America’s banking strength lies in diversity. Yet as PNC CEO Bill Demchak warns, deposit flight to megabanks will only accelerate in the next crisis. Unless addressed, a SOFR-only regime risks tilting the system further toward concentration. IOSCO-aligned credit-sensitive supplements for SOFR such as AXI and FXI offer a modern solution to safeguard stability and preserve bank diversity.
The United States has long benefited from a diverse, multi-tiered banking ecosystem – from global Wall Street institutions to regional and community banks serving local businesses and families. As Federal Reserve Chair Jerome Powell observed in mid-2023, “the U.S. has benefited from its rich, multi-tiered banking ecosystem, and that diversity should be preserved”. This diversity is not just a quaint feature; it is a cornerstone of economic resilience. Small and mid-sized banks provide roughly 60% of small business loans nationwide, fueling entrepreneurship and local growth. Preserving that diversity is therefore a matter of financial stability and broad economic health. Yet as the dust settles on the transition to SOFR, there’s a risk that the current “one-size-fits-all” interest rate regime could inadvertently undermine regional banks’ ability to lend, unless we take proactive steps. In this op-ed, we make the case that new IOSCO aligned credit-sensitive benchmarks – the Across-the-Curve Credit Spread Index (AXI) and the Financial Conditions Credit Spread Index (FXI) – can fill this gap. By reintroducing a market-based credit element to interest rates, AXI and FXI would protect the diversity of U.S. banks (particularly regionals) and enhance the resilience of the financial system as a whole, to the benefit of every bank and every American.
Unintended Consequences of a SOFR-Only World
When LIBOR was phased out, regulators and industry participants coalesced around the Secured Overnight Financing Rate (SOFR) as the primary replacement. SOFR is a risk-free rate, based on overnight loans secured by U.S. Treasuries, and it has proven to be a robust and liquid benchmark for many purposes. But SOFR’s very strength – its lack of credit risk – also means loans tied to SOFR behave very differently in times of stress than loans tied to LIBOR did. LIBOR, for all its faults, incorporated a credit-sensitive component (being based on unsecured interbank lending rates) that tended to rise when banks’ funding costs rose. In contrast, SOFR tends to fall during market stress as investors flock to safe assets. As a group of regional banks warned regulators in 2019, this divergence creates a dangerous mismatch: when banks’ funding costs climb in a crisis, the interest on their SOFR-linked loans drops. Borrowers would see cheap credit just as lenders face expensive funding, incentivizing borrowers to draw down credit lines en masse while banks are under strain. The result, those regional banks predicted, would be a pullback in credit availability – either through higher loan pricing “through the cycle” or simply lower willingness of banks to lend, especially in a downturn.
“The American financial system is strong in part because of our dynamic and diverse banking system. Large, small, and mid-size banks all play an important role in our economy. ”– Jerome Powell
Fast forward to 2025, and new empirical research appears to validate those warnings. A Journal of Finance study by Cooperman, Duffie, Luck, Wang and Yang (2025) analyzes the likely impact of the LIBOR-to-SOFR transition on banks’ lending behavior. Their finding is stark: in a SOFR-only regime, banks anticipate heavier credit line drawdowns during stress and price that risk into loans, raising the cost of credit by roughly 15 basis points and cutting total credit line commitments by about 6% on average. In other words, without a credit-sensitive element to share the pain of higher funding costs, banks respond by lending less. The authors highlight a crucial nuance: regional and mid-sized banks would likely bear a disproportionate share of the burden. Why? Because the biggest banks tend to enjoy stable or increasing deposits during crises (“flight-to-safety” inflows), whereas regional banks see less of their credit line drawdowns come back as deposits.
Cooperman et al (2025) notes that during the COVID market shock, the largest U.S. banks actually experienced deposit growth well in excess of credit draws, but “for the large regional banks, each dollar of [credit] draws generated only $0.28–$0.41 of deposits”, with the rest having to be funded externally. Thus, a regional lender is hit much harder by sudden loan usage than a money-center bank that can recycle drawn funds as deposits. Recent industry voices have echoed this concern. PNC CEO Bill Demchak observed that regional banks have steadily lost customer deposits to megabanks, a trend he warned is likely to “accelerate in the next crisis” (Wall Street Journal, September 2025). His remarks underscore the structural disadvantage facing regional lenders: when stress hits, deposits flow out just as credit lines are being drawn, leaving smaller banks squeezed from both sides.
