Posted on: 20 April 2026
Fifty billion dollars of oil not produced in fifty days. Five hundred million barrels gone from the global market. Reuters counts precisely what can be seen: Gulf jet fuel exports collapsed from 19.6 million barrels in February to 4.1 million for March and April combined, Arab production down by eight million barrels a day, Ras Laffan offline for years rather than months. Striking numbers, front page everywhere, and they are the wrong part of the story. The part that matters is who was already sitting at the table with the right cards when the game began on 28 February.
On 31 March the House Select Committee on Strategic Competition with the Chinese Communist Party published a report nobody read, and it says something simple: Beijing has accumulated roughly 1.2 billion barrels of strategic reserves by early 2026. One hundred and nine days of seaborne import cover, bought largely from Iran, Russia and Venezuela through a shadow fleet of uninsured tankers often running with tracking switched off, assembled piece by piece over the past decade while the West was looking elsewhere. The report's own phrasing deserves framing: China built this reserve "at well below market cost from the very barrels Western sanctions were designed to strand".
Read it twice, because what it tells us is that sanctions meant to choke Tehran and Moscow ended up subsidising Chinese strategic preparation for today's crisis.
I have seen this pattern before, in the music industry of the Nineties. The majors were convinced that Napster was a legal problem to be solved in court, while someone smaller was quietly building the infrastructure of what came next. That was not foresight, in my view, but careful reading of signals that everyone had in plain sight and few wanted to see.
America's Strategic Petroleum Reserve is now convalescing after the 2022 drawdown. Capacity sits at 714 million barrels, actual stocks remain below pre-Ukraine levels, and there is no real margin for another comparable operation. Europe never had an SPR worthy of the name, only IEA obligations of ninety days of cover that member states meet through creative accounting. Germany is discovering now that its Gulf hydrocarbon exposure is significantly higher than the post-Nord Stream debate had suggested, while the United Kingdom is finding that "diversification" after the North Sea decline was a reassuring adjective for a reorganised dependency rather than an eliminated one. Whitehall's energy security strategies since 2022 all use the right vocabulary, but the physical reserves in depots tell a different story.
This is not the moral pattern of a villainous China evading sanctions, which is leader-page analysis. The pattern is strategic, and it points to two entirely different approaches to planning under uncertainty.
The first approach thinks in terms of markets, efficiency, just-in-time, and assumes that the global system will remain integrated, that supply chains will self-correct, that crises will be short and resolvable through financial channels. When the shock arrives, it counts the damage in real time and hopes it ends soon. Trump says the deal will come "soon", traders bet on it and markets stay calm because they are pricing imminent resolution. A perfectly rational position within its own frame, except that the frame assumes a world that no longer exists.
The second approach thinks in terms of scenarios, redundancy, optionality, and assumes that crises are endemic to the system rather than exceptions to be managed, that strategic chokepoints are permanent vulnerabilities, that whoever reaches the next crisis with real reserves will simply still be standing when everyone else has run out of air. This approach has built Beijing's architecture over the past twenty years, slowly and quietly, while Western analysis busied itself with quarterly pricing models and DCF valuations.
There is a recent precedent that ought to be studied in business schools rather than think tanks. In 2022, when Russia invaded Ukraine, Europe discovered a catastrophic exposure to Russian gas and then took two years to patch it up incompletely. Some people learned in those years, the Chinese for instance, who accelerated purchases of discounted Russian crude, reinforced the independent teapot refineries capable of processing politically risky barrels, and kept building storage capacity.
In the first two months of 2026, while Western analysts debated whether Iran would really close Hormuz, Chinese imports of Russian crude rose by 40.9 per cent. Kpler recorded a weekly high of Iranian loadings to China in the week of 16 February, more than double the previous weekly average. Two weeks before the war. The signals were all there, readable by anyone who cared to read them.
The West could have read them the same way, but it did not. European strategic documents after 2022 contain the right vocabulary: diversification, resilience, strategic autonomy. Between the vocabulary and physical barrels in a depot lies a distance measured in billions of pounds of infrastructure investment nobody was willing to authorise, because the marginal cost looked too high against a probabilistic benefit. The benefit is no longer probabilistic, in my view, but realised, and the marginal cost of unpreparedness is materialising at a pace that makes the early savings look comically shortsighted.
Here comes the awkward part. Strategic preparation has a perverse quality: it always looks oversized until the exact moment it is needed, and at that moment it looks ridiculously undersized. During long periods of normality, those who accumulate reserves look paranoid, inefficient, poorly optimised, while optimising managers get promoted for reducing working capital and the prudent ones get criticised for tied-up capital costs. Then a crisis arrives and we find out which of the two worlds was actually looking at reality.
The mechanism is structural rather than moral, and this is where things get interesting for anyone designing systems. Quarterly incentive structures reward managers who minimise visible costs and punish those who prepare for risks not yet manifest. Shareholders applaud buybacks and disapprove of defensive capex. Analysts build models on historical volatility that by definition contain no events not yet occurred. A CFO who had proposed two billion pounds of strategic hydrocarbon reserves in 2018, against the possibility of a Middle Eastern war, would have been out within six months, while the same CFO today would be out for not having done it. There is no way to look wise in real time, because wisdom is retrospective, and those who accepted the reputational cost of being called paranoid during the good years are the only ones with room to manoeuvre now.
Beijing has enjoyed two advantages that make its model not easily replicable: state control that permits long-horizon investment without quarterly pressure, and an opaque decision-making structure that can absorb internal reputational costs without having to justify them to the market. Neither condition is native to how Western democracies operate, which creates a competitive asymmetry that cannot be resolved through direct imitation. We cannot become China, but we can stop believing that our current model, as it stands, is adequate to a world in which chokepoints matter again.
The April Kpler figures make no allowance for comfort: global onshore stocks down 45 million barrels this month alone, production outages running at 12 million barrels a day. Heavy crude fields in Kuwait and Iraq could take four to five months to return to normal levels even in the most optimistic scenario of a full Hormuz reopening, and some infrastructure, Ras Laffan first, could take years. Meanwhile China continues to import from Iran through bilateral arrangements the Strait permits even during the blockade, Russia to Beijing is up almost 41 per cent at the start of the year, and Brent swings between 93 and 125 dollars depending on what Trump says on Truth each morning.
The Reuters count of fifty billion dollars lost is, to my mind, accurate and useless, because it is a photograph taken from the wrong angle. The photograph that matters is the one taken from the position of those who decided twenty years ago that this scenario, precisely this one, would happen sooner or later, and started building the system capable of walking through it without drowning. Nobody is taking that photograph, and by the time someone does, it will probably be too late to respond in time.
Anyone reading this from a position of corporate or institutional responsibility has one useful question to sit with over the coming months. Not whether the war will end soon, nor where Brent will settle after the ceasefire. The question is: "over the next decade, on which scenario are we allocating capital today, and how much of that allocation would survive the test of a similar closure in another chokepoint we are not currently worried about?"
If the answer is "we have not thought about it", we are exactly where Europe was in 2021.