The longer a firm had to deal with a broken trade, the more expensive. It was a nightmare of operational, reputational and credit risk. Investment managers were constrained from growing their business because they didn't have accurate and timely execution information on their side, and their back offices and those of their counterparties could not keep up. By the late 1980s, big broker-dealers had already created their own branch communications networks that consolidated orders and sent them to the New York Stock Exchange (NYSE) via the exchange's Designated Order Turnaround (DOT) system and for big buy-side customers a few bulge-brackets created "dummy branches," an early version of direct market access.
This was hardly satisfactory to the biggest gorilla in the room. Fidelity lusted after Salomon Brothers' paperless program trading and order management system. After inviting other buy- and sell-side firms to join in and then seeing them drop back, Fidelity and Salomon agreed bilaterally to drive an industry messaging standard.
In 1993, Fidelity, Salomon, Goldman Sachs, Paine Webber, State Street and American Century agreed to develop a standard. The only industry consortium at the time, Swift, had started out as such an entity in 1973 with 239 banks in 15 countries that understood their lifeblood was funds transfer. By 1977, Swift had 518 banks in 22 countries. By contrast, the Financial Information Exchange (FIX) consortium was voluntary and led by technology in one country with a handful of participants.
FIX finally went into production in early 1995. "The coming out party included just two brokers and one buy-side firm-and one of the brokers (Fidelity Capital Markets) was the other half of the buy-side firm (Fidelity Management and Research). The system was hardly secure and only covered equities, but it was well-tested and solid," says Jim Leman, who managed program trading at Salomon Brothers and was one of the early FIX pioneers. In 1996, the other big buy-side gorilla, Putnam Securities, had Cambridge Technologies build them a FIX engine and whispered to the sell side: "You want to do business with Putnam? Adopt FIX."
Early adopters streamlined trade processing, moved toward straight-through processing, enjoyed greater transparency, and faster and more efficient execution. Security remained an issue, but traders and portfolio managers had greater choices in trading systems as well as venues for trading. Fidelity created its own piece of software so that small broker-dealers could send in indications of interest (IOIs) and confirmations using the FIX protocol. Alternative trading systems (ATSes) were sprouting up, but had no common industry application interface (API). Goldman's FIX committee member Sam Johnson created TransactTools, a FIX toolkit that was quickly adopted by Archipelago. Johnson is now executive vice president of NYSE and CEO of Transact-Tools, a wholly owned subsidiary thereof.
FIX standards then moved abroad to the UK and Europe from 1996 to 1999, and to Asia from 1997 to 2000. In 1999, FIX began to support options, fixed income and program trading. And newer, better, faster enhancements are coming. Firms can actually adopt new versions of FIX for new asset classes without having to upgrade the FIX versions that are already working so well for them.
FIX now underwrites the smooth execution and settlement of myriad cash and derivative instruments, is used by 85 percent of all buy- and sell-side transactors and is supported globally by 23 buy-side behemoths, 44 global securities firms, 98 technology vendors and 26 exchanges.
Leman points out that FIX enabled the systems and technology that came after, like order management and execution systems, transaction cost analysis, direct market access and algorithmic trading. And now "with Regulation NMS and the Markets in Financial Instruments Directive, regulators want a lot more data. You need the FIX protocol to do all this. Otherwise, you're standing there with two Dixie cups connected by a piece of string."
-Maureen Callahan