Treasury & Rates Regime Monitor

The yield curve (10Y − 3M) and the high-yield credit spread — the bond market's two clearest recession and stress gauges. The pills below read the curve regime; credit is tracked in the readings and its own component.

Normal10Y > 3M

Flat≈ 0

Inverted10Y < 3M

▸ Regime definitions & method

Normal · upward (spread > 0) — long rates above short rates; investors are paid a term premium to lend longer. The healthy, expansionary state.

Flat (spread ≈ 0) — short and long rates converge, usually after the Fed hikes short rates up toward long. A transitional, uncertain state.

Inverted (spread < 0) — short rates exceed long; the market expects the Fed to cut as growth slows. The strongest recession warning, leading downturns by 6–18 months.

The slope alone is not the whole read. A positive curve is produced two ways that mean opposite things: the short end falling toward neutral (easing), or the long end selling off (tightening). The summary tile therefore grades the level and the cause and reports the worse of the two — see 06 Duration & Term Premium for the decomposition and the current reading.

Credit (HY OAS) — the extra yield over Treasuries on junk corporate bonds; the market's live default-risk gauge. Complacency < 3% (late-cycle, risk mispriced) · Tight 3–3.5% · Elevated 3.5–5% · Stressed > 5%.

Current readings

10Y − 3M Spread

—

percentage points

HY Credit Spread

—

option-adjusted, %

Credit Regime

—

Components · inputs

10-Year & 3-Month Yields

The two rates that form the curve. Where the 10-year drops below the 3-month, the curve is inverted — those stretches are shaded.

HY Credit Spread (BAMLH0A0HYM2)

Corporate default-risk gauge. Dashed lines mark the 3% / 3.5% / 5% regime boundaries; below 3% is late-cycle complacency, not safety.

Synthesis · the signal

10Y − 3M Spread — Inversion Tracker

The two yields combined into one signal. Everything below zero is an inversion; sustained episodes are shaded and flagged ▸ at onset. Inversion is the recession-precursor signal.

▸ Inversion episodes (10Y − 3M < 0, sustained ≥ 5 days)

#	Start	End	Days	Max Depth

Splitting the nominal 10-year yield into its two prices — real rate + breakeven inflation — to see what is actually driving yields. The pills read whether inflation expectations are anchored.

Subdued< 1.5%

Anchored1.5–2.5%

Running hot> 2.5%

▸ Regime definitions & method

Subdued breakeven < 1.5% — expectations below target signal weak demand or deflation risk; recession-adjacent and hard for policy to fix near the zero bound.

Anchored 1.5–2.5% — expectations near the Fed's 2% target; the market trusts policy. Yield moves are driven by the real rate, not inflation fear.

Running hot > 2.5% — inflation priced persistently above target; pressures the Fed to stay restrictive and a headwind for long-duration assets.

Real rate (DFII10) — the inflation-adjusted yield and the discount rate on all risk assets. Stimulative < 0% · Neutral 0–1% · Restrictive > 1%. A fast move — |Δ90d| ≥ 0.4pp watch, ≥ 1.2pp shock — reprices risk premia violently (cf. 2022).

Nominal 10Y = Real Rate (DFII10) + Breakeven Inflation

Breakeven = DGS10 − DFII10. Velocity = real rate today − real rate 90 days ago.

Current readings

Breakeven Inflation

—

10Y expected, %

Real Rate (10Y TIPS)

—

DFII10, %

Nominal 10Y

—

DGS10, %

Components · the two prices

Breakeven Inflation

The market's 10-year inflation forecast. Dashed lines mark the 2% target and the 1.5% / 2.5% regime boundaries.

Real Rate + Shock Detector

Real rate level (left, solid) with its 90-day velocity (right, dashed). Velocity thresholds mark the ±0.4 watch / ±1.2 shock lines; the amber band is the fastest surge in the sample.

Synthesis · how they sum to the yield

Nominal 10Y Decomposition — Real Rate + Breakeven

Stacked: the real-rate band (bottom) plus the inflation band (top) sum to the nominal 10Y (dark line). Watch which band is driving yields — inflation fear is the worry, real-rate-led moves are healthier.

The Fed's policy rate, where it sits versus the market, and which way it is moving. The pills read the stance gap (Fed funds minus the 3-month market rate).

Behind the curvegap < −0.5

Neutral±0.5

Restrictivegap > +0.5

▸ Regime definitions & method

Behind the curve gap < −0.5 — the 3-month market rate sits above Fed funds; the market is pricing further hikes and the Fed may be lagging. An early-tightening signal.

Neutral −0.5 to +0.5 — Fed funds tracks the market; policy roughly in line with expectations, no strong pressure priced.

Restrictive gap > +0.5 — Fed funds above the market rate; policy is holding tight and the market is pricing cuts ahead. Typically late-cycle.

Stance Gap = DFF − DTB3 · Direction = DFF_today − DFF_{90d ago}

Positive gap = Fed above market = restrictive (easing expected). Direction: rising = hiking, flat = hold, falling = cutting.

