# Competence Map — Macro Study Program

| Topic | Status | Last probed | Notes |
|---|---|---|---|
| **Rates — term structure** | Strong | 2026-06-23 | Cal S1: spot/fwd right; par yield def wrong (confused with avg); zero-coupon qualifier missing. Session 2026-06-22: spot/fwd/par card drilled. Session 2026-06-23: L-S framework landed — PC1/PC2/PC3 decomposition applied cleanly to Taper Tantrum data |
| **Rates — real vs nominal** | Working | 2026-06-22 | Cal S1: pieces named but wrong on dominant mover — said inflation exp, actual is real yields/term premium. Session 2026-06-22: full decomposition recited cold; arithmetic clean (4.50−2.00=2.50; −50/−20=−30bp trap); Fisher vs market decomposition distinction learned |
| **Rates — curve dynamics** | Strong | 2026-06-23 | Cal S1: parallel/steepener/flattener mechanics right; missed butterfly; missed duration-vs-curve distinction. Session 2026-06-22: duration-vs-curve distinction articulated cold; butterfly structure named correctly. Session 2026-06-23: slope-vs-level framing applied to Taper Tantrum — correctly classified as level-dominated move with secondary slope (anti-curvature pattern, not PC3) |
| **Rates — term premium** | Strong | 2026-06-26 | Cal S1: def OK; called it "minor part" — empirically false. Session 2026-06-22: 250/−50/+50 swing recited by heart. Session 2026-06-23: L-S vs term-premium distinction now in Anki. Session 2026-06-25: Bernanke 2015 three explanations named; misidentified dominant driver (said savings glut, Bernanke argues term premium). Session 2026-06-26 (am): drill landed — TP compression dominant articulated, 2005 savings-glut contrast landed. Session 2026-06-26 (pm) Block 007: direction kink verified — TP high → long-end cheap (undervalued), flattener when 10y TP > 30y TP, steepener when 30y TP > 10y TP. Logged Strong |
| **Rates — PCA level/slope/curvature** | Working | 2026-06-25 | PC3 vs PC2 confusion (2026-06-23). Session 2026-06-25: PC3 mechanism landed (wings opposite to belly); Taper Tantrum decomposition clean (PC1-dominant, PC2-secondary, no PC3). Re-test next session |
| **Rates — CFTC positioning interpretation (NEW)** | Working | 2026-06-25 | Session 2026-06-25: dealer-short-cover squeeze mechanism learned; speculator length as marginal buyer (more fragile than asset-mgr length) |
| **Rates — Taper Tantrum case study (NEW)** | Working | 2026-06-26 | Session 2026-06-25: L-S decomposition clean; dominant driver wrong (said term premium; correct is Fed-policy expectations); trade structure wrong on instrument (butterfly; correct is 10s30s flattener); two-trigger rule read landed. Session 2026-06-26 (am): drill landed — anti-pattern butterfly-reflex cleared, 10s30s flattener chosen correctly, PC1+PC2 decomposition clean. Session 2026-06-26 (pm) Block 007: catalyst direction verified (dovish surprise invalidates flattener); PC3-emergence scenario partially articulated (observable pattern yes, causal mechanism partial — flagged for re-drill if surfaces). Working+ |
| **Rates — carry & roll-down mechanics (NEW)** | Working | 2026-06-27 | Session 2026-06-27 (Block 008): 10y zero-coupon trap cleared (T-Note with ~4.25% coupon); rolldown definition corrected (yield-drop-as-time-to-maturity-shortens, NOT pull-to-par); steepener rolldown signs on both legs landed (both negative); flattener asymmetry landed (Ilmanen: steepener = long-vol, flattener = short-vol); DV01-neutral sizing rule landed (face ratio = duration ratio, NOT yield ratio); dovish-pivot steepener constructed correctly with regime-shift invalidation framing; bond-price convention locked in (rally/sell-off on price, not yield) |
| **Rates — vol surface (RR/BF/vol-of-vol) (NEW)** | Working+ | 2026-07-14 | Session 2026-07-04 (Block 010): swaption-mechanics (payer = call on bond, not put) landed with correction; RR formula + sign convention + dealer flow interpretation landed; BF formula + anchoring + tail-risk premium landed; delta-space vs strike-space quoting distinction landed; vol-of-vol = dynamic measure (vs static RR/BF) landed; FOMC vol crush behavior + anti-pattern (long vol into known event) landed. **2026-07-05 cold-recall: Working+** (RR sign + dealer flow direction + BF read all clean; precision point locked: "payers are being bought, dealers short payer / long receiver, rally-hedge demand" not collapsed "dealers are short"). **2026-07-12 Block 015 cold-recall: demoted Working+ → Working** — Probe 1 slipped on (a) vol-is-symmetric-not-directional (collapsed to "expectations for a hike"), (b) dealer-flow direction in RR (1m payer vol richening does not transfer to 6m ATM long position value), (c) vol term structure vs BF (different axes — term structure is curve, BF is wings-vs-belly at single tenor). Vol-of-vol-the-size-of-vol-move-is-the-signal still firing cleanly. **2026-07-14 Session 24 Block 016 cold-recall re-drill: Working → Working+** — Probe 1: P&L direction right (long-vol, long the option), vol characterization right on smile (negative-move vol > positive-move vol = observed asymmetry), precision refinement that Black's model is symmetric (lognormal assumption) so the observed asymmetry lives in the smile/skew layer above Black. All precision points clean. |
| **Rates — Black's model + limitation (NEW)** | Working+ | 2026-07-14 | Session 2026-07-04 (Block 010): Black's model mechanics + annuity factor (A = PV01 of underlying swap) landed; single most important limitation (lognormality → no skew, no vol-of-vol, no jumps) landed; anti-pattern (positive Black vega + short vol-of-vol = lose on convexity) landed. Black prices options, not rates. **2026-07-05 cold-recall: Working+** (lognormal mechanism + trade-off framework re-articulated; precision point locked: lognormal structurally rules out negative rates and fat tails, not just "simplistic"). **2026-07-12 Block 015 cold-recall: demoted Working+ → Working** — Probe 1 part (a) — lognormal mechanism was clean, but Probe 7 part (a) showed the related precision slip (Black vs normal bp vol distinction) had not been re-anchored as a separate sub-item. **2026-07-14 Session 24 Block 016 cold-recall re-drill: Working → Working+** — covered implicitly via Probe 1 vol-surface probe (Black is symmetric by assumption, the smile/skew lives above Black = Black vega/price changes when vol shifts but doesn't capture the asymmetric tail pricing). |
| **Rates — rate-cut probability extraction (NEW)** | Working | 2026-07-12 | Session 2026-07-04 (Block 010): four-layer extraction framework landed (ATM vol → ATM receiver → 25Δ → RR/BF); risk-neutral formula (N(−d2) for ATM, d2 = −σ√T/2) landed; normal (bp) vol vs Black (%) vol rationale landed (bounded distribution makes lognormal overstate moves); CME FedWatch / SOFRWatch as cleaner alternatives to ZN vol surface for direct cut-probability extraction noted. **2026-07-05 cold-recall: Working+ with re-teach** (chain did NOT land cold; full re-teach delivered including 25Δ interpretation, annual vol vs single-meeting event mismatch, cross-asset transmission logic; risk-neutral vs real-world probability precision point added; 8 scenario cards + 1 cross-asset transmission card drafted to scratch). **2026-07-12 Block 015 cold-recall: Working (no grade change)** — Probe 7 part (b) showed the framework sequence is intact, but the probability math (P(cut ≥ 25bp) = N(0) = 50% when forward = strike) did not fire — user recalled as "cut is quite likely" when 50% is the correct read. The four failure modes from the prior cold-recall also slipped on the BF sign convention (Probe 7 part d). Re-drill next session. |
| **Rates — ZN surface rep live data (NEW)** | Working | 2026-07-12 | Session 2026-07-04 (Block 010): first applied use of live vol-surface dataset. ZN ATM 5.42-5.62% Black × 73 DTE → 1-sigma move in 10y yield = ±27bp; Black-implied probability of 50bp move = ~9% (vs empirical 20%+ = vol-of-vol premium in action). TYVIX 4.71 (52-wk range 0-5.15 = 99th percentile) with -5.4% one-day drop = vol normalizing off stress event; the drop itself is vol-of-vol in motion. **2026-07-05 cold-recall: Working+** (regime read clean: TYVIX near 52wk high = anxious/big moves priced, not direction; vol-of-vol-in-motion insight nailed: the size of the vol move IS the vol-of-vol signal). **2026-07-12 Block 015 cold-recall: Working (no grade change)** — Probe 1 sub-item: vol-of-vol-precision-point still firing cleanly (size of vol move = vol-of-vol signal); the slip on Probe 1 was in the directional/positioning layer, not the vol-of-vol layer. |
| **Rates — MBS pass-through structure (NEW)** | Working+ | 2026-07-12 | Session 2026-07-06 (Block 011): single-class pro-rata pass-through after servicer (~25bp) and agency g-fee (~50-100bp); agency guarantee covers BORROWER default (not issuer); CMO/REMIC explicitly out of scope. **2026-07-12 Block 015 cold-recall: Working+ re-affirmed** — not directly probed in Block 015 (Probe 2 covered the broader basis-trade substrate which assumes pass-through structure); pass-through substrate implicit in Probe 2 response. |
| **Rates — negative convexity mechanism (NEW)** | Working | 2026-07-12 | Session 2026-07-06 (Block 011): WAM-extends-on-rally + WAC-drift-selection; MBS underperforms Treasury of equivalent duration in BOTH directions. Prepayment option framing: borrower owns free American call on MBS strike=par, investor short the option, negative convexity = cost. **2026-07-12 Block 015 cold-recall: Working (no grade change)** — Probe 2 part (a) substrate fired (prepay acceleration mechanism on rally correct), but the direction-of-trade on the long-MBS position (you're SHORT the option, option value rising = MTM hit) was the precision gap. Mechanism intact, MTM direction slipped. |
| **Rates — MBS-Treasury basis trade (NEW)** | Working | 2026-07-12 | Session 2026-07-06 (Block 011): DV01-matched long 30y UMBS TBA / short 30y UST. Hedge ratio ~2.3x MBS face to Treasury face. Net carry ~30-50bp annualized. Convexity-adjusted breakeven = +200 PSA over hold period (refi wave doubling prepay speed). Live data: 30y UMBS 5.58% coupon vs 30y UST 4.97% = +61bp gross secondary spread; net basis negative (option-overpriced). **2026-07-12 Block 015 cold-recall: Working (no grade change, but precision slip)** — Probe 2: (a) gross spread widens on rally (WAC-drift on dollar price, not yield basis) — user said "narrows" with wrong-mechanism framing; (b) net basis = gross spread − option cost, option cost rises with prepay acceleration — partial; (c) carry decomposition including convexity bleed — user bailed. Substrate intact (long-MBS-short-Treasury, +200 PSA breakeven, net basis option-overpriced), precision on direction-of-move (gross spread widens, convexity bleed negative past breakeven) slipped. |
