The Future of Banking: AI, Automation & The Creation of Meaning-Filled Work

Brian Larson

Manager- Asset & Wealth Management Consulting | LGBTQ+ Speaker | Author

发布日期: 2020年7月9日

Introduction

The U.S. financial services industry is one of the first major industries to adopt and embrace the technological advances associated with the Fourth Industrial Revolution, a new wave of 21st-century technological innovation that includes advances in machine computing, artificial intelligence (AI), robotics and other digital technologies. Such advances include the ability to instantly trade stocks online, readily approve and on-board new credit card customers, and even automate certain parts of the trading process with little to no human intervention. Most of these recent advances are due, in large part, to the rise of artificial intelligence, big data (including machine learning), and process automation capabilities.[1] However, there are risks associated with automation in the financial services industry, including, for example, the loss of jobs in the United States. “Technological efficiencies” will lead to about 200,000 job cuts in the U.S. banking industry over the next decade, Wells Fargo reported in a recent study.[2] Can technologies associated with the Fourth Industrial Revolution, like artificial intelligence (AI) and process automation be harnessed to complement workers' skills while continuing to produce good jobs and value-filled careers? Additionally, can financial institutions and the U.S. government work together to mitigate some of the potential downsides of such technology? The answer to both questions is: yes. However, to compete in the age of AI and automation, the U.S. financial services industry will have to adapt with alacrity, scale, and efficiency.

The Past, Current, and Future State of Work in the Financial Services Industry

In order to understand where the financial services industry is heading, it is important to first examine the past and current state of work in the industry in the United States. For starters, the financial services industry has always been an early adopter of technology that helps transmit data (including stock prices, bond certificates, etc.): The first transatlantic cable, a novel innovation in 1886, carried cotton prices between Liverpool and New York.[3] In the mid to late 1970s, workers on Wall Street were beginning to use rudimentary email processes, putting them among the first to adopt personal computers outside of the sciences, academia, and home hobbyists.[4] In the 1980s, Wall Street analysts were early adopters of spreadsheet software. At the time, such groundbreaking programs included VisiCalc, the first-ever digital spreadsheet, along with Lotus 1-2-3 (an IBM company), which offered expanded capabilities in some areas and similarly boosted IBM’s personal computer business.[5] According to David Wolfe, co-director of the Innovation Policy Lab at the University of Toronto’s Munk School of Global Affairs and Public Policy, “The spreadsheet immediately started getting picked up by the financial services industry for its ability to do ‘what if’ calculations, like: If the rate changes from 1% to 2%, how will it affect my investment capital?”[6] This kind of computing power allowed financial institutions to engage in scenario analysis planning that would have taken weeks or months in just a couple of hours. VisiCalc and Lotus 1-2-3 both attracted large institutional customer bases, and, with the advent of the personal computer, both seemed poised for prolonged success.

By the early 1990s, however, VisiCalc and Lotus 1-2-3 had ceded the throne to another piece of software, which came packaged with a whole “office suite:” Microsoft Excel. At the time of release, Microsoft Excel wasn’t the best or most innovative spreadsheet software, according to most tech historians.[7] Microsoft was able to edge out competitors due to its “all in one” software suite of tools that would later include Microsoft Word, and PowerPoint among a myriad of other “open and go” applications. Over the next three decades, Excel would become ubiquitous, from Wall Street to small businesses. It’s important to note that during this early period, although the cost of personal computers was rather high, especially in purchasing power parity of today, financial service firms like JPMorgan and Wells Fargo were easily able to make such large investments, especially as it became viewed as a competitive edge in an industry where information dissemination and velocity translated into profits for the firm and its clients.[8] During this period, financial service firms traded their accounting calculators for Excel spreadsheets and left the calculations to the “machine” to do.

Throughout the 1980s and early 1990s, large stacks of Excel printouts were the “go-to item” on Wall Street during meetings, according to Will Deringer, assistant professor of science, technology, and society at MIT, and a former investment banking analyst at the Blackstone group.[9] “Spreadsheets gave this impression of objectivity, a certain accuracy because they looked complicated: there are lots of digits, you can put all these ideas and data points together and create these elaborate structures and produce what seem to be incredibly precise answers.” However, many of these spreadsheets, much like the AI-generated outputs of today, were “quite vulnerable to small errors, and very brittle, with no mechanism for telling you if they’ve gone off the rails.” [10] It took many financial service firms a long time to rectify the mindset that “the machines” could provide the right investment answers all of the time. Many financial service firms during this period began to re-think the need for human judgment in the process of calculating investments and returns and the need to “drill down” into the individual “cells” that made up a spreadsheet.

