Do High-Frequency Traders Anticipate Buying and Selling Pressure?

Abstract

This study provides evidence that high-frequency traders (HFTs) identify patterns in past trades and orders that allow them to anticipate and trade ahead of other investors’ order flow. Specifically, HFTs’ aggressive purchases and sales lead those of other investors, and this effect is stronger at times when it is more difficult for non-HFTs to disguise their order flow. Consistent with some HFTs being more skilled or more focused on anticipatory strategies, I show that trades from a subset of HFTs consistently predict non-HFT order flow the best. The results are not explained by HFTs reacting faster to news or past returns, by contrarian or trend-chasing behavior by non-HFTs, or by trader misclassification. These findings support the existence of an anticipatory trading channel through which HFTs increase non-HFT trading costs. This paper was accepted by Karl Diether, finance.

Order aggressiveness and flash crashes

“…We therefore concentrate on the case where an aggressive order is not adversely selected. This assumption is in line with the recent literature, arguing that HFTs can predict adverse selection and therefore are able to avoid falling prey to it (see Hoffmann 2014; Jovanovic and Menkveld 2016; Hirschey 2018). This assumption is critical for our framework to mimic the price evolution during a flash crash, that is, price falling significantly from the level at t 0 to t 1 .…”

“…This is indeed a plausible assumption and is underscored by the relevant literature. As an example, Hirschey (2018) shows that HFTs can obtain higher profit by anticipating the directions of future price changes from limit order books. Specifically, the study finds that HFTs can predict buy and sell pressure from non‐HFTs order flow and can gain higher returns at the expense of these traders.…”

“…The increase in profitability is economically meaningful. We find that an informed trader, that is, a trader with predictive skills such as an HFT (see Hirschey, 2018), could earn up to a cumulative return in excess of 1,482 basis points (bps) based on our analysis of a sample of flash crash‐affected stocks. This return is significantly higher than is possible during the non‐flash crash periods.…”

“…Hirschey (2018) argues that HFTs can anticipate buying and selling pressure (see also Brogaard, Hendershott, & Riordan, 2014), which could help them avoid being adversely selected when deploying aggressive trading strategies (see also Hoffmann 2014; Jovanovic and Menkveld 2016). Indeed, Hirschey (2018) finds that HFTs' aggressive sales and purchases consistently lead those of other investors. This implies that the framework we illustrate above is more likely to be successfully deployed when it is implemented at a high frequency.…”

“…Secondly, aggressiveness during both the pre‐flash and flash crash periods should impact the onset of the flash crash. Thirdly, HFTs' aggressive orders should contribute to the flash crash more than non‐HFTs' aggressive orders, because our proposed three‐stage trading strategy is more likely to be successfully deployed by HFTs due to their ability to avoid being adversely selected (see also Brogaard et al 2014; Hirschey 2018). Finally, the fraction and number of aggressive buy and sell orders during the May 6, 2010, flash crash should be higher than the fraction of aggressive buy and sell orders during the surrounding periods.…”