The Journal of Finance study concludes that given their higher funding costs and lower deposit influx in crises, regional banks are more adversely affected by a risk-free benchmark, with the reference rate transition likely having “a higher impact on the provision of revolvers by these regional banks”. In fact, the authors suggest that without a credit-sensitive element, business may migrate toward those institutions best able to weather the storm, meaning credit could concentrate even more in the largest banks. That outcome would be ironic – and undesirable – in the wake of an effort to strengthen the system. It would undermine the diverse banking landscape that underpins robust credit supply to all corners of the economy.
Why Diverse Banks Mean a Resilient Economy
These findings come at a time when policymakers are explicitly recognizing the value of bank diversity. America’s financial stability relies on having both giants and community lenders in the mix. In an October 2024 speech, then Fed Governor now Vice Chair Michelle Bowman remarked that “to function effectively, the banking system requires the presence of banks of all sizes — larger, regional, and community banks. This diversity of our financial institutions is the greatest strength of our banking system”. Governor Bowman warned that this strength “can easily be imperiled by insufficiently targeted regulation, supervision, and guidance”. In other words, one-size-fits-all rules that ignore differences in banks’ business models and risk profiles could unintentionally harm the very institutions – regional and community banks – that provide so much vitality to our economy. Treasury Secretary Scott Bessent struck a similar chord at the Economic Club of New York, pledging that “At Treasury, we will lead a comprehensive and assertive effort across the Administration to empower our nation’s banks to finance the economy’s pursuit of job growth, wealth creation, and prosperity for all Americans.”
The recent banking turmoil in spring 2023 put these issues into sharp relief. When Silicon Valley Bank and others failed, regulators and industry leaders scrambled to prevent contagion. Large banks stepped up to support a vulnerable regional bank, First Republic, with a $30 billion deposit infusion – a move coordinated by JPMorgan Chase CEO Jamie Dimon alongside federal officials – precisely to shore up confidence in the mid-sized and regional tier of banks. As a joint statement by top regulators noted, this extraordinary action “demonstrates the resilience of the banking system” and the importance of banks of all sizes working to keep the economy strong. In congressional testimony around the same time (March 2023), Fed Governor Michael Barr emphasized that the American economy “relies on a healthy and diverse banking system, one that includes large, small, and mid-size banks”. He noted that small and mid-size banks (including community banks) are responsible for 60% of small business lending in the U.S., leveraging their “specialized knowledge and relationships” in local communities. In short, if the playing field tilts so much that regional banks pull back, it’s not just a problem for those banks – it’s a problem for millions of borrowers and the overall economy.
“At Treasury, we will lead a comprehensive and assertive effort across the Administration to empower our nation’s banks to finance the economy’s pursuit of job growth, wealth creation, and prosperity for all Americans.”– Scott Bessent
This is why the potential credit supply crunch identified in the LIBOR-to-SOFR transition is more than a theoretical quibble about reference rates. It strikes at the heart of banking diversity and financial inclusion. We need our regional banks to remain healthy, competitive, and able to extend credit through the cycle. We also need to ensure regulatory capital and stress-testing frameworks don’t inadvertently sideline these banks or discourage their lending. As Chair Powell noted during a financial stability conference, there are always trade-offs in regulation, but we must “bear in mind” the benefit of our multi-tiered system and make sure that reforms don’t unintentionally push more activity to the “too-big-to-fail” institutions at the expense of smaller banks. Preserving a vibrant regional banking sector is itself a form of risk management: it prevents concentration of credit and fosters competition and innovation in financial services.
