NFCI vs VIX vs HY OAS: Three Complementary Reads of Financial Stress

Three indicators recur in every conversation about U.S. financial stress — NFCI, VIX, HY OAS — and they do not measure the same thing. Reading their divergences is more informative than choosing the best of the three.

VIX, HY OAS, and NFCI capture three facets of financial stress at different frequencies and scopes. This article maps what each captures, what it ignores, and when they tell divergent stories.

1. Three indicators, three different objects

NFCI, published by the Chicago Fed since 1971, measures U.S. financial conditions broadly: 105 weekly series covering money markets, bank and market credit, intermediary leverage, non-bank financing, aggregated into a single latent factor. VIX, published by the CBOE since 1990, measures 30-day implied volatility on the S&P 500, extracted from vanilla at-the-money option prices. HY OAS, published by ICE BofA since 1996, measures the option-adjusted average spread of high-yield bonds against Treasuries of comparable maturity.

These three measures are not substitutes. They differ on three structural dimensions: frequency (weekly for NFCI, intraday for VIX, daily for HY OAS), scope (broad for NFCI, equity only for VIX, sub-IG corporate credit segment for HY OAS), and the economic nature of the signal (cyclical position of financial conditions for NFCI, ex-ante equity risk premium for VIX, corporate credit risk premium for HY OAS).

The useful question is therefore not “which is the best indicator” — a question that presupposes they measure the same thing — but “what does each indicator tell me that the others do not”. This reading grid sits within financial conditions broadly defined and complements the reading of the NFCI as aggregate index with an external mapping rather than an internal read.

2. VIX: equity volatility, high frequency, narrow scope

VIX (CBOE Volatility Index) measures annualized implied volatility of S&P 500 options at approximately 30 days to maturity. Its calculation formula integrates all liquid strikes (methodology revised in 2003 from the original 1993 VXO) and produces a percentage value corresponding to the annualized standard deviation of anticipated S&P 500 variation.

The 1990-2024 historical VIX average oscillates around 19. Peaks exceed 80 (November 2008 record at 89, March 2020 at 82). Sustained compression periods below 12 characterized 2017-2019 and 2023-2024. The distribution is heavily right-skewed: 90% of the time VIX sits below 30, 99% of the time below 50, but rare moves above produce extreme observations.

The narrow nature of VIX — it captures only 30-day equity volatility — is both its strength and its weakness. Strength: it is extremely reactive, intraday, and gives a near-instantaneous read on equity-market anxiety. Weakness: it misses credit stress episodes without immediate transmission to equities (for example the August 2007 ABCP market stress, which tightened NFCI sharply without moving VIX for several weeks), and it over-reacts to equity shocks without systemic implications (the May 6, 2010 flash crash, the August 2024 yen carry trade unwind, which moved VIX without an NFCI signal).

VIX enters as a component of the NFCI risk sub-index, but with moderate weight (typically 5 to 10% of the risk sub-index, which itself is 25-35% of the aggregate). This dilution by construction means VIX and NFCI can durably diverge without statistical inconsistency — NFCI is on average more stable than VIX and captures information VIX ignores. An empirical chronological audit of NFCI versus VIX over the recent cycles is provided in the empirical NFCI-VIX audit over recent cycles, which complements the present comparative analytical mapping.

3. HY OAS: sub-investment-grade corporate credit risk

ICE BofA HY OAS (US High Yield Master OAS, formerly Merrill Lynch HY Master) measures the market-cap-weighted option-adjusted average spread of corporate bonds rated BB+ or below (high yield) against Treasuries of comparable maturity. The option-adjusted methodology removes the effect of embedded prepayment optionality (call options) in most HY bonds, producing a measure of pure credit spread.

The 1996-2024 historical HY OAS average oscillates around 520 basis points. Peaks exceed 1,800 bps (November 2008 record at 2,182 bps, March 2020 at 1,100 bps). Sustained sub-350 bps levels characterize late-cycle accommodative phases: pre-GFC 2007, 2014-2015, 2017-2019, and 2024-2026 (280 bps in May 2026, a historically tight level).

