Australian Review: HFT and Dark Trading Don’t Merit Further Regulation

The Australian Securities and Investments Commission (ASIC) recently released its Review of High-Frequency Trading and Dark Liquidity, which examines the impact of regulatory changes made in response to a 2012 report that identified growing concerns about high-frequency trading (HFT). They include:

• Extreme price movement rules for the Australian Securities Exchange (ASX) and its competitor, Chi-X Australia
• Suspicious activity reporting rules
• Enhanced data reporting rules
• Rules to require direct and immediate control over orders and certification of algorithmic trading systems

The report comprises two main sections — one on high-frequency trading (in equities and futures markets) and one on dark liquidity.

The broad conclusion of the report is that current levels of high-frequency and dark trading do not constitute a concern for ASIC and, as a result, ASIC doesn’t foresee the need for further regulation. Apart from this benign conclusion, the report also features some interesting observations summarized below.

The Australian HFT space is dominated by the 10 largest HFTs that account for 78% of all HFT turnover, a number that has risen from 64% in 2012. These traders are specialized desks within investment banks, proprietary trading firms, and hedge funds.

Regarding high-frequency trading, ASIC notes that:

• During a stakeholder engagement process, ASIC found that negative sentiments about HFT have levelled off, likely as a result of increased familiarity and sophistication of investors in the new high-frequency market environment. The level of HFT has remained steady at approximately 27% of total turnover.
• The concentration of HFT has increased with 30% fewer trading accounts. ASIC notes this is likely because of an increase in the cost of trading capital following the end of the US Fed’s quantitative easing policy, the implementation of the Volcker rule, as well as market structure changes, such as: price improvement and market integrity rule (requiring a minimum resting time of 500 milliseconds for small orders), introduction of tiered block-trading thresholds, cessation of priority crossings.
• HFTs appear to be trading more aggressively than in 2012 but still contribute significantly to the orders at the best bid and offer (BBO). Further, HFTs appear more sophisticated at avoiding other HFTs. They are now operating in a greater range of securities, and this may have contributed to lower transaction costs in these securities. Another observation was that additional high-frequency trading acts to dampen price movements (during nonextreme market moves).
• The cost to other users is estimated at between 0.7 and 1.1 basis points (BPS), a figure that ASIC notes is far smaller than more common trading costs (e.g., an average spread of 13 BPS). HFTs do not appear to drive transaction costs, and increasing levels of HFT for lower turnover securities appear to reduce transaction costs in those securities.
• In the futures markets, HFT has seen rapid growth and now comprises around 21% of traded volume in the common Commonwealth Treasury Bond Futures. ASIC is conducting inquiries regarding excessive order entry and cancellation.

Regarding dark liquidity, ASIC notes that:

• The lowering of block trade thresholds in 2013 appears to have encouraged the use of dark trading for its original purpose of executing block trades while minimizing market impact.
• Client demands for improved standards as well as ASIC regulations appear to have improved the fairness and transparency of crossing systems. Further, the use of crossing systems has declined, likely in response to the price improvement rule. (Read about the CFA Institute study on this rule here.)
• ASIC notes a trend for dark venues to incorporate broker preferencing rules, something it says it is unlikely to support in any form.

In broad strokes the ASIC review of trading in Australia appears to be consistent with a number of recent studies that fail to support the negative perception of modern markets that is popular in some quarters. While concerns about certain market practices do exist, it has proven difficult for researchers to demonstrate systematic market dysfunction as implied by books such as Flash Boys. In this way the findings of the ASIC review are consistent with the recent CFA Institute market liquidity report, which also found some concerns with market characteristics in the broader context of historically low trading costs.

The criticisms of HFT tend to revolve around three main issues:

• Latency arbitrage
• Phantom liquidity
• Market fragility

Given the timescales on which latency arbitrage and phantom liquidity occurs (micro- or even nanoseconds), it will likely not be possible to empirically demonstrate a market-wide problem, especially considering data of necessary quality is typically not available to researchers. Another reason may be that these issues affect mostly institutional investors who are increasingly sophisticated in their order submission strategies themselves. One can also sense a shift in the debate away from HFT itself and towards the disadvantages of such a complete fragmentation of the market, particularly in the US. It is the mismatch in communications infrastructure between the different venues that allows the existence of these unpopular strategies. With far less fragmentation in markets outside of the US, perhaps this explains the relatively benign findings of the ASIC review.

Market fragility is also a difficult concept to test empirically given the lack of natural experiments to study. Further, the idea that liquidity must necessarily exist in the midst of a major news event is not necessarily intuitive. The implication of this insatiable demand for market liquidity is that there must always be a bigger fool to take an investor’s losses. The urge to step aside during uncertainty is so fundamental that it is not obvious whether it can or should be overturned. This theme was picked up by Martin Wolf in a recent Financial Times article. The recent ‘Black Monday’ flash crash contradicts this somewhat nihilistic point of view. During this day we saw Costco trade with an 8-point spread ($117.03–­$125.14), clearly an unjustifiable price at the time. We have posited in our market liquidity report that this kind of issue may be a consequence of the homogenization of liquidity provision as HFT market makers take over from the various human liquidity providers.

If you liked this post, consider subscribing to Market Integrity Insights.

Image credit: iStockphoto.com/halbergman