Costs incurred on multi-dealer platforms, non-exchange venues which allow traders to send request for quote (RFQ) orders to multiple liquidity providers, are increasingly encouraging buy- and sell-side firms trading fixed income, currencies and commodities (FICC) to trade bilaterally, a study has suggested.
According to the research from Coalition Greenwich, the electronification of the FICC markets has increased costs, which make it more difficult for liquidity providers to return their cost of capital. As platforms charge fixed fees, however, they have not experienced the same erosion on margins. Platforms also often charge additional fees ranging from credit intermediation, ancillary service costs, straight through processing and platform administration. The research has implied that the growing costs are beginning to resonate with the buy-side and encourage more bilateral forms of trading.
In response to increased costs, direct connections between a liquidity provider and taker are increasingly desirable as they allow the liquidity provider to maintain control over margins and data, and minimise information leakage. The multi-dealer vs single dealer debate has long-been debated in FX markets in recent years, but whether that translates as heavily in credit and rates is yet to be seen.
There is little evidence to imply that buy-side credit and rates traders are pushing for bilateral connections as aggressively as their sell-side counterparts. While information leakage and data control are definite pain points for the buy-side, costs incurred through trading on venues are often absorbed by liquidity providers. In other words, it is not really the buy-side’s problem yet and offers them little incentive to alter their current tried and tested methods of execution in this market.
“I’d say for the past 18 months, there has seemed to be more of a drive from the sell-side to encourage us to adopt some of these potentially direct connections and therefore avoid going through a venue,” said Stuart Campbell, head of trading at fixed income specialist, Bluebay Asset Management.
One of the key factors driving the sell-side’s desire for bilateral connections is the control of transaction data, of which they have almost none.
“One of the most irksome unintended consequences [of electronification in the FICC markets and subsequent use of multi-dealer platforms] is that participants lose control of the data associated with their activity, which is then repackaged and sold back to the end users. Platforms take different approaches to these services, but they are frequently charged in aggregate, making it hard to attribute to individual transactions,” said Coalition Greenwich’s Thomas Jaques and Subodh Karnik, authors of the report.
Traders in the FICC trading landscape are captive to a select few multilateral trading facilities (MTFs) such as MarketAxess, Bloomberg and Tradeweb. Almost all trading volumes go through these platforms and they possess a monopoly on data. It is a key pain point across the Street, with many firms begrudged at the prospect of being charged for the data they have generated through their own supply and demand.
While bilateral connections allow traders to control who sees their data, this alone is not enough for buy-side firms to forgo multi-dealer RFQ platforms altogether and limit themselves to a smaller pool of liquidity providers.
Untrustworthy data
A direct connection with one or a select few liquidity providers is reliant on trustworthy data, which is often fragmented and requires intensive scraping. Third-party platforms are heavily integrated into buy-side workflow and implementing a new bilateral connection is a long-term and expensive endeavour. These solutions can be developed in-house or offered through APIs via an order management system (OMS) provider.
“Pre-trade data in this market is not something I want to spend money on. I would not need to go to the rest of the market and RFQ to 100 people and give away all that information leakage if I could trust that a bank is offering the best price,” adds Campbell.
“If I can believe the prices that I see on the screen that are being shown to me then I’d be more willing to engage directly with a bank. The reality is, sometimes when I engage with them, they’re not the best price at all. I don’t have enough trust in that data.”
A potential solution to the data conundrum is a consolidated tape. Similar to the TRACE system in the US, firms could access data at a significantly lower cost. Participants and trade unions alike have come together in the last few years in their calls for a standardised data offering in Europe, however, due to the enormity of the task of creating a tape that fits the needs of all participants, the road to its development has proved to be long and windy.
In response to ongoing pressure, the European Commission set out plans in early 2021 to design and implement a consolidated tape in the primary and secondary bond markets as part of a review of MiFID II.
With a standardised and reliable source of data in the market, FICC buy-side firms may be convinced that the cost of implementing a bilateral connection with a smaller number of liquidity providers could be a worthwhile investment.
“Unfortunately, there has not been many people put their hat into the ring to generate a consolidated tape, but we are hopeful that one will come. Ultimately, it’s got some negatives, but we think it will be beneficial. If we can get that then I’d be far more willing to go down a more data intensive path than we’re currently on now,” says Campbell.
The sell-side pays the price
While control of data in fixed income is a pain for both the buy- and sell-side, when it comes to the costs incurred from the services provided by trading platforms, including credit monitoring and netting, it is the sell-side that bears the brunt.
