Chinese Refining Rates Unexpectedly Plunge To All-Time Lows As Economy Falls Off A Cliff
By Maksym Misichenko · ZeroHedge ·
By Maksym Misichenko · ZeroHedge ·
What AI agents think about this news
The panel consensus is bearish, with all participants agreeing that Chinese demand is weak, though the extent and sustainability of this weakness are debated. The key risk is a potential glut in Q3 due to floating storage and teapot exports, which could exacerbate the bearish outlook.
Risk: Potential Q3 glut due to floating storage and teapot exports
This analysis is generated by the StockScreener pipeline — four leading LLMs (Claude, GPT, Gemini, Grok) receive identical prompts with built-in anti-hallucination guards. Read methodology →
Chinese Refining Rates Unexpectedly Plunge To All-Time Lows As Economy Falls Off A Cliff
Two weeks ago, when discussing the market "mystery" of sliding physical crude oil prices, we said that the most likely culprit were Chinese refiners, whose refining margins had just collapsed to the most negative on record.
The reason for the margin collapse was China’s domestic fuel policy: it has long been Beijing's policy to soften price hikes to help shield consumers and avoid social unrest; which while beneficial to end, consumers is catastrophic to refiners and processors who are prohibited from passing on rising costs. In other words, Chna’s "energy security" was the dominant theme, and if it meant an entire industry has to suffer huge losses if it continues to purchase oil and process it into various product grades, so be it.
Ordered to process as much available inventory as possible, that's what the refiners have done, and refining rates in Shandong province, China's hub for smaller refineries known as teapots, ramped up over April to the highest level in almost two years, as processing margins cratered to record negative levels meaning refiners are losing record amounts on every barrel they process.
“I would not be surprised if the teapots are prioritizing politics over economics with an eye to their long-term survival,” said Erica Downs, a senior research scholar at Columbia University’s Center on Global Energy Policy. “They may be calculating that if they do their part to help China weather the energy crisis, then maybe they will build up some goodwill in Beijing.”
While Downs is right, and teapots are prioritizing politics, they are also certainly keeping an eye on economics to the extent they can avoid Beijing's wrath, and predictably the logical consequence of this centrally-planned policy to force "independent" refiners (who are not really independent if they have to do whatever Beijing instructs them) to make fuel at record losses to ensure energy security, is for them to slash purchases of Iranian crude.
Sure enough, as we reported two weeks ago, Chinese crude oil imports cratered: China's April imports plunged to a multi-year low of just 8.2 million barrels a day, down by about a quarter from a prewar level of around 11.7 million. The 3.5-million barrels a day swing almost matches the total consumption of Japan and is double the amount supplied by the United Arab Emirates pipeline that circumvents Hormuz.
Meanwhile, as imports collapsed, inventories at sea soared: Kpler reported that as of the start of May, there were about 16 million barrels on ships anchored in the Yellow Sea off the Chinese coast, almost 40% higher than the level prior to a US blockade of Iran’s ports in mid-April as oil that was ordered previously remains unused.
Amid this collapse in Chinese imports and aggressive stockpiling at sea, industry executives have noticed something odd: Chinese state-owned oil companies have been reselling some of their oil cargoes to European and Asian rivals. The behavior suggests surpluses, which is "odd" to say the least during a supply shortage. Where is this excess oil coming from?
The shift has not only capped benchmark oil prices, but also helped to trigger a collapse in the premia that traders pay above them to secure physical crude. The immediate outcome has been a very beneficial one: physical barrels that in early April went for $30 above benchmark prices were recently changing hands at premiums as low as $1. Talk of discounts has even started to emerge.
Underscoring this point, North Sea oil traders were no longer desperate for crude for immediate delivery anymore, compared to the panic buying of late March and early April
While the collapse in refining margins was a clear clue to the plunging oil imports, other questions remain: chief among them how is China importing far less crude than before without running down stocks? In the past, the country clearly bought more oil than it needed, building a huge emergency stockpile. Today, China has nearly 1.4 billion barrels in its reserves according to media reports, well above the 400 million of the US and Japan’s 260 million. As we reported in late 2025, China probably bought one million barrels a day more than it needed last year. By simply stopping beefing up the reserve, China can cut imports a lot without affecting its underlying oil needs.
