Hedge funds, exchange rates and causality: evidence from Thailand and Malaysia

Abstract

This article contributes to the debate on hedge funds and exchange rates in Thailand and Malaysia. It provides the first empirical evidence on causal relation between hedge funds and exchange rates. Using a new Granger noncausality procedure proposed by Toda and Yamamoto (1995) and monthly data for the January 1994 to April 2002 period, two important findings emerge. First, hedge funds lead Thai baht during the 1997 crisis. Second, there is a bidirectional causality between hedge funds and Malaysian ringgit for the pre-crisis period. In all other cases, no causal relation can be established.

Notes

¹ According to an interview with Stanley Druckermiller (i.e. the former head of the Quantum Fund) there were short positions in the Thai baht and Malaysian ringgit. The Quantum's year-to-date returns climbed from 3.2% at the end of the first quarter to 22% as of 3 September 1997 (Wall Street Journal, 5 September 1997, p. C1).

² As of June 1997, the ratios of short-term liabilities to international reserves were 210.6, 162.9 and 141.1% for South Korea, Indonesia and Thailand, respectively.

³ They also found that the funds exposure were significant during the ERM crisis in 1992, the global bond rally in 1993/1994. However, the funds exposures were insignificant during the stock market crash of 1987 and the Mexican peso crisis in 1994.

⁴ In the case of Malaysia, the post- August 1998 period is avoided due to the imposition of capital control which almost completely closed the financial market to foreigners.

⁵ For a complete description of these biases, refer to Fung and Shieh (2002).

⁶ The test has a comparable performance in size and power to the likelihood ratio (LR) and Wald tests (Zapata and Rambaldi, 1997).

⁷ Assuming 3-order VAR, where HF=hedge funds; EX=exchange rates.

⁸ Specifically, we estimate the following unrestricted error-correction model, , where is a random error and Δ is a first difference operators.