Oil & The China Syndrome — Checkmate in Chinese Revisited PauloMacro — Jun 30, 2026 — Paid
When it comes to oil, I admit that I have more questions than answers today. Chalk it up to my feeble attempts to keep an open mind while I dream up the most absurd potential outcomes as the most likely ones. We must always respect Shrub's Razor: Shrub's Razor is a philosophical trading principle whereby the funniest, most absurd outcome is also the most likely one.
As some readers may recall in my comments on the chat, what started as a general uneasy feeling of a 'lack of control' back in April when equity risk was starting to rip and oil failed to maintain momentum was, with the benefit of hindsight, a sense of seeing a bullish consensus among experienced commodity players in a firm consensus that the market had stopped confirming. Something else was going on, and that was a classic "Kovner signal" to get out of the way. "What I am really looking for is a consensus the market is not confirming. I like to know that there are a lot of people who are going to be wrong." — Bruce Kovner
Of course, this sense is how many drawdowns begin, and in not listening to it, I took a nice dose. At least now I think I understand the two key visible factors I got wrong in being bullish oil into the recent collapse.
First, beyond the visible SPR and commercial inventory draws in crude and products which we have discussed for months, China stepped out of the global market beginning in April to the tune of >4mmbpd which significantly loosened the physical market. I will come back on China after addressing the second factor, as China is at the heart of my questions.
Side note for younger readers: the title of this note harkens back another crazy time in the oil market (1970s) and refers to a theoretical nuclear meltdown scenario where a reactor's core becomes so hot it melts through its containment structures and tunnels straight through the Earth, conceptually "all the way to China." The China Syndrome was immortalized by the thus-named 1979 disaster thriller featuring a television news reporter (played by Jane Fonda) and her cameraman (Michael Douglas) who accidentally film an emergency shutdown at a nuclear power plant. They soon uncover systemic corporate cover-ups and corner-cutting, while a dedicated plant supervisor (Jack Lemmon) desperately tries to expose the dangers of a potential meltdown.
The second factor was that speculators took that physical cue and swung from one of their largest net long positions to one of their lowest exposures in the span of a few weeks. The current Brent + WTI (Nymex + ICE) combined managed money net long position in notional US$ has rarely been this washed out. In fact, the current level of positioning marked significant near-term lows over the past two decades, exceeded on the downside by the 2015 shale implosion, the sudden 2018 surplus, April 2020 Covid collapse, mid-2024, and 4Q25 (when the consensus was for a 3mmbpd Superglut in 2026). If you told me a few months ago that speculators' bullishness and positioning would be swinging extremes like this in the teeth of ~5-7mmbpd global inventory draws, I would have said you probably need to talk to someone. As it turns out, I'm the one who is crazy and has been wrong. Welcome to the Upside Down.
For more context on just how wild positioning is, John Kemp at Reuters noted this weekend that WTI and Brent net positions are down to the 11th percentile going back to 2011… while the long/short ratio for Brent is now down to the 2nd percentile. In WTI crude, the non-commercial (speculative) net long position in futures & options as a percent of open interest is actually at levels below the shale blowups of late 2015, and even lower than the troughs seen in 2023-24. Only 2025 saw lower levels than this, back to 2012.
I discussed positioning in more detail two weeks ago when Brent was just below $90, so I won't rehash the full case further. As a corollary to sentiment though, around the same time I started a thread of oil bears dunking on experienced oil observers and barrel counters. The thread got some attention after my pal Kevin Muir cited it in his excellent bullish energy note two weeks ago which I definitely encourage you to read (and subscribe to him as he is excellent). To be clear, I began this thread because of a phenomenon I call The Pile On which is why, in breaking some trading rules (that clearly exist for good reason), I have actually upped my exposure again over the last two weeks of decline. The issue more fundamentally is that this bearish jubilation reflects a Soros Misconception. Good process → Good outcome = Deserved success. Good process → Bad outcome = Tough beat. Bad process → Good outcome = Lucky break. Bad process → Bad outcome = Just desserts.
Don't get me wrong: I will gladly take a lucky break over a tough beat. Yet the dunking really doesn't bother me, because I know its foundation is in Bad Process → Good Outcome. Saying things like "there's plenty of oil around" while suspending laws of physics governing volume and time is not rational. Saying that oil is pricing the present inventory when it's at $100, but then "let's price the future superglut at $70" is inherently inconsistent, and a narrative solving for price.
My priors notwithstanding, I believe it is possible that a bear trap is being set with current positioning, but for the trap to spring, the physical market needs to tighten. If a loosening near-term physical market swung positioning down, it stands to reason that speculators will have a very hard time shorting or remaining underexposed a physical market that tightens again out of nowhere. For that, I come back to factor #1: we need to see what China will do. Keep in mind, even despite China's absence, the world is still drawing >35mm barrels per week since this crisis started, with Kpler citing a -49mmbbls drawn last week (this is crazy). Still, if China were to reemerge as a buyer, the physical market would significantly tighten almost immediately.
