TED Spread¶

A historical money-market stress indicator equal to the three-month unsecured U.S.-dollar interbank rate minus the matched three-month U.S. Treasury-bill yield.

Version: v1 · 2026-08-30 · History
Domain-specific #: 2937
Origin domain: financial markets
Subdomain: money markets
Aliases: Treasury-Eurodollar spread

Core Idea¶

The TED Spread is a historical U.S.-dollar money-market indicator formed by subtracting the yield on a three-month United States Treasury bill from the three-month unsecured interbank borrowing rate. In the operational series formerly published through FRED, the private-rate leg was three-month U.S.-dollar LIBOR and the public-rate leg was the three-month Treasury-bill secondary-market rate:^[1]

\[ \operatorname{TED}_t=L^{USD,3m}_t-Y^{Tbill,3m}_t. \]

If both inputs are quoted as annualized percentage rates, the result is a percentage-point spread; multiplying that difference by 100 expresses it in basis points. A LIBOR reading of 5.50 percent and a Treasury-bill yield of 5.10 percent therefore produce a TED Spread of 0.40 percentage points, or 40 basis points. The subtraction removes much of the common short-rate level and foregrounds the extra yield on unsecured bank funding relative to a highly liquid sovereign reference.

The indicator does not isolate a single latent cause. Its widening can incorporate perceived bank credit and counterparty risk, scarcity of unsecured term funding, a flight into Treasury bills, and market-specific supply, regulatory, or benchmark effects. BIS work describes it both as the relative riskiness of LIBOR-based financing and, historically, as the basis risk between private money-market rates and Treasury-bill rates.^[2]^[3] Federal Reserve and BIS research consequently use the TED Spread as an aggregate sign of bank-funding or financial stress, not as a structural estimate of any one component.^[4]^[5]

The name predates the cash-rate implementation: “T” referred to Treasury bills and “ED” to the Eurodollar futures contract. The defining contrast survived changes in traded contracts, but the best-known modern series became specifically three-month USD LIBOR minus three-month Treasury bill. That implementation is now historical. FRED discontinued TEDRATE in January 2022 when its three-month USD LIBOR input was removed; the panel-based U.S.-dollar LIBOR system itself ended in June 2023.^[1]^[6] A spread using a SOFR-based leg can support related stress monitoring, but it is not an observation-equivalent continuation: SOFR measures secured overnight Treasury-repo borrowing, whereas LIBOR embedded unsecured term bank-funding risk.^[7]^[8]

TED Spread is therefore an autonomous domain-specific abstraction: it has a stable instrument contract, formula, interpretation discipline, and historical use across trading, financial-stability monitoring, and empirical research. Strip away the specified money-market benchmarks and only the general pattern of Baseline Deviation remains.

Structural Signature¶

Locked operation: matched-date three-month USD unsecured interbank rate − matched-date three-month U.S. Treasury-bill yield → signed yield gap in percentage points/basis points → conditional evidence about bank-funding and market stress.

Sig role-phrases:

Unsecured bank-funding leg — a three-month U.S.-dollar interbank benchmark, canonically three-month USD LIBOR in the discontinued FRED series, carries term funding, bank-credit, and benchmark-specific influences.
Sovereign reference leg — the matched three-month U.S. Treasury-bill yield supplies the safer and usually more liquid comparator, while retaining its own demand, supply, and market-microstructure effects.
Maturity and currency match — both legs must refer to comparable U.S.-dollar horizons; substituting a different tenor or currency changes the object rather than merely refreshing its data.
Signed subtraction — the Treasury yield is subtracted from the interbank rate. The ordering is part of the identity and makes widening normally positive when private funding becomes dearer relative to bills.
Basis-point scale — the result is a rate difference, not a ratio, price, probability, or dimensionless index. One percentage point equals 100 basis points.
Stress-proxy interpretation — a wider gap is treated as evidence consistent with greater banking credit/funding strain or flight to safety, subject to confounding rather than as direct measurement of a single causal variable.
Versioned benchmark contract — the exact source, fixing convention, date alignment, business-day treatment, and revisions belong to the observation. A replacement input creates a related series whose comparability must be established.
Historical-series boundary — the canonical LIBOR-based daily series is discontinued. A SOFR-minus-bill spread must be named as a successor construction, not silently appended to TEDRATE.

