The global financial crisis has undergone various phases. The build-up of the financial crisis from July 2007 was followed by a systemic, and largely generalised, outbreak at the time of the Lehman collapse in September 2008. This was followed by the systemic response of governments around the world which led to risks in the financial sector morphing into sovereign risks. Figure 1 taken from IMF (2010) exemplifies these developments based on ten-year sovereign swap spreads (see also Caceres et al. 2010).
Figure 1. The four stages of the crisis
Ten-year sovereign swap spreads, in %
Source: Bloomberg LP.
At this current juncture of the crisis phase, the paths of emerging-market countries and advanced markets (especially in the Eurozone) have diverged. On the one hand, stronger economic fundamentals in some key emerging markets, along with low interest rates in advanced countries, have led to a rebound in capital flows to emerging markets, reversing the significant drop that occurred at the height of the crisis (IMF 2011). On the other hand, the interaction between the financial sector and sovereign credit risks, especially in periphery countries in the Eurozone, remains a critical factor as the confluence of funding pressures and banking sector vulnerabilities make financial systems highly vulnerable to a deterioration in market conditions.
In this column, we focus on global market conditions and systemic risks in advanced countries by examining the equity, interbank spreads, sovereign credit default swaps, and foreign-exchange volatility across a number of these countries. As in a previous Vox column piece (Hesse and González-Hermosillo 2009), we use Markov regime-switching techniques to examine financial stress in a formal way. Given the intrinsic volatility of high-frequency financial data, especially during periods of stress, we choose the ARCH Markov-Switching model (SWARCH) by Hamilton and Susmel (1994) because it is able to differentiate between different volatility states. The Markov-switching models are estimated with daily data in first differences to account for the non-stationarity in the data. The findings are presented in the heat maps shown below, where the volatility state with the highest probability of occurrence is plotted. The data covers the period January 2008 to mid-January 2011.
The findings suggest that interbank markets (based on Libor-OIS spreads) in the US, the UK, Canada, and Australia have moved to the low volatility level that prevailed before the crisis erupted. In contrast, after a short period of low stress for the Euro Libor-OIS spread during 2009-2010, the Markov-Switching models indicate that this interbank spread has recently moved back to higher volatility states. This is consistent with increased sovereign risks in some European countries in the aftermath of the financial crises in Greece and Ireland. The findings for the foreign-exchange volatility model indicate two periods of systemic risk in this asset class, namely during the Lehman Brothers collapse and at the peak of the European debt crisis prior to the Greece bailout in late-April 2010.
Equity markets in advanced economies have generally moved back to a low-volatility state. However, many countries in Europe (notably Greece, Italy, Portugal and Spain and until recently Ireland) have remained in the medium-volatility state. Belgium has also recently moved back to the medium equity volatility state. For the countries examined here, a generalised high-volatility state occurred during the collapse of Lehman Brothers in the fall of 2008. During the period leading up to the bailout of Greece in late-April 2010, only some countries (notably in the European periphery) exhibited a high-volatility state. More recently, equity markets in most of these countries have moved to medium or low-volatility states.
The Markov-switching models for sovereign CDS markets show a similar story with countries in the European periphery such as Greece, Ireland, Italy, Portugal and Spain but also Belgium all being in a high CDS volatility state. But even in some advanced countries with lower credit premia such as Germany or the Netherlands the volatility has remained in the high volatility state, implying that markets see the risk of occurrence of spillovers across European countries. Country-specific credit default swaps Markov-switching models for Greece, Ireland, Italy, Portugal, Spain, and Belgium indicate that all these countries, with the exception of Belgium, have been in the high-volatility state prior to the peak of the European sovereign crisis at the time of the Greece bailout. Since the onset of the global financial crisis, equity volatility for these countries has predominantly only oscillated between the medium and high volatility state. Finally, with the exception of Japan, volatilities of countries in Asia Pacific (Australia, Hong Kong, and Korea) have all converged back to the pre-crisis low volatility state.
This column uses Markov-switching models for equity, interbank spreads, sovereign credit default swaps, and foreign-exchange rate volatility for a number of advanced countries in order to examine their volatility states in these asset classes since the beginning of the global financial crisis. The results suggest a recent diverging behaviour not only among asset classes but also across countries. While overall systemic stress emanating from interbank spreads and foreign-exchange volatility has subsided, there are still pockets of systemic risk, especially in the European periphery, particularly in sovereign credit default swaps and equity markets.
With the interaction between sovereign and banking sector risks intensifying, especially in the Eurozone, and hence increasing volatility and the potential for spillovers across asset markets, policy should be geared towards breaking this adverse sovereign-financial loop. Market concerns about insufficient progress to repair fiscal balance sheets and the financial system, especially for countries in the European periphery, are reflected in the heat maps shown above.
While the analysis does not capture the recent turbulences in the Middle East, the Markov-Switching equity models are likely to have been affected judging from the observed market volatility around the globe.
The views expressed in this article are those of the authors and should not be attributed to the IMF, its Executive Board, or its management. Any errors and omissions are the sole responsibility of the authors. The authors also thank Ryan Scuzzarella for research assistance.
Caceres, Carlos, Vincenzo Guzzo, and Miguel Segoviano (2010), “Sovereign Spreads: Global Risk Aversion, Contagion or Fundamentals”, IMF Working Paper 10/ 120.
Hamilton, James D, and Raul Susmel (1994), “Autoregressive Conditional Heteroskedasticity and Changes in Regime”, Journal of Econometrics, 64 (September-October):307-333.
González-Hermosillo, Brenda and Heiko Hesse (2009), “Global Market Conditions and Systemic Risk”, IMF Working Paper 09/230.
Hesse, Heiko and Brenda González-Hermosillo (2009), “Financial Crisis, Global Conditions, and Regime Changes”, VoxEU.org, 21 April.
International Monetary Fund (2010), Global Financial Stability Report, April.
International Monetary Fund (2011), Global Financial Stability Report Update, January.