Last spring, WatersTechnology first reported that a consortium of banks led by Citi and Bank of America was preparing to combine its members’ collateralized loan obligation (CLO) trading efforts into a new multi-bank trading platform under the working title Project Octopus. This month, the efforts of that consortium have come to market as Octaura.

The independent company plans to roll out an electronic trading platform this year, with functionality for syndicated loans coming first.

Citi and Bank of America were joined by Credit Suisse, Goldman Sachs, JP Morgan, Morgan Stanley, Wells Fargo, and Moody’s Analytics in the development of the platform, as well as the project’s tech partner, Genesis Global, a low-code provider for financial markets. The group’s efforts follow the trend of institutions looking to digitize the fixed-income market, especially loan trading.

As a result of Octaura’s forthcoming platform launch, Bank of America will sunset its single-dealer platform, Instinct Loan Match. Citi will end the use of the bids wanted in competition (BWIC) protocol on its internal client platform, Citi Velocity.

Octaura will roll out syndicated loan functionality first, followed by CLOs. A gradual rollout is an effort not to overwhelm the market, says Brian Bejile, Octaura’s chief executive. Most loan trading is still done over the phone, and participants on the buy and sell side will need time to adjust their workflows, he says.

Octaura will also roll out an app marketplace for datasets and analytics to solve workflow needs. Moody’s Analytics, a partner in the Citi Velocity ecosystem, had already integrated its data and analytics into Citi’s internal client platform, providing Bejile with a framework for an integration with Octaura. Data and analytics tools developed by other third parties will eventually be made available to clients and dealers through the marketplace.

The convenience Octaura aims to bring clients is modeled after ridesharing apps like Uber and Lyft. “They didn’t invent transportation, but they made it convenient and available for us,” Bejile says. “That’s how I think about interoperability. What did Uber do? They took Google Maps, integrated with a payment system, and linked with companies that have existed for a long time.”

For interoperability, Genesis Global’s low-code platform allows for APIs that can handle inbound events or transactions and outbound data. Octaura is in the process of integrating with other capability providers and hopes to present the final product in a unified, cohesive format, Bejile says.

Genesis’s initial build was shown to more than 400 buy-side participants. In it, two protocols were used: Citi’s CLO BWIC and Bank of America’s Loan Match protocol. BofA’s protocol allowed the bank to set the mid-market prices for a list of loans. During a match session, buy-side participants expressed buy or sell size interest through the platform. When two clients presented offsetting buy and sell interest or a match on a particular loan, a trade was executed automatically between those two counterparties.

The protocols act as daily scheduled events. “The question the client then asks is, ‘What do I do between some of these auctions?’” Bejile says. The feedback indicated a need to create protocols that addressed how people were trading beyond these protocols. This created the inventory protocol.

“Dealer inventory is something that is unscheduled, it happens any time. Banks are advertising, ‘Hey, I can buy this, I can sell that,” Bejile says. “Clients can come in anytime during the day and engage the banks in the inventory.”

The protocol is like a request-for-quote, but there are two key differences. In an inventory protocol, the dealer starts the process by their listing sell and buy interest, while in an RFQ, a buy-side participant initiates the process by stating either buy or sell interest and size. The dealer also states the prices at which they are willing to buy or sell. A client in RFQ does not list a price, but rather requests a price from the dealers.

When Octaura goes live later this year, the Citi CLO BWIC protocol on Citi Velocity and BofA’s loan platform will be turned off. “Octaura’s protocols make obsolete those predecessors by definition, so they will be sunsetted as part of Octaura going live,” Bejile says.

In the making

Octaura began as an internal project at Citi, which Bejile headed up. He had previously spent 18 years at the bank and was global head of CLO issuer management when the seeds of Octaura were sown. Bejile commissioned an auction system for Citi’s internal client platform, Citi Velocity, that would make the bidding process in CLO auctions more efficient than the usual method of picking up the phone and calling into auctions. The internal project undertaken by Citi allowed clients to submit bids electronically.

