摘要
In October 1996, The Dhaka Stock Exchange （DSE） adopted trading halts for individual stocks, collectively known as ＂circuit breakers＂, to reduce the stock market volatility. This paper reviews the existing circuit breakers literature and developed five hypothesis--＂Magnet Effect＂, ＂Cool off-Heating （C-H） Effect＂, ＂Information Hypothesis＂, ＂Volatility Spillover Hypothesis＂ and ＂Trading Interferences Hypothesis＂--which could be tested empirically not only in the Dhaka Stock Exchange but any stock exchanges around the world. This paper also suggests most appropriate econometric models for empirical testing. GARCH for inter day data and Event Study methodology for intra day data. Moreover, to test the robustness non-parametric tests need to use along with parametric one. Considering the stock market bubbles in 1996, it has been found that it was optimal for the regulators to adopt this trading halt, but not for the market. It failed to protect the market. However, this might be the consequences of misconceptions about the purpose and effectiveness of circuit breakers. Despite many arguments contrary to this mechanism and absence of any conclusive empirical evidence for a fragile stock exchange like DSE, it may be useful sometimes to replace the ＂invisible hand of the marketplace＂ with the ＂visible hand of the market regulators＂.