Recognition requires all eight roles. “A risky rate minus a safe rate” is only a generic credit spread. A LIBOR–OIS spread changes the comparator and more narrowly targets unsecured bank term premia over an overnight-indexed expectation. A corporate-bond spread changes the borrower class and maturity structure. A broad financial-stress index aggregates multiple inputs. None is TED merely because it widens during crises.

What It Is Not¶

Not counterparty-default probability. TED is a market-price difference whose widening can reflect credit, liquidity, flight-to-quality, regulatory, supply, and benchmark effects. Converting it into a default probability requires a separate model.
Not pure liquidity risk. Funding liquidity is an important interpretation, but neither leg is a clean liquidity-only instrument. Boudt, Paulus, and Rosenthal use TED to separate stress regimes while studying interactions between market and funding liquidity; that application does not prove TED uniquely identifies either one.^[9]
Not the LIBOR–OIS spread. Both use LIBOR historically, but OIS supplies a different reference tied to expected overnight rates. Treasury bills add sovereign-market scarcity and flight-to-quality effects.
Not a corporate credit spread. TED compares an interbank benchmark with Treasury bills at a short matched horizon; it does not compare corporate bonds with Treasuries or summarize the whole corporate default distribution.
Not a universal recession forecast. A widening spread records conditions in money and safe-asset markets. It can coincide with or predict stress in a sample, but it does not guarantee an equity decline, recession, or bank failure.
Not a timeless constant. Input definitions, submission methods, regulation, Treasury supply, market liquidity, and benchmark cessation alter the series. Historical thresholds cannot be transplanted without checking the measurement regime.
Not preserved by substituting SOFR. SOFR is secured, overnight, and transaction-based. A 90-day average SOFR-minus-bill series can be useful but does not retain the unsecured term bank-credit content of three-month LIBOR.^[7]^[10]
Not the stress event itself. TED is an indicator. It can spike during a wholesale-funding run or systemic crisis without constituting the run, the failed institutions, or the causal propagation mechanism.

Scope of Application¶

TED Spread travels literally wherever the same two benchmark legs, tenor, currency, and subtraction convention are being used. Its scope is instrument-bound rather than metaphorical.

Money-market monitoring. Traders and analysts used the daily gap to summarize the pricing separation between unsecured dollar bank funding and Treasury bills.
Financial-stability surveillance. Central-bank and policy research placed TED among market-based stress indicators and financial-conditions inputs, especially during episodes when bank funding and safe-asset demand diverged.^[5]^[7]
Crisis chronology. Historical studies use the spread to date and scale money-market dislocation. BIS research reports a peak of 457 basis points in October 2008 and relates widening to banking-sector funding strain.^[4]
Funding-liquidity research. Empirical studies use TED as a regime variable, control, or aggregate proxy rather than as the dependent phenomenon itself. The Boudt–Paulus–Rosenthal two-regime model estimated a switch near 48 basis points for its July 2006–May 2011 sample.^[9]
Hedge and basis-risk analysis. Historically, divergence between Treasury-bill and Eurodollar/private-money-market rates exposed hedgers to basis risk, particularly during flights to quality.^[3]
Financial-stress index construction. A TED-like component can enter a composite index, but its weight, transformation, replacement policy, and relation to other components belong to that index rather than to TED itself.
Historical backtesting. The discontinued FRED series remains useful for analysis over its recorded period, provided users do not imply current publication or splice unlike successor data without disclosure.

The concept does not automatically extend to other countries, currencies, maturities, or reference-rate pairs. Those may be structurally analogous bank–sovereign spreads, but they are not the U.S.-dollar TED Spread unless a source explicitly defines and governs the extension.

Clarity¶

Use a five-question test before labeling a number TED:

Is the private leg a three-month U.S.-dollar unsecured interbank rate?
Is the comparator a three-month U.S. Treasury-bill yield observed on a compatible date?
Is the calculation private rate minus Treasury yield?
Is the result reported as percentage points or basis points?
Is the benchmark regime stated—historical futures construction, LIBOR-based cash series, or a named successor rather than an undisclosed splice?

