The Oil Anomaly: What the Price Is Telling Us and Why I'm Still Not Wrong

The Question Every Honest Analyst Must Answer First

On February 28, 2026, the United States and Israel struck Iran. Within days the Strait of Hormuz was effectively closed — and has remained so for 75 days as of this writing. If you had told any serious oil economist on day one that throughput would collapse from 20 million barrels per day to just over 2 million at its worst in March, recovering only partially to approximately 3.8 million by April, they would have said Brent at $150–$200. Every serious physical market participant was saying it. The physical market briefly agreed: North Sea Dated spot peaked at $144.42 on April 8 — the highest since S&P Global Platts began publishing the measure in 1987. But the futures market — the number quoted everywhere — told a different story, and that distinction is the most important analytical correction in this paper.

Before measuring the gap, we need to be precise about which price we are talking about — because the answer changes the entire framing. There are three distinct oil prices, and most commentary conflates them. Brent spot — the EIA-confirmed daily benchmark reflecting immediate physical delivery — peaked at $138/barrel on April 7 (EIA STEO May 12, 2026). ICE front-month futures — the paper contract pricing delivery approximately one month forward — ran approximately $30/barrel below the spot price at that peak moment, meaning front-month futures were closer to $108 while physical barrels were pricing at $138. North Sea Dated Brent — what European and Atlantic basin refiners actually pay for real barrels loading now — peaked at $144.42 on April 8 (S&P Global Platts, the highest since the benchmark was established in 1987), with the IEA's own April OMR referencing physical prices near $150 at the acute moment of scarcity for specific cargo transactions. Cash Dubai — the benchmark for physical Middle Eastern crude delivered to Asian buyers — broke $170 on March 19, touching an intraday high of $176.80, the highest price recorded for the Cash Dubai benchmark since its establishment as a formal crude pricing reference — surpassing the 2008 Brent peak of $147 in nominal terms and confirmed as a record by Rory Johnston (Commodity Context) in real time. The spot-to-futures spread alone tells the story: spot ran $30/barrel above front-month futures at the April 7 peak — the EIA STEO's own language. North Sea Dated ran $35/barrel above ICE futures at the mid-April peak (IEA May OMR, exact figure). Cash Dubai ran approximately $57–$70/barrel above ICE front-month futures on March 19 — derived from the $176.80 intraday high against the ICE front-month closing price of $107.58 ($69.22 spread) and intraday high of $119.13 ($57.67 spread) per TradingView/ICE continuous contract data, representing roughly thirty to seventy times the normal $1–$2 spread in calm markets. These are not rounding errors. They are physical markets simultaneously confirming what the paper market was refusing to price: the shortage was real, it was acute, and it was hitting buyers of immediate barrels hardest. The "anomaly" is a futures market phenomenon. The paper price was anchored by resolution expectations. The spot and physical markets — all of them — had already rendered their verdict.

This paper answers that question with every data point I can find. The bear case gets a fair hearing. The bull case gets the data it deserves. And the probability distribution over the three possible interpretations — timing shift, magnitude revision, or structural change — gets stated explicitly, with falsification conditions attached.

Note: Two distinct physical-to-futures spreads tell two different stories. The $35/bbl spread (North Sea Dated Brent vs ICE futures, mid-April) is the IEA May OMR's exact figure. The $57–$70/bbl spread (Cash Dubai vs ICE front-month futures, March 19) is calculated from the Cash Dubai intraday high of $176.80 against the ICE front-month closing price of $107.58 and intraday high of $119.13 on March 19, 2026 — representing roughly thirty to seventy times the normal $1–$2 Cash Dubai-to-ICE spread in calm markets. These are not the same market pricing the same thing. They are two markets — paper and physical, West and East — diverging at historically unprecedented levels simultaneously. Sources: IEA May OMR 2026 (exact $35/bbl Dated-ICE spread) · Rory Johnston / commoditycontext.com (Cash Dubai $176.80 record) · TradingView UKOIL continuous contract March 19, 2026 · EIA STEO May 12, 2026.