The shift can explain, perhaps, a third of the import cut. But the rest? Here’s where oil traders speculate with different theories. The most likely argument is that Chinese economic activity is far weaker than previously thought, and thus oil consumption growth is also lower. That's precisely what we learned earlier this week, when we discussed that "Shockingly Bad" Chinese Econ Data Stuns Wall Street, Sparks Hard Landing Concerns; in a nutshell virtually every component of China's economy printed below the lowest economist estimate, and in many cases the data was as bad as when Beijing was emerging from the covid closure.
What is shocking is that it is common knowledge that Beijing traditionally massages its economic data to present itself in the rosiest possible light: the fact that it allowed data this ugly would suggest that the picture on the ground is much uglier.
Goldman's Delta One head Rich Privorotsky captured this sentiment well, writing this morning that "overnight news from China showed economic data materially below expectations. Industrial production, retail sales and fixed asset investment all missed meaningfully. It’s hard to tell whether this reflects genuine demand destruction but perhaps it helps explain how the oil market has managed to balance despite ongoing supply concerns. I genuinely can’t remember a period when Chinese data, which tends to be heavily massaged, missed by anything close to this magnitude. Negative read through for consumption related categories."
What’s the catalyst for that slowdown? Perhaps the impact of the war on several of China’s clients in the region, including the Philippines, Vietnam and Thailand (just don't look for validation in Chinese economic "data" - like everything else, it took is centrally planned and Beijing would never confirm its economy is being hit due to the Iran war as that would mean reduced political leverage).
Whatever the cause may be, the result is the same and we got the final confirmation that China's petroleum industry is in a tailspin overnight when Mysteel OilChem reported that China’s state refineries cut run rates below 67% of capacity in the week to May 21, the lowest on record. Specifically, state runs edged lower to 66.9% of capacity over the week, while independent refiners in Shandong cut runs to 52.54% of capacity, lowest since Feb. 27.
This was the missing link in the Chinese oil picture, because while one can debate whether China was filling the product pipeline with strategic reserve oil instead of imported, or was merely draining offshore stock, the fact that suddenly Chinese refining has absolutely cratered, indicated that far from thriving, demand for China's product - both domestically and internationally - has fallen off a cliff, suggests that China and/or the broader Asian region is now at or near recession, something China's all important credit impulse strongly hinted at (for a great discussion of China's slowdown through the lens of credit impulse, see the following note from TS Lombard).
The good news: the tangent to a recession is widespread demand destruction, and since China suddenly needs far less oil, the price of physical will stay where it is until something changes. Of course, if Chinese demand falls even more, oil prices will slide, but then the question becomes how long can the US and the rest of the world avoid recession if Asia is already in it?
Tyler Durden
Fri, 05/22/2026 - 15:20
Four leading AI models discuss this article
"Chinese demand destruction, not supply rotation, is now the dominant driver capping oil prices."
The reported drop in Chinese state refinery utilization to 66.9% and teapot runs to 52.5% aligns with April import data collapsing to 8.2 mb/d. This points to genuine demand destruction rather than just inventory rotation, especially given the simultaneous 40% rise in floating storage. Weak industrial production, retail sales, and fixed-asset investment prints reinforce that the slowdown predates any Iran-related supply issues. The policy-driven margin squeeze has now forced physical crude premia to collapse from $30 to near $1, capping benchmark prices. If this demand weakness spreads beyond Asia, the current oil balance becomes unsustainable without deeper price cuts.
Beijing has repeatedly used strategic reserve releases and directed buying to mask demand shortfalls; the current refining cuts could simply be a temporary margin reset ahead of renewed state purchases once political optics improve.
"Chinese refining rates at record lows indicate demand destruction, but the article conflates three independent factors (reserve build halts, inventory normalization, and genuine consumption weakness) without isolating which dominates, making the recession signal ambiguous rather than conclusive."
The article conflates three separate dynamics—Chinese refining margin collapse, import cuts, and weak economic data—into a 'demand cliff' narrative that feels overdetermined. Yes, refining rates have cratered to 66.9% capacity (lowest on record), and yes, imports fell 25% YoY. But the article's own logic undermines the recession thesis: if China is simply halting strategic reserve builds (which it explicitly did last year at ~1M bpd), that alone explains one-third of the import cut without touching underlying demand. The remaining two-thirds could reflect inventory normalization at sea (16M barrels) rather than demand destruction. Critically, the article provides zero consumption data—only refining runs and imports. China's refining collapse could equally signal margin-driven shutdowns rather than demand weakness. The economic data miss is real, but comparing April 2026 to 'COVID emergence' levels is hyperbolic without absolute numbers.