Current readings

Fed Funds (DFF)

—

effective rate, %

Policy Gap (DFF − 3M)

—

positive = restrictive

Direction (90d)

—

Components · inputs

Fed Funds vs. 3-Month Market

The policy rate the Fed sets against the 3-month T-bill the market trades. Where the market line sits above Fed funds, hikes are priced; below, cuts.

Policy Stance Gap (DFF − DTB3)

Positive = restrictive (easing expected); negative = behind the curve (hikes expected). Dashed ±0.5 lines mark the regime boundaries; the dot marks today.

Synthesis · direction of travel

Fed Funds Level & 90-Day Direction

Fed funds level (left, solid) with its 90-day change (right, dashed). Sustained positive velocity is a hiking cycle; sustained negative is a cutting cycle; flat is a hold.

A quantity-side cross-check: real industrial output (INDPRO), confirming or contradicting the price signals. The pills read year-over-year growth — the contraction gauge.

ContractingYoY < 0%

Slowing0–1.5%

Expanding> 1.5%

▸ Regime definitions & method

Contracting YoY < 0% — output shrinking year-on-year; the real economy is in contraction. When this confirms a stressed price-side read, the recession signal is no longer just a forecast.

Slowing 0–1.5% — growth positive but fading; the zone where price-based warnings begin to show up in the hard data.

Expanding > 1.5% — healthy output growth; the real economy is absorbing financial conditions.

INDPRO is monthly and released with a lag, so it confirms slowly — by design. It is a coincident-to-lagging check on the leading, daily price signals, not a substitute for them.

Current readings

Industrial Production

—

index, 2017 = 100

YoY Growth

—

% vs. 12 months ago

Activity Regime

—

Components · inputs

Industrial Production — Level

The raw index of factory, mine and utility output. The long-run uptrend; recessions show as visible drawdowns.

Short-Term Momentum (3M annualized)

The near-term pace of output, annualized. This leading view turns down before year-over-year growth does; the zero line is the contraction boundary.

Synthesis · the contraction detector

Industrial Production — Year-over-Year Growth

Trailing-year growth — the headline recession gauge. The zero line is the contraction boundary; the +1.5% line marks expanding / slowing. Stretches below zero are shaded (note the 2020 collapse).

How the Japanese yen, BOJ policy and the yen carry trade transmit into US Treasuries and rates — a channel the Fed weighs for financial stability, separate from inflation and the domestic economy. Japan is the largest foreign holder of Treasuries (~$1.1T), and cheap yen funds a global carry into higher-yielding US assets. The pills read the carry-unwind state — a rapid yen surge forces deleveraging that spikes yields and tightens conditions abruptly (cf. Aug 2024).

Stable · carry onΔ3M > −4%

Yen firming · watch−7…−4%

Carry unwind≤ −7%

▸ The four indicators, definitions & method

1 · Carry width — US 10Y − Japan 10Y. The yield gap that funds the trade: borrow yen cheaply, buy higher-yielding US assets (Treasuries included). Wide > 3pp (strong incentive, but large positioning building) · Moderate 1.5–3pp · Compressed < 1.5pp (BOJ hiking / Fed cutting closes the gap and pulls the trade home).

2 · Carry-unwind momentum — 3-month % change in USD/JPY. Rapid yen strength (a negative move) is the fingerprint of a deleveraging cascade: leveraged shorts cover, and forced selling of US assets — equities and, via risk-parity, Treasuries — moves rates. Watch −7…−4% · Unwind ≤ −7%.

3 · BOJ normalisation — BOJ balance-sheet YoY + Japan 10Y / policy rate. As the BOJ tightens (QT, YCC exit, rate liftoff), rising JGB yields and a shrinking balance sheet pull Japanese capital home. The largest foreign UST holder buying less — or selling — pushes up the US term premium on the long end, independent of Fed policy.

4 · FX-hedged UST 10Y yield — US 10Y − hedge cost, where hedge cost ≈ US 3M − Japan 3M. What a currency-hedged Japanese investor actually earns on a Treasury. Attractive > 1% · Marginal 0–1% · Unattractive < 0% — when it turns negative (as in 2022–24, US short rates far above Japan's), hedged Treasuries lose money and foreign demand fades.

Carry width = US 10Y − JP 10Y · Hedge cost ≈ US 3M − JP 3M

Hedged UST 10Y = US 10Y − Hedge cost · Unwind = Δ_3m USD/JPY

USD/JPY is yen per dollar — a falling USD/JPY is a stronger yen. Monthly series; Japan-rate legs (JGB 10Y, 3M interbank, policy rate) populate on the next keyed FRED refresh. Thresholds are heuristic reads, not calibrated signals.

Why the Fed cares (beyond inflation & growth) — a disorderly unwind is a market-functioning and financial-stability event: yields gap, volatility spikes, and conditions tighten faster than any hiking cycle intends. The Fed's reaction function has a stability leg, and in Aug 2024 the carry unwind — not US data — drove the repricing of near-term cut expectations.

