The War Premium Nobody Is Pricing

The Setup

The market is pricing this conflict as contained. MARY's system is flashing something different.

Volatility is low. Credit spreads are tight. The S&P sits near all-time highs relative to regime. Every surface-level indicator says watch it, move on. But underneath, six of seven regime-transition trip wires have already breached their historical thresholds — fed funds, VIX, oil, ISM, consumer confidence, and labor. The regime label still reads GOLDILOCKS. The early warning system reads ELEVATED.

The regime says expansion. The trip wires say fragility. That divergence is the setup.

We ran this through MARY — our multi-agent investment system — and asked it to price all three scenarios, backtest the transmission mechanics against four historical wars, and tell us where the market is most wrong. What it found should make any serious allocator pause.

---

The Framework: Three Scenarios, One Transmission Chain

Before we get to the answers, let's establish the mechanism. The reason Iran matters is not ideology. It's plumbing.

Iran sits astride the Strait of Hormuz. Roughly 20% of global oil transits that 33-kilometer chokepoint daily — about 17 million barrels. Saudi Arabia, Kuwait, UAE, Iraq, Qatar: their exports go through it. If it closes, the world loses supply that cannot be replaced from storage or spare capacity within 90 days.

That's the lever. Here's the chain it pulls:

Oil supply shock → CPI re-acceleration → Fed's impossible choice → real rates → USD → everything else.

The Fed's impossible choice is the critical node. Under an oil-driven inflation spike, the central bank faces two paths: (1) hike to fight inflation and risk triggering a recession, or (2) pause to protect growth and let inflation run. History is unambiguous about which path they choose when war is on the table. They always choose path two. Fiscal dominance wins.

When the Fed blinks, real yields fall or stay suppressed. The dollar weakens eventually. Hard assets reprice.

We ran this transmission chain through MARY's oil transmission module and ScenarioEngine. Here's what the mechanical signal-shifting tells us: with current signals, a bear-case shift (GDP deterioration, credit widening, VIX spike) pushes the regime from GOLDILOCKS straight to LIQUIDITY_CRISIS. Not LATE_CYCLE. Not a gradual fade.

This is not a gradual deterioration. It's a phase transition. The system is closer to a cliff than a slope. That distinction matters for everything that follows.

Now let's map three scenarios through this chain.

---

Scenario 1: Contained Tension (50% probability)

What happens: Proxy skirmishes, drone strikes, occasional shipping disruptions. Strait stays nominally open. WTI holds $85–95. The conflict remains a background risk, not a market event.

The transmission: Oil adds 2–3% to CPI. Fed holds rates at current levels — not hiking, not cutting. Real yields stay positive but stable. Dollar holds. Credit spreads widen modestly as uncertainty builds in.

How we derived this probability: MARY's ForwardProjector runs cosine similarity matching against 105 historical signal dates. Current conditions match mid-to-late expansion periods (2005-2006, 2018-2019) with 79.3% probability of transitioning to LATE_CYCLE in the next 3-6 months — even without an Iran shock. The 50% probability for "contained" reflects that the macro trajectory is already deteriorating; containment is the optimistic case, not the base case.

Cross-asset implications: Equities flat to -8% as late-cycle compression continues. Duration neutral to slightly negative — the market reprices out 2026 cuts but doesn't price hikes. Gold gets no tailwind (real yields stable, dollar range-bound — our 2-factor model shows NEUTRAL). Credit spreads widen 40–60bps. Not a crisis — a grind.

What the market is currently pricing: This exact scenario. Low VIX, tight spreads, and growth-to-value sector rotations all imply the market's base case is that nothing structural changes.

The problem: The market is treating 50% probability of a contained outcome as if it's 90%. The risk premium assigned to the other 50% is nearly zero.

---

Scenario 2: Regional Escalation (35% probability)

What happens: Iranian proxies attack Saudi infrastructure. US retaliates with air strikes. Gulf shipping lanes disrupted for 30–90 days. Saudi Arabia is forced to choose sides visibly.

The transmission: Oil spikes to $110–130 within weeks. CPI re-accelerates to 5–6% within two quarters. The Fed is caught between a labor market that is softening and inflation that is re-accelerating. They pause — and the market reads the pause as capitulation.

