Credit spreads are not merely measures of expected default risk. They are also real-time indicators of market liquidity, which is often a function of refinancing conditions, bank funding pressure, and the capacity of the financial system to transmit credit to the real economy. When credit spreads widen sharply, borrowing costs rise, refinancing becomes more difficult, market liquidity deteriorates, and credit availability can contract. For policymakers and market participants focused on financial stability, this makes short-duration credit and funding spreads especially informative. 

The transition from London Interbank Offered Rate (LIBOR) to Secured Overnight Financing Rate (SOFR) materially strengthened the integrity of US reference rates. SOFR is robust, transparent, and grounded in deep overnight secured funding markets. But SOFR is intentionally a near risk-free rate. It does not directly capture changes in bank funding costs or broader credit conditions. This distinction matters most during periods of stress, when risk-free rates may fall as central banks ease policy and investors seek safety, while unsecured bank funding costs and corporate credit spreads rise.  

That divergence creates an important challenge for lending markets. Corporate credit lines are drawn more heavily when funding markets are stressed. Recent research in the Journal of Finance by Harry Cooperman, Darrell Duffie, Stephan Luck, Zachry Wang, and Yilin Yang shows that this can raise expected bank funding costs and dampen credit supply through a debt-overhang channel. Historically, this effect was mitigated when loans referenced credit-sensitive benchmarks such as LIBOR. The transition to risk-free rates may exacerbate the friction in some settings, particularly when borrowers are more likely to draw on committed lines precisely when bank funding spreads are elevated. 

This is the financial stability context in which transaction-based credit-spread benchmarks such as the Across-the-Curve Credit Spread Index (AXI) and the Financial Conditions Credit Spread Index (FXI) should be understood. Conceived in work by Antje Berndt, Darrell Duffie, and Yichao Zhu (2023), AXI and FXI are not alternatives to SOFR. They are designed to be used alongside SOFR as transparent, transaction-based credit-spread supplements, with the USD benchmarks having been independently reviewed by Promontory Financial Group against the relevant IOSCO Principles for Financial Benchmarks. In this structure, SOFR remains the risk-free-rate foundation, while AXI or FXI can provide an additional measure of bank funding costs or broader credit conditions where appropriate. 

Figure 1. Historical AXI and FXI credit-spread levels (Source: STOXX Ltd) 

AXI primarily reflects bank funding conditions across unsecured instruments with maturities from overnight to five years. FXI extends the framework to broader corporate credit markets. Together, they are intended to help measure the credit and funding component that risk-free rates, by design, do not capture. This distinction is important: the objective is not to reduce the role of SOFR or impair SOFR market liquidity but to strengthen the post-LIBOR benchmark architecture by allowing robust credit-spread supplements to coexist with SOFR. 

The historical evidence underscores why this matters. During the Global Financial Crisis, Treasury yields declined as investors sought safety and the Federal Reserve eased policy, but credit spreads widened dramatically. Corporate borrowing costs, especially for weaker and shorter duration issuers, rose even as risk-free rates fell. The result was a sharp tightening of financing conditions despite monetary easing. 

The COVID-19 crisis revealed a similar pattern. Risk-free rates moved lower, but short-duration credit spreads widened abruptly as firms drew heavily on committed credit lines and markets feared a broad collapse in revenue and liquidity. In that episode, central bank intervention helped restore market functioning quickly. But the episode reinforced a central lesson: credit stress often appears first through liquidity demand, refinancing pressure, and balance-sheet funding costs, not only through realized defaults. 

This matters for credit supply because undrawn credit-line commitments are extremely large and are more likely to be drawn precisely when wholesale bank funding spreads are elevated. The Cooperman-Duffie-Luck-Wang-Yang research finds that, in a scenario where wholesale bank funding spreads reach Global Financial Crisis levels, drawdowns on SOFR-linked lines would be roughly 60 percent higher than on LIBOR-linked lines. That increase is priced into newly originated credit lines, raising expected drawn-credit costs and reducing the supply of committed credit. 

The impact is not uniform across banks. During COVID, many firms drew credit lines for precautionary reasons and left much of the cash on deposit, helping offset funding pressure at some institutions. In a more severe confidence shock, however, drawn funds may leave the lending bank and migrate toward larger institutions perceived as safer. This dynamic can be especially challenging for regional and mid-sized banks: credit lines are drawn on the asset side just as deposits are leaving on the liability side, increasing reliance on more expensive external funding precisely when funding spreads are elevated.  

The 2023 regional banking stress provided a targeted and instructive example. Broad corporate credit spreads widened only modestly, but bank funding instruments and financial-sector spreads reflected acute concern about deposit flight, liquidity mismatches, and confidence in regional institutions. Silicon Valley Bank’s collapse illustrated how quickly depositor confidence can weaken, with deposits moving from vulnerable regional institutions toward the largest banks, including JPMorgan. Beyond the immediate funding pressure, this dynamic has broader implications for the continued polarization of the US banking system, as stress episodes can reinforce the migration of deposits, liquidity, and market confidence toward the largest institutions. These issues are directly relevant to the Federal Reserve’s 2026 Review of the US Bank Failures.  

This is why well-designed benchmark credit spread supplements can play a useful role. They can reduce the mismatch that arises when the reference rate on a loan falls during stress while the lender’s marginal funding costs rise. In the Cooperman-Duffie-Luck-Wang-Yang framework, the funding-cost wedge depends partly on the covariance between bank funding spreads and credit-line drawdowns. A benchmark credit spread can reduce that wedge because drawdowns become less attractive when bank funding spreads are high. 

Timing is important. With the LIBOR transition successfully complete and SOFR now firmly established, the market can focus on the remaining question of how to measure and manage credit-spread risk within a SOFR-centered framework. That question is becoming more relevant as financial risks increasingly arise from multiple channels, including private credit, nonbank financial intermediation, and geopolitical uncertainty. Greater public clarity around robust, transaction-based credit-spread supplements could help market participants assess these tools consistently and distinguish well-designed supplements designed to complement SOFR from less robust credit-sensitive rates that raised concerns during the benchmark reform process. 

The policy implication is not that the market should return to LIBOR. LIBOR was no longer sufficiently robust and was rightly replaced. The lesson is different: a SOFR-centered system can be made more resilient if market participants have access to robust, transparent, transaction-based credit-spread supplements that capture the funding and credit conditions SOFR excludes. 

AXI and FXI are designed to serve that purpose. Used alongside SOFR, they can help lenders, borrowers, and policymakers observe and manage the transmission of credit and funding stress across the financial system. They can also support more resilient credit provision by improving the alignment between contractual loan economics and the real funding conditions banks face during stress. 

For financial stability, the central question is therefore not whether SOFR is the right foundation for US benchmark reform. It is. The question is whether the post-LIBOR framework should also include robust tools for measuring and managing credit-spread risk. The evidence from past crises and from recent research on bank funding risk and credit supply suggests that it should. 

Featured Author:

Alex Roever is Senior Advisor to SOFR Academy, an American financial market infrastructure company. He previously spent more than 25 years at J.P. Morgan Securities, where he served for nearly a decade as Managing Director and Head of U.S. Interest Rate Strategy. Mr. Roever represented J.P. Morgan on the Financial Stability Board’s Market Participants Group, which was charged with recommending enhancements to major interest rate benchmarks, including IBORs, following the Global Financial Crisis. He is a CFA Charterholder and has also served as a Senior Director at CFA Institute. 

All views expressed by members are their own and not reflective of the views of the Bretton Woods Committee.