Part I of my two-part blog on local currency emerging debt showed that adding this asset class to a portfolio of traditional government bonds from developed countries significantly improves the risk-return ratio. This is a good thing, of course, from a long-term perspective, but does not change the fact that German bunds are very different to government bonds from, say, Thailand. But this is probably needless to say after the recent turmoil in emerging debt markets. For this reason, I will focus in this post on some tactical aspects of investing in EM local currency debt.
A quick look at the leading benchmark, the JP Morgan GBI-EM Global Diversified Index, will tell you that it can be very helpful to have some tactical moves at your disposal when investing in local currency emerging debt . For example, countries like Malaysia, Poland and Turkey each have the maximum capped weight of 10% in this index. This is quite different from the joint weight of just 6% of these countries in the MSCI Emerging Markets (Equity) Index. So, while for equity investors, these countries are somewhat irrelevant, bond investors are forced to lose themselves in the murky depths of the government finances in these far-flung parts of the EM universe.
What state are you in?
But is this really necessary? According to recent research performed by the IMF, it depends… A study carried out by Jaramillo & Weber (2012) shows that the degree to which government finances are significant for EM local currency debt returns depends on the degree of global risk appetite. This sounds more complex than it is. Jaramillo & Weber just take the well-known and well-accessible VIX Index to measure the degree of global risk appetite. Next, they use the VIX to divide their observation period – 2005-2011 – into a tranquil period in which investors are willing to take on more risk (low VIX) and a turbulent period in which investors are risk averse (high VIX). They then apply a regression analysis to establish which factors are relevant to returns in each of the two periods.
Their analysis shows that in tranquil times, investors are interested almost exclusively in economic growth and inflation expectations in the emerging country that issues the bonds. When the VIX is low, general macro-economic factors, that are largely determined by the global economy, explain most of the emerging market debt returns. However, when the financial markets become more turbulent, driving up the VIX, investors shift their focus towards factors such as government deficits and the level of debt to GDP. So, only when panic strikes, do government finances become relevant. And, when this happens, the countries with the biggest deficits and highest debts are the ones hit hardest.
Click to enlarge
The work of Jaramillo & Weber shows that the state of government finances in all their complexity is not always the most decisive explanatory variable. However, neither can it be completely ignored since these government finances take center stage in times of high volatility. This is the main reason that, at Robeco, we also look at EM local currency debt from a different angle as well.
For some time now, we have been successfully using a quantitative model to forecast yield movements for German, US and Japanese government bonds. And recently, Duyvesteyn & Martens (2013), also tested this model for six emerging countries: Brazil, Mexico, South Africa, Poland, Malaysia and South Korea. Since we are looking at local currency emerging debt, it is important to note that this model focuses only on the yield component and not on the currency component of the return on emerging debt.
The model considers three factors:
- bond-market momentum
- equity-market momentum and
- the slope of the yield curve.
The results obtained with this model are pretty impressive for emerging bond markets as well. With only a few exceptions, a long-short strategy generates positive returns on each of the three factors. If you invest in a basket of bonds from these six countries, the long-short strategy will perform even better, measured by the risk-return ratio (information ratio) of the strategy. And that’s not all. When this basket of six countries also invests, equally weighted, in each of the three factors , the information ratio improves by a further 30%. Hence, mixing countries and factors significantly improves the robustness of the model as well as the risk-return profile of your investment. (For all the exact numbers please refer to the paper of Duyvesteyn & Martens (2013) below.)
Finally, one more important aspect is that the combined strategy can easily be used as an overlay in addition to a regular (beta 1) investment in EM local currency debt. This enables you to realize the benchmark return plus an additional yield-related alpha. So this quantitative approach offers you tools for tactical moves, not only in traditional bond markets, but also in emerging debt markets. Certainly, no superfluous help for investors as local currency emerging debt proves to be a pretty exotic asset class.
Duyvesteyn, Johan G. and Martens, Martin P.E., Emerging Government Bond Market Timing (February 14, 2013). Available at SSRN