The fifth question is now decisive. FRED’s TEDRATE is explicitly discontinued, despite the fact that a charting system can still display its history. A current chart labeled “TED Spread” may calculate something else. The analyst should disclose the exact ticker or source series, both legs, tenor convention, observation window, and treatment of missing days. “TED-like” is preferable when a secured or synthetic reference replaces LIBOR.

Interpretation also requires two layers. The measurement layer says exactly what was subtracted. The construct layer asks why the gap changed. A 25-basis-point increase might arise because the unsecured rate rose, the bill yield fell in a flight to quality, or both. The same endpoint therefore supports different narratives; decomposing the leg movements is a minimum diagnostic before assigning cause.

Manages Complexity¶

Money-market stress comprises interbank credit concerns, term-funding scarcity, collateral conditions, safe-asset demand, monetary-policy expectations, and microstructure. TED compresses their joint effect on two observable yields into one signed coordinate. The compression makes dates, regimes, and portfolios comparable without requiring the user first to estimate every latent component.

That economy is useful precisely because it is lossy. A policymaker can see that private and sovereign short rates have separated; a researcher can condition a model on high- and low-spread states; a risk manager can flag a break in normal co-movement. None needs to claim that the spread tells the whole causal story. The indicator transforms “something changed in several linked markets” into a bounded next question: which leg moved, what market mechanism drove it, and is the movement replicated in other measures such as LIBOR–OIS, commercial-paper spreads, repo conditions, or a composite stress index?

Version discipline contains a second complexity. By fixing the benchmark contract, TED distinguishes genuine historical changes from changes in the instrument used to observe them. This is especially important across the LIBOR transition, where a numerically smooth successor could conceal a conceptual discontinuity.

Abstract Reasoning¶

The subtraction licenses several disciplined inferences.

First, directional decomposition: \(\Delta TED=\Delta L-\Delta Y\). A wider spread can be decomposed into a higher unsecured rate, a lower bill yield, or both. This identity is arithmetic; causal attribution is not.

Second, relative-price interpretation: holding maturity, currency, date, and convention fixed, widening means the market price of unsecured bank term funding has increased relative to the Treasury-bill reference. It does not mean either absolute rate is high. TED can widen while both rates fall if bill yields fall faster.

Third, regime classification: a researcher may define a high-TED state and test whether relationships differ across it. Boudt and colleagues estimated a 48-basis-point switch in one model and sample, but that is an empirical threshold, not a definition of crisis for all eras.^[9]

Fourth, proxy triangulation: confidence in a stress interpretation rises when TED, LIBOR–OIS, other private-safe spreads, funding quantities, and market-liquidity measures move consistently. Divergence calls for diagnosis rather than averaging away the disagreement.

Fifth, counterfactual restraint: because both legs respond to policy and market structure, TED alone cannot reveal what the spread would have been absent an intervention. Event attribution needs a design beyond the indicator.

Knowledge Transfer¶

Within financial markets, the complete mechanism transfers directly across dates, portfolios, crisis studies, and research designs that preserve the same benchmark definition. An observation used by a trader and one used in an econometric model remain instances of TED because their roles and arithmetic match.

Across other currencies or benchmark pairs, transfer is usually shared abstract mechanism, not name identity. A private funding rate minus a government or near-risk-free reference instantiates Baseline Deviation and may function as a stress proxy, but its borrower pool, collateral, maturity, institutional regime, and safe-asset dynamics differ. The portable lesson is to make relative pricing visible by subtracting a declared reference and to audit the fidelity of the resulting proxy. The term TED Spread should stay with its Treasury–Eurodollar/U.S.-dollar lineage.

The LIBOR-to-SOFR transition illustrates transfer’s limit. The St. Louis Fed could replace LIBOR with a 90-day average SOFR in its stress index and observe high historical correlation, yet it also documented divergence and cautioned that time would reveal whether the stress signal behaved similarly.^[7] The calculation pattern transfers; the latent construct loading does not automatically transfer with it.