To review where things stand with China, my friends at HFI Research were kind enough to update crude balances for June MTD using Kpler data — you can see China is now cycling over -5mmbpd of crude imports from pre-war levels. At this point China is beginning to draw visible onshore inventories. Meanwhile global onshore inventories have never been this seasonally low, with the bulk of the draws outside China thus far.
Some have speculated that China released crude from underground SPR caverns that analysts can't see via satellite. I have not bought into this view so far, in part because conversations with traders and others close to Chinese SOEs (the only players allowed to take SPR barrels) have suggested that China has not released barrels, although some reports in recent days by other teams like Energy Aspects citing similar conversations suggested they may have been drawn from underground. Can we know for sure? No, China is a black box. The balances generally squared out between significant run cuts by Chinese refiners (especially SOEs), a modest amount of demand destruction (Chinese data has deteriorated as the PBOC drained liquidity in March-May), and product export restrictions enacted in March.
Chinese Refining — Quick Primer. China's total refinery capacity stood at approximately 19 million b/d in 2024, with the government having announced a cap of 20 million b/d by 2025. In terms of actual throughput, China's refineries processed a new annual record of ~14.8mmbpd in 2025, up ~600kbd YoY, boosted by the ramp-up of new independent refining capacity (specifically the private mega Yulong refinery in Shandong, a large-scale independent that doesn't fit the old "small teapot" mold), as well as additions at SOEs. The gap between nameplate capacity (~19–20mmbpd) and actual runs (~14.8mmbpd) reflects the chronic overcapacity that has long plagued the sector. State-Owned Enterprises (SOEs): The SOE segment is dominated by three giants: China Petroleum & Chemical Corporation (Sinopec), China National Petroleum Corporation (CNPC/PetroChina), and CNOOC — which together have over 800,000 employees and a bit under US$1 trillion in annual revenue. Sinopec is the country's largest refiner; both Sinopec and PetroChina are fully integrated companies. Together the three NOCs account for roughly ~75% of China's refining capacity (source: EIA). Teapot / Independent Refiners: Teapots have been growing in size and sophistication and now account for about one-quarter of total Chinese refining capacity. That implies roughly 4.5–5mmbpd of nameplate capacity. These teapot refineries are generally privately owned and concentrated in Shandong province, and have historically suffered from extremely low utilization rates — often as low as 35–40%. They play a strategic role because they handle discounted and politically risky crude (Iran, Russia, Venezuela, etc), while major SOEs remain more insulated from sanctions risk.
Note that % utilization rates for SOEs fell from low 80s to upper 60s since the war. If SOEs constitute 75% of the system (i.e. ~15mmbpd), then a -15% cut in utilization would suggest over 2mmbpd of lost product output. Product data suggests ~500kbpd was lost from exports. If we also assume a modest ~500kbpd in 'demand destruction', China has still drawn at least ~1mmbpd in domestic product stocks. Over the past three months, assuming just ~1mmbpd in Chinese product draws implies over 100mmbbls gone. Since most products degrade over time (particularly jet fuel), it would be unusual for an economy of China's size to be carrying something greater than 4-5 months of inventory. As swing refining capacity, teapots and excess refining capacity act as a form of SPR for products. In a world where crude can be stored indefinitely but products have a shelf life, a large crude SPR and excess refining capacity ensure insulation from embargos, Middle East wars, and other harmful geopolitical developments.
In this context, China today is the only refining center globally with excess capacity, and its exports are desperately needed with 4-5mmbpd of Middle East product exports offline. The only other significant refining center that has excess capacity relative to domestic consumption is the US, but we are already running flat out and timespreads are beginning to reflect a level of tightness that is forcing domestic gasoline prices higher in order to price out US exports. I expect this trend to continue as we are only just entering the peak of driving season which runs through Labor Day. Gasoline and diesel July-Aug timespreads suggest crude oil would normally be somewhere in the $100-120 range given this tightness. The 3-2-1 crack one month out vs WTI is a $56 crack — product prices are implying over $125/barrel. The conclusion is that the world is in a deficit of both crude and products, but products are running out faster because of Middle East and Chinese refinery export losses and lower stock shelf life.