“Is the buy-side only going into bilateral trading because of these explicit costs? No, definitely not. In general, the sell-side is making direct fee payments as part of the RFQ process for these transactions on MTFs. The buy-side is charged only in an indirect way as a part of the net price received,” Christoph Hock, head of multi-asset trading at Union Investment, explains.
“At first look, focusing on trading fees only, it might seem easier to interact directly with sell-side firms without an MTF involved. However, our impression is that an RFQ mechanism is one where we get the very best result for our investors, and that’s the reason why a decent percentage of our business goes via this process.”
Large US bulge bracket banks can pay up to $50 million each year from transactions that take place on platforms. However, price bundling means venue-incurred costs can be difficult to separate and are not always seen by the buy-side.
“The cost increases seem to be more damaging to the sell-side than the buy-side. There is maybe a slight increase on the fees that we are charged to do a trade on a venue. The price does move around,” adds Campbell.
“If the buy-side push to see exactly what the cost is on every ticket, I think more people will become aware of the fact they’re being charged for these trades. That might drive a second wave of people taking their trading offline to their own interface of API connections.”
Information leakage and competition
In sending an RFQ to multiple brokers, there is the risk of information leakage. Coalition’s study said liquidity providers uniformly highlighted the damaging impact of this on market efficiency. Although, this is dependent on the size and nature of the trade being executed and on the reliability of the data put out by liquidity providers in sensitive scenarios bilateral trading methods are preferred.
“When we are looking at a medium size order it really matters how many broker firms we ask for a price. If you blast an RFQ for such a type of order to 15 or 20 broker firms, you have a significant information leakage and get most likely a decent number of rejects,” says Hock.
“Involved broker firms realise the whole Street is aware of this individual transaction and know therefore it might take days to trade out of this position. The highly sensitive trades in the high yield world and in emerging markets for example, that is where you just ask one or maybe two broker firms for a quote to achieve the best outcome for your investors, and that’s fully in-line with best execution.”
Information leakage has the potential to damage a firm’s ability to hedge their positions. Often referred to as the ‘winners’ curse’, the firm that wins the best price must move quickly to ensure it has hedged away any risk. If too slow, it is at risk of having negative positions put out by competitor firms.
Bilateral connections certainly apply themselves well to the trading of some financial instruments. For example, when trading treasuries, buy-side firms can get comfortable with limiting themselves to five or six direct connections with major sell-side banks. However, multi-dealer platforms, and the breadth of liquidity available on them, are more applicable for the trading of others. The danger that in a bilateral connection a liquidity provider may not offer the best price remains.
“In bilateral trading, as the market structure is currently set up, it is less likely to create this highly competitive element which delivers you the most aggressive price. Therefore, the most efficient way for us to achieve the best result in the fixed income world is still the RFQ process,” adds Hock.
“Some broker firms might offer you a different price when they know that that there is no competitive element. However, in the future maybe the market will move in the direction of having a central limit order book (CLOB) for certain types of fixed income instruments, a kind of click-and-trade facility for a number of aggregated broker streams.”
While the desire for bilateral connections in FICC trading remains with the sell-side, implementing a new channel of this kind has a high initial cost, which can act as an incentive for buy-side firms to instead connect to a venue. Due to this high cost and level of complication of implementation, a ready to use plug and play solution is often preferred.
“I don’t want to spend a lot of time and money building my own technological solution for a problem that doesn’t really exist for me at the moment. I would need that to be on a plate right in front of me offered as an out of the box solution by my OMS provider and I would need to see some significant benefit to me trading bilaterally rather than through the venue,” adds Campbell. “But at that point I want to make sure I’m not a venue.”
Still a way to go
While Coalition Greenwich’s report suggests FICC traders are leaning towards bilateral connections, the largest share of credit and rates respondents, 47% and 44% respectively, plan to continue to use RFQ on multi-dealer platforms most heavily in the future.
There is not enough incentive for buy-side firms to start trading bilaterally. Costs are absorbed by the sell-side and until pricing changes from indication only to click-to-trade, it seems unlikely that the small indirect increase in fees dedicated to platform commissions will drive buy-side firms to dedicate lofty sums to developing the infrastructure needed for bilateral connections.
MiFID II reguIation has continued to encourage trading volumes onto lit venues and regulators are increasingly examining the definition of a venue. Streaming pricing from bilateral connections from multiple liquidity providers via in-house APIs leaves buy-side firms at risk of being defined as an MTF and buried in an additional layer of costs.
“If we all gravitate and take our trading off-venue there’s nothing to say that the venue may just collapse its pricing structure and so straight away its everybody back on board,” concludes Campbell.