1995年-BIS国际清算银行_The_anatomy_of_the_bond_market_turbulence_of_1994_30页_3mb

Summary of "The Anatomy of the Bond Market Turbulence of 1994"

Core Content

This working paper by Claudio E.V. Borio and Robert N. McCauley analyzes the sharp increase in bond yield volatility across major bond markets in 1994, focusing on thirteen industrialized countries. The study uses over-the-counter (OTC) data for implied bond yield volatility to assess the dynamics behind the market turbulence.

Main Points

1. Market Dynamics as Key Drivers of Volatility

The paper identifies three main market dynamics that explain the rise in bond yield volatility:

• Persistence: Volatility tends to persist over time, reverting to its mean gradually. This suggests that once volatility increases, it tends to remain elevated. The persistence parameter for most countries ranges from 0.84 to 0.97, with the US at 0.90 and Germany at 0.96.

• Impact of Market Movements: A decline in bond prices is associated with higher volatility. However, this relationship is not symmetric in all markets. For eight of the thirteen countries, volatility increases only with falling bond prices. The US and Canada are exceptions where implied volatility does not react significantly to market movements. The magnitude of this effect is substantial, with a 16% increase in long rates potentially leading to a 5 to 8 percentage point rise in volatility.

• Foreign Disinvestment: A significant portion of the volatility increase in 1994 was due to the withdrawal of foreign investments. In Germany, France, and Italy, foreign investors sold large amounts of government bonds, which coincided with sharp increases in implied bond yield volatility. For example, in Germany, foreign liquidation of over DM 13 billion in March 1994 was associated with a 4 percentage point rise in volatility.

2. Market Spillovers

Volatility in one market can spill over into others, especially during periods of heightened market activity. In 1994, spillovers were more pronounced than in previous years. Before February 1994, volatility spillovers were limited, but after the US monetary policy tightening, volatility spread more widely, with New York and London playing central roles in transmitting volatility to other markets.

3. Limited Role of Domestic Economic Factors

While domestic economic factors such as inflation performance and expectations do influence bond volatility, they played a limited role in explaining the 1994 episode. Changes in inflation and growth expectations did not correspond to significant changes in volatility. For instance, in the US, inflation expectations fell slightly, yet volatility increased. Similarly, in Japan, inflation expectations were relatively stable, but volatility still rose sharply.

Key Information

• The bond market turbulence of 1994 was marked by a significant and persistent increase in yield volatility, with the US showing a 5 percentage point rise and others seeing increases of 10 or more.
• The study uses OTC data for implied bond yield volatility, which is considered more reliable than exchange-traded data for international comparisons.
• The paper emphasizes that the rise in volatility was more attributable to market dynamics than to changes in economic fundamentals.
• The impact of foreign disinvestment was particularly notable in continental Europe, with large-scale sales contributing to volatility increases.
• The analysis also highlights the role of leverage in bond markets, especially among foreign investors, which can lead to forced selling when prices fall.
• The paper finds that the relationship between volatility and economic fundamentals is not consistent across all countries and is not a general explanation for the 1994 episode.

Conclusion

The 1994 bond market turbulence was primarily driven by market dynamics, including persistence, spillovers, and foreign disinvestment, rather than by changes in domestic economic fundamentals. The study provides a comprehensive framework for understanding how volatility spreads across markets and the role of leverage in amplifying it. While domestic factors may influence volatility, they were not the primary cause of the 1994 increase. The findings suggest that risk management systems should focus on understanding these market dynamics rather than solely on economic indicators.