| **Rates — PSA prepayment mechanics (NEW)** | Working | 2026-07-12 | Session 2026-07-06 (Block 011): 100 PSA = 6% CPR baseline, 0.2%/month ramp month 1-30, then flat 6% CPR. 300+ PSA = refi wave, 500+ PSA = aggressive, 800+ PSA = peak refi. Current 5.5% coupon pool running ~200-250 PSA in 2026-07 environment. **2026-07-12 Block 015 cold-recall: Working (no grade change)** — Probe 2 substrate fired (prepay acceleration on rally + PSA ramp mechanics implicit in the +200 PSA breakeven and 200 → 600 PSA scenario); precision points already covered under negative convexity and basis-trade items. |
| **Rates — SOFR construction (NEW)** | Working | 2026-07-12 | Session 2026-07-07 (Block 012): volume-weighted median of overnight Treasury-repo rates (tri-party + GCF + FICC-cleared bilateral), published by NY Fed ~8am ET. Secured, overnight, Treasury-collateralized only. **Pre-read gap surfaced:** ARRC Closing Report covers institutional arc but not construction — taught in-session. **2026-07-12 Block 015 cold-recall: Working (no grade change)** — not directly probed in Block 015 (Probe 3 covered the broader SOFR-OIS basis which assumes SOFR construction); SOFR construction substrate implicit. |
| **Rates — OIS structure (NEW)** | Working | 2026-07-12 | Session 2026-07-07 (Block 012): fixed-for-floating swap on geometric-average overnight rate. Used for discounting because overnight loans have near-zero credit risk. Precision locked: geometric (compounded) average = economically equivalent to rolling overnight loans. Hull Ch. 7 SOFR-OIS section landed. **2026-07-12 Block 015 cold-recall: Working (no grade change)** — not directly probed in Block 015 (Probe 3 covered the broader SOFR-OIS basis which assumes OIS structure); OIS substrate implicit. |
| **Rates — repo mechanics + GC/specific (NEW)** | Working | 2026-07-12 | Session 2026-07-07 (Block 012): cash borrower posts collateral + pays repo interest; cash lender posts cash + earns interest. Two orthogonal axes: GC vs specific (collateral flexibility) and tri-party vs bilateral vs FICC-cleared (settlement mechanics). Specific collateral on special (e.g., CTD) can repo at NEGATIVE rates. **2026-07-12 Block 015 cold-recall: Working (no grade change)** — not directly probed in Block 015 (Probe 3 covered the broader SOFR-OIS basis which assumes repo mechanics); repo substrate implicit. |
| **Rates — SOFR-OIS basis + tenor-vs-credit (NEW)** | Working+ | 2026-07-14 | Session 2026-07-07 (Block 012): basis = TERM-UNSECURED-bank-funding premium over riskless, **NOT** a credit spread (both legs are near-zero credit). Tenor risk only. The credit-spread read is term-unsecured-minus-OIS. Historical: 2008 hit 350bp+, March 2020 hit 130bp+, normal 0-15bp. **Precision locked:** SOFR-OIS = tenor; term unsecured - OIS = tenor + credit. **2026-07-12 Block 015 cold-recall: demoted Working+ → Working** — Probe 3: (a) "SOFR-OIS = tenor" precision point slipped — user used "SOFR-OIS spread" as proxy for "bank funding risk" rather than tenor premium, collapsing the distinction that was supposed to be locked; (b) quarter-end direction wrong — user said "funding component shrinks at quarter-end" when it actually WIDENS (banks pull back from repo, balance-sheet-driven scarcity). **2026-07-14 Session 24 Block 016 cold-recall re-drill: Working → Working+** — Probe 3: substrate mostly intact ("bank funding" intuition right direction, quarter-end widens right), precision gap on the tenor-vs-bank-funding mechanism explicitly re-anchored: basis = price of 2y term funding above rolling overnight, both legs near-zero credit, term premium driven by SLR denominator turnover + G-SIB surcharges + regulatory friction. Mechanism (tenor) vs. surface observation (bank funding stress) is now distinguished cleanly. |
| **Rates — DKW three-component decomposition (NEW)** | Working | 2026-07-12 | Session 2026-07-09 (Block 013): three components named correctly (expected inflation + IRP + liquidity premium) per DKW Section 2. Naive read T10YIE = expected inflation wrong by ~80-150bp wedge. DKW 2010s historical estimates: expected ~60-100bp, IRP ~30-80bp, liquidity ~30-50bp at 10y. **2026-07-12 Block 015 cold-recall: Working (no grade change, but precision slip)** — Probe 4 part (b) user said "ignores liquidity premium" as the sole reason the naive read is wrong, but DKW has THREE components (expected + IRP + LP). Substrate intact, only-one-component-named precision gap. |
| **Rates — FRED breakeven series (T10YIE/T5YIFR/DFII10/DFII30) (NEW)** | Working | 2026-07-14 | Session 2026-07-09 (Block 013): T10YIE = DGS10 - DFII10 (a spread, not a yield); T5YIFR = 5y5y forward breakeven (years 5-10 specifically), FRED formula given; DFII10/DFII30 = real-yield levels. Mechanical difference between T10YIE (spread) and DFII10 (level) clean. T5YIFR = forward measure, less contaminated by near-term cyclical noise. **Pre-read gap:** FRED methodology pages not done — taught in-session per probe-vs-teach fallback. **2026-07-12 Block 015 cold-recall: demoted Working → Working with re-teach** — Probe 4 part (a) user said "naive expected inflation is 2.3% as indicated by DFII10" but DFII10 is the REAL YIELD, not expected inflation. Real yield and expected inflation are the two legs of T10YIE, not the same thing. Naive expected inflation = T10YIE = 2.25%, not 2.30%. **2026-07-14 Session 24 Block 016 cold-recall re-drill: Working (no grade change — slip RECURRED)** — Probe 4: same precision gap as Block 015 — user labeled DFII10 as expected inflation (2.30%) and T10YIE as real yield (2.25%), which is inverted. The real-yield vs breakeven label distinction is a recurring slip across two sessions. Anki Card 7 re-drill pushed (note ID 1784060117580) but precision is not yet substrate-locked. Carried to Block 017 cold-recall pass for further verification. |
| **Rates — Liquidity premium sign convention (NEW)** | Working+ | 2026-07-14 | Session 2026-07-09 (Block 013): chain articulated — TIPS less liquid than nominals → TIPS yields bid higher (TIPS cheaper) → measured T10YIE UNDERSTATES true breakeven → liquidity premium added back as POSITIVE correction. DKW historical 10y LP estimate ~30-50bp, compressed post-2010s as TIPS market deepened. Positive LP = TIPS illiquid; LP can flip negative briefly in flight-to-quality episodes. **Precision locked.** **2026-07-12 Block 015 cold-recall: demoted Working+ → Working** — Probe 4 part (a) related precision: real yield (DFII10 = 2.30%) and expected inflation (T10YIE = 2.25%) are distinct measures; user collapsed them. LP precision itself is locked, but the related real-yield-vs-expected-inflation distinction slipped. **2026-07-14 Session 24 Block 016 cold-recall re-drill: Working → Working+** — Probe 4 LP sign chain articulated cleanly (less liquid TIPS → bid yields higher → measured T10YIE understates true breakeven → LP positive correction). LP substrate re-locked. Related real-yield-vs-breakeven-label distinction is the recurring slip and is tracked separately under "FRED breakeven series" row. |
|| **Rates — Cross-block signal-layer discipline (NEW)** | Working+ | 2026-07-14 | Session 2026-07-09 (Block 013): rates-microstructure read = three legs (Block 011 MBS basis trade-leg + Block 012 SOFR-OIS funding-leg + Block 013 breakeven inflation-layer). User chose 10s30s flattener as cleanest expression of long-end richening view in trade-implication probe (Q6), correctly distinguishing it from long-TIPS/short-nominal breakeven trade (liquidity-contaminated) and short-MBS-basis trade (mortgage-noise-contaminated). **2026-07-11 Block 014 update:** third leg (cross-currency basis) added; diagnostic power = separation of which combination of three legs moves = which layer of dollar-funding plumbing broke. **2026-07-12 Block 015 cold-recall: Working (no grade change, but sequence substrate slip)** — Probe 8 part 3 (diagnostic priority): user said "MBS-Treasury first, then SOFR-OIS, then cross-currency" when the substrate says SOFR-OIS FIRST (leading indicator — domestic bank funding breaks first in a crisis), cross-currency SECOND, MBS LAST (lagging indicator — prepayment/option, requires a rate move to actually trigger). **2026-07-14 Session 24 Block 016 cold-recall re-drill: Working → Working+** — Probe 5: SOFR-OIS first, cross-currency second, MBS-Treasury last — sequence substrate restored cleanly. Block 016 main material expanded the framework with a two-ordering diagnostic (policy-channel vs. balance-sheet-channel) where the diagnostic ORDER reveals the TYPE of shift, not just the magnitude. Working+ on the expanded substrate. |
| **Rates — Fed reaction function (NEW)** | Working+ | 2026-07-14 | Session 2026-07-14 (Block 016): dual mandate (max employment + price stability) loaded with operational vs. rhetorical asymmetry — Fed *says* price stability, *does* employment (asymmetric loss function). 2% inflation target = 2012 FOMC formalization, amended to "flexible average" in 2020 framework review (FAIT). Forward-guidance taxonomy: Odyssean (commitment, e.g., 2008-2014 "extended period") vs. Delphic (forecast, e.g., 2014+ dot plots, "data-dependent"). 2014 O→D transition trigger: forward-guidance trap (Odyssean commitment compressed long-term yields, creating asset-purchase feedback and financial-stability concerns). Three regime-change anchors internalized: Powell 2018 "long way from neutral" (Delphic, market misread as Odyssean), 2020 framework review (FAIT formalization), 2022-2023 hiking cycle (525bp in 16 months, largest since 1980s, "transitory" abandoned Aug 2022 Jackson Hole). Substrate taught inline due to Hull Ch. 30 HJM/LMM gap (read but doesn't cover monetary policy). Re-probe Block 022 (Batch 3 verification). |
| **Rates — Reaction function diagnostic (NEW)** | Working+ | 2026-07-14 | Session 2026-07-14 (Block 016): two-ordering framework — (a) Reaction function shift (Fed pivot, framework review, dot-plot surprise) → vol surface first (minutes), SOFR-OIS basis second (hours-days), breakevens last (weeks); (b) Funding/dollar crisis (March 2020, 2008, quarter-end stress) → SOFR-OIS basis first (hours), cross-currency basis second (hours-days), MBS-Treasury basis last (days-weeks). **Synthesis insight (load-bearing):** diagnostic ORDER reveals TYPE of shift — surface-first = policy channel (expectations repricing, no transmission friction); SOFR-OIS-first = balance-sheet channel (transmission friction, rate cuts won't transmit 1:1 to term funding). Live-data read mid-2026: 2y OIS 4.33% (USDSB3L2Y), 2y term SOFR 4.33% + 72bp = 5.05%, +72bp basis = elevated tenor premium, constrained band (not crisis; normal 30-50bp). Three-channel synthesis: dot plot gradual cuts 3.75%→3.0625% LR; SOFR-OIS elevated tenor premium = balance-sheet channel friction; T10YIE 2.26% slightly above 2% target = FAIT still operational. Discretionary positioning: long basis swap (receive term SOFR / pay OIS) = bet spread persists or widens; short basis = bet on cuts transmitting cleanly. Re-probe Block 022 (Batch 3 verification).
