Find us on Facebook Find us on LinkedIn Follow us on Twitter Subscribe to our YouTube channel Instagram

Three essays on the impact of electronic trading in futures markets

Luke Bortoli

This dissertation consists of three research essays that examine the impact of electronic screen trading on futures markets. The research provides new empirical evidence on an increasingly important issue, given the rapid increase in trading technology utilised by securities markets. Each essay addresses a research question with scarce or inconclusive prior research findings in order to aid researchers, regulators, exchange policy makers and systems builders as they confront issues related to electronic trading. Rare natural experiments provided by the transition to electronic trading on the Sydney Futures Exchange (SFE) are analysed.

A major factor driving the trend towards automation in futures markets is competitive pressure amongst exchanges to reduce transactions costs. Prior literature provides evidence of a reduction in bid-ask spreads following the introduction of electronic trading. The first essay extends this literature by examining the impact of electronic trading on brokerage commissions in futures markets. The introduction of electronic trading on the SFE is associated with lower brokerage commission fees relative to floor trading, even after controlling for liquidity, volatility and broker identity. New evidence on the magnitude of brokerage commissions is also documented and implies that rates charged on futures transactions are considerably smaller than those in equities markets. Finally, consistent with studies based on equities markets, order size, the probability of execution error, execution difficulty and broker identity are significant determinants of brokerage commissions in futures markets.

The second essay examines the profitability of local traders surrounding the introduction of electronic trading. Local traders are among the most publicised features of floor-traded futures markets directly affected by the movement to electronic trading. Researchers and exchange participants alike acknowledge that local traders perform an important role in the efficient functioning of a floor-traded market. Local traders have publicly objected to electronic trading, arguing that order flow-related information in a pit trading environment is necessary to their survival. As found in prior research, which suggests that locals are profitable due to superior trading skill, the introduction of electronic trading on the SFE does not coincide with a reduction in local trader profitability. Furthermore, consistent with the findings of prior research, the strategies of profitable local traders are characterised by an aversion to holding overnight positions, trading in round trips, trading frequently and trading in small quantities. With respect to these characteristics, local traders do not change their trading behaviour in response to the shift to electronic trading.

The third essay in this dissertation examines the impact of increased limit order book disclosure in an electronic futures market. The literature identifies depth information in the limit order book as a primary difference between floor and electronic trading systems. Unlike trading by open outcry, electronic trading is capable of delivering varying levels of limit order book transparency. Evidence is presented of a change in trading behaviour coinciding with a change in pre-trade transparency on the SFE. Consistent with predictions of a theoretical model based on execution risk, the study presents evidence of a statistically significant decline in depth at the best quotes; however, there is little evidence of an increase in spreads. Further, the proportion of market orders exceeding depth at the best quotes increases in a transparent limit order book, reflecting a reduction in execution risk. The findings suggest that in a transparent market, limit order traders charge market order traders a premium for execution certainty by withdrawing depth from the best quotes, but not with sufficient uniformity to reduce bid-ask spreads.


Alex Frino and Elvis Jarnecic