Slippage in futures markets: Evidence from the Sydney Futures Exchange

Abstract

This article examines the market-impact cost of trades executed in futures markets, which is commonly referred to as slippage. With the use of a unique data set provided by the Sydney Futures Exchange, this article documents that slippage costs incurred in executing packages of trades in stock-index and interest-rate futures markets are significantly smaller than market-impact costs documented previously for equity markets. Furthermore, in contrast to research based on equity markets, there is little evidence that trade packages executed in futures markets convey information, or that purchases and sales behave differently. In fact, the (2002) define slippage as the transactional risk that arises from the inability of a trader to accurately foretell the price at which an order to purchase or sell will be executed, especially for a large or complex trades. 2 Slippage is comparable to the market-impact costs in equity markets. In this article, the term slippage is used to define the variation between traded and unperturbed prices in futures markets, whereas the term market-impact cost is used to define such variation in equity markets. 3See Kraus and Stoll (1972);Holtahusen et al. (1990);Gemmill (1996);and Keim and Madhavan (1996). evidence presented in this article implies that slippage in futures markets is entirely a liquidity cost, and symmetrical for purchases and sales. This is consistent with previous work, which conjectures that there is an absence of private information in stock-index and interest-rate futures markets. Finally, analogously to previous research, there is some evidence that trade size and the identity of traders are determinants of slippage; however, these variables explain less than 10% of the total variation in slippage.

Large trades and intraday futures price behavior

“…Based on the F statistics and the t statistics, the results are all statistically significant at the 1% level. These results are consistent with the findings from the equity markets (Chan & Lakonishok, 1993) and from the futures markets (Frino & Oetomo, 2005). The regression results on the information effects are very similar to the regression results on the total price effects.…”

“…First, to document the empirical patterns of total price impacts in five futures contracts traded on the CME using four-year intraday data. Thus, we can observe whether the empirical results documented by Frino and Oetomo (2005) across all trades executed in Australian futures markets are consistent with the empirical results we obtain in the US futures markets for outside customer trades. Second, to perform tests to determine whether the economic condition hypothesis proposed by Chiyachantana et al (2004) influences price, liquidity, and information effects of block buys and sells in US futures markets.…”

“…We use opening and closing prices of a single trading day to measure the effects of a single trade, while Frino and Oetomo (2005) use opening and closing prices of more than one day to measure the effects of a trading package.…”

“…Particularly, trades are grouped into packages if they are executed (a) by the same trader, (b) in the same direction, and (c) consecutively without a one-day trading break. See Frino andOetomo (2005, p. 1133). 2 They use daily volume-weighted average prices (see Berkowitz, Logue, & Noser, 1988;Jones & Lipson, 2001) as the benchmark to calculate the total price effect of large trades.…”

“…However, Chiyachantana et al (2004) do not examine the asymmetry between the information effects of buys and sells in bullish and bearish markets by decomposing the total price impact (or market impact cost) into permanent and temporary (liquidity) price effects. Frino and Oetomo (2005) is the first study to report empirical evidence of market price impact (slippage) of trading packages in futures markets. They characterize the total, information, and liquidity effects incurred in executing package of trades in four main futures contracts (SPI 200, BAB, 3-year bonds, and 10-year bonds) on the Sydney Futures Exchange, using intraday data from July 1, 2000, to June 30, 2003.…”