In the 1990s Wall Street started using new, readily available technology to create new, more complex kinds of trading and investments. One of these new trading “inventions” was the creation of never-before-seen “bundled” derivatives. This kind of transaction, which dates back thousands of years in its earliest forms, “involves an agreed-upon value for certain resources between two parties over time, and is often used in the hopes of stabilizing markets or (perhaps more commonly) raking in percentages if the resources’ real-world value goes up.”[11] Mortgage derivatives, future derivatives, etc. began to be bundled together and assigned their own “derived” value. Although technophiles love to debate the topic, it is commonly thought that the intersection between personal computing and spreadsheets occurred with the invention of these new derivative bundles. For the better part of nearly two decades since their invention, derivatives, like the aforementioned, promulgated and largely went unchecked or regulated by both financial service firms and regulatory authorities.

The 1990s also saw the advent of the “information superhighway.” The internet forced a fundamental rethinking of information technology and communications architectures.[12] Bill Gates famously said in a 1994 speech at a Bank Administration Institute conference that “Banks are dinosaurs, they can be bypassed.” Because Microsoft was leading the charge with its Excel offering, many technophiles questioned whether or not Microsoft would enter the banking arena and disrupt one of America’s biggest industries. A 1994 Bloomberg headline read “Bill Gates Is Rattling the Teller’s Window,” but in reality, Microsoft had no plans to enter banking.[13] The internet would force banks and financial institutions to build digital capabilities in-house to keep up with new and growing customer demands (the automated teller machine, ATM, was birthed at this time as well).

In 2010, journalist Michael Lewis made waves with his nonfiction book “The Big Short: Inside the Doomsday Machine,” which gave Americans a closer look at the 2007-2008 financial crisis, including the sub-prime mortgage debacle that sparked the fire. Among other things, Lewis’ book details some of the trading behaviors and tools that drove this economic catastrophe: “By shifting much of the actual number work onto techno-tools financial firms have widened the gap between decision-makers and the real-world impacts their work achieves.”[14] It’s clear from Lewis’ findings that many of the top-level brass within financial firms simply did not understand how the tools used by employees under them worked. Put another way, many of these CEOs and CFOs understood the potential upside to leveraging new technological tools, but failed to take into account or, to borrow baking terminology, hedge-against, potential downside risk. I will explore in further sections the advent of “automated desk decision making” and, in particular, the phenomena of “flash crashes” within the United States stock market.

Over the past few years, automating internal processes through the use of artificial intelligence remains rather a “black box” at financial service firms. Prediction machines (PMs) do a fantastic job of serving up “next best actions” or recommending potential up-sell and cross-sell offers to banks’ clients.[15] However, the algorithms used to “predict” which ones to present to clients are opaque. This raises similar questions to the ones banking CEOs faced during the advent of Excel spreadsheets in the 1970s and 1980s. If top-level management at firms does not have access to the rationale or decision-making process of a particular algorithmic output, should red flags be raised? The answer is simple: yes. CEOs of major financial institutions, such as JPMorgan Chase’s Jamie Dimon, are trying to “crack the code” that fuels artificial intelligence in hopes of reaching new clients and make banking processes more transparent and equitable.[16]

The days of men (and some women) running across the trading floor with leaflets of orders to buy or sell are largely a figment of the past. The sounds that fill a trading room floor today are the purrs of data servers, not the thunderous stampeding of rowdy traders.[17] In banks across America today, teller positions have dwindled or been eliminated in favor of ATMs (automated teller machine) and other robotic technology requiring little to no human intervention. Scenes like “runs on the bank” featured in classic movies like “It’s a Wonderful Life,” and the mass shredding of paper Excel outputs like that in “Wolf of Wall Street” will serve as bonafide examples of financial institution activities pre-advent of the internet and other such innovations associated with the Third Industrial Revolution.

Job Cuts Associated with the Current State of Work in the Financial Services Industry

According to a 2019 Wells Fargo & Co. report, “Technological efficiencies will result in the biggest reduction in headcount across the U.S. banking industry in its history, with an estimated 200,000 job cuts over the next decade.”[18]. Most of these job cuts will be directly correlated to the amount of technological investment a financial institution makes. As it currently stands, the financial services industry spends roughly $150 billion every year implementing and updating new technology.[19] This kind of spending “will lead to lower costs, with employee compensation accounting for half of all bank expenses” said Mike Mayo, a senior analyst at Wells Fargo Securities LLC. Additionally, a 2019 study by Alex Pierron of Optimas, a research consultancy, predicts “66,000 new technology and data science jobs will be created by 2030 in roles such as computer programming.”[20] Opimas also found that between January and May of 2019, about 35 percent of jobs posted by 19 European and US banks and asset managers were for technology and data roles. While AI and automation will lead to job cuts, financial services firms can still focus efforts on re-skilling or up-skilling their current workforce in order to leverage the valuable human capital they already have in-house. These efforts to re-skill workforces are explored in the following sections.