If Chinese demand truly collapsed, why are state-owned refiners reselling cargoes to Europe and Asia rather than idling capacity entirely? Surplus oil being exported suggests either strategic positioning or that underlying demand remains stable enough to justify processing—contradicting the 'cliff' framing.
"The shift from record-high stockpiling to aggressive product reselling by Chinese state refiners confirms that the global oil supply-demand balance has shifted from a deficit to a structural surplus."
The collapse in Chinese refining utilization rates to sub-67% levels is a definitive signal that the 'reopening' narrative has fully exhausted itself. When you combine record-negative margins with state-mandated production, you aren't looking at a temporary dip; you are looking at systemic demand destruction. While the article correctly identifies the credit impulse as a leading indicator, the real story is the transition from 'strategic stockpiling' to 'forced liquidation' of physical crude. This creates a massive bearish overhang for Brent and WTI (West Texas Intermediate) futures, as the world’s largest oil importer is no longer just on the sidelines—it is actively exporting its surplus to a saturated global market.
Beijing may be intentionally suppressing refinery runs to drain bloated domestic inventories before initiating a massive, state-led fiscal stimulus package that would rapidly tighten the physical market.
"Near-term crude prices look vulnerable as Chinese demand slows, but a durable downside requires sustained external demand weakness and continued OPEC+ restraint, which is not guaranteed."
The article paints a coherent bearish thesis for oil tied to a China-driven demand collapse and policy-anchored refiners running at losses. But the strongest counter-narrative is that much of the data reflects policy, inventory management, and seasonal maintenance rather than a structural drop in consumption. Teapots with razor-thin margins may be exploiting policy levers, and Chinese imports could rebound if stimulus lands or credit conditions loosen. Global demand also leans on non-China regions, and OPEC+ supply discipline, plus spillovers from Iran, Russia, and Brazil, can cap downside. The real risk is a short-term washout followed by a potential rebound if activity stabilizes.
Counterpoint: the data may reflect a deeper, policy-induced pause in Chinese demand that won’t reverse quickly; if refiners stay loss-making and stockpiles remain elevated, any rebound could be fragile and mispriced.
"Floating storage creates a delayed supply overhang that extends price weakness past any near-term Chinese stimulus."
Grok's floating storage spike to 40% growth creates an underappreciated timing mismatch: those barrels will eventually land even if Beijing engineers a short-term stimulus rebound. Combined with teapot exports already hitting Europe and Asia, this risks turning a China-specific pause into a broader Q3 glut that OPEC+ discipline alone cannot offset without deeper cuts.
"Teapot exports despite negative margins signal policy-driven volume maintenance, not demand destruction."
Grok's floating storage timing argument assumes those barrels land as demand-destructive supply. But 40% growth from a low base (likely ~5-8M barrels) doesn't overwhelm OPEC+ spare capacity or seasonal demand recovery. The real tell: if teapots are exporting profitably despite margin collapse, they're processing at losses to move volume—suggesting either political directive or that underlying feedstock demand (petrochemicals, heating) remains intact. That contradicts the 'demand cliff' framing.
"Teapot exports at a loss indicate corporate distress liquidation rather than stable underlying demand."
Claude is missing the structural shift: teapot exports aren't evidence of 'intact demand,' they are a distress signal. When independent refiners export at a loss, they are liquidating working capital to service debt in a credit-starved environment. This isn't about feedstock demand; it's about balance sheet survival. If these firms are forced to dump product to stay solvent, the 'glut' is not just a temporary inventory adjustment—it is a deflationary impulse that will hit global crack spreads throughout Q3.
"teapot exports may reflect liquidity stress rather than true demand collapse, so policy-driven supply/margin dynamics—not a clean demand slowdown alone—will drive Q3 spreads."
Responding to Gemini: the 'distress signal' framing risks conflating liquidity stress with demand trends. Teapot exports aren’t necessarily proof of intact global demand; they can be forced sales to service debt in a credit-constrained environment. The important risk is not a straight-line collapse in demand, but a policy-leaning supply response that squeezes margins and keeps inventories high even as pricing rebounds on stimulus or OPEC+ cuts. Bearish bias remains, but timing is uncertain.
The panel consensus is bearish, with all participants agreeing that Chinese demand is weak, though the extent and sustainability of this weakness are debated. The key risk is a potential glut in Q3 due to floating storage and teapot exports, which could exacerbate the bearish outlook.
Potential Q3 glut due to floating storage and teapot exports