Four Things We Learned from Author, HBS Fellow, and Obama-Era SBA-Administrator Karen Mills About SMBs
Enigma gathered top female financial services executives from companies like Citizens Bank, Comerica Bank, and Amazon for an intimate round table discussion with author, HBS Senior Fellow, and Obama-Era SBA Administrator Karen Mills on the opportunities and challenges in the small-and-medium-sized business (SMB) economy. Despite the fact that “over half of the people who work in the US economy own or work for a small business,” Mills asserted that she “often had to pound the table in the West Wing to make sure the voice of small business was heard.”
We wanted to share four of the things we learned during the discussion from Mills about the challenges – and opportunities – in the SMB economy.
1. Why SMBs are under-appreciated and underserved
SMBs face a lending gap where they are less likely to receive traditional bank loans compared to larger businesses. Mills suggests the reason for this is twofold: information opacity surrounding SMBs and heterogeneity of SMBs. Information opacity, explained Mills, refers to the inability of lenders to access accurate and complete information on SMBs such as full-year revenue statements or operating status. Heterogeneity, meanwhile, is the idea that small businesses are diverse – in revenue, industry, and operating model – making it hard to have one lending model that fits their various needs.
Mills suggested that these two problems can be addressed when alternative data and technology intersect with small business lending for a solution she calls “The Small Business Utopia.” Under this Utopia, more financial institutions would have access to data like Enigma’s – near-real time revenues, event-based triggers, payment processing information, etc – that would form a more accurate picture of the financial health of the business borrower and lead to better and faster credit decisions.
“I've been very pleased to be connected to Enigma because they were early as a data aggregator,” said Mills. “People are using [this data] in various ways which turn out to be predictive.”
2. How the SMB landscape has changed since Mills first published her book in 2018
In 2018, Mills released the first version of her book, Fintech, Small Business & The American Dream: How Technology is Transforming Lending and Shaping a New Era of Small Business Opportunity. She is currently writing the second edition, reflecting on what will change the future of small business lending, including:
Technological advancements: AI, automation, and machine learning
Business health data: Revenues, fraud, technographics, etc. from providers like Enigma
Innovation entrepreneurs: Fintech innovation and disruption of traditional players
The new edition is scheduled for release in early 2024.
3. Which businesses are falling through the cracks and how can we reach them
According to Mills, the problems facing SMBs are most amplified for the smallest of small businesses who are early in their development and need technical assistance across loans, payments and more. “The very smallest and the hardest to reach entrepreneurs are the ones that are falling through the gaps,” said Mills.
Mills thought data and technology could be one solution for banks to make these sorts of loans more quickly. However, Mills warned data alone is not enough: local relationships and community banks in combination with tech and data will be key.
4. How automation is affecting which businesses are getting loans
Mills was particularly interested in how automation – where trained computers rather than humans make loans – could help or hurt small businesses in lending decisions. One concern Mills pointed out was training a computer on existing data could lead to an algorithm that reflects the biases in the current lending environment.
She referred to an important recent study looking at Paycheck Protection Program (PPP) loans – to which Enigma contributed data – spearheaded by NYU Professor Sabrina Howell. In the study, process automation was found to “reduce the racial disparities in credit access through enabling smaller loans, broadening banks’ geographic reach, and removing human biases from decision-making.” PPP loans, pointed out Mills, don’t have a credit screen and – when given out through automation rather than manually – distribution became more equitable.
Ultimately, Mills has mixed thoughts on automation: “I end up with these two sides of the equation. One possible scenario is that ‘AI will be bad and automation will be bad because it will be built on past biases.' But also ‘that automation can take away bias,' as we saw in the recent research on PPP," said Mills. “Those are the two conflicting scenarios out there.”