AI adopters encounter a more dangerous paradox

Improved productivity might be a red flag

A while back I wrote about how Jevons’ Paradox was going to surprise early adopters of AI by increasing the need to hire humans rather than make them obsolete. The Wall Street Journal recently joined the Jevons’ paradox bandwagon calling out companies using tools like Microsoft Copilot who are experiencing the paradoxical need for more human capital as a result of AI deployments.

Jevons’ paradox is a good news surprise for AI adopters – they won’t have the feared cultural backlash that comes with technology driven staff reductions. But now comes another paradox and this time the paradoxical surprise could be existential.

Sangeet Paul Choudary recently introduced Munger’s Paradox to the AI discussion. Choudary tells the story of when Charlie Munger, president of Berkshire Hathaway, and his partner Warren Buffet were presented with the opportunity to buy a new technology that would double their textile manufacturing productivity. Buffet said “I hope not. I will have to close my mills.” Munger explained the unexpected response with what is now known as the Munger Paradox:

“… the benefits of a huge productivity increase in the production of a commodity product will all go to the benefit of the buyers of the product, not the owner of the equipment. The technology commoditizes the producers’ position.”

If an enterprise is experiencing a large productivity increase in its core business, then that business is at risk because the profit center could move away from the producer. At first blush it doesn’t make sense, but of course that is why it’s a paradox. We have already seen numerous examples of technology-based productivity improvements that have not benefited the ones using the technology. Spotify struggles with profitability today even though it has mass, low-cost distribution through the internet because the music labels demand 70% of the revenue in return for access to their music. Netflix had to switch to creating content rather than just distributing it when it engaged streaming.

Fortunately, Choudary offers a set of simple questions to help AI adopters identify the risk an increase in productivity might pose to their business as well as help identify adjustments that might be made to the business to avoid losing the benefits. Let’s look at how early adopters of AI in equipment finance can analyze and defend against Munger’s Paradox.

“Are productivity gains delivered on your core production chain or in complementary functions?”

The answer in equipment finance is likely both so adaptation is both necessary and possible. Core functions of the business for which AI will drive productivity include reducing the risk to capital, underwriting (speed and efficiency), and matching borrowers to capital. Complementary functions that will benefit will be processing documentation and regulatory compliance – AI is very good at following rules. Financing overall is a complementary function to equipment OEMs so any benefit to financing from AI will benefit equipment manufacturers and their captives because they own the “content” of the deal.

“Can you change the economics in your favor by shifting profit pools?”

There are already signs of change in equipment finance to deal with the AI productivity paradox. Better underwriting combined with more sophisticated Ai-based risk management can improve margins – particularly for the more agile independents. Equipment finance as an industry can use AI to provide safer access to a riskier market to attract capital. This trend is already well underway as banks have pulled back from the market during higher interest rates. EF companies can use AI to hedge and engage risk in a way that serves the more diverse needs of private capital.

Tech-savvy lenders can also use AI to find more and better borrowers to drive scale over the fixed costs of the business. But borrowers are going to benefit from AI productivity in the form of increased competition for their business, i.e., lower prices, so better customer service before and after funding is going to become a key differentiator as the financing itself is further commoditized by AI.

“Can you leverage AI to create proprietary production advantages?”

Equipment finance is a challenging profession for differentiation. The core product, US dollars, is the ultimate commodity. Separating one funding offer from another comes down to difficult-to-defend positions like price, convenience/speed, terms or structure, and future relationship promises like renewal. But risk management, the heart of any financing business, is complex and rich in the subtleties of human behavior. Indeed, risk management in equipment finance is an effort in predicting and managing borrower behaviors and in this form is arguably the core-IP of the business.

But this is where the opportunity for AI as a part of proprietary methods thrives – predicting and managing human behavior to better outcomes. AI is already part of proprietary methods to find customers, match customers to products, and encourage customers to buy in the ecommerce world. AI is part of a new frontier of intense development in healthcare to both diagnose conditions and encourage better therapy compliance via data generated by the wealth of sensors now worn and carried by patients.

Managing risk from pre-origination throughout the term of a funded contract is an opportunity for proprietary AI advantage in equipment finance. Such developments are proprietary and defendable because much of the data upon which AI models are built is proprietary and the models themselves can be designed with methods and purposes that are not visible to outside competitors.

Indeed, there are those who worry about explaining AI credit models to regulators. AI based risk management and operational tools will be natural trade secrets. The answer to this question is strong for equipment finance.

New threats and opportunities

This leaves us to consider the consequences of the changes that AI productivity increases bring. Choudary assesses the impact of productivity in a given industry with the question “Who gains scale advantages?” A scale advantage is a reduction in cost or increase in profit that comes from using the new productivity at scale. We identified the expected changes above and the table below hypothesizes who gains advantage from each.

A paradox always creates a surprise for someone. Sometimes that surprise is an unexpected benefit, but sometimes the surprise is the end of the business. In the case of Munger’s Paradox in equipment finance, the “attentive and prepared” have the opportunity to parry the threats and thrive with new advantages of scale. Many will suffer the consequences of passive engagement, but others will “close some mills” and dramatically expand elsewhere in the profit chain.