The nature of the HY OAS signal is very different from VIX: it captures anticipated default risk on the corporate sub-IG segment, which includes roughly 2,200 issuers representing 1.4 trillion dollars of outstanding debt in May 2026. It is a fundamentally structural indicator — it moves slowly, integrates institutional allocation decisions (insurance companies, asset managers, pension funds), and reflects appetite for corporate credit risk on a horizon of several months to several quarters.

HY OAS enters as a major component of the NFCI credit sub-index, with substantial weight (typically 15 to 25% of the credit sub-index). Its contribution to the aggregate is therefore more direct than that of VIX. This explains why NFCI and HY OAS move in concert more often than NFCI and VIX — but the correlation stays imperfect, because the NFCI credit sub-index also includes SLOOS, IG OAS, and non-bank financing, all of which can diverge from HY.

4. When the three indicators converge, when they diverge

Convergence episodes are the major systemic shocks where equity risk, credit risk, and broad financial conditions move simultaneously. October 2008 illustrates this case: VIX above 80, HY OAS above 1,800 bps, NFCI above +4.0. The aggregate signal is unambiguous, and reading just one of the three indicators suffices to grasp the magnitude of stress.

March 2020 is another convergence case: VIX at 82, HY OAS at 1,100 bps, NFCI at +1.5. Again, the three indicators tell the same story — a systemic liquidity and risk rupture touching all segments simultaneously.

Divergence episodes are analytically more instructive. Three documented examples deserve attention. First, August-October 2007: NFCI moves from zero to +0.8 in eight weeks following the ABCP market freeze, HY OAS moves from 280 to 550 bps, but VIX stays contained below 30. Credit stress is underway without having yet reached equities — precisely the phase that precedes generalization of the shock in 2008.

This 2007 episode is methodologically significant. From the August 9, 2007 BNP Paribas suspension of three money market funds with subprime exposure, the ABCP market froze within weeks, and the NFCI credit sub-index began rising while VIX stayed in the 15-25 range. The signal of imminent broader stress was clearly visible in the NFCI but invisible in VIX, and only partially visible in HY OAS (which moved from 280 to 550 bps but had not yet reached crisis levels). An analyst relying exclusively on VIX would have missed the 12-month window between the ABCP freeze and the Lehman bankruptcy.

Second, February 2018 and August 2024: VIX jumps to 37 then 65 over two sessions due to micro-structural shocks (short-vol ETF volatility in 2018, yen carry trade unwind in 2024). HY OAS and NFCI stay stable. The three indicators agree these are isolated equity shocks without systemic transmission — a diagnosis confirmed by rapid VIX normalization in subsequent weeks.

Third, May 2026: NFCI at −0.55 (accommodative), HY OAS at 280 bps (tight), VIX at 18 (near historical average). The three indicators tell a coherent accommodative regime on the surface. But the NFCI credit sub-index is at −0.7 while the leverage sub-index is at +0.1, and primary dealer positioning has slowed since Q4 2025. The internal NFCI divergence tells a story neither VIX nor HY OAS captures. This tension is developed in the article on the current divergence across indicators.

5. Reading divergences rather than chasing “the best”

The analytical grid that emerges from this mapping is simple: none of the three indicators is universally superior, and each has its domain of relevance. VIX is useful to detect intraday equity shocks and the ex-ante S&P 500 risk premium on short horizons. HY OAS is useful to assess appetite for sub-IG corporate credit risk on a medium horizon. NFCI is useful to position cyclically the broad financial conditions on a long horizon.

Reading the three together brings more information than reading the best chosen one. Episodes where they converge validate the aggregate diagnosis; episodes where they diverge inform on the specific nature of the stress (equity only? credit underway without equity? structural accommodative compression?). A rigorous analysis of the NFCI internal components alongside the comparative read with VIX and HY OAS produces a more robust analytical triangulation than any of the three indicators taken in isolation.

Last updated — 19 May 2026