Interestingly, in China local refinery margins are now flipping firmly positive also, despite local prices being controlled. The world desperately needs China to ramp its refinery runs and export products at these cracks; the US simply does not have the capacity to do more. Refineries are running so hard without proper downtime that they are beginning to literally catch fire. China's excess refining capacity is not only a product SPR for China, but for the rest of the world as long as Middle East refining capacity is offline. Recent speculation that Chinese refineries may increase runs in July make sense, but oddly, the chatter seems to suggest they may slow-walk a loosening of the export ban. Significantly raising product export quotas would allow private refiners to increase runs while selling price-controlled fuel domestically. Reuters reported last week that this is looking increasingly likely, but the speculation reads like a creep rather than a real lift.
So the window is open for China to ramp crude purchases, and my understanding is that China is currently lifting remaining floating storage around Asia aggressively while also doing the majority of the buying in the current recent wave of Persian Gulf egress which has cut the oil stranded behind the Strait of Hormuz in half. However there appear to be games afoot. While the Chinese are lifting barrels, they don't seem to be ramping purchases from Saudi, Brazil, or West Africa (WAF). On the contrary, they continue to flip WAF cargos at significant -$4-8/bbl discounts to Dated Brent as recently as last week. It's the most bizarre behavior — does China want barrels, or not want barrels? Every signal shows a green light for Chinese refiners to buy crude, but they seem to be gaming the physical market and slow-walking purchases. Why?
Some analysts have suggested that China's electrification is shifting its energy mix away from oil, but transport fuels for autos only make up ~20% of China's oil consumption. While EVs as a % of new car sales has risen from 1% a decade ago to over 50% today, only ~12% of the entire Chinese auto fleet is currently EV, and China's oil demand has continued to grow by ~500kbpd each year.
So here is the big question: what is China waiting for? I can only speculate. China's strategy of not buying crude is draining domestic and global product stocks and keeping global product prices elevated. If their refiners are the world's SPR, this would be the time to use it, unless you want to throw the rest of the world into recession. Does China owe Trump a favor (some have speculated a behind-the-curtain handshake over Taiwan in the April Beijing meeting)? They only stepped out of the market around the time the US began its blockade of Iranian vessels in April. Then I find myself thinking that this whole exercise was a trial run by China to test how they could navigate an embargo along the lines of Japan in WW2. The trial seems to have succeeded: they have sent a message to the rest of the world that they are just as powerful on the demand side as Opec is on the supply side. But they still need oil, and they can't draw forever. Remember China does not have allies — they have suppliers, and they have customers.
And in all this, Iran's North Star still holds. The regime survived, and to ensure its future, they need to make this hurt. Trump getting killed in the midterms due to higher gasoline prices sends a message. Midterms are in 126 days and if Trump squeaks through with a Republican Senate, he's not a full lame duck, and he's already said he would come back for more.
Sudden calls for a 3mmbpd+ Superglut in 2027 by nearly all the sellside notwithstanding, what does make sense is the math. Crude vs inventories has never been this far away from regressed price. If Brent sticks around here for another month, we are so far "out of the lane" that it's a Covid-era discount to "inventory fair value" but we are drawing amidst an absence of buying, rather than building. We are now at the highest 3-2-1 crack spread relative to Brent going back to Bloomberg's series inception in 1988. At this rate, maybe refinery cracks will soon be $80 and crude will be $70, and the crack itself will be bigger than the oil price! Gasoline vs WTI is at the widest divergence in the history of the energy market. Timespreads in contango vs draws of onshore inventories. Cracks near all-time wides while the crude market collapses into contango.
A parting thought on China. I have started following Chinese PBOC liquidity injections more closely because I find the annihilation of an already-cheap HK equity market stunning. Other friends have been getting frustrated with gold longs, and since China has emerged as the world's biggest buyer of gold, the price of gold has become inextricably linked to Chinese liquidity as documented by Michael Howell. I asked my pal Louis Vincent Gave not long ago why China slammed the brakes on liquidity back in March. He reminded me that the answer is obvious: the PBOC is the 21st century's Bundesbank. To the PBOC, rising energy is inflationary — period. The moment the war broke out, they slammed the brakes. But here's the thing: after a significant tightening in March-May which has shown up in poor economic data… the PBOC is injecting again since around May 21st, and is showing signs of being a bit more consistent with its injections (a 'positive hum' rather than sporadic pumps). And if the US and Japan are running hot while Europe is bouncing… and China is now stimulating — why on earth would China elect to draw crude and especially product stocks rather than raise refinery runs?
China's strategy has resulted in a significant depletion of products in storage, both domestically and globally. These products have fallen to levels that if China were to reenter the crude market and raise runs, any loosening in crack spreads would result in rising crude and overall higher product prices. But China can insulate itself from this more successfully than the rest of the world. Among major oil consumers, they have tested and proven that they are least vulnerable. I just don't know what China is waiting for — but I know they can't do it forever. More questions than answers.
Wishing you all a pleasant and relaxing Independence Day holiday. As always, kindly yours, Paulo aka Cloudbear