The Barrel that Crosses Twice

MARIA: Here's a puzzle that's been running through the crude market all summer. A cargo loads at Fujairah — on the open side of the Strait of Hormuz. On paper, that barrel never went through the strait. A lot of them did. They crossed on a shuttle tanker first, went into storage, and loaded again for the long voyage. One barrel. Two ship movements. And if you count both as exports, you've just invented crude that doesn't exist. That's the trap in this market — and underneath it is a real reorganisation of how crude moves. Who crosses the Gulf. Where Atlantic cargoes get split. And who's ordering ships for a trade that may not look like this in three years. This is Current Intelligence.

MARIA: Welcome to Current Intelligence, where we take the commodity story and follow it down to the ships. Three episodes on the dry side — bauxite out of Guinea, agriculture and the Gulf, then Chinese coal. Today we cross to the wet side. Crude tankers. I'm Maria Bertzeletou, Senior Market Analyst at Signal Research, and with me is George Sakellariou, freight analyst at Signal Maritime. George, welcome.

GEORGE: Thanks, Maria. Good to be here.

MARIA: Let's start where I started. When you see a cargo loading at Fujairah — what do you actually count?

GEORGE: Once. You count the delivered barrel once, then you measure the vessel-days separately on each leg. The shuttle movement inside the Gulf is real employment — it uses a ship and it uses time. But it is not a second export.

MARIA: And the terminal label hides the upstream journey.

GEORGE: Completely. A loading port tells you where the ship was. It doesn't tell you where the crude came from, or how many vessels it took to get there.

MARIA: So let's put the arrangement on the table. At the end of July ADNOC had chartered around twenty-five crude tankers from Sinokor. Roughly fifteen were moving crude from inside Hormuz out to storage in Fujairah and Oman. The rest supplied customers directly. And I want to flag that carefully — those counts came from a source. Sinokor hasn't confirmed them.

GEORGE: Which matters, because the ship count is the part everyone quotes and the part we're least sure of. The shape of the arrangement is more reliable than the number.

MARIA: So what does the shape tell you?

GEORGE: That somebody built a dedicated shuttle system. That isn't a spot decision — it means vessel deployment, storage access at the far end, delivery arrangements with customers. You don't assemble that for a week.

MARIA: Which brings me to the question everyone is asking. When the security situation eases, does this just unwind?

GEORGE: Not automatically — and that's the assumption I'd push back on hardest. A ceasefire announcement tells you nothing about the duration of a charter. What I'd want to see is repeat cargo programmes, how long those charters actually run, terminal throughput, and the real test — a sustained return to direct Gulf liftings.

MARIA: There's infrastructure in the background too. ADNOC said in May the new West–East pipeline is fifty percent complete, targeted for 2027.

GEORGE: Fifty percent complete is a construction milestone. It is not capacity you can use this quarter. When that line runs, a connected barrel reaches the export terminal without any tanker crossing at all — but that's a 2027 conversation at the earliest.

MARIA: In August, ADNOC L&S announced an acquisition including six VLCCs. Does that replace the chartered ships?

GEORGE: We don't know, and I'd resist the assumption. Owning six VLCCs and running fifteen shuttle tankers are different jobs. The announcement doesn't establish that one substitutes for the other.

MARIA: One more number, with a health warning attached. The IEA's September assessment put Saudi crude loadings in August at 3.5 million barrels a day — down 1.1 million from July, including Red Sea and untracked movements. That's an export-loading estimate. Not production, and not Fujairah throughput.

GEORGE: And the space between loadings and production is exactly where the vessel-days hide.

MARIA: Now the demand side, because China produced a number in July that stopped me. Total crude imports, June to July: 7.12 million barrels a day up to 8.41. Plus eighteen percent. Refinery throughput over the same two months: 12.47 to 12.51. Plus nought point three. Buying up eighteen. Processing flat. George — what's happening?

GEORGE: Procurement and processing run on different clocks. A refiner changes what it buys far faster than it changes what it runs. So a number like that is telling you about purchasing and replenishment timing — not about consumption.

MARIA: So where did the barrels actually go?

GEORGE: That's what the data doesn't tell us, and I'd be honest about it. It doesn't allocate between storage, feedstock or changes in domestic supply. It establishes the difference in pace — not the destination.

MARIA: There's a second series too, and the two don't mix. Kpler estimates put seaborne arrivals at 6.93 million barrels a day in July and 7.14 in August — up three percent. That one excludes pipeline.

GEORGE: And you can't take total imports for one month and seaborne arrivals for the next and call the difference a trend. You'd be manufacturing a change the data doesn't show.

MARIA: What matters for us is where the replacement barrels came from. Reuters reported this month that independent refiners bought over twenty million barrels from West Africa, Canada and South America, replacing reduced Iranian and Russian supply — though those are reported purchases, not confirmed arrivals.

GEORGE: And origin is the entire shipping story. A Canadian Atlantic cargo and a Canadian Pacific cargo are not the same voyage. West African buying may employ completely different vessel classes depending on whether the stems stay intact or get split. You cannot read tonne-miles off a barrel count.

MARIA: Which takes us straight to West Africa. George, September sources identified sustained splitting of VLCC stems as the main recent driver of West African Suezmax demand. Explain the mechanism.

GEORGE: A VLCC cargo is around two million barrels. If a buyer can't or won't take that in one lift, you split it into two Suezmax stems of roughly a million each. Same crude, same origin, same distance — but now you've employed two ships instead of one.

MARIA: So the barrels don't grow. The fixtures do.