| **Rates — Curve regime classification (NEW)** | Working+ | 2026-07-15 | Session 25 (Block 017): four-regime taxonomy loaded with leg-movement rule (which leg moves more + which direction determines the regime). Bull steepener = 2y falls more, curve steepens (Fed pivot / cut pricing). Bear steepener = 10y rises more, curve steepens (term-premium rebuild, supply/fiscal). Bull flattener = 10y falls more, curve flattens (flight-to-quality, recession signal). Bear flattener = 2y rises more, curve flattens (Fed-policy repricing, late cycle). **Regime = move, not state** rule locked (Q5 clean): a snapshot of 2s10s alone does not classify the regime — needs reference period + yield direction. Live-data read 2026 YTD: BEAR FLATTENER (2y +79bp 3.47→4.26%, 10y +43bp 4.19→4.62%, 2s10s −32bp 0.72→0.40), HY OAS −11bp 2.83→2.72 (risk-on). Two-channel driver decomposition: short end = Fed-policy repricing (Warsh premium + slower-cut pricing), long end = real-yield channel (DFII10 +42bp 1.94→2.36%) with breakevens FLAT (T10YIE 2.25→2.25%, T5YIE 2.28→2.28%) — NOT inflation-driven. **Pre-block probe correction:** Q1 + Q3 had leg-assignment slips ("10y falls more" treated as steepener, but it's a flattener — long end moving more is steepener, short end moving more is flattener); corrected via the four-box table. **Pre-existing card audit (Session 25):** 25 candidate notes scanned in the steepener/flattener cluster; 2 errors found + 1 confusing card; 1781690237391 (steepener label/direction both wrong) and 1781690237395 (bull flattener with INVERTED leg assignment) fixed in place; 1781690237399 (category-error parenthetical) deleted. 12 atomic Anki cards pushed (4 vocabulary + 8 scenario classification), all yield-anchored, no distractor-leak paths. Re-probe Block 022 (Batch 3 verification). |
| **Rates — Regime driver decomposition (NEW)** | Working+ | 2026-07-15 | Session 25 (Block 017 main material concept 1 + live-data probe): two-channel driver decomposition for the 2026 YTD bear flattener. Short end: Fed-policy repricing (Warsh premium, slower-cut pricing, hawkish dot plot, OIS path repricing). Long end: real-yield channel (DFII10 +42bp YTD = term-premium rebuild or r* repricing higher), breakevens FLAT (T10YIE 0bp YTD, T5YIE 0bp YTD) — NOT inflation. The 10y half is NOT being pulled up by inflation expectations; it's being pulled up by the term-premium / real-rate channel. HY OAS −11bp YTD = risk-on tone, no recession. **Precision locked:** 10y nominal decomposition = real yield + breakeven ≈ 4.19% → 4.61% (+42bp), consistent with all of the 10y YTD move being real-yield driven. Block 020 (ACM/KW term premium in practice) is the natural extension — once you can decompose a 2s10s move into expectations vs. term-premium components, you can build a regime trade that expresses a view on one component specifically. Re-probe Block 022 (Batch 3 verification) and Block 020. |
| **Rates — Yield-direction precision (NEW)** | Working+ | 2026-07-15 | Session 25 (Block 017 probe-form): yield-anchor verb discipline applied to user recall. The user said "the 10y falling" in Q1, ambiguous between yield-anchored (yields fell) and price-anchored (prices fell, yields rose). Substrate-correct interpretation: bull steepener has 10y yields falling. Locked at probe time with explicit yield-anchoring required in regime descriptions. **Mentor-side mirror:** Q3 mentor response had a leg-assignment slip on the bear-steepener mechanism — caught and corrected via the four-box table. Re-probe Block 022 (Batch 3 verification). |
| **Rates — FRED breakeven series (T10YIE/T5YIFR/DFII10/DFII30) (NEW)** | Working+ | 2026-07-15 | Session 2026-07-09 (Block 013): T10YIE = DGS10 - DFII10 (a spread, not a yield); T5YIFR = 5y5y forward breakeven (years 5-10 specifically), FRED formula given; DFII10/DFII30 = real-yield levels. Mechanical difference between T10YIE (spread) and DFII10 (level) clean. T5YIFR = forward measure, less contaminated by near-term cyclical noise. **Pre-read gap:** FRED methodology pages not done — taught in-session per probe-vs-teach fallback. **2026-07-12 Block 015 cold-recall: demoted Working → Working with re-teach** — Probe 4 part (a) user said "naive expected inflation is 2.3% as indicated by DFII10" but DFII10 is the REAL YIELD, not expected inflation. Real yield and expected inflation are the two legs of T10YIE, not the same thing. Naive expected inflation = T10YIE = 2.25%, not 2.30%. **2026-07-14 Session 24 Block 016 cold-recall re-drill: Working (no grade change — slip RECURRED)** — Probe 4: same precision gap as Block 015 — user labeled DFII10 as expected inflation (2.30%) and T10YIE as real yield (2.25%), which is inverted. The real-yield vs breakeven label distinction is a recurring slip across two sessions. Anki Card 7 re-drill pushed (note ID 1784060117580) but precision is not yet substrate-locked. Carried to Block 017 cold-recall pass for further verification. **2026-07-15 Session 25 Block 017 cold-recall re-drill: Working → Working+** — Probe at session open: user articulated DFII10 = inflation-indexed 10y Treasury securities yield (real yield) at 2.36%, T10YIE = 10y breakeven inflation rate (nominal 10y minus 10y TIPS yield). Distinction-pair precision locked; construction (T10YIE = nominal − TIPS) explicitly named. Recurring-slip flag closed; Anki Card 7 single-fact re-drill is sufficient for retention. |
| **Rates — SOFR-OIS basis swap structure (NEW)** | Working+ | 2026-07-25 | Session 2026-07-14 (Block 016 Q10): basis swap = receive/pay one floating rate (e.g., compounded OIS) vs. receive/pay a different floating rate (e.g., compounded term SOFR). Both legs float — pure-spread trade, no directional rate view (vs. regular swap which has one fixed leg and a directional view on the floating rate). Long basis = receive term SOFR, pay OIS = bet spread persists or widens. Short basis = pay term SOFR, receive OIS = bet spread compresses (cuts transmit cleanly, balance-sheet friction eases). **Naming audit 2026-07-25 (post-external-feedback):** "SOFR-OIS basis" historically used in the deck to refer to the bank-balance-sheet stress read — that lives in EFFR−SOFR (or BGCR−SOFR), NOT in Term SOFR-OIS (which is a fixing-convention basis, low single-digit bp normally). The actual "bank-funding stress" substrate lives in EFFR-SOFR. P&L framing also updated: basis is struck at zero PV; P&L = realised basis minus forward-implied basis at entry, NOT "quoted basis level." Anki Card 8 (note 1784060117585) deleted; replacement cards added (forward-P&L mechanic + two-basis taxonomy). 7 Category-B cards renamed "SOFR-OIS" → "EFFR-SOFR" to reflect substrate. |
|| **FX — CIP / UIP** | Working+ | 2026-07-23 | Cal S1 (2026-06-21): explicit gap — said "I have forgotten"; foundational concepts not internalized. **2026-07-11 Block 014:** Mechanism clean (USD at forward discount when USD rates > JPY rates, not forward premium); basis = residual after rate-differential-implied forward; dealer-intermediation vs arbitrage distinction new (the basis is the dealer's spread for warehousing the imbalance, not an arbitrage failure). Upgraded Weak → Working. **2026-07-12 Block 015 cold-recall: Working (no grade change)** — Probe 5 part (a) substrate fired (basis is the residual after stripping out the rate-differential-implied forward), instrument-confusion anti-pattern (Cbonds swap points vs basis) handled cleanly — user did not treat swap points as the basis. **2026-07-23 Block 023 (Session 31): Working → Working+ (Target 1 CIP mechanics main material landed)** — no-arbitrage frame locked (forward = interest rate differential, when basis breaks someone absorbs a cost); negative basis = balance-sheet cost to dealers (SLR/leverage-ratio regulatory constraint, not arbitrageur capture); hedger A/L framing locked (Japanese pension fund sells USD/JPY forward, fair hedge cost = US Treasury yield − JPY rate, negative basis = 25bp extra cost = dealer compensation). |
|| **FX — cross-currency basis regime classification (NEW)** | Working | 2026-07-12 | Session 2026-07-11 (Block 014): 3m USD-JPY basis three regimes with bp thresholds. Normal = -10 to -30bp (dealer capacity abundant, carry trades work). Constrained = -30 to -80bp (dealer capacity tightening, dollar strength, EM weakness, basis-trade unwinds). Crisis = wider than -100bp (dealer capacity exhausted, Fed swap lines likely active; 2008, March 2020). **2026-07-12 Block 015 cold-recall: Working (no grade change, but precision slip)** — Probe 5 part (c): cyclical driver is regulatory calendar + stress events (not generic "supply/demand"); -75bp is at the UPPER EDGE of constrained (-30 to -80bp), not "approaching crisis" (which would be > -100bp). Substrate intact, threshold precision slipped. |
|| **FX — DTV term structure of cross-currency basis (NEW)** | Working+ | 2026-07-23 | Session 2026-07-11 (Block 014): DTV 2018 finding — basis is largest at short tenors (1m, 3m) and shrinks monotonically toward zero at 5y. Shape is fingerprint of balance-sheet/regulatory constraint — short-tenor basis trades cycle through bank balance sheets more frequently under SLR / leverage ratio denominator; long-end basis near zero because longer holding periods let balance-sheet constraints amortize. **2026-07-12 Block 015 cold-recall: Working (no grade change)** — not directly probed in Block 015 (Probe 5 covered regime classification + cyclical/structural decomposition, both of which assume DTV term structure substrate). **2026-07-23 Block 023 (Session 31) Probe 4: Working → Working+ (mechanism refinement)** — user said "velocity in moves" (price-volatility framing); corrected to velocity = dealer balance-sheet *turnover*, not price volatility. The constraint binds where the dealer has to roll the position most frequently (short tenor, high frequency), not where price moves most. Substrate firing on the binding-tenor frame, mechanism precision locked. |
|| **FX — dealer-intermediation vs CIP arbitrage (NEW)** | Working+ | 2026-07-12 | Session 2026-07-11 (Block 014): CIP basis trade = dealer-intermediation trade, balance-sheet-constrained (the bank warehouses the imbalance, earns the basis as required return for committing scarce balance sheet). CIP arbitrage = textbook opportunistic trade, price-only, leverage-dependent. **Precision locked:** the basis is the dealer's spread for warehousing the imbalance, not a textbook arbitrage failure. Post-2008 the basis is fundamental, not a footnote — global dollar funding runs through dealer balance sheets, FX swap is the plumbing. **2026-07-12 Block 015 cold-recall: Working+ re-affirmed** — Probe 5 part (b) substrate fired (dealer-balance-sheet constraint is the dominant driver of the post-2008 basis, consistent with DTV's empirical finding that basis is largest at short tenors). |