To put the above-mentioned investment spend into perspective, and as it stands today, “Funds run by computers that follow rules set by humans account for 35% of America’s stock market, 60% of institutional equity assets and 60% of trading activity.”[21] Additionally, due to advances in machine learning (the ability for a machine to analyze data, detect patterns and make inferences with little to no human intervention), new artificial intelligence programs are also writing their own investing rules. The human counterparts of such programs have little insight into the decision-making process of the algorithms drawn upon to produce such recommendations because the algorithms are auto-generated based on a wide-swath of rules and processes. The concept of a financial advisor relying solely on “word of mouth” or daily stock market “check-ins” to produce portfolio investment strategies is no match for computer-generated algorithms that use big data to compile portfolios of greater risk diversity and return. The local neighborhood financial advisor/planner that will survive this revolution is the one who embraces data-backed decisions to evaluate portfolio risk and stock evaluation while simultaneously leaning in on his/her soft skills to generate new business and gain trusted-advisor status among clients.

For example, consider Ant Financial, a spin-off of China’s Alibaba Company, which is now the most valuable unicorn and the largest financial technology (fintech) firm in the world.[22] “At the core of Ant Financial’s success is its ability to leverage data to learn about users’ needs and respond with digital services to address them. The data is assembled into a powerful, integrated platform that uses AI to power such functions as application processing, fraud detection, credit scoring, and loan qualification.”[23] Ant Financial employs fewer than ten thousand people to serve more than 700 million customers. By comparison, Bank of America employs 209,000 people to serve 67 million customers with a much more limited scope of services and products.[24] Does this mean that Bank of America can suddenly or overnight shrink its workforce to the same fraction as Ant Financial? This is likely not the case for two reasons: First, many of Bank of America’s products including home mortgages, small business loans, etc. still require a lot of manual human input.[25] For example, to obtain a home mortgage today a customer would still need to interact with a loan officer and an onboarding customer service representative. Second, relationship banking is still a major staple of the United States financial services industry. For example, a small business owner who takes out a loan likely has built a relationship with a local banker or loan officer with whom he/she maintains contact with over the lifetime of the loan. It’s for these reasons that employment at superstar firms like Bank of America will likely not shrink to the proportion seen at firms like Ant Financial. However, superstar firms like Bank of America and Fidelity must move their business models to compete in the age of AI and become an AI-first company.

Rise of Challenger Banks or So-Called “Neo-Banks”

Challenger Banks or so-called “neo-banks” are the digitally-first players entering the financial field at the moment. These banks are typically customer-obsessed, application powered (few have physical storefronts), and usually represent a disruptive force to the more traditional financial institutions.[26] Most of the traditional banks have been in business for many, many decades (Citi- 207 years, Wells Fargo, 168- you get the picture). These ‘legacy institutions,’ as the name may imply, are not usually quick to change and, more often than not, have antiquated, siloed data systems that are both costly and cumbersome to run.

BankMobile, Chime, and SoFi all fit the mold of a challenger or neo-bank. These challenger banks all offer varying features, but some common features include fee-free ATM access, online bill pay, P2P mobile payments, little or no account minimums, and, no-fee checking/savings accounts.[27] With these offerings and lower barriers to access, these neo-banks can ‘bank’ (a term that means providing banking services or funding/capital) to the under and non-banked portions of society. These digitally-first banks have even been appealing to high-wealth clients who view the ease of doing business online or through their phone as reason enough to open a new account and transfer funds.

Large financial institutions that have big digital banking footprints and understand their digital-first clients will have a leg up on their competitors in the months and years ahead. With near-zero and below zero interest rates, banks and other financial institutions will all be clamoring to diversify their asset mix and many, if not all, will feel the pressure to boost their mobile and digital banking experiences in order to have a leg up on their competitors and attract more clients. Many banks and other financial institutions have already begun ramping up their customer service, application features, and, all-around ease of doing banking. As banks feel the squeeze to attract low-cost capital reserves, consumers will be the victors. Consumers today are just a couple taps on their iPhone away from trading stock, managing their retirement portfolio, and even refinancing their home; Why shouldn’t they expect a banking experience that is both inexpensive and user-friendly?

A “Second Wave of Automation and AI” is Approaching

Financial institutions in the U.S. and abroad are using technologies associated with the Fourth Industrial Revolution to deliver superior customer service to their clients while minimizing the fully loaded cost of a full-time employee. McKinsey sees “the second wave of automation and artificial intelligence emerging” in the next few years, in which machines will do “up to 10 to 25 percent of work across bank functions, increasing capacity and freeing employees to focus on higher-value tasks and projects.”[28] To capture this opportunity, banks must take a strategic approach and remove cross-department silos and encourage the dissemination of client data across departments (while, of course, keeping within the lines of federally mandated laws and regulations around data usage). In some cases, they will need to design new processes that are optimized for automated/artificial intelligence work, rather than for people. They will also need to couple together specialized domain expertise from vendors with in-house capabilities to automate and bolt in a new way of working. R. Martin Chavez, an architect of Goldman Sachs Group Inc.’s effort to transform itself with technology, said that “all traders will soon need coding skills to succeed on Wall Street.”[29] It is clear that the skill sets of employees in the financial services industry will need to be re-tooled to meet the changing demands of the employers and customers alike.