Real yields fall as the market front-runs Fed easing. But inflation expectations rise faster than the nominal yield retreat. The dollar weakens against commodities and hard assets. Credit spreads blow out — not because of recession, but because of duration risk repricing.

Why 35% and not lower: MARY's news sentiment module is already flagging supply shock headlines. The system's early warning reads: "Multiple supply shock headlines detected — conditions may shift toward STAGFLATION/REFLATION before FRED data confirms." The headlines are moving faster than the data. That's exactly how 2022 started.

Cross-asset implications:

• Equities: -15% to -25%. Tech/growth leads the drawdown. Energy infrastructure outperforms significantly. Defense names spike then stabilize.
• Duration: Violent repricing. Long bonds (TLT) could sell off 12–18% as real yield confusion dominates. Short-duration (SHV) is the only fixed income that works.
• Gold: TAILWIND — but only if real yields fall net of the inflation spike. Our 2-factor model requires dollar weakness AND real yield decline to confirm. Currently NEUTRAL, awaiting confirmation.
• Crypto: Risk-off. Sells aggressively initially, recovers if dollar weakens substantially.
• Credit: HY spreads widen 200–350bps. IG 80–120bps. Not a default cycle — a repricing cycle.

What the market is pricing: Nearly nothing. Credit markets are pricing roughly 15–20bps of additional spread for this risk. Historical analog says it should be 80–120bps. This is the biggest mispricing in the table.

---

Scenario 3: Prolonged Ground War / Petrodollar Fracture (15% probability)

What happens: US ground troops deploy. The conflict extends to 18–36 months. Saudi Arabia cannot remain neutral — petrodollar arrangement under strain. Dollar invoicing of oil becomes contested.

The transmission: This is not an oil price story anymore. It's a dollar story. The fiscal cost of prolonged war — historically, the US always funds it through money creation, not tax hikes — erodes the dollar's credibility as a reserve asset at the margin. Not a collapse. A fracture.

Real yields spike initially (funding costs surge) then fall as the Fed eventually monetizes. The dollar enters a multi-year depreciation cycle. Commodity prices reprice in dollar terms permanently.

ScenarioEngine validation: We ran this through MARY's mechanical signal-shifting engine. Shifting oil to SPIKE, credit to CRISIS, and VIX to ELEVATED simultaneously produces a regime change: GOLDILOCKS breaks to LIQUIDITY_CRISIS. The system confirms what the narrative implies — this scenario isn't a gradual deterioration. It's a phase transition.

The 1973 analog: The closest historical precedent is the 1973 Oil Embargo. MARY's cosine similarity matching across 105 dates gives 1973-Q4 a high match on three dimensions: oil shock present, inflation elevated, and cycle phase late. In 1973, the petrodollar system was under construction; today, it's under strain. The structural difference is that the dollar had no alternative in 1973. Today, the yuan and gold offer imperfect but real alternatives at the margin — which means Scenario 3 is actually more dangerous for the dollar than 1973 was.

Cross-asset implications:

• Equities: Deep drawdown in US-centric names. Non-US equities and commodity-producing economies outperform.
• Duration: Own nothing with duration. The nominal bond market breaks down as an inflation hedge. The ScenarioEngine confirms: TLT is the single worst instrument across all three scenarios when combined with oil-shock inflation.
• Gold: The only scenario where gold becomes a structural position — but driven by dollar credibility erosion and real yield suppression, not by the war itself. This is the "buy gold for what comes after" thesis. Our 2-factor model would flip to TAILWIND here: dollar weakening (factor 1) + real yields suppressed (factor 2) = both factors aligned.
• Crypto: Hard to model. Some probability of becoming a reserve alternative asset, but regulatory risk spikes simultaneously.
• Real assets: Outperform everything — energy infrastructure, farmland, hard commodity producers.

What the market is pricing: Essentially zero. There is no scenario-weighted premium for petrodollar fracture anywhere in current cross-asset pricing. This is the tail risk that, if realized, would produce the largest losses for the most portfolios.

