Wall Street Jobs …
… Won’t Be Spared from Automation
by Thomas H Davenport
Harvard Business Review (December 14 2016)
I participated in a cognitive technologies conference a few days after the November 8 election, and much of the talk at breaks (and some on the stage) was about the election results and the reasons behind them. There was a general feeling that the “Rust Belt” had been largely responsible for Donald Trump’s victory, and a growing, if belated, understanding of the economic plight of some citizens from that Midwestern region.
Some conference participants were concerned that this beleaguered region might grow. In fact, one attendee – an old friend who strategizes about technology for a big New York bank – commented that perhaps Wall Street would become “the new Rust Belt”. His concern was that automation of the finance industry would hollow out jobs in that field in the same way that robotics and other technologies have reduced manufacturing employment.
This is a sobering prospect, but there is plenty of evidence that it’s a real possibility. Key aspects of the finance industry have already been automated to a substantial degree. Jobs in the New York finance field have been declining for several years. According to data from research firm Coalition Ltd, more than 10,000 “front-office producer” jobs have been lost within the top ten banks since 2011. Coalition also suggests that global fixed-income headcount has fallen 31% since 2011.
Floor trader, of course, has long been the archetypal job on stock exchanges. But there are precious few left of them. As the Financial Times notes, in 2000 there were over 5,500 floor traders on the New York Stock Exchange; now there are fewer than 400. Many of the remaining traders work only part-time. Most trading jobs have been taken over by servers running trading algorithms.
Wall Street staffing reductions thus far have come largely from traditional automation technologies, but new cognitive technologies are likely to accelerate them. Much of Wall Street’s back office operations involve the performance of relatively structured tasks. Many of these could probably be taken over by tools like robotic process automation, which can reach into multiple systems for needed data and apply rule-based decision logic.
Regulatory compliance has been one of the few growth areas in recent years on Wall Street, but a variety of cognitive tools are attacking that area as well. Systems from Digital Reasoning are automating internal fraud investigations. IpSoft’s Amelia is focused in part on facilitating compliance in customer conversations. Narrative Science automates the creation of anti-money laundering reports. RAGE Frameworks automates the extraction of data for credit and wealth management, and can create automated compliance reports on the process. My guess is that the number of compliance employees has peaked and will begin to diminish.
Many entry-level jobs on Wall Street involve combing through data to make a case for a particular financial transaction. But those tasks are also beginning to be automated, too. Kensho, for example, is a startup that analyzes data on markets and generates reports on their implications. A Fortune article noted that
Kensho has the potential to replace the Street’s trove of market strategists, and its ability to crunch data and offer advice should make investment bankers nervous too.
Another common task of financial analysts and attorneys is to prepare disclosure data on a company’s financial history for potential investors. But another startup, iDisclose does that automatically.
Another common role in the finance industry is to provide investment advice. While the traditional Wall Street firms (those that still exist) haven’t yet embraced the “robo-advisor” concept for their high-end clients, automated advice is becoming pervasive at the lower end of investing. Vanguard, Charles Schwab, and Fidelity have all taken some steps in that direction, and startups like Betterment, Wealthfront, and Personal Capital are pursuing Millennial customers with money to invest. The capabilities already exist for higher-end versions of robo-advice, and a few banks like UBS have begun to explore them. Investment advice is complex, data-intensive, and rapidly changing, so it seems very likely that there will be substantially fewer human investment advisors in the future.
There are other finance-oriented tasks that will be performed by automation, including some new ones involving ongoing financial management for consumers that should have been done by banks long ago. The key point, however, is that these are tasks, not entire jobs. Wall Street won’t look like it had a mass exodus anytime soon. But job after job will be whittled away over time. Entry-level jobs will probably be the hardest hit; if you can teach a recent college graduate to do a task, you can probably teach a machine to do it.
If there is any good news here, it’s that there will be a substantial number of jobs that involve working alongside machines. If you’re already familiar with key financial processes, you’ll have a much better chance of keeping your job if you learn to work alongside smart machines that perform key aspects of those processes – monitoring the machines, fixing them, and picking up the ball when they drop it. Or you can become skilled at overseeing them, understanding when the financial world has changed and when the algorithms are no longer well-equipped to deal with it. Finally, of course, there will be many jobs involved in building intelligent finance systems. The fintech industry is one of the fastest-growing areas of technology, and much of its focus involves automated decisions and capabilities.
Who knows what political and economic upheavals will be caused by the continued decline in Wall Street and financial sector jobs? But the presidential election of 2016 might provide some clues. People who no longer see economic opportunity don’t always react immediately, but they do eventually react. Wall Street employees may express their dissatisfaction differently than the typical Rust Belt manufacturing worker. But they will surely not let their livelihoods vanish without protest.
Thomas H Davenport is the president’s distinguished professor in management and information technology at Babson College, and cofounder of the International Institute for Analytics. He also contributes to the MIT Initiative on the Digital Economy as a fellow, and as a senior advisor to Deloitte Analytics. Author of over a dozen management books, his latest is Only Humans Need Apply: Winners and Losers in the Age of Smart Machines (2016).
This article is about ANALYTICS
While regulatory compliance can be automated on Wall Street (and indeed, should be), the challenge of how to leverage that investment into something yielding competitive advantage still lies firmly in the human domain. This is one of many examples where, done correctly, automation should not reduce the workforce, rather promote it to focus on tasks of higher mental faculty. The challenge for Wall Street, and the finance industry in general, is to properly delineate between execution by automation, and innovation by (human) consideration. This requires re-skilling of large swathes of the workforce from ‘do-ers’ to ‘thinkers’. Some will adapt, some will not. However, this is evolution. And evolution is good overall.