Leading the Charge: Fidelity and the Case for Re-Skilling

Accenture’s 2018 Future Workforce Survey (of 100 banking CEOs and 1,300 bank employees) found, on average, only 1 in 4 senior bank executives is ready to work with AI. And while most cite the growing skills gap as the number-one factor influencing their workforce strategy, only 3 percent plan to significantly increase their investment in reskilling programs in the next three years.[30] Abby Johnson, Fidelity Chairman and CEO, has been leading her firm’s charge to become an “AI-first” firm.[31] Abby along with Vipin Mayar, an executive vice president of Fidelity’s Data, Insights, and Analytics group, made this initiative starting in 2011 one of the firm’s top priorities.[32] At the beginning of Fidelity’s journey, the firm “hired top data scientists and expanded its recruitment efforts to attract talent that is drawn to technology companies or Silicon Valley.” Mayar commented, “Our use cases, culture, and data were a big draw for this talent, and we have now built a world-class team here. It helped that this was a top priority for Abby.”[33] While statistics around the number of data scientists hired is not publicly available, it is clear that Fidelity was moving in the direction of adopting a more agile, fin-tech backbone in an effort to harness the power of its data and meet growing customer demand for more digital product and service offerings.

In an effort to not lay-off some of its workforce back in 2011, Fidelity chose to encourage the emergence of new types of digital skills within the firm through re-skilling and up-skilling efforts including micro-credential programs and learning and development initiatives.[34] One of these skills included data and AI-focused product management, “wherein experts looked across functions with a keen eye for the business impact of analytics and led the agile teams in the identification and deployment of new applications.”[35] By the end of 2012, Fidelity had devised an integrated data strategy and “provided software developers and data scientists with the tools to build, train, and deploy machine-learning models quickly.” As Marco Iansiti and Karim Lakhani note in their book, “Competing in the Age of AI,” perhaps even more important than the technologies advances during this period were “the organizational and cultural shifts toward the adoption of agile methods in order to offer a company of Fidelity’s size the agility and decision-making speed of a smaller one.”

Interestingly, Abby and Vipin also recognized the need to get senior business leaders on board with deploying AI and machine learning applications within their respective departments. To meet this need, “a comprehensive education effort, with hundreds of business leaders was established and is still running, stronger than ever, to this day.”[36] These hundreds of business leaders not only attended classes to drive these capabilities, but they also gained a better understanding of how other leaders within the different lines of business were leveraging new technology more broadly and deeply across the firm. Additionally, the firm has also offer micro-credential programs, free of charge, to every employee at Fidelity. These micro-credential programs, typically 3-4 months in length per module, allow employees to gain valuable, potable skills that they can then deploy within Fidelity and even potentially leverage to receive pay or benefit increases.[37] Some of the programs commonly completed by employees at Fidelity include cloud computing, machine learning, and programming modules. By allowing employees the ability access to such re-skilling and up-skilling programs empower workers to gain valuable skills for their career journey at Fidelity and beyond!

While roles within Fidelity have certainly changed and evolved from 2012, Abby’s initiative to re-skill or up-skill workers to meet the demands of the Fourth Industrial Revolution proved vital in Fidelity’s long-term success and financial health. Fidelity teams are now “driving its increasing data- and AI-centric operating models across its multiple lines of business, working hard to enable a huge range of processes from portfolio analysis to customer service.”[38] Vipar did, however, note, “While Fidelity will never entirely lose its human touch, its investment advisers remain a vital part of the business, AI is playing more of a role in improving the company’s performance and delivering exceptional customer experience.” Ironically, Vipar also notes that Fidelity is seeing a dramatic increase in jobs requiring cross-functional policy governance, cybersecurity, and privacy.[39] Thus, while some jobs will undoubtedly be lost to AI and automation, firms that can model Fidelity’s example will be in the best position to leverage its current workforce to meet demands of the Fourth Industrial Revolution while fully utilizing and empowering its human capital.