Examples¶

Canonical¶

Suppose three-month USD LIBOR is 5.50 percent and the three-month Treasury-bill yield is 5.10 percent on a compatible observation date. Then

\[ TED=5.50\%-5.10\%=0.40\text{ percentage points}=40\text{ basis points}. \]

The unsecured bank-funding leg is 5.50 percent; the sovereign reference leg is 5.10 percent; the maturity and currency match is three months and U.S. dollars; the signed subtraction is private minus public; and the basis-point scale gives 40. The result says only that unsecured bank funding is priced 40 basis points above the bill yield. To interpret a change, retain both component series: a move from 20 to 40 basis points caused only by falling bill yields has a different immediate decomposition from the same move caused only by rising LIBOR.

Mapped back: unsecured bank-funding leg; sovereign reference leg; maturity and currency match; signed subtraction; basis-point scale; stress-proxy interpretation.

Applied / In Practice¶

During the 2007–2009 financial crisis, researchers tracked the TED Spread as banking funding markets strained. A BIS working paper defines it as three-month LIBOR minus the three-month Treasury rate, reports that it widened after uncertainty about mortgage-backed securities in August 2007, and records a 457-basis-point peak in October 2008 after failures of financial institutions.^[4] The episode maps the indicator’s proper use: an extreme relative-price gap marked a stressed regime and supported investigation of loan-supply responses. It did not, by itself, prove which bank would fail or partition the gap into credit and liquidity components. Boudt and colleagues later used TED to distinguish regimes in a model of funding and market liquidity rather than treating it as either latent variable itself.^[9]

Mapped back: versioned benchmark contract; stress-proxy interpretation; historical-series boundary; signed subtraction; basis-point scale.

Structural Tensions¶

T1: Compression versus causal identification. One spread condenses several linked market forces into a legible stress coordinate. That makes monitoring and conditioning efficient, but the same compression discards the information needed to distinguish bank credit fear, unsecured-funding scarcity, bill-market flight to quality, policy expectations, and benchmark distortions. Diagnostic: Is the spread being used to flag a state, or is a causal claim being made that requires additional instruments and leg-level evidence?

T2: Safe reference versus active market. Treasury bills are used as a safer comparator, yet their yield is not inert. Scarcity, issuance, regulation, collateral demand, and flight-to-quality buying can move the reference and widen TED even without a proportional deterioration in the unsecured leg. Diagnostic: Did the private rate rise, did the Treasury yield fall, or did both move?

T3: Historical continuity versus benchmark integrity. A long series is valuable for comparison, creating pressure to splice a SOFR-based successor onto LIBOR TED. But secured overnight repo and unsecured term bank borrowing load on different risks. Preserving continuity can destroy construct fidelity. Diagnostic: Has the replacement’s equivalence been demonstrated for the intended use, or merely assumed from correlation in an earlier sample?

T4: Stable threshold versus regime dependence. Fixed basis-point thresholds make dashboards actionable. Their meaning changes with monetary-policy regimes, regulation, Treasury supply, market structure, and the benchmark definition. A threshold estimated in 2006–2011 is not a timeless crisis constant. Diagnostic: Was the threshold calibrated and validated in the same instrument and regime in which it is being applied?

T5: Domain autonomy versus reduction to a parent prime. TED is a named, recurrent instrument with a precise money-market contract; reducing it to Baseline Deviation loses the borrower class, sovereign reference, maturity, units, interpretation, and historical benchmark boundary. Yet only Baseline Deviation travels intact outside finance. Diagnostic: Does the task require recognition of the historical financial indicator, or only the general operation of reading a signed gap against a reference?

Structural–Framed Character¶

TED Spread is structural-leaning but domain-specific. Its evaluative weight is low: subtraction itself is neutral, and “wider” is mathematically defined before an analyst judges it concerning. Its human-practice dependence is substantial because both LIBOR and Treasury-bill yields arise from designed markets and conventions rather than an observer-free natural process. Its institutional origin is decisive: benchmark administrators, central banks, contract markets, tenor rules, and publication systems determine what the legs mean.