Restoring Lost Functionality: AXI and FXI
How can we capture the benefits of LIBOR’s credit sensitivity without reviving its flaws? This is where the Across-the-Curve Credit Spread Index (AXI) and Financial Conditions Credit Spread Index (FXI) enter the conversation. AXI and FXI are newly developed, market-driven indices that measure bank and non-bank credit spreads, respectively, over the risk-free rate. In essence, these indices reflect the extra yield above SOFR that banks (for AXI) or corporations and non-bank financial institutions (for FXI) must pay to borrow in unsecured markets. They are built entirely on real transaction data: for example, AXI pulls from large volumes of trades reported to FINRA’s TRACE and other post-trade data sources across various maturities from overnight to 5-years. Unlike the notorious LIBOR, which was based on banks’ self-reported estimates and became prone to manipulation, AXI and FXI are transparent, verification-friendly, and anchored in actual trades. An independent review by IBM Promontory has found that these indices comply with IOSCO’s best-practice principles for financial benchmarks, meaning they meet high standards for integrity and reliability.
Crucially, AXI and FXI are designed to move with credit conditions in a dynamic yet robust way. When credit risk is rising – say, investors grow nervous about banks’ health – AXI will rise (widening the spread over SOFR), directly indicating higher bank funding costs. When markets are calm, AXI stays low, reflecting easy funding conditions. This behavior makes them valuable tools for aligning loan pricing with actual funding realities. If banks use AXI as an add-on to SOFR in loan contracts, the interest rate on those loans would automatically include a credit-sensitive adjustment. During good times, that credit spread would be minimal (keeping loans affordable). But if stress hits and unsecured funding becomes costly, AXI would spike and loan rates would rise accordingly, before borrowers draw down en masse. In effect, AXI and FXI can serve as an “automatic stabilizer” for lending – much like LIBOR did, but via a cleaner mechanism. By discouraging opportunistic drawdowns (since loan rates go up when funding is tight), a credit-sensitive benchmark reduces the debt-overhang problem that otherwise makes banks pull back credit. Indeed, in the Journal of Finance paper’s calibrated model, the welfare-optimal reference rate had about 70% of LIBOR’s credit sensitivity – suggesting that an intermediate solution (not pure LIBOR, but not pure SOFR either) maximizes credit availability.
From a policy perspective, incorporating AXI/FXI could mean loans that better reflect true economic risk, allowing banks to lend more confidently. Banks that choose to supplement SOFR with a credit-sensitive index may actually experience less earnings volatility in a crisis, since their loan income rises alongside funding costs. Over time, widespread use of such benchmarks in lending could make the system safer by keeping credit flowing during stress – the very opposite of the pro-cyclical credit crunch we fear from a SOFR-only approach. As an added incentive, if a bank’s portfolio is buffered by a credit-sensitive supplement, regulators could recognize that in capital planning – for example, such a bank might warrant lower stress capital buffers on credit line commitments, because the loan rates themselves would help absorb some stress losses. In short, AXI and FXI offer a modern, data-driven way to reintroduce credit-sensitivity into the system, without returning to the opaque LIBOR era. They blend seamlessly with SOFR (they are quoted as a spread over SOFR), ensuring no disruption to the broader rate ecosystem. And because they are transaction-based, they maintain market discipline and transparency.
Shared Interest: Why Big Banks Are Supporting Supplements for SOFR
It might be tempting for the largest banks to shrug off AXI and FXI. After all, megabanks can fund themselves cheaply even in rough waters, and they don’t necessarily need a credit index to price their loans – they have plenty of data and sophistication to do that internally. Moreover, many large institutions hedge their risks via derivatives tied to SOFR, and they may not see an immediate balance-sheet need for an index like AXI. However, supporting AXI and FXI is not about the needs of any single large bank – it’s about bolstering the ecosystem that we all rely on. The events of 2023 made it clear that when regional banks come under threat, the whole system can tremble. In the case of First Republic Bank’s near-collapse, it was Jamie Dimon and other big-bank CEOs who coordinated a rescue deposit package, precisely because they knew a cascading loss of confidence in regional lenders could spread and cause far broader damage. As The Washington Post reported, this “bank-to-bank intervention” – 11 banks depositing $30 billion – was unprecedented in modern U.S. banking and reflected the acute concern among both federal officials and Wall Street executives about shoring up the system’s weakest links. In announcing the effort, the Treasury and Federal Reserve explicitly welcomed the “show of support by a group of large banks” as a sign of confidence in the system’s resilience.