A Macro View of Employment in the Financial Services Industry

It is important to note that banking employees are not necessarily those traders on Wall Street or the familiar-faced bank teller; rather, a macro picture of financial services employment includes employees located in the thousands of call centers around the United States and the corporate employees located in the hundreds of bank headquarters spread across the nation. These kinds of back and front office employees are expected to be slashed “by about a fifth to a third over the next few years.”[40] However, jobs related to technology, sales, advising, and consulting will be less affected, according to the Wells Fargo study. It would appear that jobs less affected by changes in technology within these fields tend to rely more heavily on “soft skills” like communication, leadership, problem-solving, and negotiation. After all, many financial institutions still pride themselves on face to face financial advising and putting a human face to their investment recommendations (even if the actual recommendations are done by “the bots” in the back-office).

Part of the estimated 200,000 job cuts in the financial services industry includes those call center (“back office”) employees who are commonly associated with the customer service operation of banks and other financial institutions. These jobs are ripe for the insertion of automation and augmentation by artificial intelligence. According to Michael Tang, a Deloitte partner who leads the firm’s global financial services innovation practice, “We’re already seeing signs of it with chatbots, and some people don’t even know that they’re chatting with an A.I. engine because they’re just answering questions.”[41] The kind of back-office support will feel the effects of automation the most initially as financial institutions consolidate their human capital and invest in technological advances that automate and streamline sometimes mundane and routine processes.

Removing the Mundane/Routine and Creating Meaning-Filled Jobs

The routineness of activities like answering customer inquiries that involve direct, non-obfuscated answers can allow call center employees to focus on delivering value to the customer on the other end of the phone, chatbot, or even teller window. Employees can now focus on customer experience and have control over more value-added activities like up-selling and cross-selling.[42] For example, when a bank customer calls to report a missing or stolen credit card, the back-end operations of the bank can automate canceling the credit card and reissuing a new one, allowing the customer service representative (CSR) to not waste time opening a myriad of applications to complete the same task(s).

Keeping with the above example, as an added benefit to the customer/client, for example, the customer service representative may notice (or receive an alert on its desktop) that the client has been looking at obtaining a new home mortgage and can chat with him/her about the process and requirements. On the flip side, the customer service representative may notice that the client is likely to default on a current line of credit and can offer strategies and potentially provide more lenient terms for the client to make payment and remain in good standing with the financial institution.[43] The point of these examples is to illustrate that technological advances within the financial services industry can automate certain parts of a particular job, freeing a call center employee or a bank teller to provide higher value-added interactions with a client. These examples also highlight that besides removing the monotony associated with routine tasks and processes, technology can contribute to overall employee morale and create more satisfying long-term careers by re-focusing the job(s) on creating meaning-filled, customer-centric value.

The Role of the U.S. Government in Stimulating Job Growth in the Financial Services Industry

The pace of technological innovation within the financial services industry has left regulators and lawmakers scrambling to better understand the technology and implement laws to govern the usage of such technology to prevent its improper use. Traditional regulatory change in the United States tends to be extremely slow and bogged down in a system of bureaucratic red tape. Thomas Kochan, a professor of work and employment relations at the Sloan School of Business at the Massachusetts Institute of Technology (MIT), notes, “Most of our laws and the regulations and procedures used to enforce them still reflect the earlier era [the New Deal Era of the 1930s]. The task of updating our policies, business strategies, and workplace practices to suit our knowledge-driven economy and the diverse labor force is huge, essential, and long overdue.”[44] Professor Kochan is correct in his diagnosis of the woefully inadequate system of regulation in place now in the United States. Policies that may have seemed progressive in the 1930s are far too antiquated today and will need complete revamping to make sure such technological innovation is used for good.

A recent article in Forbes entitled, “A Regulation Revolution In Financial Services” notes if regulators hold still today, they’re actually “accelerating backward.”[45] The Treasury Department, to not accelerate backward, issued a fintech report in 2018 that calls on the federal financial agencies both to innovate and to coordinate.[46] This has given rise to regulatory technology (“regtech,” for short). Regtech relies on interagency collaboration and private sector innovation to increase the speed at which federal and state laws can help detect and prevent abuses in the financial services industry. Regtech has also given rise to a slew of new jobs including those in the public (cyber public policymakers, for example) and private sectors (fintech jobs like engineers, programmers, and other private watchdog groups).

The United Nations reports that less than 1 percent of global financial crime is apprehended because of technology which is out of date and unscalable.[47] Traditionally, financial crimes have “data typologies,” distinctive patterns that become easy to spot and can be consolidated and analyzed quickly. Today’s machine learning (ML) tools can find such patterns, while new encryption techniques can make it safe to share data much more widely while safeguarding privacy. Anti-money laundering, or AML, is one of the most advanced regulatory technology (“regtech”) use cases. Technology like regtech can also fix the AML “Know-Your-Customer” (KYC) rules, which currently block millions of innocent people from financial access because they lack traditional identity documents (government-issued identification, etc.). A crude example behind the idea of KYC is to prevent a terrorist from obtaining a credit card. New digital identity techniques can screen nearly everyone cheaply, accurately, and prevent those with malintent from obtaining precious financial resources. The same technology can be used to help with issues regarding “redlining” or the systemic discrimination of someone because they live in an area deemed to be a poor financial risk.[48] Banks can begin to extend credit to those underrepresented areas and take into account different considerations (like on-time utility payments, savings rates, etc.) when deciding on creditworthiness and develop new terms for financial inclusion. Clearly, there are huge societal benefits associated with the adoption of “regtech.”