Its vocabulary does not travel intact. “Three-month USD LIBOR,” “Treasury bill,” “Eurodollar,” and “basis points” remain financial-market terms; replacing them with generic signal and baseline changes the named object. Recognition within the home domain is literal because the same inputs and formula recur across monitoring and research. Import outside the domain occurs by analogy unless another market explicitly defines the same U.S.-dollar pair.

The portable skeleton is Baseline Deviation: choose a reference, subtract it from an observation, and treat the signed departure as information. Measurement and Proxy–Target Fidelity govern how the instrument is built and interpreted, but the single skeleton is the reference-relative gap. Its character: mathematically sharp and structurally legible, yet inseparable from historically contingent financial benchmarks and the risks those benchmarks carry.

Structural Core vs. Domain Accent¶

This section explains why TED Spread is a domain-specific abstraction rather than a prime.

What is skeletal. A focal observable is compared with a declared reference on a common scale, and their signed difference becomes the object of analysis. That relation can recur in quality control, physiology, forecasting, and many other substrates as Baseline Deviation. The general reasoning—match comparison conditions, preserve sign, inspect both components, and avoid overinterpreting a proxy—is genuinely portable.

What is domain-bound. TED fixes the focal observable as three-month unsecured U.S.-dollar bank funding, canonically LIBOR; the reference as a three-month U.S. Treasury bill; the units as percentage points or basis points; and the interpretive habitat as money-market and bank-funding stress. It also carries an institutional history from Treasury and Eurodollar instruments through a LIBOR-based cash series to discontinuation during benchmark reform. Remove those named legs and the result may still be a spread, but it is no longer TED. Change LIBOR to SOFR and the secured/unsecured distinction changes the risk loading. Change the Treasury bill to OIS and the indicator becomes a different named spread.

Why this does not clear the prime bar. Cross-domain recognition does not preserve TED’s vocabulary or mechanism. Calling a difference between two biological measurements a “TED Spread” would be metaphorical and would add no analytic content beyond Baseline Deviation or Comparison. Even within finance, corporate, swap, sovereign, and term spreads have distinct contracts. The cross-domain reach belongs to Baseline Deviation, Measurement, and Proxy–Target Fidelity; TED contributes the concrete financial implementation and its interpretive cautions. That bounded residual is precisely why the entry is autonomous and domain-specific.

TED Spread strictly instantiates Baseline Deviation. It declares a short Treasury yield as the reference, subtracts it from an unsecured bank-funding observation, and makes the signed departure a first-class fact. This is the minimal structural parent.

It is also a Measurement in ordinary encyclopedia prose: a defined data procedure maps two market rates to a value with units, timing, source, and convention. Measurement is not proposed as a second parent because Baseline Deviation already captures the more discriminating universal genus and additional ancestry would add little placement information.

Proxy–Target Fidelity governs interpretation. TED stands in for banking credit, funding liquidity, or broad stress only imperfectly; the mapping can change when one leg’s market structure or benchmark changes. Risk and Liquidity identify important constructs carried by the signal, but TED neither measures a known outcome distribution nor isolates ease of conversion. Comparison and Contrast are too broad to be useful additional parents.

Relationships to Other Abstractions¶

Current abstraction TED Spread Domain-specific

Parents (1) — more general patterns this builds on

TED Spread is a kind of Baseline Deviation Prime

TED Spread strictly instantiates Baseline Deviation.

Hierarchy path (1) — routes to 1 parentless root

TED Spread → Baseline Deviation → Comparison → Self Checking

Neighborhood in Abstraction Space¶

TED Spread sits in a sparse region of the domain-specific corpus (80^th percentile for distinctiveness): few abstractions share its structure, so a faithful description tends to retrieve it precisely.

