Emerging market fixed-income represents a distinct asset class, not only because the assets of which it is composed share common characteristics, but also because treating it as such is necessary to solving the problem of minimizing risk at any given rate of return.
We believe the class embodies special risk characteristics, such as intermittent market access, financial fragility, legal complexity, political volatility, and opaqueness that preclude the long-term success of a generalist approach. Unlike the more established markets, emerging market fixed income is a high-returning asset class that moves through relatively short boom-bust cycles. It takes expertise and experience to understand which way the cycle is turning.
Enough common and distinct characteristics run across emerging market fixed income to conclude it is a separate asset class. Emerging market fixed income includes both hard and soft currency bonds and notes. Brady bonds are the best known, but they now constitute a minority of the investment universe, exceeded in size by local currency instruments and beginning to be rivaled by Eurobonds. These assets combined now amount to approximately $1 trillion.
What are these characteristics that unite Nigerian promissory notes, Polish treasury bills and dollar-denominated Eurobonds of Mexican cement companies and separate them from other global bonds and from the U.S. high-yield market?
First, the issuing countries are considered "emerging" and generally have a gross domestic product per capita of $9,385 (World Bank definition) or less. Countries with low GDP/capita ratios have relatively few participants in the financial system and this implies a shallowness in domestic capital markets and a certain degree of financial fragility.
Second, the international issuance of these countries is generally in currencies other than their own. The Germans issue in deutsche marks, the Japanese in yen, and the French in francs. Emerging countries issue in dollars, mainly, but also in yen, sterling, and marks because their local currencies are not ones that would find ready acceptance on the international markets, save, perhaps, during periods of great speculative enthusiasm. The Financial Times lists exchange rates for more than 200 such currencies, including the Laos New Kip, the Nicaraguan Gold Cordoba, and the mighty Vanuatu Vatu. We would be surprised to see international issuance in these currencies, so when these countries need international capital, they issue in dollars or some other hard currency.
A third shared characteristic is that emerging countries tend to benchmark their currencies to some hard currency, usually the dollar. Bulgaria has set the lev at 1500 lev to the deutsche mark. The Hong Kong dollar is set at 7.74 per U.S. dollar. The Argentine peso is set at 1 to the U.S. dollar. Other countries use "crawling pegs" or descending bands by which they depreciate their currencies at some rate reflecting an inflation differential, as is the case in Colombia, Russia and Mexico. Contrast that with the policy of the U.S. Federal Reserve, which has explicitly stated it does not consider exchange rates when setting policy. Benchmarking is necessary for these countries because it creates the stability needed to access international financial markets. The price of hard currencies that must be purchased with local currency in order to meet international financial obligations needs to be predictable.
A fourth shared characteristic is that investment in emerging market fixed income is increasingly done through specialized teams and trading desks in New York, London, Boston and elsewhere, serving specialized teams of portfolio managers and research analysts. The investment community increasingly is treating it as a separate asset class.
Is this to assume that it is an asset class just because Wall Street says it is? Not entirely, but just as quantum theory tells us that observation of reality affects reality, so does the organization of investment firms affect the assets in which they invest. First, when a common group of specialists is dealing with a set of assets, these assets will tend to become more correlated by this very fact, as relative price changes will be subject to arbitrage. Traders are constantly comparing relative spreads and relative value and arbitraging the differences. Second, treating emerging market fixed income as an asset class also tends to give it more stability overall. Taking the case where emerging market fixed income is treated as part of the global bond universe, a bond manager might have 20% exposure or 0% exposure (all or nothing), whereas a specialist manager might have 100% exposure or 85% exposure. This is because a specialized manager is a permanent investor in the asset class; he is always there to buy or sell at some price. The third implication of this particular shared characteristic is that more research is provided on these assets.
The correlation statistics shown in Table 1 show emerging market fixed income exhibits little commonality with asset classes with which it might be associated if we were to assume it is not a separate asset class. The indexes given above are supplied by J.P. Morgan and include the Emerging Markets Bond Index and various other equity and bond indexes. In this matrix, you can see the EMBI is no more correlated with U.S. Treasuries or U.S. high yield than it is with U.S. stocks.
Emerging market fixed income is now too large to be ignored. When we first began investing in these assets in 1989, we found only $9 billion of investible instruments, excluding bank loans. At year-end 1996, the asset class added up to $940 billion, the bulk of which is local currency instruments.
