What Are Credit Spreads and Why Do They Predict Recessions?

The short answer

When a company borrows money by issuing a bond, investors demand a higher interest rate than the government pays, because companies can default while the U.S. government (in theory) cannot. That extra yield is the credit spread.

Think of spreads as a real-time fear gauge for the corporate economy. When spreads are narrow, investors are relaxed about default risk. When they blow out to several times their calm-period level, it means the market is pricing in serious economic trouble: companies going bankrupt, earnings collapsing, credit freezing up.

Unlike the yield curve, which signals future trouble, credit spreads often react faster and closer to the moment of stress. They are less about what might happen and more about what the market believes is happening now to corporate balance sheets.

→ New to fixed income? Explore the Eco3min Financial Education Hub

What the data shows

Using ICE BofA High Yield Option-Adjusted Spread (FRED: BAMLH0A0HYM2, 1996–2024), every U.S. recession was preceded by a significant widening of high-yield spreads.

Key readings from the data: unusually narrow spreads indicate complacency (historically rare and often followed by corrections). A sustained widening well above the long-run average signals stress, and a blowout to several times calm-period levels marks acute crisis conditions, as in late 2008. On the Moody’s Baa measure, a public series that runs back to 1986, the spread over the 10-year Treasury peaked at 6.16 percentage points on 4 December 2008 and at 4.31 on 23 March 2020, against a median of 2.13 since 1986 (Moody’s, via FRED, series BAA10Y).

The lead time is shorter than the yield curve: spreads typically widen meaningfully several months before recession onset, compared to 12–17 months for the 2s10s. This shorter, higher-conviction lead is the basis of the Eco3min framework reading HY OAS as a recession and equity-reversal lead. However, spreads can give false alarms during non-recessionary stress events. The 2015–2016 energy sector collapse pushed HY spreads to stress levels without triggering a broad recession (the Moody’s Baa spread, via FRED, reached 3.63 points in February 2016), though the energy sector itself experienced one.

The spread-VIX relationship adds a useful cross-check: when credit spreads widen without a corresponding VIX spike, the signal may be sector-specific rather than systemic.

Why it happens: the macro mechanism

The textbook account ties spread widening to a rising expected rate of corporate default, through monetary policy, credit cycles and corporate fundamentals. It explains the direction of the move better than its size.

The chain typically starts with central bank tightening. Higher rates increase borrowing costs for companies, especially those with variable-rate debt or near-term maturities. Simultaneously, bank lending standards tighten, reducing the availability of credit. Marginal borrowers, typically high-yield issuers, face a double squeeze: higher costs and reduced access.

As economic conditions deteriorate, corporate revenues slow. Companies with high leverage find themselves unable to cover interest payments from cash flow. Default risk rises, and bond investors demand wider spreads as compensation.

The regime matters. In a normal tightening cycle, spreads widen gradually as conditions tighten. In a liquidity crisis (2008, March 2020), spreads blow out suddenly because the market for corporate bonds itself freezes: sellers cannot find buyers at any reasonable price. This distinction between credit deterioration (gradual) and liquidity seizure (sudden) is critical for interpreting spread movements correctly.

Investment-grade spreads (IG) follow the same logic but with lower amplitude. A sustained move well beyond their usual range already signals stress.

→ Working framework: Monetary Policy · Debt & Systemic Fragilities

The bond market doesn’t hope. It prices.

Why credit spreads widen: default, liquidity and risk aversion

On bonds rated AA and A, expected default losses make up only a small part of the spread. Elton, Gruber, Agrawal and Mann (Journal of Finance, 2001) split the gap between corporate and Treasury rates into expected default, state taxes (paid by corporate bondholders, not by Treasury holders) and a residual. Expected default explained a small share, less than state taxes under the authors’ central tax estimate. The residual moved closely with the risk factors that explain equity returns, which marks it as a premium for systematic risk.

Working from credit default swap prices, Longstaff, Mithal and Neis (Journal of Finance, 2005) assign most of the spread to default risk once the premium for bearing that risk is included. Their non-default component varies over time with bond-specific illiquidity and with market-wide liquidity. The two studies fit together. The loss investors expect is modest, but the compensation they demand for carrying default risk, and for holding bonds that can be hard to sell, swings with market conditions.

Gilchrist and Zakrajšek (American Economic Review, 2012) carry the split to the business cycle by separating the part of corporate spreads that reflects firms’ expected defaults from a residual, the excess bond premium. They find that the predictive power of spreads for economic activity comes mainly from that residual, which they read as a gauge of the risk-bearing capacity of the financial sector, whose decline contracts credit supply.

What it means for different economic actors

Bond investors are directly exposed. Wide spreads mean existing corporate bonds lose value (yields up = prices down), but also mean higher yields on new purchases.

Equity investors can incorporate credit spreads as an early warning system. Stocks and high-yield bonds draw from the same economic pool: corporate earnings and solvency. When credit markets signal stress, equity markets typically follow within weeks to months. A divergence, stocks rising while spreads widen, has historically preceded episodes of correction.

Corporate borrowers face a direct cost increase when spreads widen. Companies planning to refinance or issue new debt during a spread-widening environment face materially higher costs, or may find the market closed entirely. This is the mechanism by which credit markets transmit financial stress into real economic contraction.

A common error is watching only the VIX as a fear gauge while ignoring credit spreads. The VIX measures equity option demand: a relatively narrow signal. Credit spreads reflect the actual cost of borrowing for the real economy, making them a more fundamental measure of economic stress. The thread is followed through in the most common misreadings of credit and spreads.

Practical observation

Go deeper

Related questions

• Why does the yield curve invert before recessions?
• What is the Sahm Rule and how accurate is it?
• What has historically followed VIX spikes above 30?
• Why does corporate debt at record levels matter?

Frequently asked questions

Do credit spreads widen during a recession or before it?

Mostly before, with the peak arriving later. On the Baa measure, the spread had already moved well off its lows when the 2001 and 2008 recessions began, as dated by the FRED series USREC, and in 2020 the widening started in the last week of February, days before the recession’s March start. The peak tends to fall inside the recession or after it, in early December 2008 nearly a year into the 2008–09 downturn, and in October 2002 almost a year after the 2001 recession had ended. In 2020 the lead shrank to days because the shock came from outside the credit system.

What is a “normal” level for high-yield spreads?

There is no fixed norm. Readings far below the long-run average of the ICE BofA HY OAS are historically unusual and suggest compressed risk premiums (complacency); a sustained move well above it signals elevated stress. For a long public reference, the Moody’s Baa spread over the 10-year Treasury has a median of 2.13 percentage points since 1986 (Moody’s, via FRED, series BAA10Y). These levels shift over time with the composition of the high-yield market and changes in default recovery rates.

Are credit spreads better than the yield curve at predicting recessions?

They serve different functions. The yield curve (2s10s) provides an earlier signal, typically 12–17 months ahead, about the trajectory of monetary policy. Credit spreads provide a shorter-lead, higher-conviction signal about actual economic stress. Using both together (yield curve for early warning, spreads for confirmation) produces the most robust framework.

Do credit spreads predict stock market crashes?

Not crashes specifically, but prolonged drawdowns. Sharp spread widening has preceded every major equity bear market since 1996. However, the correlation is strongest during credit-driven crises (2008) and weakest during valuation-driven corrections (2000 dot-com) where credit conditions remained relatively healthy even as stocks declined.

Last updated — 3 November 2026