Current readings

Carry Width (US−JP 10Y)

—

percentage points

Yen 3-Month Move

—

USD/JPY Δ3M

FX-Hedged UST 10Y

—

what a hedged JP investor earns

Components · inputs

Carry Width & the Yen

US−Japan 10Y differential (left, the carry incentive) against USD/JPY (right). A wide gap with a weak yen is peak carry; a narrowing gap and a firming yen is the trade unwinding. The 10Y gap populates on the next data refresh.

BOJ Policy — Balance Sheet & JGB Yield

BOJ total assets, year-over-year (left) — expansion is carry fuel, contraction is normalisation — with the Japan 10Y JGB yield (right). A rising JGB and a shrinking balance sheet pull Japanese capital home, pressuring the US long end.

Synthesis · the transmission to US rates

FX-Hedged Treasury Yield & Carry-Unwind Episodes

The hedged UST 10Y (left) — foreign demand fades when it drops below zero — with the 3-month yen move (right, dashed); its ±4 / −7 lines mark the unwind-watch / unwind thresholds. Months of rapid yen strength (Δ3M ≤ −4%) are shaded: those are the deleveraging windows that transmit into US yields and Fed expectations.

The long end — the part of the curve the Fed does not set. Tabs 01–03 read the price of money at the short end and the slope between the two; this one reads the 30-year, the tenor ladder that connects it back to policy, and whether a move in long yields is being paid for in real rates or in expected inflation. The pills read how far the long end has out-run policy over the last year.

Calm< +0.75pp

Watch+0.75 … +1.50pp

Shock≥ +1.50pp

▸ Regime definitions & method

The spread is built from two legs, so its change decomposes exactly: Δ(10Y − 3M) = Δ(10Y) − Δ(3M). Shape — steepening or flattening — is the sign of that change. Tone has three cases, not the textbook two:

Bear — both legs up. Bull — both legs down. Twist — the legs move in opposite directions. The twist is the case this tab exists for: the long end rising while the policy leg falls is the market repricing duration against the direction of policy. Forcing it into bull or bear by whichever leg is larger would have labelled the year to Aug 2026 — long end +0.35pp, policy leg −0.44pp — a bull move, an easing, on a 0.09pp margin.

What is graded — not the size of the long-end move but the divergence, Δ(10Y) − Δ(3M) over 52 weeks, and only when the 10-year has actually risen. Size alone is the wrong measure: a 10-year up 0.35pp is the 69th percentile of its own history and unremarkable when the Fed is hiking. The same 0.35pp against a policy leg 0.44pp lower is something else entirely.

Calm divergence < +0.75pp, or the long end not rising — duration is not being repriced against policy.

Watch +0.75 to +1.50pp — the long end is out-running policy. 11.7% of days, 1990–2026.

Shock ≥ +1.50pp — a full term-premium repricing. 1.1% of days; the dates are 1990–91, 2001–02, 2009–10, 2021 and 2024–25.

Where this is computed — in 01_liquidity/analysis/analyze.py, published on Treasury_Rates.json as _meta.rates_verdict, and read by this page and the homepage. Neither re-derives it. The band also carries a signed vote into the pillar's liquidity regime score, so a long-end tightening now moves the Money verdict instead of leaving it untouched.

Why it matters for equities — a rise paid for in real rates lifts the discount rate on long-duration cash flows with no offsetting revenue growth. A rise paid for in breakeven inflation at least carries nominal pricing power with it. The synthesis chart splits the move into those two legs.

Current readings

30-Year Yield

—

DGS30, %

Divergence · 1Y

—

percentage points

Curve Move · 1Y

—

Components · inputs

The Tenor Ladder — One-Year Change by Maturity

Every tenor the Fed controls against every tenor the market prices. Bars right of zero are yields that rose (a tightening at that maturity), bars left are yields that fell. Where the sign flips along the ladder is where policy stops setting the price.

The Long End vs. Policy

The 30-year and 10-year against the fed funds rate. When the long lines rise while the policy line falls, the market is repricing duration independently of the Fed — the divergence, not the level, is the signal.

Synthesis · who is being paid, and for what

Drivers of the 10-Year — Real Rate vs. Breakeven, One-Year Change

The one-year change in the nominal 10-year split into its two legs — the change in the real rate (TIPS, shaded) and the change in expected inflation (violet, dashed). The two legs sum exactly to the black line, because the breakeven is the nominal yield less the real yield. A move carried by the real leg is a duration and term-premium repricing; one carried by the breakeven leg is an inflation repricing. They call for different equity responses, and the nominal yield alone cannot tell them apart.

Cross-market · is this a US-only event

Global Long Rates — 10-Year Government Yields

A US long-end repricing is rarely a US-only event. The marginal buyer of Treasuries prices them against the domestic alternative at home, so a rising Bund, Gilt or JGB weakens the structural bid for US duration. Issuers not yet fetched are listed below the chart rather than silently omitted.

Issuer	10Y now	12m change	24m change	Highest since	As of

▸ Duration episodes (long end rising and out-running policy by ≥ +0.75pp/1y, sustained ≥ 20 days)

#	Start	End	Days	Peak divergence	Long leg at end	Policy leg at end