Context
Each month I have taken the opportunity to reflect on the war in Iran and consider what might happen next. This is the fourth piece in that series.
Eighty-nine days after Operation Epic Fury began, the war is closer to its end than it has been at any prior point. The Mexican standoff I described on 1st May appears to be resolving in substance, with Trump taking a Bath, which the May piece called as the most likely outcome. Trump has not yet signed and surrendered. But the question is less whether he will and increasingly: when he will.
On 23 May US time, Trump posted on Truth Social that a deal had been "largely negotiated, subject to finalization." The deal as described in regional press includes an official declaration of the war's end, a 30 to 60 day window for nuclear talks, gradual reopening of the Strait of Hormuz with Iran continuing to manage access on a fee basis, and US ending its blockade of Iranian ports. Iran's foreign ministry, through spokesperson Esmail Baghaei, publicly described the emerging text as a "framework agreement." Iranian Parliament Speaker Mohammad Bagher Qalibaf travelled to Qatar to take part. Fars News, the IRGC-affiliated outlet that had spent two months denying that negotiations were occurring, confirmed implementation details and corrected Trump's characterisation by stating publicly that Iran would continue to manage the waterway.
This is largely the same deal Iran put on the table on 28 April. The terms have not moved in Trump's favour. The position has been publicly acknowledged by Iran for the first time. The signature has not yet arrived.
Recap of the May Piece
The 1st May piece (sixty-one days into the war) found a conflict that had settled into a structure neither side wanted to call by its real name. The shooting had mostly stopped under a Pakistan-mediated ceasefire from 8 April. The Islamabad Talks of 11-12 April had produced nothing. Both sides had imposed blockades: Iran on the Strait of Hormuz, the United States on Iranian ports. Brent had touched \$126 overnight on the Wednesday before the piece, the highest level since June 2022. Goldman estimated Hormuz exports had fallen to 4% of normal. The Swan and Haberman reporting in the New York Times on 8 April had documented the decision-making behind the war: a Situation Room in which Vance was the lone dissenter, Hegseth said "we might as well do it now" as the substitute for strategic analysis, and Trump's epistemology was captured in his pre-war reassurance to Tucker Carlson that things would be OK "because it always is." Trump had spent two months announcing breakthroughs that did not exist, producing what the piece called a hit rate low enough to constitute evidence rather than anecdote.
The piece identified three paths from there.
- The Long Grind: the standoff continuing because the alternatives were worse, sustained by "on the verge of a deal" Truth Social posts that bought news cycles.
- Taking a Bath: effectively a US surrender, dressed in the language of a great deal, that defers the nuclear question and looks like the deal Iran had been offering since late April.
- Another Vietnam: kinetic escalation to ground operations with a low best case and a worst case of the second Iraq war.
The analytical key was Iran's 28 April proposal: reopen the strait, lift the blockade, end the war, defer the nuclear programme. Iran was willing to give up everything except the latent nuclear capability. Trump could not accept that offer because the war was justified on nuclear grounds. Iran could not offer more because more would mean conceding the only thing the regime believed guaranteed its survival. Both positions were internally rational. The intersection was politically infinitesimal. The piece concluded that the Bath was the most likely resolution, with Trump's political clock the binding constraint, and that by Northern autumn the Long Grind would be politically brittle enough to force the choice. Underneath the standoff, the May piece flagged the structural damage to the transatlantic alliance and the broken coalition that Trump had not built in February and could not summon in March.
Those three paths still hold. The Bath remains the most likely. The Vietnam tail has thinned. The Grind is not endless. What follows is what the past month adds to each.
The False Siren Calls Continued
The May piece documented Trump's hit rate on public statements about the war and noted that the next Witkoff-Kushner trip to Pakistan would almost certainly produce another "on the verge of a deal" post regardless of what happened in Islamabad. That prediction has held with some force. The 5 May post pausing Project Freedom cited "Great Progress" toward a deal. The 23 May post announced the deal was "largely negotiated." Both posts were followed by extended periods in which no actual signature was produced and the underlying terms remained as Iran had set them.
What the past month has clarified is what these posts are doing analytically. The May piece read them as expectation management buying news cycles for the Grind. That reading was correct as far as it went. The deeper reading the data now supports is that the posts are doing internal work for Trump. Each "Great Progress" post is a small step in the cognitive process of accepting that the deal is going to happen and that he is going to sign it. The repetition is the bridge from the war he thought he was fighting to the quasi-surrender deal he is going to sign. He cannot look at the deal directly. He looks past it. He says he will "stand firm" on dismantling Iran's nuclear programme even as the negotiating text defers that question. He insists Iran will not control Hormuz even as Iranian state media publicly confirms Iran will.
