Smallwood says that on the sales and trading side, banks hold contracts that govern how they can interact and transact with clients.
“there's contracts that govern the way that banks can interact and transact with their clients.”
Person
Lee Smallwood of Citi is a banking professional who spoke on a podcast about AI in financial services, covering how natural language processing (NLP) could be used in banks' back-end and front-end work. He discussed derivative contracts, the LIBOR replacement, Bloomberg chat, compliance and trade booking.
Smallwood said banks once focused on revenue, while the cost of regulatory non-compliance now weighs heavily on the bottom line. He estimated revenue-focused NLP tools could appear in two to five years rather than ten. He also said customer privacy limits how much can be drawn from client conversations.
1 episodes · 15 quoted claims
Smallwood says that on the sales and trading side, banks hold contracts that govern how they can interact and transact with clients.
“there's contracts that govern the way that banks can interact and transact with their clients.”
Smallwood says banks' focus 10 to 15 years earlier was on revenue generation, while the downside of regulatory non-compliance now weighs heavily on the bottom line.
“whereas maybe rewind 10, 15 years ago, a lot of the focus would be on massive revenue generation opportunities, and that's still the case.”
Smallwood says an ISDA is a contract that governs how a bank and a client interact, particularly in derivative trading.
“there's a contract called an ISDA that governs the interaction, particularly of derivative contracts.”
Smallwood says millions of derivative contracts across the industry have prices that depend on LIBOR, which is being replaced.
“And think about the millions of contracts across the industry that are governed based on derivatives that have the price contingent on LIBOR.”
Smallwood says many industry working groups are figuring out how to transition away from LIBOR from a business and regulatory perspective.
“There's a lot of working groups to figure out how the industry will transition and sort of a business and regulatory perspective.”
Smallwood cites margin requirements for non-centrally cleared derivatives as an earlier regulatory change that altered how banks trade with certain clients.
“things like, without getting too technical, margin requirements for non-centrally clear derivatives. That was a big change in the way that we go in the interaction of certain trades with certain clients”
Smallwood says most communication in markets and sales and trading happens through Bloomberg Instant Messenger.
“most of the markets and sales and trading industry communication happens via Bloomberg Instant Messenger.”
Smallwood says Bloomberg terminals, with black screens and orange text, are found on nearly every trading floor.
“any trading floor you step on, you'll see people with Bloomberg terminals. When you see the black screens with the orange text”
Smallwood says Bloomberg is starting to build NLP-based software, for example for municipal bonds, that can show how many clients requested a type of bond.
“I believe municipal bonds is a good example, where Bloomberg is starting to build NLP-based software that can tell you that this many clients have requested this type of bond”
Smallwood says bank compliance departments are generally huge and use many layers of filters to analyze chat messages.
“Compliance departments in banks are generally huge. And the level of layers and number of funnels and filters that they go through in terms of analyzing chat is quite tremendous.”
Smallwood estimates revenue-focused NLP applications could appear in about two to five years rather than ten.
“maybe more like two to five years than 10 years until you start to see some really interesting applications really focused on moving the top line.”
Smallwood says spoken language differs from written language, so NLP on voice transcripts needs better transcription and models adapted to speech.
“the more you realize the way that we speak is not the same way that we write necessarily.”
Smallwood says speaking a trade into a microphone and having booking and clearing completed automatically would save a lot of effort.
“trade booking, if you could speak that into a microphone, I bought or sold this, and all the booking is done, and it goes through the clearing, then that saves a ton of efficiency”
Smallwood says big institutions take customer data and privacy very seriously, which limits how much can be extracted from conversations with clients.
“big institutions take customer data and privacy very seriously. And so it's how much can you extract from one side of the conversation?”
Smallwood suggests analyzing only the bank side of banker-client communications to judge which sales approaches lead to deals, without identifying the clients.
“can you look at the communications between a banker and a set of clients where you don't necessarily know who the client is”