Speech by Mark Steward, Executive Director of Enforcement and Market Oversight at the FCA, delivered at the 19th Annual Institute on Securities Regulation in Europe event in London.
Speaker: Mark Steward, Executive Director of Enforcement and Market Oversight
Event: Practising Law Institute’s 19th Annual Institute on Securities Regulation in Europe, London
Delivered: 6 February 2020
Note: this is the speech as drafted and may differ from the delivered version
- We work with a very high degree of precision and collaboration with both US regulators and EU authorities on issues that affect our markets. We need to avoid fragmentation in our markets which will provide wrongdoers with arbitrage opportunities for misconduct.
- The market cleaniness metric was first published in 2008 when it found that approximately 30% of takeovers showed abnormal price movements 2 days prior to an announcement. Our latest figure published in 2019 is currently at its lowest score at around 10% and we have developed a new measure which has produced a lower figure of 6.4%.
- We have taken a strategic and integrated approach to our market integrity efforts encompassing enforcement, primary and secondary oversight and surveillance, and wholesale supervision which could explain improvements recorded by the market cleaniness metric and abnormal trading volume ratio.
- Would-be abusers should be aware of our increased detective power which materially ratchets up the risk of abusers being found out.
It is a pleasure to be here this morning to open these two days of discussions on the confluence of EU, UK and US law and practice in our securities markets.
I use the word ‘confluence’ deliberately. While there can be debate about technical differences in law or interpretation or approach between jurisdictions, UK, EU and US approaches in our securities markets are not differences of principle nor ones in which materially different regulatory outcomes are sought or expected. Indeed, UK and EU law here remain, for now, entirely aligned.
The common underpinnings are important because they enable us to work, with a very high degree of precision and collaboration, with both US regulators and EU authorities on issues that affect our markets.
For example: the FCA has worked closely with European authorities for some time on a story that appeared this week about the seizure of a property in London as part of proceedings against a trader allegedly involved in a dividend stripping tax avoidance schemes that have operated in Denmark, Germany, France and Italy.
The FCA has been investigating substantial and suspected abusive share trading in London’s markets that has allegedly supported these schemes. These investigations are now very close to their conclusion and decisions about action are imminent.
We are also working with both EU regulators and US authorities on several market abuse investigations involving trading in EU, London and US markets.
This is only possible because the underpinnings of our laws and practices are built on common principles in which equivalent or compatible outcomes can be achieved and there is frankness and intimacy in global regulatory cooperation and collaboration which ensures we are able to tackle the most insidious and organised cross-market abuse.
This is necessary if we are to ensure our markets remain places that work well where genuine forces of supply and demand can operate safely and efficiently.
We do not want to be regulating fragmented markets in which wrongdoers are able to escape detection or liability either because we lose the sight of the common principles and standards that underpin our laws.
Indeed, fragmentation will only create arbitrage opportunities for misconduct, affecting the integrity of all markets.
Let me turn to our approach to market integrity in our securities markets.
In 2019, the FCA received 9.8 billion MiFID II transaction reports – an increase of 17% on 2018. About 10% of these reports were received from European Economic Area National Competent Authorities (EEA NCAs) via the EU’s Transaction Reporting Exchange Mechanism or TREM. The FCA, in turn, shared about 70% of the reports received from UK firms with at least one other EEA NCA.
In addition, as part of a deliberate strategy on our part to increase the granularity of our data and the accuracy of our detection processes, we now also ingest the FTSE 300 order book which generated around 150 million order reports per day. Our systems are now able to consolidate the order book so we can aggregate orders in the same stock across different platforms.
The third integer in this powerful database is the product of the market’s own concerns and suspicions with around 6,000 Suspicions Transaction & Order Reports (STORs).
The consolidation of this data gives us the capacity to examine what is happening in our markets with increasingly bright lights in close to real time. This would not be possible otherwise. Little in the cash market is now unobservable or undetectable.
This data has also enabled us to think of new ways to measure what is happening in our market.
Many of you will be familiar with our annual market cleanliness metric (the MC metric). The MC metric was based on the percentage of abnormal movements in price in the two days preceding a takeover announcement.
The MC metric was first published in 2008 and it found that approximately 30% of takeovers showed an abnormal price movement in the 2 days prior to its announcement.
It was perceived to be a proxy for insider dealing in the market, or, at least, the amount of insider dealing that might be occurring in the market ahead of takeovers.
The MC metric remained at around 30% for a few years. By 2013, it had fallen to around 15%. However, it then climbed again and by 2016 it had risen to around 22% before falling back to its lowest score of around 10% in last year’s report. This was the last figure we published.
This might suggest there is some volatility either in the MC metric itself or in the incidence of suspicious activity. Both may be the case. There are also limitations to the MC metric:
- The focus on takeovers means the sample size is relatively small which means to is prey to all the problems that arise from an extrapolation from a small sample.
- Secondly, because takeover announcements are not always unexpected, legitimate speculation rather than market abuse explains some of the abnormal price rises.
In 2018, we committed to reviewing the MC metric and developing some new metrics, largely to overcome these limitations. I envisaged a very difficult task.
But this did not factor in the richer data set that we are now ingesting into our Markets Data Processor, including MiFID II data, which provides data across a broader spectrum.
With the richer data we now collect, we have developed a new additional metric, which we call the Abnormal Trading Volume ratio or ATV.