|| **FX — BIS data (NEW)** | Working | 2026-07-12 | Session 2026-07-11 (Block 014): BIS Triennial 2022 — global FX turnover $7.5T/day (down from 2019 peak $8.0T post-COVID, still 1.5x 2010). FX swap = $3.8T/day largest single instrument (the basis-trade plumbing). USD appears on one side of ~88% of all FX trades, structurally stable across 2010-2022 surveys. EUR ~31%, JPY ~17% (one-sided, overlapping). **2026-07-12 Block 015 cold-recall: Working (no grade change)** — not directly probed in Block 015 (Probe 5 covered the live-data / basis decomposition which assumes BIS data substrate). |
|| **Cross-asset — Rates ↔ FX (NEW)** | Working | 2026-07-11 | Session 2026-07-11 (Block 014): cross-currency basis as the bridge — a *dollar-funding-stress* read priced in FX swap markets. Phase 3 block 073 will do the full Rates↔FX cross-asset treatment; Block 014 is the canonical rates-side foundation. Re-probe Block 073 (Phase 3) |
| **FX — CB swap-line mechanism (two-step, jurisdictional primary) (NEW)** | Working+ | 2026-07-27 | Session 33 (Block 024 pre-read side-session, McCauley + Schenk 2020): Two-step intermediation Fed → foreign CB → foreign private banks. **Primary reason = regulatory jurisdiction** (Fed has no legal authority to supervise or lend to foreign private banks). Credit risk insulation is secondary. Foreign CB is the only entity with (a) legal authority in its jurisdiction, (b) supervisory information on stressed banks, (c) operational plumbing to on-lend at scale. Substrate landed after one re-anchor — user initially cited credit risk alone, missed jurisdictional primary. |
| **FX — modern swap network history (2007 origin, 2008 unlimited, 2013 standing, 2020 EM expansion) (NEW)** | Working+ | 2026-07-27 | Session 33 (McCauley + Schenk 2020). Dec 2007: Fed opens $20bn to ECB and SNB, first at scale, modern network origin. Oct 2008: expanded to all major CBs, lines lifted to unlimited, peak GFC. Feb 2010: unlimited lines wound down (14+ month unwind). Oct 2013: lines reactivated with capped amounts, only 5 CBs (ECB, BoE, BoJ, SNB, BoC) — standing network born. Mar 15 2020: Fed cuts standing rate from 50bp OIS to 25bp, drawings explode (ECB ~$150bn in weeks). Mar 31 2020: network expanded to 9 more CBs including some EM, first EM standing access. |
| **FX — asymmetry / public-good argument (Fed only dollar LOLR) (NEW)** | Working+ | 2026-07-27 | Session 33 (McCauley + Schenk 2020). Fed is the only CB that can print dollars. No other CB can be the dollar-supplier of last resort. The Fed's swap line is therefore a public good to the global banking system, not a bilateral favor. "Exorbitant privilege" applied to funding, not to currency. Standing network exists *because* of the GFC — pre-2008 Fed treated dollar-funding stress as foreign; post-2008 Fed treats it as systemic (foreign CBs' domestic banks have dollar liabilities to US banks). |
| **FX — signal stack diagnostic (foreign CB repo ops → basis → draw → Fed announcement) (NEW)** | Working+ | 2026-07-27 | Session 33 (McCauley + Schenk 2020). Four signals, earliest to latest: (1) foreign CB dollar repo operations — operational telegraph, (2) cross-currency basis move — real-time continuous, (3) foreign CB formal draw request — intermediate, (4) Fed announcement — confirmation, 2-3 week lag. The basis is the *earliest continuous* signal; the foreign CB repo ops are the *earliest operational* signal. PM edge: position before the Fed acts; basis is the entry trigger; Fed tool choice is the confirmation. |
| **FX — stress-telegraph read (size + tenor + bid-to-cover) (NEW)** | Working+ | 2026-07-27 | Session 33 (McCauley + Schenk 2020). Three signals that identify the telegraph: (1) size step-up — >2x trailing 4-week average; normal ECB $ repos ~$50-200m per op, weekly/biweekly; stress: multi-billion daily, (2) tenor lengthening — O/N/1-week normal, ≥2-week stress; longer tenor = CB pre-funding future stress, (3) bid-to-cover > 2x — quantity-rationing signal. Operational read: "ECB stress index" = weekly $ repo notional + tenor vs trailing 4-week average. >2x + ≥1-week = telegraph. |
| **FX — quantity-rationing mechanism (bid-to-cover > 2x = CB balance sheet is binding) (NEW)** | Working+ | 2026-07-27 | Session 33 (McCauley + Schenk 2020). Bid-to-cover > 2x on a foreign CB dollar repo = the binding constraint is *quantity*, not price. CB is rationing dollar supply; its balance sheet (or legal authority) is the limit. Raising the price doesn't expand quantity; only the CB can expand quantity, and only via the Fed swap line or another facility. The next move is a swap-line draw request, not a price hike. Mechanism landed clean on first probe attempt. |
| **FX — path-vs-direction discipline (speeches move magnitude/path, not direction) (NEW)** | Working+ | 2026-07-27 | Session 33 (McCauley + Schenk 2020). Process rule: central bank speeches during a basis trade move *magnitude and path*, not *direction*. Only the foreign CB's formal draw request flips direction. Worked example: BoE governor says "we are seeing global dollar funding pressures but UK markets are functioning orderly" — this is *bullish* for a long-USD/EUR basis trade (BoE out of queue, Fed response targeted at ECB, faster, ECB-specific). Substrate landed after one re-anchor — user initially read BoE speech as ambiguous or negative for thesis. |
| **FX — time-horizon discipline (broader stack = wider stop, not tighter) (NEW)** | Working+ | 2026-07-27 | Session 33 (McCauley + Schenk 2020). Process rule: when signal stack broadens (multiple CBs flagging stress, not just one), direction stays same but path shifts — slower Fed action, larger magnitude, longer time in trade, **wider** stop-loss (not tighter), **wider** profit-take. Tightening stop while extending horizon is a category error — it cuts the upside before the Fed acts. Substrate landed clean on first probe attempt. |
| **FX — global financial cycle framework (Borio) (NEW)** | Working+ | 2026-07-27 | Session 33b (Borio et al. 2016). Global financial cycle is real and measurable: VIX, cross-border bank flows, dollar funding, asset prices co-move, driven primarily by US monetary policy. Not a metaphor. Three implications: (a) CBs respond to the *same* global cycle, not independently; (b) CB's reaction function is *conditioned on* the Fed's stance; (c) FX intervention is most credible when *augmenting* the global cycle, not fighting it. Substrate landed clean on first probe attempt. |
| **FX — trilemma → dilemma for small open economies (NEW)** | Working+ | 2026-07-27 | Session 33b (Borio et al. 2016). Standard trilemma (independent MP / free capital / FX stability) collapses to a *dilemma* for small open economies because the global cycle transmits through capital flows regardless of what the small CB does. Small CB's actual choice: FX stability (via intervention / capital controls) OR accept the global cycle transmission. Substrate landed on first attempt. |
| **FX — intervention as signal, not flow (NEW)** | Working+ | 2026-07-27 | Session 33b (Borio et al. 2016). Intervention volume is tiny relative to daily FX turnover (e.g., SNB $50bn over 3 months ≈ $500m/day ≈ 0.1% of daily JPY turnover). Mechanically too small to move the rate. Works instead by signaling CB intent and changing positioning. PM read: watch intervention as a signal of CB intent and constraint, not as a market-moving flow. Substrate landed on first attempt. |
| **FX — Borio-aligned defined-horizon fade (default) (NEW)** | Working+ | 2026-07-27 | Session 33b (Borio et al. 2016). Process rule: when a CB intervenes against the global cycle, the Borio-aligned default is a defined-horizon fade (days to weeks, pre-decided exit). NOT a multi-month carry trade. Same direction, different horizon, different sizing, different entry trigger. Substrate landed after one re-anchor — user initially framed as carry trade (right direction, wrong population/horizon). |
| **FX — capitulation bet upgrade path (NEW)** | Working+ | 2026-07-27 | Session 33b (Borio et al. 2016). Process rule: upgrade from Borio-aligned defined-horizon fade to capitulation bet requires (a) constrained CB intervention capacity (declining reserves, political pressure) AND (b) absence of real-economy justification for the level (e.g., not PPP-overvalued, no terms-of-trade support). Both missing → stay with defined-horizon fade. Outlier: SNB 2011-2015 held 1.20 floor for 3.3 years because EUR/CHF at 1.20 was PPP-overvalued (real-economy justification) AND CHF was only European safe haven (no constrained capacity). Substrate landed after user correctly identified Borio-aligned fade as default. |
| **FX — language-frame ladder (CB swap-line communications) (NEW)** | Working+ | 2026-07-27 | Session 33b (CB swap-line communications). Three frames in CB swap-line communication language, in order: Frame 1 = "standing arrangement / precautionary" (routine plumbing, no stress language); Frame 2 = "address dollar funding pressures / support smooth functioning" (acknowledged stress, specific problem named); Frame 3 = "extraordinary measures / backstop / persistent" (systemic stress, network expansion). Watch the *escalation* of language as a leading indicator. Worked examples: 2013 reactivation (Frame 1), March 15 2020 rate cut (Frame 1-2), March 31 2020 EM expansion (Frame 3). Substrate landed clean on first probe attempt. |
| **FX — statement vs draw distinction (NEW)** | Working+ | 2026-07-27 | Session 33b (CB swap-line communications). Process rule: a foreign CB swap-line *statement* is preparation; a *draw* is action. Statements can be issued for many reasons (routine review, internal policy, pre-meeting signaling, political cover) without a stress event. A draw is unambiguous; a statement is ambiguous. The PM edge sits in the gap between statement and draw, but the statement alone is not enough to fade or follow. Substrate landed after one re-anchor — user said "Fed hasn't ratified the statement" (direction right, mechanism off — Fed ratifies *draws* not *statements*). |
| **FX — narrow vs broad draw distinction + time-horizon tie-in (NEW)** | Working+ | 2026-07-27 | Session 33b (CB swap-line communications). A single small CB draw (e.g., RBNZ) = narrow signal (isolated event, Fed response targeted). A major CB draw + broadening within a week (e.g., BoJ) = broader signal consistent with the time-horizon discipline rule (broader stack = longer time in trade = wider stop, not tighter). The RBNZ draw was an early-narrow signal; the BoJ draw is the broadening confirmation. Substrate landed on first attempt. |
