In the early days of the emerging markets (EM) asset class, lumping countries together in an index might have made sense because the EM universe was small. Today, however, it is hard to argue that widely used EM bond indices serve the best interests of investors. We believe the EM asset class now needs to be re-examined as the underlying regions it covers diverge in their fundamentals. Asia and Latin America, for instance, have different characteristics. Asia tends to benefit when commodities prices are low as net importers of commodities, while Latin America tends to benefit when commodity prices are high as net exporters of commodities. Most index providers, however, use a market capitalization-weighted approach to create a fixed income benchmark. This means that most EM bond benchmarks do not lead to optimal portfolios, in our view.
How did a group of disparate countries come together to be known as emerging markets? When I started my career in the early-1990s as an analyst on the emerging markets desk of a major Wall Street investment bank, emerging markets were first being considered as an asset class. Most banks established an emerging markets group to trade a new class of bonds, known as “Brady bonds” (named after then-U.S. Treasury Secretary Nicholas Brady), that were restructured loans of lesser-developed countries. In order to get these loans off of their balance sheets, major banks needed to give them a catchier name. “Lesser-developed country bonds” did not sound enticing. As a result, they became known as “emerging market bonds.”
These emerging market countries were a mixed bag, ranging from commodity-driven countries such as Brazil and Russia to countries that had strong historical and political ties to the United States, like Panama and the Philippines. They had little in common, except for the fact that they all had defaulted on U.S. dollar loans. Descriptions of all the bonds, including their coupons, principals and structures, fit neatly into a single booklet. To firmly establish emerging market bonds as an asset class, the largest, most liquid of these bonds became part of the first emerging market index, known as the J.P. Morgan Emerging Markets Bond Index (EMBI), in 1999.
Nearly 20 years later, many fixed income investors have dedicated allocations to emerging markets. An emerging markets index based on the largest, most liquid issues made perfect sense for bond traders. The larger the issue, the higher the liquidity—and the higher the revenue they generated for their trading desks. But with several decades of hindsight, can we say that investors are making the optimal allocations to emerging markets? If benchmarks offer a clue to allocations, we believe the answer is no.
Benchmarks Fall Short as Allocators
A major shortcoming of bond benchmarks is that they typically are weighted by market capitalization. This often means that the most heavily indebted countries and companies have the highest weights in an index. As a result, investors who are using passive, benchmark-based fixed income products could end up lending the most money to countries that often can least afford to pay it back. Following the most commonly used benchmark for EM hard currency sovereign debt (J.P. Morgan’s EMBI), for instance, would have led to an approximately 42% allocation to Latin America, but only a 23% allocation to Asia as of December 31, 2016. As a driver of over 50% of the world’s GDP growth between 2007 and 2017, Asia likely deserves a greater allocation.1
Has a much higher allocation to Latin America relative to Asia served investors well? History indicates that Latin America generates a lot of volatility for the returns it produces. When we consider the three-, five- and 10-year returns of Asia versus Latin America across investment grade, high yield and local currency, Asia has consistently outperformed.
In fact, Asia has outperformed in every case from a risk-adjusted perspective. It has only slightly underperformed Latin America’s local currency bonds over the past 10 years on an absolute return basis. Benchmarks giving a higher weight to Latin America would have left investors with lower returns and higher volatility in their portfolios. Anyone who would have invested in Asian high-yield—rather than Latin American high-yield—over the past three years, five years, 10 years, or since inception of their respective benchmarks would have had higher returns with lower volatility.
Emerging Market Benchmarks—the Right Construction?
Benchmarks for emerging market bonds have evolved from their birth as all-sovereign, U.S.-dollar (USD)-denominated debt to now include corporates and local currency bonds as well. But there are still distinct benchmarks for each type:
No single benchmark, however, captures the entire investable universe.
Another major issue with benchmarks is the frequency of their rebalancings. Almost all commonly known benchmarks rebalance on a monthly basis, based on their inclusion criteria. The reasoning: since benchmarks should be replicable and representative, monthly rebalancings enable the benchmark to keep up with new issues and eliminate matured securities to provide investors with the most updated snapshot of the investable universe. While this makes sense from a benchmark-construction perspective, it might not make sense for a long-term investor who is constructing a portfolio with a multidecade investment horizon. Unfortunately, most passive strategies, “smart beta strategies” and even active managers with low tracking error mandates are incentivized to rebalance at the same frequency as their underlying benchmark to minimize their tracking error. The result could be a portfolio that engages in excessive trading to minimize tracking error. This leads to high transaction costs in an asset class that typically has higher bid-offer spreads of 10 basis points (0.10%) to 2%. Compounded over the long run, the trading costs might detract from long-term performance.
How Should Investors Allocate?
How can fixed income investors overcome the adverse selection introduced by benchmarks? We believe one approach might be to allocate regionally to emerging market bonds. Decades of history since the dawn of emerging markets have taught us that emerging market fundamentals are diverging. To understand why, consider that Asia’s GDP per capita has grown at an average of 7% per year over the past 20 years, while Latin America has grown at only 1.5% per year over the same period2. The fast-paced economic development in Asia created a supportive environment for corporate profit and credit spreads, leading to lower default rates. At the same time, the rise in economic productivity also lowered inflation and interest rates. In the past 20 years, Asia has produced a favorable fixed income investment climate and we expect this to continue.