Why the Price Hasn't Moved Higher — Five Structural Explanations

A note on what this section explains and what it does not. The five structural buffers below explain why the general price level is not $200 — they describe the forces that have absorbed supply shock pressure. They do not fully explain the specific futures-physical spread ($35–$70/barrel between ICE futures and physical cargo prices). That spread exists for a separate reason: futures markets price forward delivery with a resolution discount — the probability-weighted expectation of diplomatic resolution compressed into a single forward price. Physical markets price immediate delivery with no resolution discount. The buffers suppress both prices; the resolution expectation creates the gap between them.

The single most important fact that my original model underweighted: the world entered this crisis with more oil in storage than at any point since February 2021. And China — the player whose strategic behaviour I described in Paper 2 — had been building inventory for 12 months at a rate that now looks like deliberate preparation for exactly this scenario.

1. The Pre-War Inventory Cushion

The IEA's March 2026 report confirmed global observed inventories stood at over 8.2 billion barrels at the start of the crisis — the highest level since February 2021. This is the first number that explains everything that follows. A net supply loss of 12.8 mb/d — confirmed by the IEA May OMR after all bypass routes, Iran's continued exports, and non-OPEC surges are accounted for — hitting a system already running near-record inventories takes far longer to reach operational stress than the same shock hitting a lean system. The elevated starting inventory is the primary reason the price anomaly exists at all.

My original modelling assumed inventories were in the normal operating range. They were not. They were elevated. This was a genuine analytical miss — not because the data was unavailable, but because I weighted the supply shock magnitude too heavily against a demand-side framework when I should have started with the inventory level as the primary variable. The lesson: in a supply shock, inventory is the clock. Everything else is a multiplier.

2. China's 1.4 Billion Barrel Inventory Fortress — The Precise Breakdown

This number requires precise disaggregation, because the composition matters as much as the total. China entered the crisis with an estimated 1.4 billion barrels in total oil inventories as of December 2025 — but this figure is the EIA's deliberately broadened definition combining two distinct layers. Government-held strategic reserves (the true SPR equivalent): approximately 360 million barrels — roughly comparable to the US SPR of 413 million barrels. Commercial crude oil inventories at refineries and tank farms: an estimated 1 billion barrels — against only 411 million barrels held commercially in the United States. The EIA combines both layers into the 1.4 billion figure specifically because China's NOCs were directed in 2024 to treat commercial stocks as emergency reserves, making them functionally strategic. China's real asymmetric advantage over the West is therefore not the government SPR layer — which the US roughly matches — but the additional 1 billion barrels of state-directed commercial inventory that has no OECD equivalent. China's 1.4 billion total exceeds the combined government-held strategic reserves of all 32 IEA member nations (approximately 1.2 billion barrels). Beijing does not need to buy a single barrel on the open spot market. It is drawing inventory while the rest of the world competes for scarce supply. And it is not releasing reserves to help the global market — it is keeping them as strategic currency for precisely as long as the crisis serves its interests.

3. The IEA's 400 Million Barrel Emergency Release — And the US SPR Drawdown

In March 2026, the 32 IEA member nations coordinated a 400 million barrel emergency reserve release — the largest coordinated release in the IEA's history. The US component alone is authorised at 172 million barrels. The drawdown has accelerated sharply, as the table below shows — all figures from EIA Weekly Petroleum Status Reports:

The correct framing: the SPR releases provided a critical buffer that delayed when operational stress thresholds are reached. They did not solve the problem. Each week of releases is one week closer to the point where the buffer is exhausted and the physical depletion path reasserts itself without the cushion. The JPM chart's operational stress threshold of 7.6 billion barrels is being approached not because the releases failed — but precisely because they succeeded in masking the underlying depletion long enough to keep prices from spiking to $200.

4. The Bypass Routes and the Full Supply Balance

The IEA April OMR provides the most precise single measure of the export shock: Gulf exports across all routes — Hormuz plus all bypass pipelines — plunged by 15.8 mb/d month-on-month to 8.7 mb/d in March. That is the starting point. Everything that follows explains the components behind that number and what happened to them.