The ATV overcomes some of the innate limitations in the MC metric because:
- the ATV takes into account a much broader and richer set of events and is not limited to takeovers, using over 1,000 unexpected, potentially price-sensitive announcements;
- secondly, it includes products not considered before, such as contracts for difference and spread bets, which have been used by insiders in some of our more complex enforcement cases; and
- thirdly, the ATV considers changes in volume as well as changes in prices, so the new test brings a new and additional signal to the measurement.
The ATV is calculated by observing movements in trading volumes of relevant securities that are the subject of an unexpected price sensitive announcement where there are no other potentially price sensitive announcements.
The observation period is divided into two parts, a benchmark period, which begins 30 trading days prior to the announcement (which we call the Observation Period) ending 10 days prior to the announcement.
We then compare volumes in the Observation Period to those in the period that is 10 days before the announcement (the Announcement Period). The ATV then measures whether trading volumes significantly change with a 5% variation qualifying as significant.
Last year, the ATV’s first year, we tested trading before 1070 announcements and found, on average, that 6.4% of them were associated with an abnormal price increase using this methodology.
It is important to point out that the ATV metric is not intended to be a necessarily sufficient measure of market cleanliness: we do not call it a market cleanliness measure, for example, nor do we consider it a replacement for the old MC metric (as opposed to an additional view).
Instead, we see both the MC metric and the ATV as different proxies that enable us to develop and measure trend lines as well as being statistically reliable and rational.
Despite its limitations, the reduction since 2008 to 2018 from 30% down to 10%, in the case of the MC metric, is a significantly positive movement, with the trend fortified by last year’s initial ATV result of 6.4%.
What might explain these positive outcomes?
The FCA published an Occasional Paper in 2014, asking the question why had the score fallen so dramatically since 2009, concluding
‘We recognise that, at this stage, we do not know whether the FSA/ FCA’s actions are driving the improvement in the measure but the increase in the FSA’s enforcement activity and educational agenda and decrease in the market cleanliness statistic are roughly contemporaneous’
An increase in enforcement is likely to be a significant factor here. Or perhaps to put it more fairly, I suspect the reduction would not have occurred at the same rate and to the same extent without:
- the FCA’s work creating a broader understanding that insider dealing is in fact a type of fraud;
- the FCA’s continuing investment in tackling market abuse; and
- FCA prosecutions becoming successful with substantial terms of imprisonment and financial sanctions being imposed.
But I doubt these factors alone fully explain the improvement in the MC metric or the even lower ATV metric.
I suspect the result is the consequence of the sum of our total activity, encompassing not only enforcement but also primary and secondary market oversight and surveillance and wholesale supervision. And this approach has become more integrated in the last few years, concentrated into a single strategy with a single aim: ensuring our markets uphold and demonstrate the highest examples of market integrity.
This involves a dynamic combination both supervision and enforcement working together in a single strategy not only on detection and enforcement but also on ensuring market participants are behaving properly in preventing abuse, protecting inside information from leakage and misuse and reporting suspicious activity.
With this approach, good regulatory outcomes depend less on selection of the so-called ‘right regulatory tool’ than on the strategic integration of concurrently deployed regulatory functions and powers, focusing on the same outcome.
This dynamic interaction across our own capabilities and harnessing the product of the market itself not only raises the integrity bar, it has significantly improved our radar for detecting market abuse
This strategy also depends on market participants also playing their role which explains why we have taken strong action against firms that fail to implement appropriate systems and controls to carry out their own trade surveillance, to report obviously suspicious activity or to report transactions accurately at all.
This dynamic interaction across our own capabilities and harnessing the product of the market itself not only raises the integrity bar, it has significantly improved our radar for detecting market abuse.
This is a significant deterrent itself and would-be abusers should be aware of our increased detective power which materially ratchets up the risk of abusers being found out and should force a recalculation of the wrongdoers’ ‘risk of being caught vs reward’ ratio.
Let me say a few words now about manipulation, which often escapes attention because of the focus on insider dealing.
Market manipulation and the use of false or misleading statements to create false markets or otherwise illicitly profit from markets are equally corrosive of market integrity, especially if left undetected.
We made a strategic decision to ingest the equity order book into our Markets Data Processor because it is very difficult, if not impossible, to detect trading manipulation with transaction data alone. Indeed, I cannot imagine how there could be any sensible attack on such manipulation without an ability to surveil the order book in a consolidated way in tandem with transaction reports, which is what we are now able to do.
Now we can look for manipulative trading more easily and the proportion of investigation work is now split 60:40 between insider dealing and manipulation. This is a big change from relatively recent days when our wholesale investigation caseload was almost exclusively based on suspected insider dealing.
And our reach and ambition is not confined to the cash market as we develop equally broad and sensitive radar for other traded asset classes and markets, especially where insider dealing is less of a threat than manipulation.
We also have a smaller number of very significant investigations on foot where it appears false and misleading statements have contributed to the destruction of considerable shareholder value:
- In one case, which for the moment must remain anonymous, we are investigating the demise of a listed company which we suspect was surreptitiously stripped of assets while its controllers fed the market with false and misleading information to boost the share price, which they may also have profited from.
- In another case, we are looking at action against 2 directors of another failed listed company whom we allege orchestrated a complex fraud to falsify the firm’s revenue and profits and to mislead the market.
These types of cases are complex and more difficult to investigate than insider dealing or other types of market abuse because they are not transactional or involve opportunistic trading; perpetrators often work in groups, in an organised way, using sophisticated techniques, over extended time periods; these schemes generally involve elaborate cover-up strategies that need to be unpicked and the volume of documents involved often run into the millions of digital pages.
Taken from the FCA Website