As technology has made trading more advanced, it has also put more pressure on firms to manage and store all that data. Firms that rely on spreadsheets to do that are overwhelmed by the high volume of trading data today.
Jim Mullen, founder and CTO of Firm58, a trading and back-office provider, illustrated the problem in an interview with ThinkAdvisor.
“As trading came off the floor and went electronic, you saw what used to be a single trade of, say, 1,000 shares get broken down into 10 trades of 100 shares, or get broken down even further. In addition to that, those trades can then get placed across many marketplaces so in the last 10 to 15 years, a proliferation of new markets have come out so basically you can trade anywhere you want,” he said.
That has resulted in two new realities for firms, he said. “One is I have physically a lot more data. I have to store it and house it just like any other trade.”
The other problem is the complexity of determining fees. “I’ve added complexity to any calculations around fees and commissions because I’m involved in multiple parties. A single trade of 1,000 shares can now go to many places. Each of those places might have different fee structures in place,” he said.
As the SEC and FINRA have become more aggressive, he said, the industry has realized it can’t run compliance from a spreadsheet anymore.
“Spreadsheets allow you really only to load so many physical trades or orders,” he said. “A lot of what we’re talking about in the realm of compliance is order based. Let’s say you do 10,000 physical trades per day. You might do 10 times that in the number of orders you place because you can cancel those orders, other things can happen.”
Spreadsheets are good at sorting data, he said, but “what they’re not good at is finding patterns, and that’s where the next generation of compliance is at.” For example, he said, “if I’m wanting to trick you, as another trader in the market, I might place an order to buy something that’s below the bid price. That might trigger you to actually match that particular bid and place another buy order in the marketplace, in which case I would turn around and cancel my buy order and do a sell against your buy. What I did was actually manipulate the market; I tried to trick you by putting something out there that I had no intention of fulfilling. That’s the kind of sequence of events that we try to hunt for, and they’re only getting more complex. The complexity gets even worse as you add more data.”
One of the reasons some traders still rely on spreadsheets rather than more advanced databases is because they don’t have the skill set. “They use what they know, even though it might not be the best tool. I’m pretty good with a hammer and I can use the claw of a hammer to screw in a screw, but that doesn’t mean it’s the right tool.”
He recommended advisors who don’t have the skills to manage a trading database on their own contract with a vendor who has the “compliance and the surveillance to look through this information and at least put them in the ballpark, hopefully close to home plate, where they might be able to identify problems and more quickly resolve them.”
“The whole idea,” he added, is that “as your business grows, as your volume and physical number of transactions increases, you don’t want to have to be continuously adding more people to your compliance group. You’d like to have a solution where that’s more scaled; where I can keep the same group, let the system do the work and I’m just looking for various problems.”
— Check out FINRA Hires SEC Data Analytics Guru Kurtas for New Post on ThinkAdvisor.