Pre-war baseline — why three different figures exist. Depending on what products are included, different sources give different pre-war Hormuz flow figures. OIES/Kpler put February 2026 Hormuz exports at 18 mb/d (crude + condensates + refined products). The IEA's broader figure of ~20 mb/d includes LPG and NGLs. Pre-war bypass routes carried a further 3.9 mb/d (IEA April OMR). The IEA's own export accounting implies total pre-war Gulf exports across all routes of approximately 24.5 mb/d — simply 8.7 mb/d post-war plus the 15.8 mb/d stated drop. These figures are not contradictory; they measure different product baskets. The 18 mb/d figure is used for Hormuz-specific comparisons throughout this paper; the 15.8 mb/d export drop and 12.8 mb/d net supply loss are the IEA's own anchor figures.

The bypass routes — what they achieved and what they cannot do. When the war began, Saudi Arabia immediately scaled its east-west Petroline from approximately 2 mb/d toward its maximum. Kpler data shows Yanbu loadings reaching 4.0–4.2 mb/d in the week of April 6 — constrained by port terminal capacity at Yanbu, not pipeline capacity. The UAE redirected via ADCOP to Fujairah. The ITP pipeline from Iraq to Ceyhan contributed approximately 250,000 b/d in March — broadly consistent with its pre-war run rate, meaning Iraq provided minimal incremental bypass capacity above pre-war levels. Combined: the IEA April OMR confirmed alternative route exports reached 7.2 mb/d in April — up from 3.9 mb/d pre-war, a net increase of 3.3 mb/d. There is however a structural limitation that matters enormously for product markets: the bypass pipelines carry crude oil only. Refined products, LPG, and NGLs that previously flowed through Hormuz have no pipeline alternative whatsoever. This is why product crack spreads — distillates, jet fuel, naphtha — reached historic extremes even as crude futures appeared relatively contained. The crude bypass worked. The products bypass does not exist.

Iran's continued exports to China. The factor most analysts missed in early modelling. Iran exported approximately 2.16 mb/d in February 2026 — its highest level since July 2018 — entirely to China. After the war began, Iranian crude continued flowing: TankerTrackers confirmed at least 11.7 million barrels transited the Strait to China in the first two weeks (approximately 0.84 mb/d). For the full month of March 2026, UANI's tanker tracker recorded total Iranian physical exports of 35.7 million barrels — averaging 1.136 mb/d — a 45% drop from February's pre-war peak but still a meaningful continued flow into one market. After the US naval blockade of Iranian ports began on April 13, exports declined further — shadow fleet operations, ship-to-ship transfers near Singapore, and overland rail routes mean exports continued at an estimated 0.5–0.8 mb/d post-blockade.

Non-Gulf supply surge. The IEA May OMR confirmed Atlantic Basin output increased by approximately 1.6 mb/d above pre-war levels, with notable gains from the United States, Brazil, Canada, Kazakhstan and Venezuela. Venezuela's shipments rose from approximately 800,000 b/d in December 2025 to 1.2 mb/d in April — the highest monthly level since late 2018. As Rice University's Francisco Monaldi noted: "Venezuela helps; every little bit helps. But in the grand scheme of things, it doesn't change the equation. There is no medium-term solution other than reopening the Strait." Guyana's production reached 926,550 b/d in February 2026, with exports flowing one-third to the US and two-thirds to Europe. Argentina showed no confirmed crisis-response surge.

The net supply loss — the anchor number. After the bypass routes, Iran's continued exports, and the Atlantic Basin surge are all factored in, the IEA May OMR confirms the net global supply loss: 12.8 mb/d. Gulf production stood 14.4 mb/d below pre-war levels. The EIA STEO May 12 independently confirms production shut-ins of 10.5 mb/d in April. The gap between the 15.8 mb/d export loss and the 12.8 mb/d net supply loss reflects two opposing forces: partial offsets — bypass route expansion (+3.3 mb/d above pre-war) and Atlantic Basin production surges (+1.6 mb/d) — working to reduce the net impact; and additional losses from Gulf production shut-ins, supply imprisoned in Gulf floating storage, and infrastructure damage — working to worsen it. The precise magnitude of each cannot be cleanly isolated from public data and is not independently calculated here. The 12.8 mb/d is the IEA's own confirmed bottom line. The world was receiving 12.8 mb/d less oil than before the war, after everything the market threw at the problem.

