How could advancements in machine learning, data analysis, and artificial intelligence impact the future of financial services – and their everyday customers?

That was the key question economists and financial experts explored at the Federal Reserve Bank of Boston’s 68th Economic Research Conference, which was held at the Bank on Friday and Saturday. The event was titled, “The Future of Finance: Implications of Innovation.”

Participants covered a range of topics tied to technological innovation, including its potential impacts on financial inclusion and small businesses, and what it might mean for bank regulation and monetary policy. The conference’s keynote address focused on the possible benefits and challenges of financial service providers using AI to maximize profitability.

Boston Fed President and CEO Susan M. Collins said in her welcoming remarks that, given the Federal Reserve’s roles in the U.S. monetary and financial systems, it’s “essential” that researchers explore both the opportunities and risks associated with financial innovation.

“Changes are occurring very rapidly. The technology discussed here will not stand still and, in all likelihood, will only grow in importance,” she said. “We want to explore the potential impacts on traditional and nontraditional financial intermediaries, and, more broadly, on the public’s access to financial services.”

Papers and presentations shared during the conference

Researchers presented four papers during the conference, and two panel sessions explored technology in bank regulation and cryptocurrency. The paper titles, links to all the papers and presentations, and video recordings of each panel session are listed below and available on bostonfed.org:

• “Fintech, Financial Inclusion, and The Future of Finance,” Emily Williams (Harvard Business School)
• “Regulation Meets Technology: Evolution of Small Business Lending in Underserved Areas since 2007,” Mattia Landoni (Federal Reserve Bank of Boston) and J. Christina Wang (Federal Reserve Bank of Boston)
• “FedNow and Faster Payments in the US,” Charles M. Kahn (University of Illinois)
• “Big Tech, Financial Intermediation and the Macroeconomy,” Fiorella De Fiore (Bank for International Settlements and the Centre for Economic Policy Research)
• “Decentralized Finance (DeFi), Blockchain/Distributed Ledger Technology (DLT), and Smart Contracts: Hope or Hype?” panel session with Lin William Cong (Cornell University), Hanna Halaburda (New York University), and Christine A. Parlour (University of California, Berkeley)
• “How Can Technology Make Regulation and Supervision Smarter and Regulators More Efficient?” panel session with Slavka Eley (European Labour Authority), Jermy Prenio (Bank for International Settlements), and Lawrence J. White (New York University)

How are “new” tech lenders impacting access to credit for small businesses?

In a session on their paper, Boston Fed senior economists Wang and Landoni discussed how small businesses who had existing relationships with banks were able to access Paycheck Protection Program, or PPP, loans more quickly during the COVID-19 pandemic.

Wang said fintech companies and merchant cash advance lenders – many of which provide immediate loans with high interest rates – became important lenders, especially in underserved counties, after banks generally reduced their lending from 2007 – 2019 following the global financial crisis.

But both banks and these newer credit sources scaled back on lending during the height of the pandemic in 2020 – 2021, Wang said. Then, when businesses looked to secure emergency loans through the Paycheck Protection Program, small businesses who had existing relationships with banks not subject to stress testing got loans earlier than those who used fintech companies, Landoni and Wang found.

A longer wait for credit can hurt small businesses depending on loans to pay employees, so technological advancement in lending needs to be better understood and government programs may need to better take new lenders into account, Wang said.

Keynote: What are the benefits and challenges of using AI to assess credit risks?

The conference’s keynote address was delivered by Adair Morse, a professor of finance at the University of California, Berkeley’s Haas School of Business.

In her presentation, “AI Innovation for Credit: Frontiers of Benefits & Red Flags,” Morse said lenders can use AI to maximize profitability in several ways, including targeted marketing, authenticating data, and creating more efficient “underwriting” models, which financial institutions use to assess credit risks.

But Morse said the use of AI in credit also poses several “red flags,” such as the potential for deception, price collusion, and discrimination. She said that if AI technologies are used to gather and analyze detailed data about a particular community, they could essentially become “local lenders,” replacing community banks. That could create problems for small businesses, which account for roughly half of all private employment in the U.S., she said.

Morse said small businesses often depend on long-term relationships with local banks to access loans, especially during economic downturns. Will AI lenders support a community’s small business needs, she asked, or will they prioritize profits?

“We’ve got to think holistically about AI,” she said. “We’ve got some red flags, and we should pay attention to those red flags … so that we can move safely into this next era.”

Learn more about the conference participants, read the papers, and watch the presentations on bostonfed.org.