Skeptical readers might object that this must be a sign of stressed markets, a value trap ready to snap shut. Aren’t the “fragile five” economies facing an impending rout of their currencies? This question brings us to the second and third factors supporting emerging market bonds. The volatility of local currencies is undeniably substantial; the currency risk swamps the volatility of the underlying bonds. Nonetheless, the currency exposures may provide an incremental boost to long-run expected return. Because of the differences in productivity growth, in the long run we expect emerging market currencies to appreciate relative to the U.S. dollar and other major developed world currencies.
The expected real return for investing in foreign bonds is the real yield on the bonds, less defaults, plus any movements in the real exchange rate.2 So what drives movements in the real exchange rate? One of the primary long-term factors is differing levels of productivity increase between two economies. Productivity matters because, even in open economies, not all goods are tradable. This drives higher inflation and thus appreciation in the real exchange rate in the economy with faster-growing productivity.3
Think of sports cars and hospital visits. Because cars are freely shipped across borders, their prices should be equalized. For example, suppose the nominal exchange rate of interest is 13 Mexican pesos to 1 U.S. dollar. Disregarding transportation costs and tariffs, a sports car selling for $100,000 in the United States should sell for about 1.3 million pesos in Mexico. There are very few “bargains” to be found by shopping abroad for tradable goods.
Hospital services, however, are a different story because they are non-tradable goods. As we age, we expect to spend less of our consumption budget on sports cars and much more on health care. Yet a Mexican hospital is not a viable substitute for my health care needs, if for no other reason than proximity. It’s easy to drive across the border and save a few thousand dollars on a car, but not on an emergency appendectomy! This opens up the possibility for non-tradable goods of all sorts to have dramatically different real prices. An emergency medical procedure may cost half a sports car in California but only a tenth of the same sports car in Mexico. I, for one, am not willing to arbitrage away that price discrepancy.
This relationship between productivity and the real exchange rate is known as the Balassa–Samuelson effect.4 Emerging markets’ higher rates of productivity growth drive higher wage growth and, therefore, higher price inflation in non-tradable goods. This propels an increase in the emerging market consumption basket relative to the developed market consumption basket, resulting in a rising real exchange rate. In fact, the long-run expected increase in the real exchange rate between two countries can be approximated by the difference in productivity growth rates (estimated by real per capita GDP growth rates). Depending on which countries you include in the analysis, the expected real exchange rate appreciation in a basket of emerging market currencies should be about 50 bps annually. For perspective, the expected currency appreciation alone is slightly higher than the expected real yield on a 10-year U.S. Treasury bond. Of course, these expected currency gains are small compared to the real yield on emerging market bonds. But it is comforting to know that the local currency exposure should add a long-term incremental return, rather than eat away at the attractive real yield through depreciation.
The next question for most investors is: What about the increased volatility associated with local currency exposure, particularly in the case of fragile emerging market currencies? This has certainly been true historically; for instance, the volatility of emerging market currency returns soared during the East Asian financial crisis of 1997 and the devaluation of the ruble in 1998. But would it surprise you to learn that the volatility of the Mexican peso to U.S. dollar exchange rate over the last five years has equaled that of the Norwegian krone to U.S. dollar (both at 12.3%)? Or that the most volatile of the fragile five, the South African rand, has delivered but a tad higher risk than the Australian dollar (15.2% to 14.9%)? In fact, as shown in Figure 3, the trailing five-year average volatility for emerging market and developed market currencies 5 has converged substantially over recent years. Taking on any foreign currency exposure certainly adds volatility. But, at least for the past five years, a basket of emerging market currencies has contributed no more volatility to an international portfolio than a basket of developed currencies. And as emerging markets continue to do just that—emerge—the convergence of currency risks may be expected to continue.