5. Demand Destruction Arrived Faster Than Modelled

The IEA estimates demand has contracted by 420,000 barrels per day for full-year 2026, with Q2 2026 showing the sharpest quarterly demand contraction since Covid-19 — approximately 2.45 million barrels per day. S&P Global Commodity Insights expects refinery crude runs to decline 5.2 million barrels per day in Q2 year-on-year (S&P Global Commodity Insights, April 2026 Oil Market Outlook). Asian petrochemical producers have already cut operating rates sharply as feedstock supply dried up. Airlines are reducing schedules. Industrial users are switching fuels.

Demand destruction is not a neutral variable that balances the supply loss — it is part of the mechanism by which the market clears at ~$106 rather than $200. But it comes at a macroeconomic cost: the destruction is real GDP, not just a statistical offset. When oil prices eventually normalise, the destroyed demand does not fully restore. The sectors that cut deepest — petrochemicals, aviation, manufacturing — take quarters, not weeks, to restart. This matters for the CPI transmission thesis because it means the stagflation scenario is not contradicted by demand destruction. The sectors destroying demand are not the same sectors transmitting inflation. A South Korean petrochemical plant shutting down does not lower the price of trucking flour from Kansas to Cleveland. The diesel-freight-food corridor is structurally inelastic in the short run. It transmits regardless.

For India specifically, the demand destruction argument misses something critical: India has no bypass option for LPG. Cooking gas for 300 million households flows almost entirely through Hormuz. The LPG price shock is not an abstraction. It is a direct tax on every kitchen in the country — hitting in the same month as food prices rise from elevated freight costs, and services inflation begins. This is the human transmission mechanism that CPI models understate.

What the Buffers Cannot Change: The Inventory Cliff

The five buffers above explain the current price. They do not change the destination. They only determine when the destination is reached — and how fast.

Here is the critical distinction that current market pricing misses. The buffers explained above describe why the price is ~$106 rather than $200. They do not address what happens when those buffers are exhausted. The JPM inventory chart and the Morgan Stanley price regime distribution below say the same thing in two different ways: the inventory depletion path leads to a cliff that is now four to six weeks away, not twelve.

The Inventory Depletion Arithmetic — Updated and Verified

The IEA May 2026 OMR provides the complete picture. Global observed inventories drew by 129 million barrels in March and a further 117 million barrels in April — a combined draw of 246 million barrels (129 + 117 = 246 mb; the IEA May OMR rounds this to 250 mb in its headline summary — "drawn down by 250 mb over March and April, or 4 mb/d"). The pre-war starting level per the IEA March OMR: "more than 8.2 billion barrels, the highest level since February 2021." Applying the confirmed draws: 8,210 − 129 − 117 = 7,964 million barrels as of May 1 — still 364 million barrels above the JPM operational stress threshold of 7.6 billion barrels. At the EIA STEO May 12 confirmed Q2 draw rate of 8.5 mb/d: 364 mb ÷ 8.5 mb/d = 42.8 days from May 1 = approximately June 12. The arithmetic is transparent and independently verifiable from three primary sources: IEA March OMR (starting level), IEA May OMR (March and April draws), EIA STEO May 12 (Q2 draw rate). The Aramco CEO confirmed cumulative gross supply losses exceeding 1 billion barrels since February 28.

Confirmatory evidence — gasoline demand destruction is not happening. Within the total liquids draw, US gasoline inventories fell by 4.084 million barrels in the week ending May 8 to 215.7 million barrels — nearly 50% larger than the 2.85 million barrel draw forecast — and are now 5% below the five-year seasonal average. Three consecutive weeks of larger-than-forecast gasoline draws directly falsifies the demand destruction narrative at the consumer level: households are not reducing gasoline consumption at a rate that offsets the supply shock. The diesel-freight-food corridor remains under pressure even as the macro narrative prices in resolution.