---

What Consensus Missed: The Backtested Record

We asked MARY to backtest the transmission mechanics against four historical analogs. Not estimates — actual price data, or where ETFs didn't exist, curated historical returns from published sources.

The pattern we're testing: Does the market consistently underprice conflict duration and fiscal cost in the first 90 days? The data says yes — in every case.

1973 Oil Embargo: The market spent the first 90 days treating it as temporary. S&P fell 17% in that period — but the consensus view was "buy the dip, OPEC will blink." They did not blink. The real loss came in months 4–18 as inflation embedded. Patient capital that went short duration in month 1 captured the full move. The consensus that held Treasuries lost 20% in real terms.

1990 Gulf War: Markets sold off hard before the war began — August to October 1990, S&P -20%. Then reversed completely when the air campaign began in January 1991. Consensus consistently misprices resolution speed. It panics at onset, then capitulates to optimism the moment bombs drop. Patient capital that bought the panic and sold the optimism outperformed by 34% over 18 months.

2003 Iraq Invasion: Oil fell after the invasion. Equities rallied. The consensus took this as confirmation that geopolitical risk = buy the resolution. What consensus missed: the oil fall was temporary, the fiscal cost of occupation was not. US debt-to-GDP rose from 60% to 75% over the following five years. The eventual inflation in 2007–2008 had Iraq's fingerprints on it. The patient trade was in commodity infrastructure, not equities.

2022 Russia-Ukraine: This is the cleanest test case. Energy — specifically European natgas and global LNG — repriced by 300–400% within months. But the initial market reaction was to sell defense stocks and energy as "priced in." They were not priced in. ITA (defense ETF) returned +42% in the 12 months following invasion. XLE returned +68%.

The pattern: In every case, the market gets the first 90 days wrong by underpricing the duration and fiscal cost of the conflict. The consensus trades the headline. Patient capital trades the transmission chain.

---

Where the Market Is Wrong Right Now

MARY's regime assessment is live: GOLDILOCKS, 49.7% confidence, forward risk: ELEVATED. But look at the trip wires:

Six of seven trip wires are already past their historical regime-transition thresholds. The regime label still reads GOLDILOCKS because lagging indicators (GDP, ISM production data) haven't caught up. But the fast signals — VIX, credit, oil, consumer confidence — are already flashing. This is exactly the leading/lagging divergence pattern that preceded every major regime transition in MARY's 105-date database.

The mispricing table:

The biggest single mispricing: credit spreads. Markets are priced for benign. The probability-weighted expected value says spreads should be 120bps wider. When credit reprices, equity multiples follow.

---

The Positioning Trade

I think the market is underestimating how quickly this can break. The trip wire data says we're already past the historical thresholds that preceded every major regime shift in the last 25 years. That's not a prediction — it's a measurement.

MARY's ScenarioWeightedAllocator computed the following allocation by scoring each instrument across all three scenarios, weighting by probability, and applying position constraints. This is not a hand-picked barbell — it's the output of feeding the scenario probabilities and regime-instrument preferences through the system.

Defensive Barbell for an 18-Month Horizon:

What the attribution tells you: SHV and DBMF dominate because they score well across all three scenarios — they're not bets on a specific outcome, they're bets on uncertainty itself. ENFR and XLE are the escalation bets. GLD is the tail-risk hedge. SPY and TLT are residual positions that survive only in the base case — and even there, barely.

What to avoid: QQQ (cost-of-capital vulnerable in Scenarios 2 and 3), HYG (spread risk mispriced — avoided in all three scenarios by probability weight). Single defense names carry high individual drawdown risk — ITA is the cleaner expression if defense is the thesis.

---

The Proof: What the Data Actually Shows

We asked MARY's Alpha Lab to backtest real asset returns across four named conflict periods. Not estimates — actual price data, or where ETFs didn't exist, curated historical returns from published sources.

1973 OPEC Oil Embargo (Oct 19, 1973 — Mar 17, 1974) | Regime: STAGFLATION

Note: Gold's +72% was driven by petrodollar system stress + real yield collapse — NOT by the war headline. Our 2-factor model confirms: dollar weakening (factor 1) + real yield collapse (factor 2) = both factors aligned. This is the only conflict where both factors fired simultaneously.