| **FX — Regime 1 vs Regime 2 Fed response distinction (Target 1) (NEW)** | Working+ | 2026-07-27 | Session 34 (Block 024 Target 1). Fed always meets a request. Regime 1 (currency-specific): Fed ratifies the draw only — no rate cut, no network expansion. ECB has to ask on its own. Regime 2 (global): Fed ratifies *plus* preemptive rate cut (e.g., March 2020 OIS+50bp → OIS+25bp) *plus* possible network expansion (e.g., March 31 2020 +9 EM CBs). The Fed always meets a request; the regime determines whether the Fed *preempts*. Substrate landed after one re-anchor — user initially said Fed "less likely to meet" in Regime 1 (wrong; Fed always meets, just doesn't preempt). |
| **FX — Systemic criteria for swap-line activation (4 criteria) (NEW)** | Working+ | 2026-07-27 | Session 34 (Block 024 Target 1). Four criteria together define the Fed's diagnostic of "systemic" swap-line activation: (1) Multiple foreign CBs drawing simultaneously (cross-section break). (2) Draw sustained over multiple weeks (not quarter-end flush). (3) Draw size large relative to CB's normal dollar operations. (4) Accompanied by Frame 2/3 public statements. Three or four criteria met = systemic. One or two = currency-specific or local. Substrate landed clean on first probe attempt. |
| **FX — Three-signal-layer diagnostic tie-back (Block 014 + 023 + 024) (NEW)** | Working+ | 2026-07-27 | Session 34 (Block 024 Target 1). Tie-back of CB swap-line diagnostic to Block 014/023 three-signal-layer. SOFR-term-vs-OIS wide + cross-currency flat = US-domestic bank funding stress → Fed uses standing repo (NOT swap lines). Cross-currency wide + SOFR flat = global dollar funding stress → Fed uses swap lines + possibly rate cut + possibly network expansion. Both wide = systemic (March 2020) → Fed uses everything. CB swap-line diagnostic is specifically the cross-currency basis layer. Working+ (re-affirmed via Block 024 Target 1). |
| **FX — Post-2013 PM edge (read delta, not event) (NEW)** | Working+ | 2026-07-27 | Session 34 (Block 024 Target 1). Pre-2013: swap-line activation = crisis signal. Post-2013: swap-line architecture is a standing tool. The Fed can change rate/scope/terms without creating new architecture. PM reads the *delta* (parameter change), not the *event* (activation itself). A 25bp rate cut on the standing network is calibration. A network expansion to 9 new CBs is escalation. Substrate landed clean on first probe attempt. |
| **FX — Frame vs regime independence (independent substrate layers) (NEW)** | Working+ | 2026-07-27 | Session 34 (Block 024 Target 1). Frame (Source 3 language ladder) and regime (Probes 1/2 market signature) are *independent* substrate layers. Frame 2 statement does NOT imply Regime 2. Regime is determined by cross-section signature ONLY (multiple currency pairs wide = Regime 2). PM combines: frame for language telegraph, regime for market structure. Same direction (e.g., fade) but different trade construction depending on regime. Substrate landed after slip — user conflated Frame 2 (language) with Regime 2 (market). Direction right, layer wrong. |
| **FX — CB inaction 5-explanation framework (NEW)** | Working+ | 2026-07-27 | Session 34 (Block 024 Probe 3). Five distinct explanations for CB persistent inaction on a wide basis: (1) Containing via own reserves — own dollar ops step up when basis wide. (2) Judging level acceptable — own ops flat + "monitoring/new normal" statements. (3) Institutional constraint — own ops flat + "lacks authority" statements. (4) Benefiting from wide basis — own ops flat + non-committal statements, policy goals aligned with wide basis (e.g., BOJ wanting weak JPY). **User-added in this session.** (5) Debt-stability vs currency-stability trade-off — own ops flat + policy stance accommodates weak currency to keep imported inflation supporting easy policy (high-debt sovereigns: Japan, US, Italy, UK). **User-added in this session via national-debt mechanism question.** Discriminating operational signatures per explanation. Substrate landed after user-added explanations 4 and 5. |
| **FX — National debt / FX basis mechanism (high-debt sovereigns) (NEW)** | Working+ | 2026-07-27 | Session 34 (Block 024 Probe 3 follow-up). For high-debt sovereigns (Japan, US, Italy, UK), wide cross-currency basis → weak currency → imported inflation rises → CB tightens policy → domestic yields rise → debt service rises (lag 6-24 months). Japan has ~¥1,100T debt; 100bp JGB yield rise = ~¥11T annual debt service (~2.5% of total tax revenue). Creates a *binding* debt-stability vs currency-stability trade-off for the CB. Compressing basis (intervention, rate move) can worsen long-term fiscal position by removing imported inflation pressure that allows the CB to keep policy easy. **User-added in this session.** |
| **FX — central bank reaction functions** | Working+ | 2026-08-02 | Cal S1: explicit gap — said "I forgot"; no recall of the framework. **2026-07-27 Block 024 Sessions 33-34 (pre-reads + Target 1):** McCauley + Schenk mechanics (two-step structure, regulatory jurisdiction primary), 4 systemic criteria (multiple CBs drawing, sustained, large draw, Frame 2/3 statements), 5-explanation CB inaction framework (containing / judgment / institutional constraint / benefiting / debt-stability vs currency-stability trade-off), Frame vs regime independence, post-2013 PM edge (read delta, not event). **2026-08-02 Block 024 Session 36 (Targets 2-4):** Borio cycle-position framework locked (augmenting global USD cycle = works, fighting = default fail) + velocity-vs-direction refinement + sub-cycle vs global cycle distinction (SNB 2011-2015 case = fought EUR sub-cycle while augmenting global, held 3.3 years) + commitment size moderates holding window not direction of failure. Tier 1 vs Tier 2 architecture: (B) moral-hazard/political = primary driver, plumbing asymmetry at 2013 enabled tiering, BIS data shows EM FX swap depth has *grown* since 2013 (plumbing exists now, Fed still keeps Tier 2 episodic = political not plumbing). Speed ≠ architecture (March 2020 6-day expansion was 9 parallel episodic processes, not standing architecture). Coordination = (cycle stress exists) × (Fed deems action appropriate) — 2011 Eurozone (EUR-specific but coordinated because ECB Tier 1) and 2013 taper (shared EM stress but uncoordinated because Fed deliberately chose taper) are counter-cases that refine (A) cycle-stress to necessary-but-not-sufficient. Draw-size + draw-velocity = cleanest leading indicator of whether fade-basis trade wins or fails after swap-line activation. **Diagnostic cap (load-bearing):** cross-currency basis is a *diagnostic*, not a *trade* — direction comes from Fed tool read, not basis level. 33 atomic Anki cards in FX deck total. |
| **FX — BIS data (NEW)** | Working+ | 2026-08-08 | Block 025 fully closed at Working+ across 4 targets across Sessions 38-39. Session 38 pre-block probes (5 vocab + 1 transition): sales-desk methodology (dealer-side booking not end-client flow, Singapore-jurisdictional attribution), USD translation effect (USD appreciation → *understates* USD turnover, real growth > 28% headline), turnover → volatility causal direction (vol drives turnover via 4 channels, not reverse), FX swap share decline decomposition (mechanical + supply-side Block 023 dealer BS + demand-side April 2025 institutional ex-post hedging into forwards), CNY rise within USD system (95%+ vs USD, not displacement), GBP collapse as Tier 1 test (Tier 1 ≠ vehicle currency share), USD share FX-strength artifact (USER CATCH), spot+forward vs FX swap structural differences (USER CATCH — regulatory/NSFR, margin/collateral, settlement timing enable partial substitution), institutional-investor forward surge = coincident-to-lagging indicator, layered dollar-funding architecture (Fed → dealers → non-bank intermediaries). Target 1 (BIS methodology): flow ≠ dealer capacity, PM sizing needs flow + outstanding + dealer BS utilization (three-layer framework). Target 2 (instrument composition): 2022→2025 growth rates FX swap +5% / spot +42% / forwards +60% / options +120% / currency swaps +30%; structural-shift discriminator = divergence across instruments vs across-the-board tracking. Target 3 (currency participation): EUR decline from trade-finance structural shift (USER CORRECTION — not carry-trade demand, low EUR rates = MORE EUR funding demand not less), CNY rise does not prove USD displacement (USD share at new high 89.2% = MORE USD-centric system not less), CHF rise to 6.4% (sixth place up from eighth), Singapore jurisdictional gain 9.5%→11.8% (4th largest hub). Target 4 (PM implications): BIS-regime-discrimination (structural widening vs transient dislocation = fade works/fails), Fed-tool-trigger for fade entry (Frame shift leading + cross-section coincident + Fed action lagging), trade construction differs by regime (fast/tight vs slow/wide). **Diagnostic cap (load-bearing, ties Block 023+024):** in structural-widening regime, wide basis is new equilibrium = fade into trend, fails. In transient-dislocation, basis mean-reverts, fade works. 11 atomic Anki cards pushed in Block 025 (Session 38: 5, Session 39: 6). |
| **FX — drivers / regime shift** | Working | 2026-07-11 | Cal S1: named interest rates / money supply / trade; missed liquidity/positioning; missed regime-shift in carry-unwind. **2026-07-11 Block 014 update:** carry unwind / global dollar funding read as the regime shift mechanism locked in. **2026-08-08 Block 025 update:** BIS-regime-discrimination substrate (structural widening vs transient dislocation), trade construction by regime (fast/tight vs slow/wide), Fed-tool-trigger for fade entry (Frame shift leading + cross-section coincident + action lagging). Re-probe Block 030 (Phase 1 FX closeout). |
| **Credit — HY/IG spreads, OAS** | Working | 2026-06-21 | Cal S1: BBB-/BB+ threshold right; spread ranges approx; OAS concept partial but missed negative convexity in callable |
| **Credit — default cycle** | Weak | 2026-06-21 | Cal S1: skipped / unknown; default cycle lag (6-9mo spread lead) and trade implication not internalized |
| **Equities — index construction** | Working | 2026-06-21 | Cal S1: market-cap weighting + concentration risk right (mag-7); mentioned equal/price weight; missed float-adj + fundamental weight |
| **Equities — sector rotation, valuation** | Weak | 2026-06-21 | Cal S1: 3 valuation methods named (PE/FwdPE/PB); forward PE reasonable but missed bias + Shiller CAPE; sector rotation explicit gap |