Potential Risks to Society At-Large (Flash Crashes & Wealth Concentration)

As previously mentioned, physical “runs on the bank” (people fearing an economic collapse “running” to their neighborhood banks in the hopes of withdrawing their funds in cash) are likely an event of the past. However, a digital version of a “run on the bank” is raising eyebrows across the financial services industry. This kind of digital “run on the bank” is called a “flash crash.” A flash crash is defined as “an event in electronic securities markets wherein the withdrawal of stock orders rapidly amplifies as price declines. The result appears to be a rapid sell-off of securities that can happen over a few minutes, resulting in dramatic stock price declines.”[49]

A flash crash, like the one that occurred on May 6, 2010, is exacerbated as computer trading programs, not human traders, react to changing conditions in the market.[50] These programs recognize heavy selling in one or many securities and automatically begin selling large volumes at an incredibly rapid pace to avoid losses.[51] The results are utter depletion of entire stocks and a re-distribution of “risk” across different stocks and portfolios. Because many of the algorithms that trigger such rapid selling don’t have call-back options, many of the sales are permanent and unable to be recalled or canceled. Recently, major stock exchanges like the NYSE, have started to implement circuit breakers that halt trading until buy and sell orders can be matched up evenly. The idea behind financial circuit breakers is much like a circuit breaker one may find in his/her own home. The use of a circuit breaker is intended to stop a “surge” of power from short-circuiting and potentially leading to combustion (in the example of a flash crash, this would be epitomized in the “burning up” of a stock”). The risks associated with simply automating buy and sell orders come with the potential to bring mayhem to the global order of financial service institutions and exchanges.

Artificial intelligence and automation also can exacerbate a growing divide in the US between the rich and the poor. In a report titled, “Artificial Intelligence and Its Implications for Income Distribution and Unemployment,” the National Bureau of Economic Research concluded that artificial intelligence will increase the disparity of wealth between the rich and poor, otherwise known as the income inequality gap.[52] Economists in charge of the study, Anton Korinek of Johns Hopkins University and Joseph E. Stiglitz of Columbia University believe economic inequality “is one of the main challenges posed by the proliferation of artificial intelligence and other forms of worker-replacing technological progress.”[53] A report from Credit Suisse found the richest one percent own half of the world’s wealth.[54] “In the United States, for example, the expected life spans of the poor and the wealthy have diverged significantly in recent decades, in part because of unequal access to healthcare and ever more costly new technologies that are only available to those who can pay,” Credit Suisse states. The idea is that the Fourth Industrial Revolution will create “winners and losers” and the “losers” will be those who are already on the lower end of the wealth-income gap and don’t necessarily have the skills (soft and hard) to compete with those who have the skills (and the wealth, for that matter).[55] This compounding effect of sorts that is created with artificial intelligence and automation will contribute to the income inequality gap and, for society-at-large, creates monumental issues around public welfare.

Korinek and Stiglitz believe we’re heading on a similar path to the one that led to one of the worst periods in American history: The Great Depression. During this period, rapid technological advances in the agricultural sector meant that fewer workers were needed to tend to crops on farms across America (back then the U.S. economy was more dependent on agricultural output). “There are clear parallels to the situation today in that a significant fraction of the workforce may not have the skills required to succeed in the age of artificial intelligence,” they wrote.[56] Research conducted by Deloitte Consulting on the Future of Work found that artificial intelligence will create more jobs overall, but a lack of relevant skills will mean a vast majority of the workforce are unprepared to fill them.[57] This then begs the question: is universal basic income (UBI) the answer to helping the “losers” of the Fourth Industrial Revolution or is the answer in providing subsidies to those workers who are actively engaged in reskilling opportunities that are judged to meet the skillset required of jobs working in tandem with artificial intelligence? The old verbiage, “Give a man a fish, and you feed him for a day. Teach a man to fish, and you feed him for a lifetime,” may be applicable here.[58]

Another interesting topic of research relating to the income inequality gap is the possibility of the wealthy themselves “being augmented using artificial intelligence to provide them with an economic advantage.”[59] This kind of “economic advantage” is related to the compounding effect mentioned above, and still rather under-researched today due to the infancy of many of the mentioned innovations associated with the Fourth Industrial Revolution. “If intelligence becomes a matter of ability‐to‐pay, it is conceivable that the wealthiest (enhanced) humans will become orders of magnitude more productive – ‘more intelligent’ – than the unenhanced, leaving the majority of the population further and further behind,” Korinek and Stiglitz write.[60] While presented as a dystopian view, there’s reason to consider this outcome given the growing wealth divide. Research makes it clear that government action is needed to make sure that such tools and working knowledge of them are made available to be accessed by both the rich and the poor as to not further exacerbate and compound the current income inequality gap.

