At a panel hosted by SunGard and moderated by The Economist, Wells Fargo's Steve Ellis encouraged firms to look at compliance with a glass-half-full perspective.
In a utopian world, better risk tools would result in less regulation. As Alexander Fleiss of Rebellion Research pointed out, we do not live in a utopian world. The president and chief risk officer of Rebellion, a New York-based investment advisor and tech firm, didn't specify whether that meant risk management will never be that good, or that regulators will never back off. Wells Fargo's head of wholesale services, Steve Ellis, joining Fleiss on a webcast discussion that focused on the crossroads between quant trading, risk, and regulation, urged companies to be proactive anyway. Because like the New York State Lottery motto says: hey, you never know.
For service providers, regulation isn't always bad for business, anyway. Ellis pointed out that compliance tools can be made into a competitive advantage, since compliance is universal and the time pressure to meet regulator guidelines can be significant.
Jeff Bell, CEO of New York-based technology firm Lime Brokerage and head of clearing and technology at parent company Wedbush Securities, says his tools are used by sell-side firms that traditionally compete with Wedbush. "But they're facing the same challenges we are and they don't want to take the time or energy or investment to make that happen." Regulatory pressure helped lead to the creation of Lime's pre-trade risk system, which has registered risk checks as fast as 200 nanoseconds─an industry-best, according to Bell. At the same time, reviews and audits by regulators pulls resources away from other projects, stifling innovation.
He emphasized that risk management is a big part of the offerings from Lime, known more for its high-frequency tools. "A lot of people think what we're building is just technology platforms to make trading fast," he said. "The art is making it fast and reducing the risk while you're doing that. It's a series of buying power-type checks and regulatory checks. But the key is being able to do those checks and that validation very quickly."
Rebellion Research's flagship offering is a machine-learning tool that predicts global economic trends. Fleiss singled out Argentina's 2012 troubles, Greece's 2009 collapse, and Japan's 2013 surge as events that his software indicated before conventional analysts. He hastened to add, more than once, that machine-learning economists will never replace their human counterparts.
"I see our technology as cable versus satellite," he said. "One isn't going to replace the other. They're just different options in the same market. The benefit of machine learning is that we're able to cover 46 countries without having to hire five analysts per country. I think we'll always need humans, I just think that machine learning will become a much larger part of the market."
The financial industry has barely scratched the surface of artificial intelligence, he argued, calling this not the dawn but the pre-dawn of that technology.
"A lot of people think it's much farther along than it is," added Ellis. "We're just starting to learn how to think of structured data and unstructured data, how to pull it together and find some interesting insights."
He mentioned fraud prevention as a way that Wells Fargo is using big data and machine-learning technology to scan massive datasets and pick out new kinds of unusual activity that humans couldn't have.
For original version, please click here.