AI: The End of the Inertia Premium
How AI agents are about to abolish the hidden subsidy behind global banking
Hi Fintech Futurists —
Today’s agenda is below.
AI: AI agents could change the fundamental economics of banking as consumers are more easily able to dynamically move money across accounts
Analysis: NASDAQ and NYSE enter the $1B+ Onchain Capital Markets Race
PODCAST: How Polygon Became the Payments Chain Moving $2.3T in Stablecoins, with CEO Marc Boiron
CURATED UPDATES: Machine Models, AI Applications in Finance & Investment Outlook
To support this writing and access our full archive of newsletters, analyses, and guides to building in Fintech & DeFi, subscribe below (if you haven’t yet).
We are especially excited about the AI primer from last week here.
Our Ecosystem: AI Venture Fund | AI Research | Lex Linkedin & Twitter | Sponsors
The end of the Inertia Premium
The traditional banking sector is primarily built around the concept of net interest margin (NIM). This is the difference between the interest Banks earn on loans and the interest they pay back to their depositors, relative to the size of their asset base.
Simply put, it measures how efficiently a Bank is managing loans relative to their funding costs. Successfully growing NIM is therefore both a function of optimising the lending book (efficiently pricing risk, duration matching, and pre-empting changes to market conditions) and keeping funding costs to depositors low.
The latter is particularly common in retail banking where everyday consumers are inefficient with their deposits. This is known as the “inertia premium”, a higher NIM resulting from consumers failure to compare rates, switch accounts, and redeem rewards points before they expire.
AI agents acting on behalf of consumers are now positioned to eliminate it.
The size of the Problem
The national rate (weighted-average on all insured deposits) that banks pay on interest-bearing checking accounts in the US was only 0.07%, and 0.38% on savings accounts, while top online savings rates exceeded 4%.
The gap is roughly a tenfold differential available to any consumer willing to open an online account.
Gap over time - national savings rate vs. top-of-market (2021-2026)
Non-interest-bearing demand-deposit account balances grew at a CAGR of 28% in the last five years, compared with only 3% for interest-bearing equivalents. Retail banks have been accumulating a funding base built on customers who are arguably underserved. We speculate that deposits have so far been sticky because of switching costs, FDIC insurance, or simply because consumers don’t know where to look.
Of the global total of $70 trillion in consumer deposits, $23 trillion sits in checking accounts with near-zero rates. In a scenario where agents help the deposit base reprice to 80% of top-of-market rates, what does NIM look like? All else equal, Mckinsey estimates the global banking profit pools of $1.2 trillion could shrink by as much as 10% over the next five to ten years, enough to bring average returns below the cost of capital.
The credit card business shows similar customer inertia:
Credit cards generated $234B in US revenue in 2024, through a blend of interest income from revolvers, interchange fees, annual and penalty fees, and unredeemed rewards. Much of this depends on consumer inertia.
Annual Total Fees Charged (United States)
More than one in five cardholders had not redeemed any rewards over 2024. Annual forfeiture incidence — the share of accrued rewards points lost through account closure or expiry — runs at between 3% and 5%.
75% of all US revolving credit card balances belong to prime or super-prime consumers who have access to lower-cost alternatives but consistently fail to act. Despite the relative ease of switching, humans are not acting.
Ant Financial offers a precedent
This disruption has already played out before, and Western banking analysts largely missed it at the time.
In 2013, Alibaba’s Ant Financial embedded a money market fund directly into the Alipay mobile wallet. The minimum investment was one yuan, approximately sixteen US cents. Access required a single tap inside an application hundreds of millions of consumers were already using for payments. The fund offered an annualised yield of 3.96% against Chinese banks’ ordinary demand deposit rate of 0.35%. The rate differential was roughly eleven times, comparable to the current gap between US national savings averages and top-of-market online rates. Alipay’s Yu’e Bao money market fund attracted 150 million clients and $93 billion in assets within 18 months of launch. By 2017, Yu’e Bao had become the world’s largest money market fund, having grown at a CAGR of 125% since its inception.
Tianhong Yu’e Bao money market fund assets under management
Customers flocked to Yu’e Bao because it was so easy to use, and they could get better interest rates than their bank accounts offered, and around a third or more of China’s population came to use it. Chinese banks found their margins under pressure as competition from Yu’e Bao intensified.
Consumers in China knew higher yields were available before 2013. Capital moved when accessing those yields required a single tap inside an application already on their phone. Yu’e Bao illustrates how fast the trend can grow when friction is removed entirely.
How agents could optimise customer balances
An agent responding to a simple instruction to maintain a checking account balance at $1,000, sweep excess cash to a high-yield savings account, and replenish from savings when the balance dips below a threshold. Before long, the agent will notice payment or investment opportunities via email, text, or app alert, act within pre-authorised parameters, and report back automatically.
On credit cards, AI agents can automatically direct spending to the best rewards card in real time, trigger new applications to secure a better deal, and roll balances to another card before promotional rates expire.