Good News and the Creation of AI-Assisted Jobs

Finance industry job listings requiring skills in artificial intelligence, machine learning, and data science increased by almost 60% over the last year.[61] This means that jobs are not necessarily “disappearing” within the financial services industry; rather, they are requiring workers to learn new skills (commonly referred to as learning and development) to augment their current work capabilities. According to David Autor, a professor of economics at the Massachusetts Institute of Technology, “There is no fundamental economic law that guarantees every adult will be able to earn a living solely based on sound mind and good character.”[62] Autor furthers, “The issue is not that middle-class workers are doomed by automation and technology, but instead that human capital investment must be at the heart of any long-term strategy for producing skills that are complemented by rather than substituted for by technological change.” Indeed, this is what we see in the financial services industry in the United States: new jobs that require a mixture of both old and new skills.

According to Glassdoor data, “Some of the most common job openings in artificial intelligence and finance are for machine learning engineers and data engineers, among other highly specialized software engineering roles.”[63] New job openings for workers who can help navigate the artificial intelligence landscape, including consultants and researchers are also on the rise. Not all new job functions are rooted in computer science or engineering, however. For example, chatbot copywriters (those who write conversational answers to technical questions customers ask on websites’ “chat” functions), product strategists, and technical sales representatives are also in demand. Those who have a business or communications background may be better suited to these roles.[64] Those workers who are willing to learn more about artificial intelligence (and work in what Deloitte has dubbed a “superjob,” or a job that combines a traditional job with the power of smart machines, data, and algorithms”) will likely have a “leg up” on their peers.

Final Conclusions

The U.S. financial services industry has a long track record of embracing new technology, from the early advent of Excel spreadsheets to the current revolution in artificial intelligence and automation. However, as the Fourth Industrial Revolution kicks into high gear, there are inevitably risks associated with automation in the financial services industry, including the loss of jobs in the United States. As explored, artificial intelligence and automation in the financial services industry will likely remove mundane and routine tasks and create more meaning-filled jobs and careers across the industry. Yes, some jobs will be lost, but many new jobs will be created, many of which will likely involve a healthy dose of soft skills like creativity, communication, and problem-solving. Artificial intelligence and automation in the U.S. financial services industry promise tremendous benefits, as seen in the rise of so-called Neo-banks, and even might help solve some pressing societal problems like mass shootings. Technological advances associated with the Fourth Industrial Revolution in the financial services industry in the United States promise tremendous benefits (both within and outside the scope of the industry) and it’s up to us to harness their power correctly for the sake of creating more meaning-filled work and further push the bounds of human (and robot) ingenuity.

[1] https://www2.deloitte.com/us/en/pages/consulting/articles/transforming-financial-services-with-robotics-and-cognitive-automation.html

[2] https://www.bloomberg.com/news/articles/2019-10-02/robots-to-cut-200-000-u-s-bank-jobs-in-next-decade-study-says

[3] https://www.economist.com/leaders/2019/10/03/the-rise-of-the-financial-machines?utm_source=newsletter&utm_medium=email&utm_campaign=newsletter_axiosam&stream=top&utm_source=morning_brew

[4] https://www.princeton.edu/~ota/disk3/1984/8411/841104.PDF

[5] https://www.princeton.edu/~ota/disk3/1984/8411/841104.PDF

[6] Neil Bradford, “Learning through Governance: Innovation in the Knowledge Economy,” The Elgar Companion to Innovation and Knowledge Creation, Editors: Harald Bathelt, Patrick Cohendet, Sebastian Henn and Laurent Simon. (Edward Elgar: Cheltenham, Northampton (MA), 2017). pp. 723-38.

[7] https://www.exceltrick.com/others/history-of-excel/

[8] https://www.computerworld.com/article/2578842/banking-on-technology.html

[9] https://gizmodo.com/how-the-invention-of-spreadsheet-software-unleashed-wal-1837177232

[10] https://gizmodo.com/how-the-invention-of-spreadsheet-software-unleashed-wal-1837177232

[11] https://www.investopedia.com/ask/answers/12/derivative.asp

[12] https://www.institutionalinvestor.com/article/b16wr87f6tn2n6/what-the-1980s-can-teach-us-about-wall-street%E2%80%99s-survival

[13] https://www.bloomberg.com/news/articles/1994-10-30/bill-gates-is-rattling-the-tellers-window

