Overnight & Global
THE 10-YEAR BROKE 5%, AND THE WHOLE MOVE IS REAL — BREAKEVENS WENT THE OTHER WAY. THAT IS TIGHTENING WITH CREDIBILITY, AND IT IS WORSE FOR EQUITIES, NOT BETTER.
Take the yield move apart, because the decomposition is the whole story. The 10-year is at 5.016%, through the round number for the first time this cycle. But the 10-year real yield is 2.624% and making new highs, while the breakeven FELL to 2.379% and the 5y5y forward FELL to 2.306% from 2.321%. Nominal yields are rising because real yields are rising, and inflation compensation is moving the other way. That is not an inflation scare — it is the market believing the Fed. It is also, counter-intuitively, the worse outcome for equities: a de-anchoring would at least inflate nominal earnings, whereas a pure real-rate rise raises the discount rate against unchanged cash flows. Note where the damage landed — EM worst at -1.4%, Europe -1.0%, the US -0.5%. Dollar real rates set the global cost of capital.
Session Read
Four threads: a 5% handle on the 10-year; the move decomposed as entirely real; a correction to our own equity-cushion number; and gas moving through Hormuz again.
The 10-year through 5% — a level, not a number — The UST 10-year printed 5.016%, having run 4.973% on the 11th and 4.987% on the 14th. Round numbers matter here for a specific reason rather than a superstitious one: the external strategy read we track places the zone in which the equity-yield correlation is at risk of flipping at 5-5.5% on the 10-year. We have just entered the bottom of that band. Below it, rising yields and rising equities coexist because both reflect growth. Inside it, yields start to compete with equities outright. We are no longer arguing about whether the level matters.
CORRECTION — our equity-cushion number, on the right anchor — We published a CAPE-versus-real-yield gap of -0.19% yesterday, described as negative for the first time and a series low. Both claims were wrong. We discounted at the raw 10-year TIPS yield; the Delphic Valuation Standard §2 specifies the Cleveland Fed model real rate — inflation-risk-premium-stripped — and says so explicitly in preference to the TIPS. The 38bp between them is that premium, and it flipped the sign. On the house anchor the gap is +0.20% on CAPE and +1.59% on trailing earnings: thin at the 8th percentile since 1982, but neither negative nor a record — the true low is -1.14% in February 2000. The TIPS series starts in 2010, which is how a sixteen-year window masqueraded as a record.
What survives the correction — and it is the part that mattered — The decomposition holds. Since end-2021 the gap compressed 232bp, of which 93% is the real rate and only 17bp the earnings yield. The cushion was not spent on an equity melt-up; it was crowded out by the risk-free alternative. That is why an ordinary correction does not repair it — on cyclically-adjusted earnings equities are priced roughly where they were at end-2021, and everything that happened, happened in the discount rate. And note what this number is not: it is an aggregate yield gap, not the Delphic ERP. The house ERP is name-specific, measured by inversion (frameworks/erp_compression.md), with the market-implied aggregate near 4.3%. Different quantities.
Hormuz — the gas near-zero is gone, and we withdraw it — Our SoH file has carried for two months that no LNG carrier had transited since 11 July and that gas had no ship-to-ship workaround. Both are wrong. The Qatari carrier Al Maruna cleared the Strait around 7-8 September, and three LNG cargoes moved by STS in August. What replaces it is better: the gas shock is two shocks with opposite half-lives — a removable shipping premium, and 12.8 mtpa of damaged liquefaction on a three-to-five-year repair that no ceasefire fixes. Brent at $107.44 still carries the premium; the capacity loss is a level.
Regime & Technical Radar
Archetype unchanged — late-cycle, tightening-dominant — but the composition moved: growth has crossed BELOW ZERO while monetary prints a cycle high.
Archetype — NEUTRAL / TRANSITIONAL — tightening-dominant late cycle
Growth — -0.05 — BELOW ZERO for the first time this cycle
Monetary / Liquidity — +0.64 — a CYCLE HIGH; hike ~88% priced
Inflation — +0.48 — EASED; 5y5y de-anchor receding
Risk / Stress — -0.15 — benign, but VIX +7.9% to 17.10
UST 10y — 5.016% — through 5%, into the 5-5.5% flip zone
10y real — 2.624% — new high; carries the entire move
CAPE real gap (house) — +0.20% — 8th pctile since 1982, not a record
Brent — $107.44 — energy tail still live
The engine's composition changed this week even though the archetype did not. Growth crossed below zero to -0.05, its copper-momentum component at -0.70; Monetary printed +0.64, a cycle high, with the real-yield and nominal-level components both above +0.94; and Inflation EASED to +0.48 as the 5y5y de-anchor component came off. Read those three together and the sentence is: the tightening is intensifying while growth rolls over and inflation expectations recede. That is a real-rate squeeze.
Theme of the Day
A THIN CUSHION MEETS A 5% TEN-YEAR THE DAY BEFORE A HIKE — AND WE FIX OUR OWN NUMBER.
Three facts belong in one sentence. One: the 10-year broke 5% and the move was entirely real — breakevens and the 5y5y forward both fell. Two: the equity cushion is thin — a CAPE real gap of +0.20%, the 8th percentile since 1982 — and 93% of its compression is the discount rate, not the multiple. Three: the growth axis crossed below zero. The discount rate is rising, the cushion is thin, and the cash flows it discounts are no longer improving. We corrected the second of those today — yesterday we said the gap had gone negative on a record low, using an anchor our own standard rejects. It has not. The direction of the argument is unchanged and the decomposition is intact, but the alarm was overstated and the correction belongs on the front page rather than in a footnote. Own the things that do not need a cushion: real earnings, low leverage, energy, short duration.
Risks
- The cushion is thin, not absent. At +0.20% on CAPE — 8th percentile since 1982 — there is little between a real-rate move and the index, but the 2000 low was -1.14% and we should not have implied we were through it.
- A hawkish SEP. The hike is priced; the 2027 dot is not. A revision toward 2.6% core PCE would extend the real-rate move into the part of the curve that does the damage.
- The corridor re-escalates. A vessel was struck near Limah on 12 September. Brent at $107 already carries a premium that a single incident can widen.
- We are wrong about the decomposition. If the real-yield rise is term premium rather than expected short rates, this is a risk-premium event and our benign Risk/Stress reading is mis-set. We cannot currently split those two — a real gap, honestly stated.
- Rebalancing flow. With allocators at their most overweight equities versus bonds since 2002 on external estimates, the marginal seller may be price-insensitive and indifferent to our factor preferences.
The Call
THE CALL — fade nothing, hedge the discount rate. The hike is priced; the absence of a risk premium is not.
We do not fade the hike and we do not fade the oil. The Fed question was settled by the August core; tomorrow is about the dots. The live question is the one the tape asked on Monday: what does a synchronised DM hiking cycle do to an asset market whose risk premium has gone to zero? Our answer is that it does not crash it — credit is calm and nominal growth is robust — but it removes the market's ability to absorb rate moves, which is exactly what the growth axis crossing below zero on a 2bp real move is telling us. Positioning is unchanged and now better supported: quality, low leverage, real earnings, energy; duration light. What would change our mind: a breakeven-LED yield move, which would be a materially worse regime and which RECEDED this week; or real yields rolling over, which is the only benign repair path for the premium. What we withdraw this week: the Hormuz gas near-zero, in full, and with it any framing of the LNG shock as a single removable premium — 12.8 mtpa of damaged liquefaction is a level.