“to function effectively, the banking system requires the presence of banks of all sizes — larger, regional, and community banks. This diversity of our financial institutions is the greatest strength of our banking system.”– Michelle Bowman
The lesson is that our largest banks have a deep stake in the health of smaller banks. A stable, inclusive banking system expands the economic pie – there are more businesses to lend to, more consumer spending, and fewer crises to mop up. Large banks already recognize that a scorched-earth, “big-fish-eat-little-fish” environment ultimately isn’t in their interest. For instance, Governor Barr’s holistic review of bank capital requirements in 2023 took pains to propose rules that mainly focus on banks with assets over $100 billion, aiming to calibrate risk without choking community banks with undue burden. Even so, Vice Chair Bowman and others have cautioned that cumulative regulatory weight can overwhelm smaller institutions, and rules should be appropriately tailored to avoid imperiling community banks by accident. In that context, AXI and FXI should be seen as system-wide public goods. They equip regional banks to manage funding risk and continue lending, which in turn reduces the likelihood of panics that require heroic, ad-hoc interventions by regulators or big-bank consortia. Put simply, if AXI and FXI help keep a Fifth Third or a Zions healthy, that stability accrues to the benefit of JPMorgan, Citi, and Bank of America as well – not to mention to the benefit of countless communities that rely on the regionals.
Large banks also stand to gain from the market predictability that will come with adoption of credit-spread supplements for SOFR. Today, in a stress scenario, everyone must guess how far funding spreads will blow out and how much each bank will retrench. If credit-sensitive benchmarks were broadly used, much of that adjustment would happen automatically and transparently via rates, potentially lessening the need for emergency measures. Moreover, supporting these indices aligns with the industry’s commitment to innovation and best practices. It signals that big banks are not only focused on their own profiles but also on strengthening the financial market infrastructure that underpins lending. In fact, the Across-the-Curve Credit Spread Index was introduced in 2020 by academics (Berndt, Duffie & Zhu) as an improved measure of bank funding costs and launched in 2022. Embracing such innovations can only burnish the credibility of U.S. financial markets.
Strengthening Resilience: Policy Measures and the Path Forward
So how do we operationalize AXI and FXI to realize these benefits? One immediate step is to integrate credit-spread scenarios into regulatory stress testing and supervision. In August 2025, we submitted a proposal to the Federal Reserve to enhance the Fed’s stress testing framework in light of the LIBOR-SOFR transition. The proposal pointed out that current stress tests – while rigorous – do not explicitly model the dynamic we’ve been discussing: namely, that in a severe stress, interest rates (like SOFR) might fall at the same time bank funding costs spike, prompting surging credit drawdowns. To address this, we recommend that the Fed consider adding a credit spread shock or SOFR–funding cost divergence scenario to its bank stress tests. Concretely, that could mean assuming a widening of AXI in the scenario and checking that banks (especially regional ones) have enough liquidity and capital to handle large credit line usage under those conditions. By explicitly baking this into supervisory exercises, regulators would encourage banks to plan for such events well in advance. It would also validate the use of tools like AXI/FXI: banks that incorporate these indices in their loan pricing would perform better under the stressed scenario, providing an incentive for adoption. The Fed’s 2025 stress test results were overall reassuring for bank resilience, but adding this dimension would ensure that even if the next crisis doesn’t look like the last, banks are tested against the risk of funding-market frictions and liquidity hoarding – the very scenario that a credit-sensitive supplement for SOFR is meant to mitigate.
Policymakers could also consider recognizing credit-sensitive loan structures in capital regulations. For example, if a regional bank demonstrates that a substantial portion of its commercial loans are tied to SOFR + AXI (thus automatically repricing in stress), regulators could factor that into lower required loss absorbency for those exposures. This idea, hinted at in the SOFR Academy proposal, aligns with common-sense risk management: a loan that will charge a higher rate during a crisis is less risky to the bank’s solvency than one locked at a low rate. Such an approach would not be about giving favors to smaller banks, but about accurately measuring risk and not unduly penalizing banks that take prudent steps to safeguard their balance sheets.