| **Equities — positioning** | Weak | 2026-06-21 | Cal S1: knows COT exists but not what extremes mean; missing margin debt, hedge fund exposure, sentiment, short interest framework |
| **Commodities — term structure, inventories** | Weak | 2026-06-21 | Cal S1: contango/backwardation defs muddled (backwardation described as "higher"); mechanism not explained; inventory framework missing |
| **Commodities — energy vs metals vs ags** | Weak | 2026-06-21 | Cal S1: explicit gap — no recall of energy/metals/ags distinction, drivers, or precious-vs-base metals split |
| **Cross-asset — Rates ↔ FX** | Untested | — | Probe in calibration |
| **Cross-asset — Real yields ↔ Gold** | Untested | — | Probe in calibration |
| **Cross-asset — Oil ↔ Equities** | Untested | — | Probe in calibration |
| **Cross-asset — Credit ↔ Equities** | Untested | — | Probe in calibration |
| **Cross-asset — USD ↔ Everything** | Untested | — | Probe in calibration |
| **Cross-asset — Reflux vs correlation** | Untested | — | Probe in calibration |
| **Framework — regime identification** | Weak | 2026-06-21 | Cal S1: explicit gap — no recall of growth/inflation/liquidity/policy axes or regime classifications |
| **Framework — cycle position** | Weak | 2026-06-21 | Cal S1: explicit gap — no recall of cycle indicators (yield curve, unemployment, PMI, credit spreads) |
| **Framework — positioning & reflexivity** | Working | 2026-06-21 | Cal S1: core reflexivity (two-way causality) + USD example correct; missed boom-bust overshoot nuance |
| **Framework — liquidity plumbing** | Weak | 2026-06-21 | Cal S1: explicit gap — no distinction between market liquidity vs central bank / global dollar liquidity |
| **Framework — historical case studies** | Untested | — | Probe in calibration |
| **Process — decision process rubric** | Working | 2026-07-01 | Cal S1 (2026-06-21): explicit gap — can't articulate own process.md gates. Session 10 (2026-07-01) Block 00a: cold list of 8 sections landed (Working+); trade walk (gold short) Working with §5 risk 250bp = 3.3x framework ceiling (process-grade D); trade ID generated (Strong, with spread-convention refinement); two-trigger rule with current example (QQQ) Working+ — both triggers confirmed via real COT data (non-commercial historically high at time of trade). Rubric edits filed: process.md §4.1 Technical→Entry trigger; signal-spec.md v0.1→v0.2 trigger-type taxonomy (A/B/C/D); _templates/thesis.md §2 aligned. Re-probe Block 086 (verification + edge cases for non-rate trades) |
| **Process — anti-patterns** | Working | 2026-07-02 | Cal S1 (2026-06-21): explicit gap. Vault gap resolved 2026-06-21 (`anti-patterns.md` v0.1 created with 7 patterns). Session 10 (2026-07-01) Block 00a surfaced at least 3 anti-patterns in real-time on the user's own positions (gold short size drift, QQQ invalidation looseness, narrative-without-pricing check missing). Session 11 (2026-07-02) Block 00b first teach: 5/7 cold (pre-block), 7/7 cold (post-teach). Pre-trade checks articulated (prohibition form landed; positive-form question-form identified for willpower-depleted state). Three open positions audited: GC-S (size pre-process, narrative-without-pricing acknowledged, pricing-gap carry-forward), SLV-S (correlation leverage unaccounted for — reframe decision pending), QQQ-S (pre-process artifact, deferred rewrite). Quarterly audit protocol structured: management-log spine, journal/research periphery, table+aggregations output, 3-trigger change process (recurrence 3+/quarter, severity single → -5% book, doc gap). 7 atomic Anki cards on gold pricing-vs-narrative point added via AnkiConnect (note IDs 1782975741400-1782975742086). Re-probe Block 087 (anti-patterns verification + edge cases, Phase 5). |
| **Process — postmortem grading** | Working | 2026-06-21 | Cal S1: skill-vs-luck insight correct; missed process grade A-F (right process/bad outcome = B) per process.md §4.4 |
| **Process — drawdown protocol** | Weak | 2026-06-21 | Cal S1: gave wrong numbers (-15/-25) vs own risk-framework.md (-5 halve, -10 flat, -15 hard stop); needs drill |
| **Process — behavioral biases** | Untested | — | Probe in calibration |
| **Instrument — ES/MES (S&P 500)** | Weak | 2026-06-21 | Cal S1: explicit gap — can't recite contract specs (ES $50/pt, MES $5/pt); traded in universe per process.md §2 |
| **Instrument — GC (Gold)** | Working | 2026-06-21 | Cal S1: real rates dominant correct; cross-asset signal wrong (should be TIPS real yield not nominal 10y); second driver should be USD not global liquidity |
| **Instrument — CL (Crude Oil WTI)** | Weak | 2026-06-21 | Cal S1: explicit gap — no recall of OPEC+/shale/geopolitics supply drivers or EIA inventory series |
| **Instrument — ZN (10y UST)** | Weak | 2026-06-21 | Cal S1: explicit gap — can't recite contract specs ($100k face, 1/64 tick = $15.625); no recall of typical daily range |

> **Calibration history (2026-07-12 refactor):** Previously a 161-line section duplicating full session entries from `session-log.md`. Removed. Per-session probe results + status transitions are the canonical record in `session-log.md` (see also `session-log-archive-2026.md` for pre-Aug entries). The competence-map table above is the live state view; session log is the session-by-session history.
### 2026-07-17 — Session 26 (Block 020 — ACM/KW term premium in practice — complete)

| Rates — Term premium definition (NEW) | Working+ | 2026-07-17 | Session 26 (Block 020): TP = 10y nominal yield MINUS expected avg level of short-term Treasury yields over next 10y. Residual that pure expectations theory cannot explain. Compensation for duration risk that expectations don't reward. Can be negative (safe-asset demand, dealer balance-sheet capacity, regulatory forced duration buying; 2010s US 10y TP was negative for most of post-QE era). **2026-07-20 Session 28 re-probe:** substrate fired clean (avg-short-rate-not-chained-forward, three-driver negative TP); precision refinement that the wedge IS the gap in expected-rate paths (algebraic identity). Re-probe Session 29. |
| Rates — ACM-KW methodological difference (NEW) | Working+ | 2026-07-17 | Session 26 (Block 020 Q3): ACM anchors expected-rate path to the FORWARD CURVE (lets expectations move freely with curve). KW constrains expected-rate path to SURVEY MEDIANS (SPF or Blue Chip). Same conceptual decomposition, different assumption on how expected path forms. **2026-07-20 Session 28 re-probe:** substrate fired clean (surveys slow). Re-probe Session 29. |
| Rates — ACM-KW wedge diagnostic (NEW) | Working+ | 2026-07-21 | Session 26 (Block 020 Q2): wedge widening = curve has repriced faster than surveys updated. Tells you which model's expected-rate-path assumption matches the marginal buyer's view. Correlates with dealer-balance-sheet stress, QT pace changes, fiscal supply surprises. NOT just measurement error — it's a regime-shift indicator. **2026-07-19 Session 27 cold-recall demoted:** direction slipped — user said marginal buyer is closer to ACM (correct is closer to KW, the higher-expected-path model). **2026-07-20 Session 28 re-probe:** same sign slip recurred; right conclusion ("KW as it has the lower TP") but wrong reason ("lower TP" vs "higher expected-path"). Three sign slips now on this substrate. **Session 28 rebuild teach:** anchored wedge as the gap in expected-rate paths (algebraic identity), wedge interpretation as price-formation (marginal buyer closer to higher-expected-path model, NOT lower TP), wedge widening as balance-sheet-channel signature. **2026-07-21 Session 30 cold-recall re-probe (Probe 1, Card C 1784631449041): CLEAN at Working+.** Recurring-slip escalation Rule 3 resolution signal fired (2 clean probes in same session on same cluster). **Escalation CLOSED.** Wedge cluster substrate now locked. |
| Rates — 5y5y forward decomposition (NEW) | Working+ | 2026-07-17 | Session 26 (Block 020 concept 3): 5y5y forward = ((1+y10)^10 / (1+y5)^5)^(1/5) − 1 = market-implied avg short rate years 5-10 + 5y5y TP. Decomposition isolates the long-end LR signal (years 5-10) from path-dependence of years 0-5. Re-probe Block 022. |
| Rates — Market-Fed LR wedge diagnostic (NEW) | Working+ | 2026-07-17 | Session 26 (Block 020 concept 3): 5y5y forward minus dot plot LR median (3.0625%). Positive = market prices higher LR than Fed signals (Fed dovish on its own LR path). Negative = market prices lower LR. Trade: short 5y5y forward to bet on wedge compression. Live data 2026-07-15: 4.97% 5y5y forward vs 3.06% dot LR = +85bp wedge. Re-probe Block 022. |
| Rates — TP can be negative (NEW) | Working+ | 2026-07-17 | Session 26 (Block 020 Q1 reverse): negative TP = investors accept less yield than pure expectations implies. 2010s post-QE era. Causes: safe-asset demand exceeding duration supply, QE-era dealer balance-sheet expansion, foreign reserve asset reallocation, regulatory forced duration buying. Re-probe Block 022. |
| Rates — ACM dataset columns (NEW) | Working | 2026-07-17 | Session 26 (Block 020 main material): ACMY10 = ACM-fitted 10y Treasury yield. ACMTP10 = 10y term premium (residual in %). ACMRNY10 = expected avg short rate over next 10y. Identity: ACMY10 = ACMRNY10 + ACMTP10. 2026-07-15: 4.60% = 3.92% + 0.685%. **Slip this session:** mentor initially framed TP at +90bp (from Block 017's pre-existing framing); actual data showed +68bp (76th percentile post-2010) and TP moved -11bp YTD, NOT +60bp. Substrate is data-column identification, not trade interpretation of TP level. Re-probe Block 022 with substrate lock on current-state interpretation (elevated vs extreme TP for trade construction). |
| Rates — Block 016 / Block 020 cross-block (NEW) | Working+ | 2026-07-17 | Session 26 (Block 020 concept 4): widening ACM-KW wedge = model-level signature of balance-sheet-channel event (QT, fiscal supply, regulatory shift). Policy-channel events move ACMRNY10 in BOTH ACM and KW similarly → TP estimates agree. Wedge direction = diagnostic of which Block 016 channel is driving the curve. Re-probe Block 022. |
| Rates — TP three drivers of negative TP (NEW) | Working | 2026-07-20 | Session 27 (Block 021 cold-recall Probe 4): "treasury oversupply" was wrong (sign inverted — high issuance pushes TP UP, not down). **2026-07-20 Session 28 re-probe:** sign slip recurred — user said "forced selling by LDI / hedge funds during margin calls compresses TP" (margin-call cascades push TP UP, not down; that's the TP expansion substrate, not the negative-TP substrate). Correct drivers: (a) structural buyer duration absorption (pension LDI, foreign reserve managers, insurance — slow persistent price-insensitive bid), (b) issuance scarcity relative to demand (Treasury running down balance sheet, dealer capacity constrained), (c) Fed QE / balance-sheet absorption (Fed removes duration from market, private-hand scarcity). All three converge on private-hand scarcity of long-end duration. **Session 28 rebuild teach:** anchored substrate that negative TP isn't about price-sensitive buyers or sellers — it's about absence of price-sensitive sellers meeting price-insensitive bid. Re-probe Session 29. |