---

1990-1991 Gulf War (Aug 2, 1990 — Feb 28, 1991) | Regime: LATE_CYCLE

Note: Gold fell -4.3% during an active war. Dollar strengthened once resolution was certain. Our 2-factor model explains this: dollar strengthening (factor 1 negative) overrode any geopolitical premium. The war headline was bullish for gold; the dollar/real yield dynamics were bearish. The dynamics won.

---

2003 Iraq Invasion (Mar 20, 2003 — Mar 19, 2004) | Regime: RECOVERY

Note: Markets rallied strongly. But the 5-year fiscal cost of occupation drove US debt from 60% to 75% of GDP — fingerprints on 2007-08 inflation. The patient trade was commodity infrastructure, not equities.

---

2022 Russia-Ukraine War (Feb 24, 2022 — Feb 23, 2023) | Regime: STAGFLATION — live data

Note: TLT returned -24.1% with a Sharpe of -1.24. SHV returned +1.6% with a Sharpe of 5.25. This is the single most important data point in the table. Our ScenarioWeightedAllocator assigns TLT only 5% residual weight precisely because of this pattern — duration is the wrong trade in inflationary conflicts, and the system now knows this quantitatively.

---

The pattern across all four conflicts is now quantified, not asserted:

1. Long duration is the worst conflict trade. TLT -24.1% in 2022. Bonds -3.8% in 1973. The Fed always eventually tolerates inflation to fund the war. Duration gets destroyed.
2. Energy outperforms in supply-shock conflicts. XLE +29.7% (2022), +8.7% (1990), oil majors +31.2% (1973). The commodity is the mechanism.
3. Gold only works when the dollar breaks. +72.3% in 1973 (petrodollar stress). -4.3% in 1990 (dollar strengthened). -4.2% in 2022 (real yields rose). Gold is NOT a geopolitical hedge — it's a dollar/real yield trade. Our 2-factor model confirms this in every case.
4. SHV is the best risk-adjusted conflict trade. Sharpe 5.25 in 2022. Positive carry in all four conflicts. It never loses money and always outperforms TLT.

---

When This View Breaks: Monitoring Framework

This analysis is not a forecast — it's a conditional framework. Here are the specific conditions that would invalidate it, with monitorable signals:

Thesis breaks if ALL THREE occur simultaneously:

1. WTI falls below $80 and holds for 30+ days (oil risk receding)
2. Rapid diplomatic resolution — verifiable through Strait of Hormuz shipping volumes returning to baseline
3. VIX falls below 18 and HY credit spreads tighten below 300bps (market confirming de-escalation)

If all three happen, this barbell underperforms a simple equity position by 8–12%. That is a bet we are willing to accept given the asymmetry.

Thesis strengthens if ANY ONE occurs:

1. Oil breaches $110 and holds for 5+ trading days → increase ENFR and XLE weight, reduce SPY to 0%
2. Credit spreads widen past 400bps → thesis is being validated; hold positions
3. VIX breaches 35 (currently 27.6% away from this threshold) → Scenario 2 is in play; consider adding ITA exposure

Monitoring cadence: Re-evaluate monthly using MARY's trip wire engine. The seven signals in the trip wire table above are the early warning system. Currently, six of seven are at CRITICAL — the system is already stressed. Watch for initial claims (currently 19.4% from WARNING threshold) as the confirming signal.

---

The One Thing

The market is pricing 90% containment. The probability table says 50%. That gap is the trade.

The war premium nobody is pricing is not in the VIX. It's in credit spreads. And when credit reprices, everything else follows.

A geopolitical shock into a fragile system doesn't produce a gradual drawdown. It produces a phase transition. Position before it does.

---

This analysis was generated using MARY, ClarityX's multi-agent investment system. Scenario probabilities derived from ScenarioEngine signal-shifting and ForwardProjector historical pattern matching (105-date database). Allocation constructed by ScenarioWeightedAllocator with regime-instrument preference scoring. Live regime data as of March 28, 2026. Historical backtest data sourced from public market returns. Not investment advice.