|Researchers:||Mario Bellia, Loriana Pelizzon, Marti Subrahmanyam, Jun Uno, Darya Yuferova|
|Category:||Financial Markets, Systemic Risk Lab|
Topic & Objectives
In the past decade, High Frequency Traders (HFTs) have become dominant players in stock markets and the object of robust debate as to whether they use their speed advantage to help or harm the fairness and efficiency of financial markets. These issues have been investigated in the literature, mostly for the continuous trading phase, but have been ignored for the pre-opening phase and the opening auction, where it is believed that HFTs have a small chance of exploiting their speed advantage.
In this paper, we aim to investigate whether this common perception is confirmed in the data. We fill this gap in the literature in the context of the pre-opening phase of the NYSE Euronext Paris Exchange, using data from the Base Européenne de Données Financières à Haute Fréquence (BEDOFIH).
- First, we find that HFTs “come early to the party” despite the absence of immediate execution. The majority of HFTs start their quoting well before the 9.00 a.m. opening auction. This behavior indicates their desire to observe and learn from the pre-opening order flow before making their order submission decisions. As soon as a large amount of information arrives in the market from large brokerage firms, equity derivative markets and from news providers, HFTs start posting orders to exploit the time priority option. Interestingly, HFTs post many orders in the pre-opening phase that are highly unlikely to be executed, we call these “flash-crash orders”, and they do so for “fishing”, i.e., gaining time priority on orders that would be triggered only under extreme market movements. Therefore, they come early to the party to (i) acquire information during the pre-opening phase, (ii) benefit from the priority option for “flash crash” orders, and (iii) benefit from the priority option in the opening auction.
- Second, we find that HFTs generally make profits on executed orders during the opening auction that were submitted in the very last minute of the pre-opening phase, and make losses on other orders.
- Third, we find that the presence of HFTs does not deteriorate market quality in the pre-opening phase, in terms of liquidity provision, while contributing substantially to the price discovery process.
In summary, HFTs do behave strategically by posting and cancelling orders to extract information and make profits or by posting “flash crash orders” to benefit of the priority option, however, HFTs do not have special privileges by virtue of their speed advantage, relative to other market participants.
Our results are likely to be of interest to stock exchanges, regulators and investment management professionals. Broadly speaking, levelling the playing field across market participants is a common objective for all these constituencies, being of utmost relevance for fair investment management practices. Our analysis highlights that in terms of:
- Market quality: As a group, HFTs neither improve nor harm liquidity provision in the opening auction and consistently lead the price discovery process throughout the pre-opening phase. Our findings are important for designing proper opening mechanisms in the presence of HFT participation.
- Market fairness: We document similar profit patterns for HFTs and NON-HFTs alike, suggesting that speed is not a necessary condition to make profits, at least in the context of the fixed time of the opening auction. In fact, both HFTs and NON-HFTs do have an information advantage that allows them to significantly contribute to price discovery and to make profits. The key difference is that HFTs extract information from the order book by posting and cancelling orders, while NON-HFTs possess mostly fundamental information.
This project was supported by:
|182||Mario Bellia, Loriana Pelizzon, Marti Subrahmanyam, Jun Uno, Darya Yuferova||Coming Early to the Party||2017||Financial Markets, Systemic Risk Lab||High-Frequency Traders (HFTs), Proprietary Trading, Opening Auction, Liquidity Provision, Price Discovery|