بهبودی بعد از آشفتگی بازار اوراق قرضه: شانس خوب و سیاست خوب هر دو عوامل موثری بوده اند

This paper focuses on the resolution of bond market crises. Episodes of bond market distress are identified using secondary market sovereign bond spreads. Duration models are used to assess the role of the global environment, domestic policy, IMF programs and political events in explaining the length of distress episodes. We find a rich set of interactions between favourable external conditions, sound macroeconomic policies and the presence of an IMF program which contribute to shorter bond market crises.

Do emerging market economies recover from financial crises thanks to sound macroeconomic policies or favourable external conditions? Much of the discussion on financial crisis recovery is organized around the following question: “Is it good luck or good policy?”. Yet, the speed of financial crisis recovery could depend on both good policy and good luck. And even more interestingly, there could be significant interaction effects between various explanatory factors. Suppose that a country experiences a bond market crisis, leading investors to require a very large risk premium. This country responds with a strong macroeconomic policy. The effect of this policy could be perceived differently by investors depending on the international financial context. It could be that even the best macroeconomic policy will not affect the speed of the recovery process when world interest rates are very high (an unfavourable external environment), but will be quite effective when world interest rates are low. This paper focuses on the duration of the crisis recovery process and looks at the effect of various interactions between domestic macroeconomic policy, the external environment and IMF programs in explaining how long these crises last. We first identify episodes (spells) of bond market distress. Pescatori and Sy (2007) show that distress can be identified on the basis of secondary market sovereign bond spreads. Sovereign spreads typically reflect investor perceptions of sovereign risk and are thus widely used by researchers to provide information on the external financing conditions faced by emerging markets. Next, we study the determinants of these crisis spells in a duration framework. Apart from domestic macroeconomic policy, the external environment and IMF programs, we also control for other standard determinants taken from the vast literature on secondary market spreads and debt crises. Our econometric analysis shows that the global economic environment, the quality of domestic policy and IMF programs affect the duration of bond market distress. Moreover, we find a rich set of interactions between these determinants. The effect of domestic policy on the duration of bond market distress becomes stronger as the global environment is more favourable, and weaker when an IMF program is in place. Good external conditions affect the risk perceptions of investors more positively when a country pursues sound macroeconomic policies. Finally, the contribution of IMF programs to crisis resolution increases as the state of domestic policy worsens. These significant interaction effects should not be ignored when evaluating the commitment of domestic policymakers to address episodes of bond market distress. The remainder of this paper is organized as follows. Section 2 provides a short review of the existing literature. Section 3 outlines the strategy used to identify spells of bond market distress and discusses the estimation of duration models. Section 4 focuses on the data and discusses our explanatory variables. Section 5 presents our results and Section 6 concludes.

This paper delves into the dynamics of bond market distress and crisis resolution. It provides evidence about the role of domestic policy, global financial conditions and IMF programs for the process of rebuilding confidence. Bond market distress episodes are identified on the basis of secondary market sovereign spreads. Duration models are then used to identify the determinants of the length of the recovery process. Several factors affect the duration of bond market distress. Favourable international conditions are associated with shorter crises. Lower international interest rates, a stronger global risk appetite, a stronger world business cycle and stronger US stock market performance are all associated with shorter spells of turbulence. Sound macroeconomic policies and IMF programs also raise the probability of exiting a crisis. We find significant interactions between global financial conditions, domestic policy and IMF programs. Our results also suggest other interesting directions for further research. It could be of interest to disaggregate some of the data that we have used, for example on IMF programs. Could it be that particular elements of IMF programs interact with particular facets of domestic policy in different ways during crisis episodes? Another issue that would merit greater exploration is the effect of political events on crisis resolution and management. Generally speaking, political events have received short shrift in resolving financial crisis situations. Perhaps, as anecdotal evidence will confirm (such as the Indonesian case during the Asian financial crisis), they hold the key to successful and quick crisis resolution.