| Rates — ACMRNY10 = expected avg short rate (NEW) | Working | 2026-07-20 | Session 27 (Block 021 cold-recall Probe 5): same class of error as DFII10/T10YIE recurring slip — substrate fires on the value (3.92% in Session 26 data), slips on what the series measures. ACMRNY10 = ACM-fitted expected average short rate over next 10 years. NOT real yield. Identity: ACMY10 = ACMRNY10 + ACMTP10. **2026-07-20 Session 28 re-probe:** said "risk neutral rate" — closer than "real yield" but still wrong substrate. ACMRNY10 = expected avg short rate (expectations leg of decomposition), not risk-neutral pricing. **Session 28 rebuild teach:** anchored decomposition identity explicitly (ACMY10 = ACMRNY10 + ACMTP10) and the ACMRNY10 substrate as expectations leg of the decomposition, not real yield, not risk-neutral pricing. Three series fully named (ACMY10 = fitted 10y yield; ACMRNY10 = expected avg short rate; ACMTP10 = ACM TP estimate). Re-probe Session 29 with explicit pairing card consolidation. |
| Rates — Leverage-unwind cascade mechanism (NEW) | Working+ | 2026-07-19 | Session 27 (Block 021 case-study probes Q1-Q4): levered-investor margin-call cascade = structural-buyer reversal via forced selling. LDI pensions, basis trades, levered hedge funds suffer MTM losses on duration → margin calls on short leg → forced to sell the long-duration asset. Acts as accelerant in TP expansion events: 2018 Q4, March 2020 dash-for-cash, 2022 UK gilt, 2023 SVB HTM. Speed signature: 50bp+ in days, not weeks. Anki card note 1784465882317 (session-27) covers the mechanism + cross-case-study examples. Re-probe Block 022. |
| Rates — Historical case studies substrate (NEW) | Working (CONTAMINATED) | 2026-07-19 | Session 27 (Block 021 case-study probes Q1-Q4): four case studies decomposed cleanly — 2018 Q4 (Phase 1 TP spike + real-yield rise / Phase 2 flight-to-quality Fed-pivot), March 2020 (Phase 1 FTQ TP compression / Phase 2 dash-for-cash TP expansion / Phase 3 Fed intervention TP compression), 2022 LDI/gilt (LDI leverage unwind = forced-seller cascade, 250bp in 3 days), 2023 SVB (10y -42bp / 2y -100bp bull steepener, three-leg diagnostic ordering SOFR-OIS → cross-currency → MBS-Treasury). **⚠️ CONTAMINATED 2026-07-20 (Session 29 verification): the Session 28 rebuild teach on this entry cited ACMTP10 peak 150bp in early November 2018 as the "extreme TP" anchor. Verified NY Fed data shows ACMTP10 in Q4 2018 was -45bp (Aug) → -22bp (Oct peak) → -59bp (Jan) — TP COMPRESSED, never crossed zero, never approached 95th-percentile post-2010 (214bp). The "150bp peak" was FABRICATED. The "fade-the-TP trade worked" claim was unverified. Substrate needs cold re-anchor at Session 30 with verified 2010-2011 (the actual canonical post-2010 TP-extreme episodes, peak 257bp Feb 2010). Do not drill on 2018 Q4 TP framing until Session 30 re-anchor. ⚠️** |
| Rates — Elevated vs extreme TP rule (NEW) | Working+ (decontaminated, rule re-anchored) | 2026-07-21 | Session 28 (Block 021 rebuild teach): three-condition rule for distinguishing elevated vs extreme TP for trade construction. (a) Historical percentile: ACM TP > 95th post-2010 = extreme; > 90th = elevated. **(REVISED 2026-07-21 Session 30 verification: post-2010 percentiles are 75th=63bp, 90th=147bp, **95th=210bp** (not 214bp as previously stated), 99th=253bp, max=257bp Feb 2010.)** (b) Cross-asset confirmation: HY OAS widening + equity vol rising confirm regime-level TP expansion; if TP is rich but HY tight + VIX low, expansion is idiosyncratic (foreign-demand shock, not regime signal). (c) Dealer balance-sheet capacity constrained: primary dealer positions turned negative, Treasury supply outstripped demand at the long end. **The 3-condition rule is a permission-to-trade GATE, not a sizing input** (Session 30 user pushback: checklist-driven sizing is a process anti-pattern; sizing-on-top comes from trade EV + portfolio risk budget, not re-counting the checklist). **CONTAMINATION CLEARED 2026-07-21 (Session 30):** the 2018 Q4 worked example (ACMTP10 150bp peak) was fabricated; verified 2018 Q4 ACMTP10 was -22bp Oct peak, never positive, never close to 95th-percentile post-2010. **Re-anchored on verified 2010-2011 (peak 257bp Feb 2010) — the actual canonical post-2010 TP-extreme episode.** Worked-example card 1784631778967 pushed on verified substrate. |
| Rates — Cross-asset transmission quadrants (NEW) | Working (cause-interpretation cards pushed, re-probe pending) | 2026-07-21 | Session 27 (Block 020 Concept 1c) substrate partially anchored: rising TP + risk-on = supply/demand repricing (idiosyncratic, fade); rising TP + risk-off = recession/fear (regime-shift signal, hold or add); falling TP + risk-on = nothing special (status quo); falling TP + risk-off = flight-to-quality bid (structural bid for duration, regime-shift hedge). **2026-07-19 Session 27 cold-recall Probe 8c:** direction slipped on two of the four quadrants. **2026-07-20 Session 28:** probe deferred. **2026-07-21 Session 30:** 2x2 case-first teach delivered. Three card-shape iterations failed (regime-label → action → directional+verify) before landing on cause-interpretation shape. **4 cards pushed (1784633381092/1127/1143/1171)** with verified anchors: Q1=2010-Q1 post-QE1 unwind, Q2=March 2020 dash-for-cash, Q3=2014-2019 post-QE normalization, Q4=March 2020 early-March FTQ phase (or 2022 UK LDI). **Probe 12 (Q2): CLEAN at Working+.** Q1, Q3, Q4 not yet probed. Re-probe next session. |
| Rates — Block 021 case studies decomposition (4 case studies) | Working+ | 2026-07-19 | Session 27 (Block 021 case-study probes Q1-Q4): 2018 Q4 (Phase 1 TP spike / Phase 2 Fed-pivot FTQ), March 2020 (Phase 1 FTQ TP compression / Phase 2 dash-for-cash TP expansion / Phase 3 Fed intervention TP compression), 2022 LDI/gilt (LDI leverage unwind = forced-seller cascade), 2023 SVB (bull steepener with three-leg diagnostic ordering). All four at Working+. Cross-case-study insight = levered-investor margin-call cascade as common accelerant in TP expansion events. Re-probe Session 30 (Block 022 verification). |
| **Rates — Pairing-card consolidation substrate (NEW)** | Working+ (RESOLVED via drivers-based card) | 2026-07-21 | Session 28 (Block 021 rebuild teach) substrate for the pairing card that consolidates four substrates that slipped across Session 27 + 28: (1) ACMRNY10 = expected avg short rate (NOT real yield, NOT risk-neutral pricing); (2) ACMTP10 = ACM 10y TP estimate; (3) ACM-KW wedge sign = marginal buyer closer to higher-expected-path model (NOT lower TP); (4) Fade-the-TP-at-extreme trade = short duration when ACM TP > 95th + cross-asset + balance-sheet. **2026-07-21 Session 30 (Session 29.5 refactor + Session 30 cleanup):** the original non-atomic pairing card (1784548986964, 5 facts in 110-word back) was deleted and replaced with 5 atomic cards plus 1 drivers-based pairing card. **Drivers-based pairing card 1784632327592 is the new consolidation card** — recall target is the *drivers* (expectations leg = FOMC path / surveys / forwards; TP leg = supply/demand mechanics), not the series codes. The wedge card (1784631449017) anchors both expressions of the algebraic identity (expected-rate-path primary, TP-side as equivalent). The 2010-2011 worked example (1784631778967) locks the canonical post-2010 anchor on verified substrate. **Card design lesson:** pairing/consolidation cards should test the *drivers*, not the *codes* (Session 30 process rule). |
| **FX — CIP negative basis = balance-sheet cost to dealers (NEW)** | Working+ | 2026-07-23 | Session 31 (Block 023 Target 1 CIP mechanics main material): no-arbitrage frame locked (forward = interest rate differential exactly; when basis breaks, someone absorbs a cost). Negative basis = dealer's required compensation for SLR / leverage-ratio balance-sheet cost, NOT arbitrageur capture of a mispricing. **Precision locked:** the premium in the basis is the *cost of regulatory intermediation*, not a textbook arbitrage failure. Hedger A/L framing: Japanese pension fund sells USD/JPY forward to eliminate FX risk on USD Treasuries; fair hedge cost = US Treasury yield − JPY rate; negative basis means the actual hedge costs 25bp more than fair value, with the difference being dealer balance-sheet cost. Re-probe Block 023 Targets 2-4 reinforcement. |
| **FX — Post-GFC structural basis = 24/7 constraint (NEW)** | Working+ | 2026-07-23 | Session 31 (Block 023 Probe 2): user correctly identified the absence of quarter-end spike pattern in the persistent -20 to -30bp basis range; refined to "permanent signal + constraint binds always." **Mechanism locked:** post-GFC regulatory regime (SLR, leverage ratio, G-SIB surcharge) means dealer balance sheets are *permanently constrained*, so quarter-end window-dressing no longer adds meaningfully to a constraint that already binds every day. The basis is structural, not technical. Pre-GFC, dealer balance sheets had slack, so quarter-end shrinkage moved the basis because that was the marginal binding window; post-GFC the constraint binds 24/7 and the quarter-end spike *disappears* because the technical is already permanent. Re-probe Block 023 Target 2 (CIP deviations) for historical case-study anchoring. |
| **FX — Cross-currency basis spread = relative carry-trade scale + regulatory backstop (NEW)** | Working+ | 2026-07-23 | Session 31 (Block 023 Probe 3): user correctly identified JPY as the dominant structural carry-trade funding currency (BOJ ZIRP duration, GPIF/pension USD hedging demand). **Mechanism locked:** JPY basis wider than EUR basis because JPY-funded carry into USD assets is the canonical global carry trade (BOJ at ZIRP/NIRP for two decades, JGB yields near zero, JPY as the funding currency); Japanese institutional investors (GPIF, pensions, lifers) hold massive USD asset allocations that they FX-hedge → structural demand for USD in FX swap market. EUR basis narrower because EUR-funded carry is less structural (ECB not at same ZIRP extreme for the same duration as BOJ) and EUR/USD basis is influenced by ECB USD liquidity operations (swap-line network backstop) that JPY doesn't have to the same degree. **Spread magnitude = relative scale of structural carry trade + relative regulatory backstop.** Direction-of-hedging framing: foreign investors HOLD USD assets and want to HEDGE FX risk by selling USD forward; basis is the premium they pay to hedge. Re-probe Block 023 Target 4 (basis diagnostic) for cross-currency diagnostic usage. |