Finally, regulators and the industry should continue to communicate and collaborate on reference rate reforms. The success of SOFR itself was a triumph of public-private cooperation (through the Alternative Reference Rates Committee). In that same spirit, incorporating credit spread indices will require buy-in from multiple stakeholders: banks large and small, borrowers (who must understand that a slightly higher rate in tough times is ultimately in their interest if it keeps credit available), and regulators who set expectations. There may be technical adjustments needed – for instance, banks will want clarity on how using AXI or FXI interacts with hedge accounting, or how loans tied to these indices should be documented. None of these are insurmountable issues, and early adopters can help pave the way. What’s important is the policy signal: acknowledging that credit sensitivity has a place in a safe and sound banking system. As Bowman noted, the key is avoiding blanket approaches that inadvertently push out smaller players. By targeting the issue of credit spread risk directly, tools like AXI/FXI allow for a more tailored approach – one that bolsters the system at its pressure points instead of layering more burdens across the board.
Conclusion: Stability Through Inclusivity
In sum, the transition from LIBOR to SOFR was a necessary move for transparency and stability, but it left a gap that we now have the opportunity to fill. The Across-the-Curve Credit Spread Index (AXI) and Financial Conditions Index (FXI) represent smart, modern solutions to ensure that credit risk is not ignored in our lending benchmarks, but managed in a transparent way. Embracing these indices would help keep regional and community banks competitive, by reducing the pressure for them to contract lending when times get tough. It would also enhance overall financial system resilience, by aligning incentives – borrowers, banks, and investors all see the true cost of funds, and banks of all sizes can sustain lending without courting hidden risks.
This is not a zero-sum game between big and small banks. On the contrary, it’s about creating a more robust ecosystem in which all banks can thrive and serve their customers. The largest banks, which might not need AXI or FXI for their own books, should nonetheless champion them as part of a stable and inclusive system that ultimately benefits everyone. In the words of Vice Chair Barr, “the American financial system is strong in part because of our dynamic and diverse banking system”, with large, small, and mid-size banks all playing important roles. If we want that statement to hold true in the coming decades, we must adapt our tools and rules to preserve that diversity. Leveraging credit-spread indices is a sensible and data-backed step in that direction. It ensures that regional banks can continue to do what they do best – relationship lending and supporting the real economy – without being inadvertently hobbled by the post-LIBOR landscape.
The Federal Reserve and other regulators have often highlighted the need for continuous improvement in risk management and supervision. Here is a chance to do exactly that: incorporate the lessons of rigorous research and the voices of practitioners to fine-tune our financial system’s plumbing. The result can be a banking system that remains diverse, competitive, and resilient, even under stress – a system in which Jamie Dimon’s JPMorgan and a Regional Bank in the mid-west each have their role, and each support the other’s stability. That vision is well worth striving for. By taking up AXI and FXI as part of the regulatory toolkit, we can help ensure that credit flows freely to Main Street in good times and bad, and that America’s broad constellation of banks continues to shine as a source of strength.
ABOUT THE AUTHOR
Marcus Burnett is the Chief Executive Officer of SOFR Academy ([email protected]).
This note is provided for informational purposes by SOFR Academy, Inc. (Sofr.org), a financial engineering firm. This note is not designed to be taken as advice or a recommendation for any investment decision or strategy. Readers should make an independent assessment of relevant economic, legal, regulatory, tax, credit, and accounting considerations and determine, together with their own professionals and advisers, if the use of any index is appropriate to their goals. Neither the USD Across-the-Curve Credit Spread Index (AXI), nor the USD Financial Conditions Credit Spread Index (FXI) are associated with or sponsored by the Federal Reserve Bank of New York or any regulatory authority. Additional information about SOFR Academy, AXI and FXI can be found here.
Copyright 2025 SOFR Academy, Inc. All rights reserved
