Emerging market bond spreads: The role of global and domestic factors from 2002 to 2011

• We examine global versus domestic factors in determining EMBI spreads.
• Principal components are used to derive a measure of global risk aversion.
• Our model (estimated using PMG techniques) explains EMBI spreads quite well.
• Most of the 2008 rise in EMBI spreads was due to the risk measure.
• A model of the risk measure is also estimated.

We address the importance of external versus domestic conditions in determining emerging market bond (EMBI) spreads. Using principal components, we derive a measure of global risk aversion, which is shown to have a significant and, when interacted with a country's foreign debt to GNI ratio, nonlinear effect on these bond spreads. Our model, estimated using Pooled Mean Group techniques, which also incorporates country-specific variables (foreign debt, fiscal policy, debt servicing and political risk), is able to track developments in emerging market bond spreads over the period May 2002 to October 2011 quite well. From mid 2002 to mid 2007, the model suggests that just over two thirds of the decline in these spreads on average reflected improved fundamentals, with the rest due to easy credit conditions. During the 2008 crisis, virtually all of the run-up in emerging market spreads was due to the large increase in our measure of risk aversion. A model of the measure of risk aversion is also estimated, which identifies as its key drivers, the outlook for growth in the major OECD and large non-OECD economies as well as US credit supply conditions.