This sharpens the May piece's reading of the Long Grind. The Grind is not Trump running a strategy of sustained pressure in hopes Iran will break. It is Trump in denial about taking the Bath. The May piece had to project forward to estimate whether Trump would choose between the three paths as a deliberate decision. A month of additional data suggests the choice is being made by passive deferral rather than by active selection. Trump is not weighing alternatives and choosing the Grind. He is deferring the moment of signature because the signature is the moment the denial breaks. The Grind is the denial in operation. The Bath, when it arrives, will be the denial breaking and immediately reconstituting itself as a celebration of Trump's achievements.
The piece could project the end of the Grind based on this reading. The denial cannot sustain indefinitely. Each "largely negotiated" post that does not produce a signature degrades the credibility of the next one. The political clock continues to run. The Grind will not be endless. The question is whether the denial breaks in weeks or months, and the answer depends on which clock binds first: the political clock that has an interest in the delay ending, or the denial that has an interest in the delay continuing.
Project Freedom and the Vietnam Tail
The May piece kept Another Vietnam alive as a tail risk on three mechanisms: the room around Trump, an incident that forces escalation, and Trump's own reading of the dice. The personnel signals were noted: Vance staying home from the Pakistan trip, Witkoff and Kushner travelling as the eager-buyers from the original decision. The piece read Vance's absence as the climbdown cover being held in reserve and concluded that the Vietnam option was being kept warm.
The past month has tested the Vietnam path and found it less available than the May piece estimated. On 4 May US time, Trump launched Project Freedom, a US Navy operation to escort merchant ships through the Strait of Hormuz. CENTCOM described the operation as "just begun." Within 24 hours Iran struck shipping off the UAE, including the HMM Namu. Within another few hours Trump paused the operation, citing "Great Progress" toward a deal. The pause came while Hegseth was still publicly briefing the operation.
This was the operational test of whether Trump can sustain kinetic escalation when his political environment is squeezed. The answer was no. A president who launches a major naval operation and pauses it within 24 hours is not a president who is going to commit to a ground operation in Iran. The 25 May "self-defence" strikes and the IRGC's announcement that it had downed an MQ-9 drone and fired on an F-35 and an RQ-4 fit the same pattern: brief kinetic action, immediate diplomatic post, ceasefire formally holds. The pattern is stable enough across multiple instances to constitute the operating mode of the war's closing phase rather than a temporary expedient.
Vance has also reappeared in the negotiating circle. The May piece read his absence as the personnel signal for the Vietnam option being kept warm. His return alongside Witkoff and Kushner is the same signal in reverse: the room around Trump is being staffed for the Bath, not for further escalation. The personnel signal is now the signature signal.
The Vietnam tail has not closed entirely. It has thinned. The probability mass now sits mostly on incident risk, in which an Iranian or American miscalculation produces a casualty Trump cannot ignore. The kinetic theatre involves live ammunition, real Iranian air defence, and continuous threshold-probing by both sides. The probability that one of these probes produces a forced escalation is not zero in any given week and accumulates over the months until signature. But the deliberate-Trump-choice version of Vietnam, which the May piece flagged as the room-driven escalation scenario, has lost ground. The room is not driving that direction now. Project Freedom was the last test of whether it might. It failed.
The Standoff Was Less Stable for Trump Than We Knew
The May piece argued that the standoff was not stable for Trump because his political clock ran faster than Iran's. The data of the past month confirms this and clarifies the mechanism more sharply than the May piece could.
Iran has been running a continuous calculation. The Iranian objective is to do the minimum required to avoid restarting major kinetic operations, while continuing to extract concessions through delay. The decision variable is the level of accommodation Iran offers. The constraint is that the level must stay just above whatever Trump would need to see in order to resume bombing. Everything else is slack to exploit. When the kinetic threshold is approached, Iran does not retaliate against US forces directly. Iran retaliates regionally. The HMM Namu strike was the operational expression of this. Iran proved the strait was still Iranian to disrupt without giving Trump a target he could effectively hit back at. The Gulf states absorbed the cost. Trump paused within a day. Iran resumed the negotiation with the floor of acceptable terms slightly lower than it had been the day before.