Family — Macroeconomic Traps & Financial Fragility (29 abstractions)

Nearest neighbors

Computed from structural-signature embeddings · 2026-09-08

Not to Be Confused With¶

LIBOR–OIS spread. This historically subtracted an overnight-indexed-swap rate from term LIBOR, reducing but not eliminating differences in expected overnight rates and term premia. TED uses Treasury bills, whose market can experience safe-asset and supply effects. Tell: Is the comparator a Treasury bill or an OIS rate?
SOFR–Treasury-bill spread. This compares two secured or near-risk-free rate constructions and can be a stress-index component, but it does not preserve LIBOR’s unsecured term bank-credit component. Tell: Does the private leg contain unsecured panel-bank term funding, or secured Treasury repo?
Commercial-paper–Treasury-bill spread. Sometimes called a paper-bill spread, it measures private short-term corporate funding relative to bills rather than interbank funding. Tell: Is the borrower pool banks in the interbank benchmark or commercial-paper issuers?
Credit spread. The superordinate family includes corporate, sovereign, swap, and other risky-minus-reference yield differences. TED is one historically specified member, not a synonym for the family. Tell: Are both the three-month USD interbank and Treasury-bill legs fixed?
Wholesale-Funding Run. A run is a dynamic withdrawal or nonrenewal of short-term funding that forces asset sales and propagation. TED may signal the conditions but is not the run mechanism. Tell: Is the object a price gap or a funding-withdrawal process?
Funding Fragility. Fragility is the structural dependence that makes a funding system susceptible to disruption. TED is an observed market indicator and can remain low before fragility is activated. Tell: Are we diagnosing dependency architecture or observing a contemporaneous rate spread?
Basis-Risk Failure. Treasury/private-rate divergence historically made imperfect hedges fail, but the failure also requires an exposure, hedge instrument, and decoupling. TED is the gap that can reveal the basis. Tell: Is a hedge failing, or is only the market spread being measured?
Financial-stress index. A composite combines several rates, spreads, volatility measures, and sometimes prices using normalization and weights. TED can be one component. Tell: Is the output a single two-leg difference or an aggregation?

References¶

[1] Federal Reserve Bank of St. Louis. “TED Spread (DISCONTINUED) (TEDRATE)”. The series notes define the calculation as three-month USD LIBOR minus the three-month Treasury bill and explain discontinuation after the LIBOR input was removed from FRED. registry ↩a ↩b

[2] Remolona, Eli M.; Lekkos, Ilias; and Wooldridge, Philip D. The Valuation of US Dollar Interest Rate Swaps. BIS Economic Papers No. 35, 1994. registry ↩

[3] McCauley, Robert N. “International banking and financial market developments: Benchmark tipping in the money and bond markets.” BIS Quarterly Review, March 2001, pp. 39–45; see also Wooldridge, Philip D. “The bond benchmark continues to tip to swaps.” BIS Quarterly Review, March 2017. registry ↩a ↩b

[4] Cornett, Marcia Millon; McNutt, Jamie John; Strahan, Philip E.; and Tehranian, Hassan. Liquidity Risk and the Credit Crunch of 2007–2008: Evidence from Micro-Level Data on Mortgage Loan Applications. BIS Working Papers No. 473, 2014. registry ↩a ↩b ↩c

[5] Aramonte, Sirio; Rosen, Samuel; and Schindler, John W. “Assessing and Combining Financial Conditions Indexes.” Federal Reserve Board Finance and Economics Discussion Series 2013-39; and Hubrich, Kirstin and Tetlow, Robert J. “Financial Stress and Economic Dynamics: The Transmission of Crises.” FEDS 2012-82. registry ↩a ↩b

[6] Financial Conduct Authority. “The US dollar LIBOR panel has now ceased.”, 3 July 2023. registry ↩

[7] Kliesen, Kevin L.; McCracken, Michael W.; Trần Khánh Ngân; and Werner, Devin. “What Are Financial Market Stress Indexes Showing?” Federal Reserve Bank of St. Louis, 24 May 2022. registry ↩a ↩b ↩c ↩d

[8] Federal Reserve Bank of New York. “Secured Overnight Financing Rate Data.” Defines SOFR as a broad transaction-based measure of overnight borrowing collateralized by Treasury securities. registry ↩

[9] Boudt, Kris; Paulus, Ellen C. S.; and Rosenthal, Dale W. R. “Funding Liquidity, Market Liquidity and TED Spread: A Two-Regime Model.” Journal of Empirical Finance 43 (2017): 143–158. registry ↩a ↩b ↩c ↩d

[10] Alternative Reference Rates Committee. A User’s Guide to SOFR, 2019. registry ↩