| **FX — Term structure short-end widest = dealer balance-sheet turnover velocity (NEW)** | Working+ | 2026-07-23 | Session 31 (Block 023 Probe 4): user correctly identified the short-end as where the constraint binds; mechanism refined from "velocity in moves" (price-volatility framing) to velocity = dealer balance-sheet *turnover*, not price volatility. **Mechanism locked:** short-end FX swap market is where velocity is highest (banks rolling 1m/3m USD funding constantly) — that's where marginal balance-sheet consumption happens minute-to-minute. Long-dated FX swaps are mostly balance-sheet-stable (5y JPY/USD swap sits on books for years without re-pricing). The constraint binds on the velocity-intensive end (short tenor, high frequency), not where price moves most. **Diagnostic read:** short-end basis blowout = funding-desk stress (March 2020, 2008 quarter-ends); long-end basis blowout = positioning-desk stress (rare, means someone taking large directional view). Re-probe Block 023 Target 2 (CIP deviations) for term-structure-as-diagnostic use in 2008 + March 2020 anchors. |
|| **FX — BOJ exit second-order on basis = demand-side compression (NEW)** | Working+ | 2026-07-26 | Session 31 (Block 023 Target 1 sub-probe): user initially said rising JPY rates → basis expands (carry trade unwinds). Corrected: rising JPY rates don't directly move the basis in steady state (existing positions sticky, hedgers don't immediately change behavior). **Second-order mechanism locked:** higher JPY rates → JPY assets more attractive → JPY institutions reduce USD asset allocation → less hedging demand → less structural short-dollar demand → basis *compresses* (not expands). Mechanism: demand-side asset-allocation shift, NOT carry-trade side (carry trade is unhedged USD positioning; basis is hedged USD positioning — different populations). **Carry-trade vs hedge-trade population distinction is the load-bearing substrate.** The basis move on BOJ exit depends on whether BOJ exit causes JPY rally (basis compresses) or JPY fall (basis could blow out first then compress). **2026-07-26 Session 32 Block 023 Target 2 forwarding:** Substrate reinforced via diagnostic probe (USD/JPY basis -22→-45bp, EUR -9bp, JPY 1m -60bp / 5y -15bp) — user correctly identified JPY-specific acceleration + cross-section break + term-structure shape = velocity/balance-sheet regime, not price-driven. **Upgraded Working → Working+.** |
| **FX — CIP deviations: three signatures (NEW)** | Working+ | 2026-07-26 | Session 32 (Block 023 Target 2 main material): three signatures of CIP basis opening loaded. **(1) Acceleration:** basis moves in days, not weeks. Velocity signature. **(2) Cross-section break:** basis moves unevenly across currency pairs (JPY-specific vs EUR broad vs systemic). **(3) Term-structure distortion:** short-tenor leg blows out vs anchored long end = velocity/balance-sheet regime; whole curve moves with similar shape = price-driven regime. The COMBINATION of these signs tells you the type of shift. Diagnostic cap: basis is a *diagnostic*, not a *trade* — the trade comes from the read of WHY the constraint is binding, not the level. Anki cards 1785099069095 (three signatures) + 1785099069144 (basis diagnostic vs tradeable) pushed. Re-probe Block 022 verification. |
| **FX — Carry trader vs hedged USD investor population distinction (NEW)** | Working+ | 2026-07-26 | Session 32 (Block 023 Target 3 sub-probe): the User initially conflated the two populations on a Fed-hike unwind scenario. Re-anchored the population distinction: **(1) Carry trader (unhedged)** = long USD asset, short JPY funding, no FX hedge. P&L: USD bond yield + FX delta. Earns rate differential directly. **Does NOT trade the basis.** **(2) Hedged USD investor** = long USD asset, short USD/JPY forward hedge. P&L: USD bond yield − basis (the hedge cost). FX-locked. **Basis is the cost paid by this population.** The basis does NOT affect existing positions (locked at inception); it affects NEW hedging decisions. Carry-trade unwind does NOT directly move the basis (Treasury market move + side-effect basis blowout via vol spike + dealer capacity contraction + general dollar strength). Diagnostic: carry trade unwind + basis blowout together = systemic; divergence = normal. Reference: `~/.hermes/skills/finance/macro-mentor/references/carry-vs-basis-population-distinction.md`. Anki cards 1785099068969 (two populations), 1785099068994 (carry trader doesn't trade basis), 1785099069023 (basis existing vs new), 1785099069044 (systemic vs normal diagnostic). Re-probe Block 022 verification. |
| **FX — UIP empirical failure: carry trade puzzle (NEW)** | Working+ | 2026-07-26 | Session 32 (Block 023 Target 3 main material): UIP predicts USD with higher rate should depreciate; empirical failure: high-yield currencies tend to *appreciate*. Three competing explanations: (a) risk premium for crash risk (carry trades tail-crash in stress), (b) peso problem (rare disasters dominate unconditional expected return), (c) behavioral / order-flow (yield-chasing in low-rate world, self-reinforcing on unwind). **PM use:** UIP failure is why carry EXISTS as a strategy, not a signal to violate UIP directly. Carry trade unwind = funding-currency-weakness regime shift. Tied to Block 023 substrate: carry trade (Population 1) and basis (Population 2) are different populations — diagnostic power is in divergence/convergence. Anki card 1785099069116 (UIP failure mechanism) pushed. Re-probe Block 022 verification. |
| **FX — 2008 vs March 2020 cross-currency basis comparison (NEW)** | Working+ | 2026-07-26 | Session 32 (Block 023 Target 2 main material): 2008 GFC vs March 2020 diagnostic comparison. **2008:** all cross-currency bases blew out (cross-section absent — systemic). Term structure inverted at peak. Acceleration acute. Mechanism: dealer balance-sheet collapse + foreign real-money flight-to-USD. Fed response: dollar swap lines. **March 2020:** partial cross-section break (JPY + EUR blew out, EM more). Extreme term-structure distortion (1m -100bp+, 5y near normal). Acute acceleration. Mechanism: dash-for-cash, foreign real-money selling USD assets + maintaining hedges. Fed response: swap lines + standing repo + unlimited QE. **Diagnostic insight:** order of Fed facility activation = which layer of plumbing broke first. Anki card 1785099069166 (2008 vs 2020) pushed. Re-probe Block 022 verification. |
| **FX — Cross-currency basis diagnostic shape (NEW)** | Working+ | 2026-07-26 | Session 32 (Block 023 Target 4 + diagnostic probe): the SHAPE of the basis curve distinguishes balance-sheet-driven from price-driven constraints. **(1) Normal short-end binding (1m -60bp, 5y -15bp)** = balance-sheet/velocity regime. Short tenor widest because dealer rolls the position most frequently; constraint binds there first. **(2) Inverted (5y wider than 1m)** = price-driven regime. Long-end TP shock, long-end USD asset selling pressure, or credit/long-tenor event. Established in diagnostic probe (USD/JPY -22→-45bp, EUR -9bp, 1m -60bp, 5y -15bp) — user correctly identified non-inverted shape with normal short-end binding at uniform stressed levels = JPY-specific balance-sheet tightening. **PM use:** shape distinguishes balance-sheet from price-driven regimes. Anki card 1785099069067 (term-structure shape) pushed. Re-probe Block 022 verification. |
| **FX — Three-signal-layer diagnostic (Block 014 + 023 tie-in) (NEW)** | Working+ | 2026-07-26 | Session 32 (Block 023 Target 4): three-signal-layer diagnostic carried forward from Block 014, applied to CIP basis. Three legs: SOFR-term-vs-OIS (domestic bank funding), cross-currency basis (global dollar funding), MBS-Treasury (mortgage market). **Combinations identify which layer broke:** all three expand = systemic crisis; cross-currency only = foreign demand shock; SOFR-OIS only = domestic bank funding; MBS only = prepayment; SOFR-OIS + cross-currency = bank + global funding; cross-currency + MBS = foreign demand + refi wave; SOFR-OIS + MBS = bank + mortgage (early-cycle). Order of Fed facility activation = which layer broke first. Anki card 1785099069195 (three-signal-layer diagnostic) pushed. Re-probe Block 022 verification. |
| **FX — March 2020 scenario diagnostic (NEW)** | Working+ | 2026-07-26 | Session 32 (Block 023 close probe): USD/JPY 3m -100bp + SOFR-OIS +30bp + MBS-Treasury -50bp + DXY +5% + EM FX -5-10%. User correctly identified: (1) systemic global dollar funding crisis — all three legs blowing out + DXY up + EM down. (2) Position: long basis = bet Fed response compresses; cleaner expression: long vol in FX (JPY basis indicates JPY stress). Trade expressed as bet on Fed response, not level. (3) Fed facility: cross-currency basis broke first March 9 → swap lines activated March 15. Standing repo + QE followed. Order of activation = which layer broke first. **SOFR-OIS naming audit noted:** legacy "SOFR-OIS" replaced by "SOFR-term-vs-OIS basis" or "EFFR-SOFR" depending on context (anchored for future cleanup pass). Re-probe Block 022 verification. |

## Session 32 (Block 023 — full close) — process changes

- **2026-07-26 — Two-population conflation pattern (Session 32, Block 023 Target 3):** when a partially-right probe answer has the *direction* right but the *population* or *sign* wrong, the partial-right response is itself data that the population distinction wasn't anchored. The carry-vs-basis population distinction is the Block 023 worked example. Codified in `~/.hermes/skills/finance/macro-mentor/references/carry-vs-basis-population-distinction.md`.
- **2026-07-26 — Diagnostic probe re-anchor rule (Session 32, Block 023 Target 2):** open diagnostic probes that aggregate prior probes' substrate need a 2-3 sentence re-anchor of the substrate the probe is aggregating (per Session 31 process rule). Re-posed the USD/JPY -22→-45bp / EUR -9bp / 1m -60bp / 5y -15bp probe with prior substrate anchored; substrate landed cleanly with the velocity-vs-shape distinction as the diagnostic.
- **2026-07-26 — SOFR-OIS naming audit (Session 32, Block 023 close):** legacy "SOFR-OIS" used in the deck is imprecise. Future substrate should use "SOFR-term-vs-OIS basis" (for the tenor/funding-friction read) or "EFFR-SOFR" (for the bank-funding-stress read). Standardization cleanup pass queued for next session.