to turn high at the wrong time: when equities are in distress and investors look to their bond portfolios to provide stability.
 * Returns are earned in a tax inefficient manner.
 * Implementation costs are high. Mutual funds are needed and they are expensive.

The authors advise seeking emerging market exposure by investing in emerging market stocks; and state that emerging market bonds should be considered an allocation to the equity asset class (i.e. drawn from the portfolio allocation to equities, and not the portfolio allocation to bonds).

Index funds and ETFs
As of 2011, the only indexed portfolios of emerging market bonds available to U.S. investors are ETF portfolios. Four broad market ETF's, all introduced within the past four years, are included in the table below. Two ETF portfolios, the iShares JPMorgan USD Emerging Markets Bond Fund and the PowerShares Emerging Markets Sovereign Debt Portfolio ETF invest in U.S dollar denominated securities. Each fund holds less than 100 bonds. Two of the ETFs listed below, Market Vectors Emerging Markets Local Currency Bond ETF and the SPDR Barclays Capital Emerging Markets Local Bond ETF, invest in local currency bonds. These two ETFs hold between 100-500 bonds. All four ETFs invest in a wide number of emerging market countries, which is prudent, given the long history of sovereign default risk among emerging market countries. To further diversify against default risk, investors would also be well advised to make sure that the portfolios are not concentrated among a small selection of countries.

The ETF's have lower expense ratios than emerging market bond mutual funds or closed end funds, although ETF transaction costs (commissions, spreads, discount/premium) can significantly add to investment costs.

Active funds and closed end funds
There are a large assortment of actively managed emerging market bond funds. Most of these funds are load funds. No-load funds are available at lower costs with the lowest cost funds registering expense ratios running between 0.88% and 0.99% per annum.

Emerging market bond funds are also available as closed-end funds. These funds are traded on stock exchanges, often trade at discount or premium to net asset value, often employ leverage, and often utilize managed distribution policies. Most have higher than average expense ratios.

Active funds add to the risks inherent in emerging market bonds by introducing manager risk, the chance that poor security selection or focus on securities in a particular sector or category will cause the fund to underperform relevant benchmarks or other funds with a similar investment objective.