At the current drawdown rate of approximately 8.5 million barrels per day in Q2 (EIA STEO May 12), the JPM operational stress threshold of 7.6 billion barrels is reached in approximately 42 days from May 1 — approximately June 12. The operational floor of 6.8 billion barrels — below which pipelines cannot maintain pressure and refineries cannot operate — arrives by September 2026 under the no-resolution scenario.

The Floating Storage Signal

The IEA data on floating storage and inventories requires careful reading — the numbers describe two opposing dynamics happening simultaneously, and conflating them produces the wrong analytical conclusion.

Inside the Gulf — floating storage rising: The IEA April OMR confirmed that floating storage of crude and oil products inside the Middle East rose by 100 million barrels in March — tankers loaded and ready but unable to exit through Hormuz. Onshore crude stocks inside the Gulf rose a further 20 million barrels in the same month. This is supply physically imprisoned. It has not reached any refinery outside the region.

Outside the Gulf — total inventories falling sharply: Stocks outside the Middle East Gulf fell by 205 million barrels (−6.6 mb/d) in March per IEA April OMR. This figure covers total observed inventories outside the Gulf — both land-based stocks and oil on water combined — as refiners and consumers scrambled to replace lost Gulf supply. In April, on-land stocks fell a further 170 million barrels (−5.7 mb/d) while oil on water increased by 53 million barrels — meaning the inventory draw in April was concentrated entirely in land-based stocks, while floating storage globally continued to build, reflecting continued Gulf tanker congestion rather than any recovery in supply availability.

The net picture: Global observed inventories drew by 129 million barrels in March and a further 117 million barrels in April — a combined draw of 246 million barrels (129 + 117 = 246 mb; the IEA May OMR rounds this to 250 mb in its headline summary). The IEA's confirmed Q2 draw rate of 8.5 mb/d (EIA STEO May 12) reflects this depletion trajectory. The supply imprisoned inside the Gulf in floating storage is deferred supply — when the Strait reopens those barrels will eventually reach refineries. But the land-based inventory draw already executed is permanent damage to the global buffer. The Aramco CEO confirmed the tanker fleet is "mixed up — in the wrong places" — a physical logistics constraint that adds 3–5 weeks to the restoration timeline even in a best-case diplomatic scenario.

The Morgan Stanley Price Regime Distribution: Where $106 Futures Actually Sits

Morgan Stanley Research published a distribution of inflation-adjusted Brent crude prices since 2007. The structural insight from that work: oil prices do not distribute normally — they cluster into two distinct regimes with a trough between them. The Supply Destruction Regime (~$52–$100) describes prices at which the market is roughly balanced or oversupplied: low enough that marginal producers get squeezed and supply growth slows, but not yet high enough to force buyers to stop buying. The Demand Destruction Regime (~$124–$182) describes prices so high that airlines cancel routes, factories switch fuels, and consumers stop driving — demand itself breaks. The no-man's land between them (~$100–$124) is the historically unstable trough where prices do not linger — they resolve toward one regime or the other. The $124 demand destruction entry point is independently confirmed by JPMorgan Commodities Research. Historically, prices have spent almost no time in this gap. They resolve — one direction or the other.

Here is the critical nuance for 2026: the current ~$106 ICE futures price sits in that gap — but not because the physical market agrees with the futures market. It does not. Cash Dubai hit $176.80 in late March. North Sea Dated peaked at $144.42 on April 8. Japanese refiners paid $140–$200 per barrel (Japan Ministry of Economy, Trade and Industry (METI), average refiner acquisition cost confirmed via Reuters, April 2026). The physical market was firmly inside the Demand Destruction Regime. The futures market is pricing something different: the probability-weighted expectation of diplomatic resolution. The gap between physical prices and futures prices is the same gap you see in the distribution chart — measured two different ways. The paper market is in no-man's land. The physical market already crossed into demand destruction territory and came back as inventories were released. The question the inventory clock answers is which regime futures converge toward next.

The Two-Phase Price Resolution: Where $106 Goes — And What the Debt Structure Changes

The Morgan Stanley framework tells us where $106 sits. History and the current debt structure of the economy tell us where it goes from here — and why the resolution mechanism is more violent and faster than in any prior oil shock, regardless of which path the Fed is forced onto.