This is what the May piece's "every week, the price of any deal Trump could plausibly accept goes up" meant, in operational practice. Every week Trump's discomfort is publicly visible, Iran's position improves and Trump's deteriorates. Every "deal is imminent" post Trump writes is a unilateral concession delivered in public. Iran does not need to demand the concessions. Trump volunteers them. The Iranian negotiating team prices each post into the next round and refuses to sign until further posts have moved the floor further. The not-closing is the value-extraction phase.
The political indicators the May piece flagged have continued to deteriorate. April US CPI came in at 3.8% year-on-year, the highest since May 2023. Gasoline is 28.4% above year-ago levels. The AAA national average sat at \$4.55 over the US Memorial Day weekend. Trump's approval is at 37% in the New York Times poll, 36% in Reuters/Ipsos. The CBS poll has 66% disapproving of his handling of Iran. The Republican disapproval on inflation has risen from 28% to 40% in two months. The threshold the May piece projected for the Northern autumn brittleness of the Grind is closer than the calendar suggests. Trump's subjective pain is approaching the subjective threshold that will force him into the room with the pen. It is not yet binding. It may be within the next few weeks. Almost certainly within a few months.
The Shift That Confirms the Call
The single most informative development of the month is one the May piece did not anticipate but that confirms its central reading. Through March and April Iran consistently denied that substantive negotiations were happening. Qalibaf publicly contradicted Trump's claims about talks on multiple occasions. Iranian state media described US proposals as "ridiculous spectacles." The denials were strategic. They forced Trump to negotiate against an absent counterparty in public, which let the floor of acceptable terms continue to move while Iran remained uncommitted to any position.
That pattern broke in May. Baghaei publicly described the emerging text as a "framework agreement." Qalibaf travelled to Qatar to participate. Fars News confirmed implementation details. Pezeshkian was photographed with Pakistan's army chief Field Marshal Asim Munir in Tehran. The Iranian foreign ministry is now discussing the deal's contents in substantive terms rather than denying its existence.
Iran does not publicly acknowledge a negotiation it has not won. The move from denial to acknowledgment is the market signal about who is winning. Iran is now publicly invested in the deal happening because the deal as currently shaped is acceptable to Iran. The May piece called the Bath as the most likely outcome on the structural logic. The Iranian acknowledgment is the participants' confirmation of the call.
The negotiating coherence on the Iranian side is itself a substantive finding. The regime the 11 February Situation Room briefing described as on the verge of collapse is, three months later, running a more disciplined negotiation than the United States is. Pezeshkian's executive wing, the IRGC, the foreign ministry, and Fars News are now speaking with effectively one voice externally. Whatever internal alignment problems Iran had earlier in the war have been quietly resolved while the kinetic theatre absorbed most of the attention. The May piece's reading of Iran's structural position has held up. The May piece's projection that Iran would not move on its red line has held up. What the May piece could not see was how disciplined the Iranian execution of that strategy would turn out to be. Tehran has played a stronger hand than the analytical model assumed it could.
What the Market Is Saying
The May piece tracked Brent and gasoline as the visible economic indicators of the war. The current data extends those readings in a direction the May piece flagged but could not fully develop.
The crude market has moved meaningfully in the direction of pricing the Bath. Brent crude was at \$114 on 30 April and touched \$126 overnight in the days before. Today Brent is around \$94 and WTI around \$91. That is roughly \$20 off the wartime highs. The decline is not pricing the war out completely. The level remains 55 to 60% above pre-war. The central case in the market has rebalanced toward resolution rather than continued conflict. The May piece was forecasting against the price action; this piece is forecasting with it. The market has come to where the analysis was.
The refined product market is telling a more interesting story. Singapore gasoil (diesel) traded at \$80 per barrel before the war, at a \$17 premium to Brent. It peaked above \$230 in early April at a \$120 premium to Brent. It now sits at \$156, with a crack of about \$61. That is a partial recovery from the peak crisis, but the crack is still nearly four times its pre-war level. The forward curve declines from \$156 today to \$96 by April 2029 but does not return to pre-war norms at any point. The market is pricing the war ending soon while also pricing a permanent shift up in the distillate baseline.
Petrol tells a different story. Singapore Mogas 92 is at \$75, only modestly above pre-war, and its crack against Brent is currently negative. The market is pricing petrol returning roughly to normal as the deal closes. The diesel and jet fuel cracks will narrow more slowly and incompletely.
