Comment
Kevin Cowan: The main empirical fact motivating this interesting empirical paper by Alicia García-Herrero and Alvaro Ortíz is summarized in figure 2. The figure highlights an extremely high contemporaneous correlation between the emerging market bond spread (the EMBI spread) and the spread between the returns on Baa-rated corporate debt in the United States and U.S. Treasury bills.
This correlation has three possible explanations. First, shocks originating in developed economies could be driving the spreads of emerging market bonds. The canonical example of a developed economy shock is the Enron crisis in the United States. In this case, Wall Street was a source of instability for emerging markets. The second explanation is that shocks originating in one emerging market are transmitted to other emerging economies and the U.S. corporate market. Several recent papers have discussed this mechanism. One set of papers argues that contagion may occur because investors' risk appetite falls as a result of the negative wealth effects of financial crises, leading investors to reduce the overall share of all risky assets in their portfolios, including other emerging market bonds and U.S. corporate bonds.1 An alternative explanation is that changes in the perception of the risk of emerging markets' bonds leads to a portfolio shift away from all risky assets.2 In both cases, Wall Street is a carrier rather than the source of instability. The third explanation is that common macroeconomic events drive both spreads, so that there is no causality in the relation, only a correlation of variables. In particular, a large literature links corporate spreads to the U.S. business cycle, which in turn could be affecting the risk on emerging market debt.3 [End Page 149]
Several earlier papers also identify the correlation shown in figure 2.4 García-Herrero and Ortíz's main contribution is to propose a framework that controls for the common macroeconomic variables discussed in the third point above. In addition, the paper takes a stand on the first versus the second point, arguing that changes in the spreads of emerging market bonds are driven by events in the financial markets of developing countries themselves. This last distinction is probably minor from a policy perspective: in either case, external shocks to the supply of funds to a particular emerging economy are responsible for the variance in the spreads. The distinction is more important from an empirical perspective, however.
Emerging markets face highly volatile borrowing costs. Over the period 1994–2005, for example, the variance of the EMBI yield was three times higher than that of the U.S. Treasury bill yield. García-Herrero and Ortíz argue that part of this volatility is due to events in international financial markets, rather than events in the emerging market economies themselves. As such, this paper forms part of a growing literature that emphasizes the role of international financial markets in originating or amplifying shocks to emerging market financing. This leads to two key policy implications. First, emerging economies need to foster policies that reduce their vulnerability to these shocks, in addition to ensuring that domestic policies are not sources of volatility. Second, policymakers in emerging economies need to monitor fluctuations in U.S. and international financial markets closely.
The baseline results of the paper are based on the spread between Baa corporate bonds and U.S. Treasury bills. This spread includes three types of risks, which have different economic implications. The first component of the Baa-Treasury spread is the prepayment premium to investors for the risk that if interest rates fall in the future, corporate borrowers will retire old debt and replace it with new debt at a lower rate. The second component of the Baa-Treasury spread is a liquidity premium, which compensates investors for the greater liquidity of Treasury bonds, especially in times of financial turbulence. Finally, the Baa-Treasury spread also includes a default premium on corporate bonds. These three risks can be split empirically by breaking down the Baa-Treasury...