Phase 1 and Phase 2 are sequential, not contradictory. The March 2026 physical market was a miniature preview: Cash Dubai spiked to $176.80 (Phase 1), then partially retreated as SPR releases provided temporary Door 1 relief — demand destruction and buffer supply absorbing the acute pressure. When the SPR is exhausted, Phase 1 reasserts.

Door 1 — Hold rates: A more leveraged system breaks faster and at a lower sustained oil price than any prior cycle — the 719% all-sector debt/GDP load means the economy is acutely sensitive to demand destruction. What appears to be a controlled recession is not controllable in a system built on this much leverage. Falling asset prices trigger margin calls. Margin calls trigger forced selling. Forced selling triggers further declines. Private credit gates. Corporate debt rolls over at punishing rates. Unemployment rises. Tax revenues fall. The deficit widens. Treasury issuance increases precisely when Japan and G20 creditors are already selling. The spiral is reflexive and self-reinforcing — each step amplifies the next. This is not a recession. This is 1929 — where the Fed's refusal to accommodate transformed a supply shock into a debt deflation spiral that lasted a decade. Door 1 in 2026 is that same refusal, dressed as monetary discipline. The oil price falls. The economy does not recover cleanly.

Door 2 — Print: The Fed monetises the debt, choosing inflation over depression. In nominal terms this looks like relief — rates fall, markets rally, credit flows again. But oil is priced in dollars. When the Fed prints, the dollar buys less oil, not more. Every barrel that previously cost $150 now costs $160, then $170, as dollar purchasing power erodes against the hardest physical commodity on earth. The petrodollar architecture — 50 years of pricing oil in US dollars, recycling those dollars into Treasuries, suppressing US borrowing costs — now runs in reverse. Dollar weakness sends oil higher. Higher oil sends CPI higher. Higher CPI forces the Fed to choose again: print more or hold. If it holds after having already printed, credibility is gone and the bond market does not recover. If it prints again, the spiral accelerates. The gold/oil ratio — historically mean-reverting around 15–20 barrels per ounce (World Gold Council / EIA historical data, 50-year average approximately 17 barrels per ounce of gold; current ratio at ~$4,539 gold (May 16, 2026 spot) / $106 oil ≈ 43 barrels per ounce, more than double the historical mean — the widest divergence since the 1970s oil shocks, suggesting the dollar is being debased against hard assets and oil is significantly underpriced in real terms relative to gold) — reasserts itself as the market reaches for a non-sovereign unit of account. Door 2 has no 2008 analogue. It leads somewhere the modern debt-laden system has never been — and where Bitcoin's role as the only mathematically scarce non-sovereign savings technology becomes impossible to ignore.

At ~$106, futures are in the gap — pricing neither normal supply-demand balance nor the demand destruction that physical buyers have already been experiencing. The gap is unstable by definition. Prices resolve toward one regime or the other. The JPM inventory depletion path — operational stress approximately June 12, operational floor by September — is the mechanism that determines which direction. If inventories reach operational stress thresholds before a credible resolution, futures converge toward the demand destruction regime. If resolution arrives and the SPR releases hold prices down through the restoration lag, futures converge back toward the supply destruction regime. There is no stable equilibrium at $106. The inventory clock is the deciding variable.

The Bear Case for Oil — And Why It Has Real Arguments

Intellectual honesty requires this section. The bear case is not stupid. It has specific, data-grounded arguments that could make the "oil price stays contained" scenario run further than I modelled. Here they are, stated as strongly as possible.

Why $100–$110 ICE Brent Futures Is Still Enough to Validate the Thesis

Here is the critical point that the oil price anomaly conversation obscures. My CPI thesis does not require oil at $200. It never did. The three-stage thesis was calibrated for oil in the $100–$152 range — the base-case CPI projection of 5.5–6.2% applies at $100–110 futures, while the upper-bound projection of 6.8–7.5% (Paper 9's original range) requires oil reaching $148–152 as modelled in Part II. ICE Brent futures have been above $100 since March 12. The physical market averaged $120.36 in April per North Sea Dated. The transmission is not delayed — it is running exactly as modelled. And the fact that futures markets are discounting physical reality by $3–$35/barrel depending on the day is itself confirmation of the thesis: paper markets are pricing resolution; the physical market is pricing the actual shortage.