This asymmetry matters. American voters see petrol prices. American truckers, farmers, and airlines see diesel and jet fuel prices. The visible inflation in petrol terms is the smaller part of the actual economic damage. The bigger damage is happening in freight, agriculture, food, and aviation, where the diesel pass-through works through with a lag of weeks to months. The deal solves the petrol pump in time for the Northern autumn campaign season. It does not solve the broader cost pass-through that will keep voters frustrated through the autumn regardless. Mike Johnson telling reporters on the weekend that a deal would bring gas prices down and be "a big thing" for the midterms is empirically defensible from the petrol crack. It is much less defensible as a claim about diesel-driven inflation in food and freight, which the deal does not solve on the same timeline. Trump tracks the petrol pump with high salience and tracks the diesel crack not at all. The deal will be timed to relieve his visible problem. The invisible problem will be left to unwind on its own schedule.
The Coalition That Wasn't, Now Paying
The May piece flagged that Trump had not built a coalition in February and could not summon one in March. The Gulf states had quietly worked against him. The UAE was leaving OPEC. China and Russia were providing Iran with the financial and political space to wait Trump out. None of these states would pay any cost Trump could credibly impose on them. The piece argued that this absence shaped the resolution path in real time and would shape the post-war reckoning for years.
The latest data sharpens the financial component of this picture. The Gulf states are split by the war. Saudi Arabia and Oman are the relative winners. Saudi Arabia in particular is making windfall money on the war at the macro level, with Brent well above its fiscal breakeven and the East-West Petroline allowing 5 million barrels per day to bypass Hormuz to Red Sea terminals. The Q1 fiscal deficit headline of \$33.5 billion is real but reflects defence spending and delayed revenue recognition. Goldman estimates only a marginal 1 percentage point worsening of Saudi's fiscal position. Saudi pressure on Trump to settle is strategic rather than financial.
The losers are bleeding. Qatar's economy is forecast by the IMF to contract by nearly 9% in 2026. The Strait of Hormuz is its only route for LNG exports. Ras Laffan production has been halted since early March. Qatar has issued \$3 billion in private placement bonds since the conflict began. The UAE has lost most of its annualised fiscal surplus and Goldman estimates a 6 percentage point swing in its fiscal position. Kuwait is in the worst position structurally, with an oil-export-dependent budget that was already running a \$32 billion deficit before the war began against a fiscal breakeven of \$90.50. Kuwait has borrowed roughly \$13 billion in new debt since the war started.
None of these states is in immediate financial crisis. All have substantial sovereign wealth. All can run elevated deficits for many months by liquidating positions. None has unlimited tolerance for doing so. The mediator pressure on Trump and Iran is being driven partly by this financial bleed, in a way the May piece flagged in general terms and the latest data lets me locate in specific fiscal positions visible in the IMF analysis and the public borrowing records. Qatar wants the deal closed. Kuwait wants the deal closed. Iran is writing the deal to deliver enough to each of them that they sign on as guarantors of the regional architecture emerging from the war. The Gulf mediators are paying for the privilege of helping to write that order.
The broader coalition damage the May piece flagged continues to accumulate. The transatlantic relationship has not been repaired and probably will not be. Trump has continued to threaten reduced US troop levels in Europe, citing disagreements with allies over the Iran war. The lesson Europe is taking from this war, that the United States cannot be relied on to consult before acting and cannot be relied on to forgive non-participation after the fact, outlasts any ceasefire. The May piece's reading on this has held up; the latest data simply extends the timeline.
Where the Three Paths Lead
The Bath remains the most likely resolution. The May piece called it on structural logic. The latest data has confirmed the call from multiple directions: Iranian acknowledgment, market pricing, personnel signals around Trump, the operational test of Project Freedom, the continued degradation of Trump's political indicators. Probability mass roughly 75 to 85% in the next four to sixteen weeks. The variable is timing, not substance. The substantive terms have been roughly stable since late April. They are Iran's terms with marginal American adjustments at the package level.
The Long Grind remains a path but a shrinking one. It is the late tail of the Bath distribution, not a separate outcome. Trump's denial defers the signature; the signature arrives when the denial breaks. The Grind is not endless. Each week of delay imposes additional accumulated costs that the May piece identified and that the latest data confirms in detail: continued diesel-driven inflation working through the system, continued Iranian floor-lowering through public concessions, continued political bleeding for Trump and the Republican Party, continued financial pressure on the Gulf losers. The Grind will probably end within weeks or two to four months. Whether it ends in weeks or in a couple of months depends on which clock binds first inside Trump's head.