Another logical choice for investors is to stay away from passive strategies and to invest with active managers who are benchmark-agnostic. Studies have found that active might outperform passive in fixed income investing, specifically in emerging markets3. The more benchmark-agnostic the manager, the higher the likelihood the manager is to select securities for their merits rather than for the sake of minimizing tracking errors. Allocating with active, benchmark-agnostic managers, however, might be easier said than done. Many institutional asset managers focus on information ratio as a key measure of skill. Since the information ratio is the ratio of the manager’s outperformance relative to the portfolio’s tracking error, some managers can deliver a high information ratio by minimizing tracking errors instead of maximizing returns. Overall, too much fixation on the information ratio could lead to picking mangers who are just “closet” indexers, exacerbating the index problems they were hired to mitigate.
For investors who are looking to invest in emerging markets, we think it is imperative to invest with active managers who are most focused on total return, can deliver high active share and are benchmark-agnostic. In our opinion, this can enable investors to overcome the benchmark index biases that often lead to suboptimal exposure and too much turnover.
1Source: International Monetary Fund; World Economic Outlook Database, April 2017
2Source: Bloomberg; Data as of December 31, 2016
3Jones, Robert C., and Russ Wermers “Active Management in Mostly Efficient Markets” Financial Analyst Journal Volume 67, Number 6, 2011.
Volatility is the standard deviation of returns.
Sharpe ratio is a risk-adjusted measure calculated by using standard deviation and excess return to determine reward per unit of risk.
Tracking error measures the risk in an investment portfolio that is due to active management decisions made by the portfolio manager.
Information ratio is a measure of the risk-adjusted return of an asset or portfolio.
The J.P. Morgan Emerging Markets Bond Index (EMBI) is a set of three bond indices to track bonds in emerging markets operated by JP Morgan. The indices are the Emerging Markets Bond Index Plus, the Emerging Markets Bond Index Global and the Emerging Markets Bond Global Diversified Index.
The J.P. Morgan Corporate Emerging Markets Bond Index (CEMBI) is a global, liquid corporate emerging markets benchmark that tracks U.S.-denominated corporate bonds issued by emerging markets entities.
The J.P. Morgan Government Bond Index-Emerging Markets (GBI-EM) indices are comprehensive emerging market debt benchmarks that track local currency bonds issued by emerging market governments.
The J.P. Morgan Asia Credit Index (JACI) tracks the total return performance of the Asia fixed-rate dollar bond market. JACI is a market cap-weighted index comprising sovereign, quasi-sovereign and corporate bonds and is partitioned by country, sector and credit rating. JACI includes bonds from the following countries: China, Hong Kong, India, Indonesia, South Korea, Malaysia, Philippines, Thailand and Singapore.
The Markit iBoxx Asian Local Bond Index tracks the total return performance of a bond portfolio consisting of local-currency denominated, high quality and liquid bonds in Asia ex-Japan. The Markit iBoxx Asian Local Bond Index includes bonds from the following countries: South Korea, Hong Kong, India, Singapore, Taiwan, Malaysia, Thailand, Philippines, Indonesia and China.
The J.P. Morgan CEMBI Broad Latin American High Yield Index is the Latin America High Yield portion of the JPM CEMBI Broad Index. The CEMBI is a global, liquid corporate emerging markets benchmark that tracks U.S.-denominated corporate bonds issued by emerging markets entities. The CEMBI Broad Index is more comprehensive and includes smaller issues to cover a wider array of corporate bonds.
The J.P. Morgan CEMBI Broad Latin American Index is the Latin America portion of the JPM CEMBI Broad Index. The CEMBI is a global, liquid corporate emerging markets benchmark that tracks U.S.-denominated corporate bonds issued by emerging markets entities. The CEMBI Broad Index is more comprehensive and includes smaller issues to cover a wider array of corporate bonds.
The J.P. Morgan GBI-EM Global Latin America is the Latin American portion of the JPM GBI-EM Global Index. The GBI-EM Global indices are investable emerging market debt benchmarks that track local currency bonds issued by emerging market governments.
The views and information discussed in this report are as of the date of publication, are subject to change and may not reflect current views. The views expressed represent an assessment of market conditions at a specific point in time, are opinions only and should not be relied upon as investment advice regarding a particular investment or markets in general. Such information does not constitute a recommendation to buy or sell specific securities or investment vehicles. Investment involves risk. Investing in international and emerging markets may involve additional risks, such as social and political instability, market illiquidity, exchange-rate fluctuations, a high level of volatility and limited regulation. Past performance is no guarantee of future results. The information contained herein has been derived from sources believed to be reliable and accurate at the time of compilation, but no representation or warranty (express or implied) is made as to the accuracy or completeness of any of this information. Matthews Asia and its affiliates do not accept any liability for losses either direct or consequential caused by the use of this information.