The Transmission Chain Is Running On Schedule

The CPI transmission from an oil shock follows a fixed sequence with known lags. Every stage is now either confirmed by data or precisely dated for arrival. The projections below are built from base effect arithmetic using BLS-confirmed 2025 monthly figures — not assumptions. All calculations assume oil remains at $100–110 (current ICE futures). The $152 peak scenario modelled in Part II would produce materially higher prints — see the 'upside scenario' note below.

Step 1 — Transmission lag sequence

Step 2 — Base effect arithmetic (month-by-month, all sourced)

The Demand Destruction Complication

Here is where honest analytical tension lives. Demand destruction is real — the IEA estimates 420,000 barrels per day of net demand has been removed from the full-year 2026 global system as an annual average, but the Q2 acute figure is far larger: approximately 2.45 mb/d of demand contraction in Q2 alone — the sharpest quarterly demand destruction since Covid-19. The annual average is diluted by Q1 (pre-war) and the partial Q3–Q4 recovery; the Q2 figure is the crisis-period reality. The S&P Global Q2 demand decline is more than twice the worst quarter of the 2008–2009 recession. Two things are simultaneously true and must be held together: 2.45 mb/d is historically large as a demand contraction — it would constitute a severe recession in isolation. It is simultaneously insufficient to close a 12.8 mb/d supply gap — it addresses less than a fifth of the supply loss. The first statement explains why economists are not alarmed. The second explains why the market is wrong. Both are true. This has two effects that partially offset each other in the CPI model:

On one hand, demand destruction reduces the inflationary pressure: less economic activity means less demand for goods, services, and energy. This is deflationary. On the other hand, demand destruction is itself an economic contraction — higher unemployment, lower corporate earnings, more Fed pressure to cut. Benjamin Cowen's ITC Business Cycles Composite (M2-Normalised) is descending — when it peaks and rolls over it has historically preceded the end of the business cycle (full formula and methodology in Plain Sight Paper 1; source: app.intothecryptoverse.com). The Sahm Rule is approaching — the Sahm indicator (0.5 percentage point increase in the 3-month average unemployment rate relative to the prior 12-month low) has historically triggered at the onset of every recession since 1970; current trajectory confirmed in Paper 9 modelling. This is the mechanism that forces the Fed's hand — not a contradiction of the thesis, but the confirmation of it.

The net effect: the CPI peak may arrive at the lower end of the 5.5–6.2% corrected range rather than the upper end if demand destruction is larger than modelled. This is the primary honest uncertainty in the thesis — not the oil price direction, but the demand destruction offset magnitude.

The reconciliation between elevated demand destruction and sustained CPI transmission is this: demand destruction and inflation are not mutually exclusive — they are the definition of stagflation. The sectors experiencing demand destruction (petrochemicals, aviation, manufacturing) are not the same sectors driving CPI transmission (food, freight, services). A South Korean petrochemical plant cutting runs does not reduce the cost of trucking groceries in Ohio. The transmission chain operates through the diesel-freight-food corridor, which is inelastic in the short run regardless of whether industrial demand is contracting. This is precisely why the 1970s produced both severe recession and 10%+ inflation simultaneously. The demand destruction complication does not weaken the macro thesis — it strengthens it. The Federal Reserve will face a deteriorating labour market and stubborn above-4% CPI simultaneously. It cannot hike without breaking employment. It cannot cut without re-igniting inflation. There is no clean exit from that position. That is the stagflation trap — and it is the condition that ultimately forces the Fed's hand.

The Causal Chain: How the Oil Price Gap Resolves Into a Fed Crisis

Part II described what happens to the oil price — the acute spike, the forced Fed choice, the two doors. This section describes what that oil price does to the economy and the Fed — the CPI transmission sequence, the probability-weighted policy paths, and the three-stage thesis that follows. These are sequential layers of the same argument, not repetition.

