Bonds can no longer hide from reality by claiming 'We are different.'
Ashok Krishnan’s colleagues at Bank of America initially brushed off his pitch to automate trading. He recalls how some barely acknowledged him in the halls. But before long their aversion gave way to requests for help.
As head of electronic trading in the bank’s global markets division, Krishnan is among a group of Wall Street equities veterans -- including Phil Allison at Morgan Stanley and Mark Goodman at UBS Group AG -- promoted in recent years to take on a delicate assignment: Build machines to handle trading in other markets such as bonds and currencies. They’re convinced it’s possible after watching computers take over stocks.
In an interview at Bank of America Corp.’s headquarters in New York, Krishnan, 48, ticked off technologies such as algorithmic and mobile trading that are giving machines a greater role in making complex markets.
Persuading humans is something else.
The long-predicted automation of bond and currency trading has been slow to arrive, after traders voiced skepticism that happened to protect their business and jobs. They argued clients want to talk through complex transactions and that automation weakens relationships. Even shifting simple bets to electronic platforms would reduce interactions with customers, missing opportunities to arrange other deals and “cannibalizing” the business. Yet investors are demanding more advanced platforms to cut costs, as well as new tools providing a clearer view into markets.
“If you don’t do this, your P&L goes down because the client just does business somewhere else,” Krishnan said, invoking a Wall Street acronym for earnings.
The stakes are huge for Wall Street banks competing in the $22 trillion market for U.S. Treasuries and corporate debt and the $5.1 trillion-a-day foreign-exchange market. Handling fixed-income products remains one of industry’s biggest revenue generators. Now firms must disrupt the old model by building out their electronic platforms, or they risk getting sidelined in the future.
Equities migrated to computers first because they’re standardized and trade frequently, making it relatively simple to match buyers and sellers. The same goes for parts of bond and FX trading, such as Treasury futures and spot currency transactions. That’s left stock veterans with the technical experience to tackle more-complex products -- as well as the confidence needed to overcome years of resistance.
“It started with the equity guys saying, ‘Let’s make these markets look like our markets,”’ said Kevin McPartland, head of research for market structure and technology at consultant Greenwich Associates. “From the bond side it was: ‘No, no, no. I’ve been doing this for 20 years. I know my clients. There’s no way you can get a machine to replicate the knowledge I have.’”