A large-scale crash in financial markets due to automated trading is inevitable and MiFID's policy-makers are still out of step with how financial markets technology works, a senior academic claimed at high-frequency trading event in London yesterday.
Dave Cliff, professor of computer science at Bristol University, helped author the comprehensive UK Government-commissioned Foresight report on high-frequency trading (HFT) and said an event like the 6 May 2010 US flash crash will definitely happen again.
An absence of understanding of how technology influences the markets by both market participants engaged in HFT activity and regulators attempting to craft rules were the overarching reasons, according to Cliff.
"We can expect something on the scale of the flash crash soon and it will not be a surprise," Cliff told an audience in London's Canary Wharf, agreeing with predictions made earlier yesterday morning by US market structure expert Joseph Saluzzi, co-founder of Themis Trading.
Professor Cliff added that market participants must ensure their systems were ready for such a disruptive event, as presently regulation was not sufficient to stem behaviour that could act as a trigger.
A major factor cited by Cliff was the process of deviance normalisation that, in his opinion, has occurred alongside the growth of automated trading in financial markets, whereby a fear that a disastrous event will happen is replaced by an expectation it will not take place simply because it has not yet done so.
"This process of normalised deviance is occurring in financial markets as we get used to computers interacting with each other in such a way that you get adverse market events," he said.
During his talk, Cliff compared a potential market crash to the failed launch of US space shuttle Challenger, which crashed minutes after take-off when freezing conditions caused rubber seals to break. Initial fears about this risk had subsided with experience, explained Cliff, and became standard practice over prior launches.
Last year's Foresight Report, 'The Future of Computer Trading in Financial Markets', used market data and other evidence to conclude HFT was not itself detrimental to the market, and that draft rules in MiFID II were too far-reaching and would be ineffective in policing poor HFT behaviour.
Key concerns included a proposed 500-millisecond resting time for all orders and the abolition of maker-taking pricing models for exchanges and multilateral trading facilities. Such a resting time would inhibit arbitrage between markets, which is the core of many HFT strategies, and the efficiency of price discovery would thus be diminished, the report found.
"The problem with most financial regulation and in particular MiFID II is that it has been done very badly because it is run by politicians. They need to be held to account and educated by people like [the financial services community]," he said.
The European Parliament and Council of the European Union will in coming months begin to negotiate a final text in the trialogue stage, which includes input from the Commission. Implementation will likely to occur sometime in 2015.