How current models misread the deposit base
Current NIM models using deposit assumptions from a pre-agent world are modelling the wrong market structure. US banking industry NIM in Q4 2025 was 3.39%, the highest level since 2019. Future growth is forecasted to come from a combination of both NIM and non-interest income. But agents threaten margins in both segment simultaneously. They could autonomously open new savings accounts and move money on consumers’ behalf to find the best rates, or optimise credit card lending balances and take advantage of zero balance-transfer offers.
Revenue growth forecasted to come from non interest income
What banks can do
One path is to own the agent layer. Bank of America’s virtual assistant Erica has surpassed 2.5 billion client interactions, handling requests and providing proactive insights for 20 million customers. However, this is far from the cross-institutional financial agents that could be built.
A second option involves competing on rates, accepting that agent-mediated comparisons are coming, and pricing deposits at or near the top of the market to ensure that products are seen by agents.
A third path involves competing on depth, concentrating investment on credit decisions informed by proprietary transaction data, business lending relationships, and advisory services for segments where the bank’s data advantage holds. Financial institutions own three things AI cannot easily replicate
Regulatory licences
Customer trust
Proprietary data
These are durable competitive advantages in segments where bank data informs materially better credit decisions. Whether they remain advantages in retail deposit gathering is the open question.
👑Related Coverage👑
Analysis: NASDAQ and NYSE enter the $1B+ Onchain Capital Markets Race (link here)
We discuss the evolution of onchain capital markets, where tokenized assets are led by fixed income and commodities, with equities just reaching $1B+ in value.
Here are two data points that stick out like a sore thumb.
These charts show traditional financial assets that are tokenized onchain. You can see that low-risk fixed income is the most popular asset, followed by commodities and asset-backed credit. Equities are just starting to sneak into the picture at around $1B in total value. We examine how institutions are responding, with NYSE partnering with Securitize and Nasdaq working with Kraken to build compliant tokenized equity infrastructure.
🎙️ Podcast: How Polygon Became the Payments Chain Moving $2.3T in Stablecoins, with CEO Marc Boiron (link here)
It’s our 200th podcast episode! A huge thank you to all that continue to enjoy these podcasts as much as we enjoy making them!
In this episode, Lex chats with Marc Boiron — CEO of Polygon Labs. Marc shares his journey from law to blockchain, discussing the challenges of navigating crypto’s evolving legal landscape and the complexities of structuring compliant DeFi projects. He explains Polygon’s strategic pivot to focus on stablecoin payments, leveraging its proven blockchain and global partnerships.
Marc highlights Polygon’s real-world adoption, competitive edge, and vision to become the leading platform for on-chain payments. The episode offers insights into regulatory hurdles, industry trends, and Polygon’s mission to transform digital money movement.
Curated Updates
Here are the rest of the updates hitting our radar.
Machine Models
Ethical and Bias Considerations in Artificial Intelligence/Machine Learning - Matthew G. Hanna & Liron Pantanowitz & Brian Jackson & Octavia Palmer & Shyam Visweswaran & Joshua Pantanowitz & Mustafa Deebajah & Hooman H. Rashidi
A Critical Field Guide for Working with Machine Learning Datasets - Sarah Ciston & Mike Ananny & Kate Crawford
AI Applications in Finance
⭐ AI-Driven Payment Systems: From Innovation To Market Success - Merve Ozkurt Bas
The Rise Of Generative Ai Agents In Finance: Operational Disruption And Strategic Evolution - Inesh Hettiarachchi
Financial Modeling in Corporate Strategy: A Review of AI Applications For Investment Optimization - Olufunmilayo Ogunwole & Ekene Cynthia Onukwulu & Micah Oghale Joel & Ejuma Martha Adaga & Augustine Ifeanyi Ibeh
Investment Outlook
⭐ Private Equity Outlook 2025: Is a Recovery Starting to Take Shape? - Bain & Company
⭐ Global Venture Capital Outlook: The Latest Trends - Bain & Company
⭐ Global Private Markets Report 2025: Braced for shifting weather - McKinsey & Company
🚀 Postscript
Sponsor the Fintech Blueprint and reach over 200,000 professionals.
👉 Reach out here.Check out our new AI products newsletter, Future Blueprint. (Don’t tell anyone)
Read our Disclaimer here — this newsletter does not provide investment advice
Contributors: Lex, Laurence, Matt, Farhad, Daniel, Michiel, Luke
For access to all our premium content and archives, consider supporting us with a subscription. In addition to receiving our free newsletters, you will get access to all Long Takes with a deep, comprehensive analysis of Fintech, Web3, and AI topics, and our archive of in-depth write-ups covering the hottest fintech and DeFi companies.
The inertia premium concept is a sharp way to frame what incumbents have been monetizing. Banks didn't win on the best product or the best price, they won on switching friction, which AI agents will systematically eliminate by comparing rates, terms, and fees in milliseconds without the cognitive overhead that makes humans stick with suboptimal providers.
Where this connects to x402 and machine payments is interesting: agents that can transact autonomously will also be able to move capital programmatically, which means they can constantly optimize treasury positions, payment routes, and yield sources in ways no human treasury team can match. The inertia premium doesn't just disappear in retail banking, it gets arbitraged away across the entire financial stack. That's the more radical version of the thesis worth watching.
really enjoyed this one. 150M users in 18 months just because the friction disappeared is the whole case study!