[14] Lewis, Michael. The Big Short: Inside the Doomsday Machine (Introduction) (WW Norton, 2010)

[15] Agarwal, Ajay. Prediction Machines: The Simple Economics of Artificial Intelligence: Chapter 2 & 7 (HBR, 2018)

[16] https://www.thestreet.com/investing/stocks/jpmorgan-ceo-jamie-dimon-us-companies-should-do-more-to-address-social-issue-14916936

[17] https://www.mckinsey.com/industries/financial-services/our-insights/the-transformative-power-of-automation-in-banking

[18] https://www.bloomberg.com/news/articles/2019-10-02/robots-to-cut-200-000-u-s-bank-jobs-in-next-decade-study-says

[19] https://www.pwc.com/gx/en/financial-services/assets/pdf/technology2020-and-beyond.pdf

[20] https://www.opimas.com/research/472/detail/

[21] https://www.economist.com/leaders/2019/10/03/the-rise-of-the-financial-machines?utm_source=newsletter&utm_medium=email&utm_campaign=newsletter_axiosam&stream=top&utm_source=morning_brew

[22] https://hackernoon.com/the-story-of-ant-financial-4t2aq3zh8

[23] https://www.fastcompany.com/company/ant-financial

[24] https://www.macrotrends.net/stocks/charts/BAC/bank-of-america/number-of-employees

[25] https://about.bankofamerica.com/en-us/who-we-are/our-businesses.html

[26] https://medium.com/crowdfundup/what-is-a-neo-bank-and-how-are-they-disrupting-traditional-banking-models-3c1b2fa5b8e1

[27] https://thefinancialbrand.com/80002/challenger-banks-legacy-institutions-mobile-banking/

[28] https://www.mckinsey.com/featured-insights/digital-disruption/whats-now-and-next-in-analytics-ai-and-automation

[29] https://www.economist.com/leaders/2019/10/03/the-rise-of-the-financial-machines?utm_source=newsletter&utm_medium=email&utm_campaign=newsletter_axiosam&stream=top&utm_source=morning_brew

[30] https://www.accenture.com/us-en/insights/banking/future-workforce-banking-survey

[31] https://www.cnbc.com/2018/09/28/fidelity-the-tech-company.html

[32] https://investorfieldguide.com/fidelity/

[33] Competing in the Age of AI. Harvard Business Review Press. Cambridge, MA. 2020

[34] https://www.fidelity.com/bin-public/060_www_fidelity_com/documents/brokerage/financial-times-fidelity-profile.pdf

[35] Competing in the Age of AI. Harvard Business Review Press. Cambridge, MA. 2020

[36] https://investorfieldguide.com/fidelity/

[37] https://jobs.fidelity.com/page/show/opportunities-career-paths

[38] https://investorfieldguide.com/fidelity/

[39] https://investorfieldguide.com/fidelity/

[40] https://www.mckinsey.com/industries/financial-services/our-insights/the-transformative-power-of-automation-in-banking

[41] https://www.bloomberg.com/news/articles/2019-10-02/robots-to-cut-200-000-u-s-bank-jobs-in-next-decade-study-says

[42] https://www.pega.com/industries/financial-services/onboarding

[43] These examples are taken from personal experience during my time working for a company called Pegasystems. Pegasystems is a business process management (BPM) platform that assists Fortune 500 companies, including financial institutions, redevelop their processes to more customer-centric and reduce the back-office work commonly associated with call center and customer service employment.

[44] Kochan, Thomas. Shaping the Future of Work: What Future Worker, Business, Government, and Education Leaders Need To Do For All To Prosper. 2015. MIT Sloan.

[45] https://www.forbes.com/sites/lawrencewintermeyer/2018/12/02/a-regulation-revolution-in-financial-services/#1aadaed34fb1

[46] https://home.treasury.gov/news/press-releases/sm447

[47] https://www.forbes.com/sites/lawrencewintermeyer/2018/12/02/a-regulation-revolution-in-financial-services/#1aadaed34fb1

[48] https://www.investopedia.com/terms/r/redlining.asp

[49] https://www.investopedia.com/terms/f/flash-crash.asp?utm_source=morning_brew

[50] https://www.theguardian.com/business/2015/apr/22/2010-flash-crash-new-york-stock-exchange-unfolded

[51] https://www.investopedia.com/articles/active-trading/101014/basics-algorithmic-trading-concepts-and-examples.asp

[52] https://www.nber.org/papers/w24174

[53] https://artificialintelligence-news.com/2018/01/11/ai-wealth-inequality/

[54] https://www.credit-suisse.com/about-us/en/reports-research/global-wealth-report.html

[55] https://artificialintelligence-news.com/2017/09/08/research-despite-popular-opinion-ai-creating-jobs/

[56] https://www.nber.org/papers/w24174