Adjusted NFCI vs S&P 500

The Chicago Fed publishes two financial conditions indices weekly. The headline NFCI aggregates 105 individual indicators (credit spreads, funding rates, equity volatility, leverage measures) into a single weekly z-score where zero equals the historical average. The Adjusted NFCI runs the same input panel through a regression that removes the predicted contribution of GDP and inflation, leaving a residual that is loose or tight relative to where macro fundamentals would predict.

The 2017 Chicago Fed Letter introducing ANFCI made the case explicitly: pre-recession periods often show NFCI tightening alongside slowing growth, but if the tightening is exactly what slowing growth would predict, it carries no incremental information. ANFCI isolates the surprise component. The April 24, 2026 reading of -0.49 means current financial conditions are 0.49 standard deviations looser than what the current GDP path and PCE inflation rate would predict, which is the configuration that supports risk-on positioning rather than warns against it.

The 2008 Global Financial Crisis saw ANFCI breach +1.0 in August 2007, four months before the December 2007 NBER recession start, and peak above +5.5 in October 2008 during the post-Lehman acute phase. SPY (and the underlying S&P 500) was still within 5% of its all-time high when ANFCI first crossed zero. The credit-equity divergence was the textbook leading signal: financial conditions were tightening relative to fundamentals while equity prices were rallying. The S&P 500 lost 56% peak-to-trough by March 2009.

The February 2018 vol-mageddon episode showed ANFCI rising from -0.55 to -0.10 in three weeks while SPY fell 10% from January 26 to February 8. The reading never crossed zero, which correctly flagged the episode as a positioning unwind rather than a credit-driven event. The 2022 hiking cycle saw ANFCI move from -0.65 in January to +0.40 in October as the Fed hiked 425bp in nine months. SPY lost 25% peak-to-trough across the same window. ANFCI led the SPY peak by approximately seven weeks.

ANFCI is published weekly with a one-week lag relative to the underlying data. The Chicago Fed releases the index every Wednesday at 8:30am ET on the Bank's website and FRED. The leading-relationship-to-equities is therefore mechanical: high-frequency credit-market repricing shows up in ANFCI before it shows up in trailing earnings revisions that drive equity multiples.

The April 24 reading of -0.49 decomposes into risk indicators contributing -0.34, credit indicators contributing -0.15, leverage indicators contributing -0.05, and the macro adjustment factor contributing +0.06. Risk-side dominance (volatility and credit-default swap pricing) pulling the index lower while leverage indicators are roughly neutral is the classic late-easing configuration: market participants are not building dangerous leverage but are pricing risk premia tightly. The risk-credit subindex divergence is what cross-asset desks watch for the early signal that something is shifting.

Chicago Fed working papers and subsequent academic replications find that S&P 500 12-month forward returns vary systematically by ANFCI starting quartile. In the loosest quartile (ANFCI below approximately -0.30), SPY 12-month forward total return averaged 12.3% across 1973-2024. In the tightest quartile (ANFCI above approximately +0.30), SPY 12-month forward total return averaged -2.6%. The middle two quartiles sat around the unconditional 8% average.

The asymmetry matters for sizing. The current -0.49 reading sits inside the loosest quartile, which historically supports risk-on positioning. The configuration to fear is not negative ANFCI rising; it is ANFCI crossing zero and then accelerating. The 2007 episode, the 2015 high-yield wobble, and the late-2018 reflexive sell-off all began with the same pattern: ANFCI rising through zero while S&P 500 was within 5% of an all-time high. The 2026 configuration does not match that pattern.

The structural reason is informational asymmetry. Credit holders are senior in the capital stack and price probability of default on every funding window: weekly auctions, monthly refinancing, quarterly bond maturities. Equity holders sit junior and price conditional residual value, which is dominated by earnings expectations published once a quarter. The information density per unit time is much higher on the credit side, and that is what ANFCI captures across 105 indicators.

The 2007 episode is the clearest illustration: investment-grade OAS widened from 100bp to 165bp between February and August 2007, two-year swap spreads breached 70bp in early August, and auction-rate securities began failing in February 2008, all before the S&P 500 broke its uptrend. None of this stress showed up in the equity multiple until October 2007. The Cleveland Fed's 2022 economic commentary on financial conditions and recession risk explicitly cited the 2007 ANFCI signal as the cleanest historical example of how the index leads.

ANFCI at -0.49 with SPY rallying back toward $720 is the configuration that supports holding pro-cyclical equity exposure rather than reducing it. The risk-indicators sub-component contributing the bulk of the looseness reflects high-yield OAS sitting at roughly 290bp (near 25-year tights), moderate VIX at 18.7, and absence of dollar-funding stress in the FX swap basis.

The defensive triggers to watch are specific. ANFCI rising above -0.30 with SPY still rising would be the first warning. ANFCI crossing zero with credit-side sub-components leading the move would warrant initial defensive rotation, historically two to three months before equity peaks. ANFCI above +0.30 sustained four weeks has been the signal to materially reduce equity beta in every cycle since 1973 except the 2018-2019 episode where the Fed pivoted before the cycle resolved. The April 2026 reading is far enough below those thresholds that the cleanest read is to use ANFCI as confirmation of the current risk-on regime rather than a contrarian warning.