“We saw a 50% spike in the number of bids submitted to Citi by the buy side during the first week of launch of the feature,” Bejile says. “It was clear that the functionality was both intuitive and much needed in the space due to that significant response.”

Citi’s senior management told Bejile to find another bank willing to partner on a multi-dealer platform. He reached out to Bank of America, which operates a single-dealer platform called Instinct Loan Match. The two banks joined forces, with Citi bringing its CLO expertise and Bank of America offering its experience in loan trading and platform building.

A June 2021 report from Coalition Greenwich found that e-trading in the syndicated loan market was growing after 18 months of an “e-trading tailwind” in the fixed-income market, spurred by office and trading floor closures caused by Covid-19.

Audrey Blater, a senior analyst in the market structure and technology group at Coalition Greenwich, says consortium support is uniquely important to the success of new fixed-income trading platforms unlike in equities markets.

“You need the backing of the dealer community,” Blater says. “If they have skin in the game, so to speak, they’re going to be more willing to make prices, make markets, and use their networks to find the other side of the trade.”

Non-consortium platforms may have great technology but never get enough clients and dealers to make an impact in the market, she adds.

Sources told WatersTechnology last April that in the case of Octopus, the move was a bid to prevent existing fixed-income platforms, Tradeweb and MarketAxess, from cornering the nascent electronic CLO market and charging what the banks see as high fees for trading and market data.

Bejile says the strategy was to start small, as opposed to consortiums that have started with multiple founding members. Today’s leading fixed-income trading platforms were also born from consortiums. Tradeweb was founded in 1997 with investment from four banks. MarketAxess was formed in 2000 with support from JP Morgan and other large financial institutions, including Bear Stearns. Eventually, the banks ceded their investments. Thomson Corporation bought Tradeweb in 2004 and sold stakes to 10 major dealers in 2008. Tradeweb went public in 2019. MarketAxess went public in 2004.

Need for speed

A missing piece in the project’s physical iteration was a technology partner. Citi enlisted Genesis Global, in which the bank had made a strategic investment in 2020 for an undisclosed amount. Citi’s directive to Genesis was to build a multi-dealer platform that could support broad industry participation.

Genesis CEO Stephen Murphy says the provider was able to build a proof-of-concept in two weeks and a minimum viable product in three months. Bejile says that the swift nature of building using low-code modules and tools made the vendor stand out among other tech providers the bank considered.

Proponents of low- and no-code software say that while such solutions certainly contain code, the user need not bother with it, as apps are primarily built visually, or “Lego-style,” by custom dragging and dropping pre-configured components and widgets onto a blank slate.

The appeal of low-code and no-code development is its advantages in building speed. Every new programming language is built, to some degree, on top of older ones. In programming’s beginnings, there was machine code (binary ones and zeros) and assembly code (a low-level, plain-text version of machine code). These are most easily digested by a processor. An app written in a high-level code—say, Java—will get compiled down to machine code to run on a processor. No-code and low-code are extensions of this, which creates an additional level of abstraction that makes software building easier, faster, and more accessible to those with non-programming backgrounds.

Octaura isn’t Genesis’s only project in fixed income. Last week, the provider announced a new, expanded partnership with Neptune Networks, a fixed-income pre-trade market utility, on data distribution.

Murphy says Genesis has built generic technical components that can be applied across asset classes. In equities and foreign exchange, data needs to be distributed with low latency, so the provider built a component that could be configured to serve either of them. That same component was applied to the loan platform. Other generic components include authentication authorization and a transaction handler.

“If you log onto a system, it will look at what permissions you are allowed to see, if you’ve been authenticated, how are you authorized to do anything when you’re on the system,” Murphy says of the authentication authorization module. He says there are more than 50 of these kinds of components.

Octaura, through its name and logo, pays homage to its earlier name, Project Octopus. The company’s new logo features an infinity symbol, representing interoperable workflows and data. But when turned on its side, the logo resembles an octopus.