Another Vietnam remains possible but has thinned. Probability roughly 7 to 12%, lower than my May estimate. Project Freedom demonstrated that the deliberate-choice version of Vietnam is not available to a Trump under sustained political pressure. The Vietnam tail now consists mostly of incident risk, in which a miscalculation produces an American casualty that forces escalation Trump did not choose. The probability is real and accumulates with each additional week of kinetic theatre, but it is the smallest of the three paths and has fallen rather than risen over the past month.
The signature could be quick if Trump simply signs the framework as currently shaped. It probably will not be. Iran has reasons to haggle the package now that the framework is agreed. The structural terms are settled. What Iran is extracting now is the detail: sanctions relief depth and pace, frozen asset release size and conditions, banking access, reconstruction funding, implicit acceptance of Iran's regional position post-war. Each of these is negotiable. The detail is where Iranian leverage continues to translate into concrete gains. The Pakistani Prime Minister's comment about hosting "the next round of talks very soon" is the mediator confirming the haggle is in progress. A signature within two to three weeks is possible but improbable. A signature in July or August is more likely. September remains inside the probability mass.
Where This Points
The May piece closed by noting that the cost of the decision made in the Situation Room on 11 February was still being totted up, and that most of it had not yet been paid. A month later this remains true and the latest data lets me say more about which costs are reversible and which are not.
The visible economic costs of the war (US petrol at \$4.55/gallon, headline US CPI at 3.8%, equity volatility) will mostly reverse over the next six to eighteen months. The forward curves price this. The market expects 75 to 80% of the wartime price spike to unwind as the Bath plays out. Trump will get to claim this reversal as evidence the deal worked, and on the petrol pump indicator he will be empirically correct. The invisible economic costs will not reverse. The forward curves price a permanent shift up in the distillate baseline of perhaps 20 to 25%. This reflects some combination of structural damage to Gulf refining capacity, a permanent Hormuz risk premium on tanker transit, and the demonstrated weakening of the regional energy architecture. The Hormuz risk premium and transit tolls will live in every refined product price for years.
The structural damage to the transatlantic alliance the May piece flagged remains the war's largest political legacy outside the Middle East. The latest data adds a second permanent legacy in the refined product market itself and a third in what the May piece called the demonstration effect. Russia and China have watched a smaller, sanctioned state successfully outwait the United States by combining patient floor-lowering with regional response capability that the United States could not effectively neutralise. That demonstration is now banked, available for use, and will shape Russian planning on Ukraine and Chinese planning on Taiwan in ways that are not yet visible but are not reversible either.
Iran's red line on enrichment held. The war was justified on nuclear grounds. The deal defers the nuclear question to a 30-60 day window that almost certainly expires without resolution. The 440 kilograms of 60-percent enriched uranium at Isfahan remain intact, entombed but accessible in IAEA Director-General Grossi's phrase. The reconstitution risk the war was meant to eliminate has been guaranteed rather than solved. Some future administration will inherit it.
Iran did not break. Iran read Trump correctly, calibrated its kinetic responses to keep the United States from restarting major operations while extracting concessions through delay, converted regional response capability into asymmetric leverage, and used the time bought to assemble a deal designed to be acceptable to a Trump in denial about taking it. Tehran did not need to win. Tehran needed only to make winning unaffordable for Washington while keeping the deal acceptable in four Gulf capitals and to its own hardliners. It has done all of these things. The signature is the moment Iran gets paid.
This is not the war Trump thought he was starting. The decision room of 11 February 2026 produced a war it expected to wrap in four to six weeks. The war is now in its thirteenth week. The four-to-six-week framing has been quietly retired. Vance was the lone dissenter in February. Vance is now in the negotiating circle for the deal. The room around Trump is staffed for a Bath, not a war. The personnel signal is the signature signal.
The three paths from May still hold. The Bath is the most likely. The Vietnam tail has thinned. The Grind is not endless. The standoff is resolving on its own timetable. Trump is taking longer to step into the bath than the political logic alone would predict, because he cannot quite acknowledge what he is about to sign. He will sign anyway. The cost of the delay accumulates in diesel cracks and approval polling and Republican House seats and Kuwaiti debt issuance and a refined product baseline that will sit permanently above pre-war norms. The signature ends the war. The damage the signature does not end is what the war was actually for.
This is the fourth piece in a series on the 2026 Iran War. Previous pieces:
- The Most Dangerous Scenario (17 March),
- Groupthink with a Group of One (30 March), and
- Update on the Iran War: the Mexican Standoff (1 May).
No comments:
Post a Comment