The Bank for International Settlements’ Markets Committee has released a report looking at the evolution of what it terms fast-paced electronic markets (FPMs), focusing mainly on spot FX, and the challenges this evolution has for central banks.
The report says the market structure changes have implications for central banks’ approaches to market monitoring, including the range of participants with which they engage, the types of data they collect, and the tools and technologies they utilise.
It adds that market monitoring is a core component of central banks’ mandates and many are active participants in FPMs, either for monetary policy implementation or for reserves management purposes. “As a result, being able to understand market developments and obtain an accurate view of market conditions is of paramount importance,” the report states. “And even for central banks not actively participating in these markets, the significant pace of change in recent years underlines their importance for financial stability mandates.”
The report highlights episodes of market dysfunction, including as the sterling flash event of 7 October 2016, the EUR/USD spike in December 2016 and the flash rally in US Treasuries on 15 October 2014, to reinforce the importance of close monitoring and timely analysis of underlying drivers.
To enhance central banks’ capacity to monitor FPMs, particularly the foreign exchange market, the Markets Committee created a study group to explore aspects of structural change that have immediate relevance for monitoring approaches by central banks.
The report identifies three key structural trends that have intensified in recent years, affecting approaches to market monitoring: technological advances; a broader spread of market participants, in particular the rise of the non-bank market maker; and the surge in data usage.
The study observes that in spot FX, the share of trading volume executed electronically has almost doubled over the last decade. One result of this has been increased fragmented across a range of new venues, while the proportion of trading on so-called primary venues with transparent central limit order books has dwindled over time. Another result is that the frequency of activity and speed of information flows in FX markets has “increased drastically”.
Although the report discusses the challenges of fragmentation, it does also note that there are several potential reasons why the role of primary ECNs in price discovery has remained fairly robust, even as volumes have declined. First, it says that price discovery continues to take place on such venues because hedging activity by dealers aggregates the information from their client trades. “In other words, the tendency towards greater trade internalisation may have led to an increase in the informational content of the flow that is directed to the primary ECNs, in turn enhancing their contribution to price discovery (per unit traded),” it explains. “Second, FX dealers operate on multiple platforms simultaneously. Hence, the best quotes on primary ECNs are transparent to all market participants. Once dealers adjust their limit orders in one market, this information is almost instantaneously reflected on other venues.
“The rise of PTFs focused on electronic market-making and exploitation of short-lived arbitrage opportunities across different electronic platforms simultaneously would have further contributed to such integrated price discovery across otherwise fragmented venues,” it continues. “Finally, the primary ECNs provide a crucial backstop during periods of market stress. Hence, FX trading activity typically reverts onto the primary venues during episodes of stress as these platforms provide the highest concentration of liquidity and pre-trade anonymity.
“While dealers can internalise large FX flows and quote narrow spreads to their clients in normal times, their need to hedge inventory risk rapidly in the inter-dealer market rises sharply when volatility is elevated and client flow tends to come in the same direction,” the report adds. “Primary ECNs thus have an important role to play in price discovery during periods of increased FX market volatility and are likely to take on an increased importance in the ex-post analysis of periods of stress.”
For central banks, the study highlights the challenges of monitoring activity across a larger number of venues, some of which, such as single-bank platforms, are often “opaque by design”.
It also notes, “Even discounting the technological issues and financial costs associated with collecting, cleaning and storing high-frequency data, the sheer number of active venues complicates both near-time monitoring and ex post analysis. A judgment is needed as to which platforms are sufficiently representative of the state of the broader market. The inability to monitor activity across all venues increases the risk that key pieces of evidence may be missed out in analyses conducted.”
The study also highlights another impact of the apparent decline in the market share of primary ECNs and the rise of internalisation ratios of large dealers – greater uncertainty around the high or low-traded point in a volatile market move, or whether a specific price level has been reached in order to settle a derivatives contract or client or retail order. “Much more activity now takes place on single-dealer platforms, where not only the bid-ask spreads but also the mid prices can vary substantially depending on both the dealer and the client being priced,” the study says. “From the perspective of many end-users, this has also increased the demand for a wider set of market data.”
New Firms, New Techniques
As technology has changed, so too has the nature of participation in these markets, indeed the report highlights the changing nature of the business models of traditional bank intermediaries, which have come under increasing pressure. “On the one hand, liquidity provision has become more concentrated among the largest banks, which reap the benefits of a large electronic network of client relationships to internalise a large part of their customer flows,” the study group says. “Many other banks, however, have found it hard to compete and have resorted to an agency model of market-making or have exited the business altogether.”
Significantly, the report also observes that the amount of risk-bearing capacity that traditional intermediaries bring to the market has “arguably declined”.
While that is the case, however, the study does acknowledge the rise of the non-bank intermediaries, most notably principal trading firms (PTFs), which have gained a stronger footing. It adds that while many of these firms started by exploiting technology advantages on anonymous venues, a number of them have in recent years, moved to the direct, disclosed provision of liquidity to a network of clients. “How much additional risk-bearing capacity these firms bring to the market and how their trading affects overall market quality remains subject to debate,” it states, however.
Inevitably, the greater electronification of markets has led to the creation and commoditisation of large quantities of high-frequency data. The report acknowledges that this can raise costs and erect barriers to entry, however it also says it has opened up a range of opportunities for market participants. “Liquidity providers can rely on a wider range of inputs to build a quoted price for clients, and end-users can utilise new data to benchmark and improve the quality of their trade execution,” the report states. “The use of artificial intelligence and machine learning in trading algorithms, while nascent, also has the potential to introduce new market dynamics and increase complexity.”
On machine learning (ML), the report observes that central banks may need to utilise the technique to monitor markets. It also argues that broader application of ML in fast-paced trading could lead to more efficient markets, particularly the timely incorporation of diverse sources of data in market pricing. “But the application of ML could widen the technology divide, pushing lower-tier banks even further towards an agency approach to risk management,” the study warns, adding, “The application of commercially available third-party off-the-shelf technologies heightens the risk of application by less sophisticated market participants, who may not have sufficient controls and governance in place.
“Multiple participants using similar algorithms simultaneously could lead to herd-like behaviour,” it adds. “A good understanding of such developments is increasingly important for the effective monitoring of FPMs.”
A recurrent theme of the study is the need for central banks to embrace new technology and techniques when monitoring markets – indeed the paper states, “…The emergence of new entrants in key roles means that it is incumbent on the official sector to engage with and understand the impact of these participants.”
It acknowledges the need for central banks (and other market participants) to meet the challenge of collecting, storing and processing vast amounts of data, as well as keeping track of trading activity and patterns when so many techniques available to participants, such as algorithmic strategies, are aimed at masking activity, volume and even the participant’s identity. “Traditional market contacts (such as voice dealers) may have less visibility over flows that are driving short-term price formation,” it warns in a statement that highlights the need for central banks to engage at different levels of the participants, such as the head of the e-FX trading business. ?
The study group reiterates its preparedness to embrace the change, however, stating, “As market structure continues to evolve further, central banks stand ready to continue adapting their approaches to market monitoring as necessary to fulfil their individual mandates.”
In presenting the report, Jacqueline Loh, chair of the Markets Committee says, “This study is of core relevance to central banking and I believe it could become a key reference on the impact of technology in financial markets. Together with the Markets Committee’s recent report on Market intelligence gathering at central banks, it also provides unique insights into central banks’ market-monitoring activities.”