Credit spreads measure the extra yield investors demand to lend to companies rather than to the government, and they are one of the quieter gauges of how nervous bond markets are. This article explains how they work and what the US numbers showed in mid-October 2026. It describes the US dollar corporate bond market and Federal Reserve research; other regions have their own benchmarks and conditions.
What is a credit spread?
A corporate bond pays a yield. A US Treasury security with a similar maturity also pays a yield, and because the US government is treated as the benchmark borrower, the Treasury yield serves as the baseline. The difference between the two is the credit spread. It is the extra compensation investors require for the chance that the company could fail to pay, for the difficulty of selling the bond quickly, and for a general willingness (or unwillingness) to take on credit risk.
Spreads are quoted in percentage points or in basis points, where one basis point is one hundredth of a percentage point. A spread of 1.00 percentage point equals 100 basis points.
Because the spread is a gap rather than a level, it can widen in two ways: corporate yields rise faster than Treasury yields, or Treasury yields fall faster than corporate yields, which often happens when investors rush into safe government debt.
Which spread gauges are widely used?
The St. Louis Fed’s FRED database publishes two daily series built by ICE Data Indices that are often quoted in the press. Both use an “option-adjusted” spread, which adjusts for features such as a bond’s early-repayment rights so that bonds can be compared more fairly against a Treasury curve.
| Gauge | What it covers | Reading on October 15, 2026 |
|---|---|---|
| ICE BofA US Corporate Index OAS | US dollar investment-grade bonds, rated BBB or better | 0.84 percent |
| ICE BofA US High Yield Index OAS | US dollar below-investment-grade bonds, rated BB or below | 3.23 percent |
Both readings come from FRED, as of October 15, 2026 daily closes. FRED notes that, starting in April 2026, these series include only three years of observations, so longer history has to come from the data provider.
The high-yield index has additional construction rules. Bonds must carry an average below-investment-grade rating from Moody’s, S&P and Fitch, have more than one year to maturity, and be at least $100 million in size. Defaulted securities are excluded. Constituents are weighted by amount outstanding, and the index is rebalanced on the last calendar day of each month.
How do investment-grade and high-yield spreads compare?
Investment-grade companies are judged more likely to repay. High-yield companies, sometimes called junk-rated, carry a higher chance of default, so they must pay more.
The October 15 numbers can be turned into a simple comparison. The high-yield spread of 3.23 percent is 323 basis points; the investment-grade spread of 0.84 percent is 84 basis points. The gap between them is 3.23 − 0.84 = 2.39 percentage points, and 3.23 ÷ 0.84 is about 3.8, so the riskier group was paying roughly 3.8 times the spread.
That ratio and gap are worth watching over time. In calm periods the two spreads tend to sit relatively close to their lower ranges, while in stressed periods the high-yield spread usually widens much faster. The reading itself does not say whether 3.23 percent is high or low historically, because the FRED page shows only three years of data; judging that requires the longer record.
A worked example with illustrative numbers
Suppose, purely for illustration, that a Treasury security with a given maturity yields 4.00 percent. A high-yield bond with a spread of 3.23 percentage points would then yield about 7.23 percent. On a $10,000 holding, the extra 3.23 points over the Treasury is about $323 per year in additional yield. That extra income is not free: it is payment for bearing the risk that some issuers miss payments, or that bond prices fall if spreads widen.
If the spread widened by one percentage point to 4.23 percent while Treasury yields stayed unchanged, the same bond’s yield would rise to 8.23 percent, which means its price would have fallen. The mechanism is the same inverse link described in why bond prices fall when interest rates rise.
Why are tight spreads not the same as safety?
A narrow spread says investors are currently demanding little extra pay for credit risk. Several things can produce that: strong corporate earnings, abundant demand for yield, or simple complacency. Narrow spreads do not eliminate risk; they mean the cushion for taking it is thin.
Federal Reserve research gives a good illustration of the distinction. In an April 2016 FEDS Note, economists Favara, Gilchrist, Lewis and Zakrajšek split a broad corporate credit spread into two parts. One part reflects expected defaults. The remainder, which they call the excess bond premium, is described as capturing “investor attitudes toward corporate credit risk,” or credit market sentiment. It is a residual measure, an estimate rather than a direct observation.
What does the Fed research say about spreads and recessions?
The same FEDS Note tested how well these measures help predict US recessions, as dated by the National Bureau of Economic Research. Using monthly data from January 1973 to March 2016, the authors estimated the probability of a recession beginning within the next 12 months. Their findings, in the authors’ words and numbers, include:
- A 50 basis point rise in the overall credit spread was associated with about a 7 percentage point rise in the estimated recession probability.
- Once the spread was split into components, the predictive content came from the excess bond premium. The expected-default component was statistically indistinguishable from zero.
- A 50 basis point rise in the excess bond premium was associated with about a 15 percentage point rise in the estimated probability.
The note also gave a snapshot: by the end of February 2016, the model implied about a 55 percent chance of a recession over the following 12 months, against an unconditional rate of about 30 percent over the sample. The note describes this as the authors’ own estimate from a simple in-sample model, not an official Federal Reserve Board forecast.
Where the signal failed
The authors were open about errors. In 2002, spreads and the excess bond premium rose sharply after corporate accounting scandals and bankruptcies, and the model predicted a recession, but the economy did not slip into one. This is a reminder that spreads respond to events that are painful for bondholders without necessarily damaging the wider economy. The note also made clear that it is a short piece from 2016, so it shows how researchers have thought about the measure rather than what the measure says today.
Why might sentiment drive spreads?
The note offered two possible explanations, labeled as proposals rather than tested conclusions. First, investors may become complacent in good times and then overreact to bad news, which amplifies swings in spreads. Second, when large institutional bond investors such as mutual funds, insurers and pension funds pull back from supplying credit, financial conditions can tighten and slow growth. Both ideas help explain why a bond-market gauge can move before official economic data does, a theme also covered in why markets can rise even when the economy feels weak.
How can a reader use spreads without over-reading them?
A few habits help:
- Check which spread is quoted. Investment-grade and high-yield measures can tell different stories on the same day.
- Look at the direction and speed. A steady drift matters less than a sharp widening over a few weeks.
- Separate level from change. Without long history, a single reading such as 3.23 percent cannot be called high or low.
- Remember the Treasury side. A spread can narrow because corporate yields fall or because Treasury yields rise; the spread alone does not say which.
- Treat it as one signal. Alongside indicators such as the shape of the yield curve, spreads give a fuller view, though none of them is a forecast.
This is analysis, not a prediction. The confirmed facts are the published readings and the research findings above; any interpretation of what they imply for the economy or for prices is judgment. For how headlines more broadly move prices, see how global events actually move markets.
A modest gauge with a clear job
Credit spreads compress a large amount of investor sentiment into one number: how much extra yield the market wants for lending to companies instead of the US government. As of October 15, 2026, according to FRED, US investment-grade spreads stood at 0.84 percent and high-yield spreads at 3.23 percent. Federal Reserve research found that the sentiment-driven part of spreads carried predictive information about recessions in past data, but with false alarms such as 2002. Read them as context, not as a verdict.
This article is general information, not personalised financial advice; rules and rates change.
Sources: Federal Reserve Bank of St. Louis; Federal Reserve Board