The oil anomaly is not the end of the story. It is the beginning of it. The buffers that explain why futures are at $106 rather than $200 are temporary by definition — inventory drawdowns, SPR releases, and demand destruction do not restore supply, they delay the reckoning. Where the oil price resolves from here determines everything that follows: CPI trajectory, Fed optionality, private credit stress, and ultimately whether the three-stage thesis plays out on its original timeline or an accelerated one. Here is the probability-weighted path from current oil prices to Fed policy outcomes — stated explicitly, with every assumption visible.

Three interpretations of the oil anomaly carry different consequences for the macro thesis. Each is stated with its probability, its oil price implication, and its direct effect on the Fed stages. Read them as a decision tree, not a single forecast. Note: the oil price scenario probabilities (S1–S4) in the charts below measure which Brent price range materialises. The interpretation probabilities (A/B/C) here measure how the current anomaly resolves. They overlap but are not identical questions.

Note on probability layering: Interpretation A (55%) measures whether the anomaly resolves as a timing shift — whether the inventory clock wins before diplomatic resolution. S4 (43%) measures whether Brent sustains $140–$170 for 6–18 months. The ~12% gap between them represents the portion of Interpretation A scenarios where the inventory clock produces an acute price spike above $140 but the price does not sustain in that range for 6–18 months — i.e. a sharp spike followed by faster-than-expected demand destruction or a surprise resolution that brings prices back below $140 within one to two quarters. In these scenarios, Interpretation A executes (timing shift confirmed, inventory cliff reached) but S4 does not fully materialise (sustained price fails to persist). Interpretation A is the trigger; S4 is the sustained outcome. They overlap but answer different questions.

The two charts below make the causal chain visible. The oil scenarios chart (first below) tracks the conditional probability update from April 3 to May 14 — S4 (prolonged elevated, $140–$170) is now the single highest-probability outcome at 43%, up from 33% on April 3, as 75 days of unresolved crisis and confirmed 12.8 mb/d net supply loss narrow the range of plausible outcomes. The Fed stages chart (second below) shows the direct consequence. Stage 1 is now 90% — not 100% — because Warsh could legitimately treat the oil shock as a one-time exogenous supply-side event and issue hawkish forward guidance without a hike. The historical precedent exists: the Fed held rates at the 1990 Gulf War oil shock and issued guidance instead. That 10% optionality matters and is not discounted here.

What Has Changed in the Model vs. What Has Not

Let me be precise about the specific revisions and what remains unchanged.

The Answer to the Question

Why is ICE Brent futures at ~$106 (May 14) rather than $200? Because the world entered the crisis with the largest oil inventory buffer since 2021; because China built a 1.4 billion barrel inventory fortress and does not need to buy on the open market; because bypass routes increased by 3.3 mb/d above pre-war levels; because the IEA released 400 million barrels in March; because the Atlantic Basin surged 1.6 mb/d above pre-war output; because IEA-confirmed Q2 demand contraction of 2.45 mb/d reduced consumption; and because global floating storage outside the Gulf was immediately drawn down by scrambling buyers. And critically: because ICE futures markets were pricing a resolution probability that the physical market was not — the $57–$70/barrel premium of Cash Dubai over futures in March, and the $35/barrel premium of Dated Brent over futures in mid-April, were the physical markets saying the shortage is real, right now, whatever the paper says. The IEA's confirmed net global supply loss after all of these factors: 12.8 mb/d. That is the number the inventory clock is running against.

Does any of this change the thesis? Here is the precise update, component by component. The detailed revision notes are in the "What Has Changed" timeline in Part V above; this table is the executive summary.

What the market is doing is using the buffer supply to pretend the operational cliff isn't approaching. The JPM chart shows what happens when that buffer runs out. The Aramco CEO confirmed the tanker fleet is "mixed up — in the wrong places." The IEA May OMR confirmed the market remains "severely undersupplied until October even if the conflict ends next month."

While sovereign desks debate basis points and family offices reposition portfolios, a family in Mumbai is paying more for cooking gas than at any point in their lives. A truck driver in Ohio is watching his diesel bill consume his margin. A Korean refinery is bidding for barrels that don't exist at a price that doesn't show on any terminal. The macro thesis is abstract. The transmission is not.

The reserves are finite. The clock is not.