Given the small size of the asset class in 1989, we felt we would have to limit ourselves to a small fund, perhaps $200 million to $300 million. We did, nonetheless, set up a specialized department to do this kind of investing as we felt a specialist effort was necessary to provide the expertise needed to attenuate risk. We also believed the fundamental changes that were going on in the economics and politics of the emerging countries would lead to large-scale involvement in the international capital markets and that consequently it was likely the investment universe would exceed the U.S. corporate bond market by 2000. We are well on the way to seeing this vision fulfilled. And the pace is accelerating.
At the beginning of 1996, there were 25 market access emerging countries; there are now more than 50. And don't forget, there are more than 200 currencies out there, so we have a lot more to come. There are, then, two lessons to be learned as we look at EMFI in 1989 and today: first, it has graduated to asset class scale; and, second, specialization is necessary.
This is also a diverse asset class. The J.P. Morgan Emerging Market Index Plus holds, among its largest issuers, 27% Brazilian issues, 22% Argentine, 17% Mexican, 9% Russian and 8% Venezuelan.
This presents a significant change from the early years of EMFI, for those of you who can harken back as far as 1989. The first fund we had at my previous employer initially was concentrated in Mexico and Venezuela, which accounted for more than 60% of its assets.
Morgan's local market index shows considerable diversity as well, with a significant Asian representation, with issuers like Malaysia, 18%; Thailand, 15%; Indonesia, 11%; and the Philippines, 5%.
The asset class is dynamic. Just as below-investment-grade U.S. companies eventually graduate to investment grade in most cases, so countries can leave the emerging markets category. Earlier we mentioned that a shared characteristics of EMFI is that issuing countries cannot issue in their own currencies in the international capital markets. Some countries have done this to a limited degree, however, and some eventually will graduate. Among countries that have issued internationally in their own currencies but are still considered "emerging" are South Africa, Argentina, Croatia, the Czech Republic and India.
In a low-inflation world, the prospect for investors in the local currencies of emerging countries is an exciting one. As the free market, open economy model approaches universality, inflation becomes rarer and rarer. With an open economy, it is difficult to duce inflation without producing extreme instability, so governments are balancing their budgets by increasing tax revenue and reducing their spending. This is bringing down inflation.
Now to the extent that inflation rates are similar with that of the United States, the local currencies will tend to appreciate against that to reflect the differences in productivity gains. Emerging countries, starting from low bases, should have relatively rapid productivity growth, and this is what we are seeing. Inflation is dropping throughout the developing world and in many cases has fallen to U.S. levels and even below. Over the next few decades, therefore, the currencies of those emerging countries that get their economic models right will be the place to be. Argentina is an interesting case in point. Productivity is growing and inflation is even lower than in the United States. Over time, defending the fixed 1: 1 exchange rate will mean keeping the peso down rather than holding it up. In fact, some models already show the peso to be undervalued vs. the dollar.
The strong performance of local currencies in the coming years is one reason we expect emerging market fixed income increasingly will be regarded as a distinct asset class. This is a fifth shared characteristic.
Emerging markets inflate and burst regularly. In the course of the 19th century, the United States grew from a thinly populated agricultural country to the world's greatest industrial power, making it the most successful emerging country in recent history. This history was one of speculative inflations and depressions, great successes and waves of bankruptcies. Rapid growth creates instability by creating speculative booms that have to be busted up. EMFI is no exception, and can add enormous value while being subject to the violent swings that characterize emerging markets.
During the boom periods, the easy availability of international capital produces the semblance of good credit quality on the part of issuers. We are in such a period now. The head of global fixed income at a large investment bank asserted publicly recently that "an issuer of any credit quality can issue for 10 years and any investment-grade issuer can issue for 100 years." During such periods, credit quality appears better than it is because liquidity is high. At such times, expertise and credit research seem unnecessary as the investor who mindlessly takes the most risk wins. It's an Alfred E. Neuman type of market. (He said: "Most people don't act stupid; it's the real thing.") Liquidity eventually abates and credit strains appear. The alternation from boom to bust is not a gradual process, however, but a violent swing. One day credit is ample while the next it is non-existent. This happened during the Mexican peso crisis, in Thailand more recently, and in the U.S. high-yield market with the failure of the United Airlines deal in 1989. When this time eventually comes, investors will see clearly that specialized expertise is not only desirable, but also necessary.
There are no complete and objective rules for establishing an asset class. Distinctions and classifications are useful only if they work for the investor.
Clearly, size is important, because only size will lead to a critical mass of specialists, and shared characteristics are essential, or specialization would be meaningless. The case for emerging market fixed income as a separate asset class, however, rests first and foremost in the fact that it can add enormous value to investor portfolios if it is treated as such.
Lincoln Y. Rathnam is president of Schooner Asset Management Co. L.L.C., Boston.