GEORGE: Exactly. Which is why I'd separate two claims that get merged constantly — "West African exports are up" and "West African Suezmax demand is up." Only the second is well supported.

MARIA: And to be clear, these are dated broker observations — not a measured quarter-on-quarter increase in fixtures.

GEORGE: Right. They tell you the mechanism is operating. They don't size it.

MARIA: Brazil is pulling on the same ships, though.

GEORGE: The same regional list, yes. Brazilian and West African cargoes compete for Atlantic Suezmaxes — so when Brazilian activity picks up, the workable list for a West African stem tightens even if nothing changed in Nigeria or Angola.

MARIA: Brazil's destinations moved sharply in the second quarter. On Petrobras' own export shares: China down from sixty-two percent to thirty-five. India up to twenty-two. Europe up to eighteen. Is more Atlantic crude automatically more long-haul tonne-miles?

GEORGE: No — and that's the trap. Those are one company's shares, not all Brazilian exports. And a barrel going to Rotterdam instead of Ningbo is a far shorter voyage. More Atlantic supply doesn't mechanically mean more vessel-days. It depends entirely on who buys it.

MARIA: Then there's Dangote. Reuters reported at least sixteen million barrels of Nigerian crude bought for October arrival — about 520,000 barrels a day — plus two WTI Midland cargoes. Commitments, not completed deliveries.

GEORGE: And that one cuts both ways. Domestic refining takes exportable cargoes off the international market — but it keeps coastal movements, and it creates import voyages for US crude.

MARIA: So if you're fixing Suezmaxes, the question was never how much Nigeria produces. It's how many stems stay split — and whether they stay split when VLCC availability improves.

MARIA: Before the fleet, one more piece — because the Atlantic story has a next chapter, and it's all about timing. Argentina first. YPF's second-quarter shale oil averaged 213,000 barrels a day, inside total crude output of 266,000 — so roughly eighty percent shale. In August the company put a year-end shale target at 250,000. George — does that become tanker demand?

GEORGE: Only if it converts into repeat export stems of usable size and quality — that's Argentina's wildcard. And a year-end target is an exit rate, not a quarterly average and not an export commitment.

MARIA: And you can't treat the grade as interchangeable just because the country label changed.

GEORGE: No. A comparison with WTI Midland only means something for a specific blend, a specific receiving refinery, a specific loading arrangement. Brazil and West Africa are collections of grades and terminals too — country labels tell you very little.

MARIA: Venezuela's had the headlines, but on a much longer clock.

GEORGE: Much longer. The announcement at the end of August set out rights — an equity stake, an entitlement to buy twenty percent of production at cost, first refusal on the rest, concessions over seventeen fields. Those are stated terms. They don't verify an export schedule.

MARIA: And Chevron's numbers?

GEORGE: Over seven billion dollars and more than 600,000 barrels a day — but that's a 2031 ambition, separate from the agreement and from near-term liftings.

MARIA: So rights and output sit on completely different timelines.

GEORGE: And for the next cargo programme, a small connected development can matter more than a large undeveloped resource. TotalEnergies flagged Angola's Acacia-5 at around 6,000 barrels a day, within three months of discovery. Tiny — but it has a near-term timetable.

MARIA: Which is the distinction I'd hold onto. Which barrels have a confirmed loading window — and which are still production targets.

MARIA: Last piece — the fleet. Because owners are ordering at a pace that doesn't obviously match a tight market. 261 crude-tanker orders in the first half of this year. 204 of them VLCCs — seventy-eight percent. Chinese yards took 216, nearly eighty-three percent. And ninety-eight percent specified conventional propulsion. George, that last number surprises people. What does it tell you?

GEORGE: That this ordering wave is not a fuel-transition story. If it were, you'd see dual-fuel specifications. Ninety-eight percent conventional says owners are buying future transport capacity and replacing ageing tonnage.

MARIA: And the age profile?

GEORGE: This is where it gets interesting. On the August numbers, the VLCC orderbook is 304 ships against a fleet of 907 — about a third. Ships over twenty years old: 166, roughly eighteen percent. So in both VLCC and Suezmax, the orderbook is materially larger than the over-twenty fleet.

MARIA: Which breaks the one-for-one replacement story.

GEORGE: It complicates it. Other ships age before those deliveries arrive — lead times are around three and a half years, so an order placed today is a 2030 ship. And retirement isn't automatic: it depends on employment, condition and whether charterers accept the vessel.

MARIA: One caution I'd add — the age and sanctions categories overlap. You can't add 166 over-twenties to 167 sanctioned VLCCs and call it 333 removals.

GEORGE: No. And that mistake gets made constantly.

MARIA: So how does heavy ordering coexist with a tight market?

GEORGE: Because they address different windows. A five-year-old VLCC is trading at roughly a sixteen percent premium to newbuilding parity — that's the market paying for availability now rather than in 2030. Ordering is a 2030 decision. Tightness is a this-quarter problem.

MARIA: And there's a sale-and-purchase story underneath. Xclusiv data put 365 tanker transactions in the first seven months — Greek interests sold 70, bought 47, about nineteen percent of recorded sales. All tanker types, and only July from the third quarter.

GEORGE: And the key point is that a sale is not a removal. The ship changes owner and keeps trading. Look at the twenty-year VLCC assessment against recycling — roughly 3.4 times, and on Suezmax 4.4. While an older ship is employable, selling it beats scrapping it by a wide margin.

MARIA: Which is why I'd be careful with any "fleet renewal" headline. Frontline sold two Suezmaxes in Q2, took two VLCC newbuildings in June and July straight into one-year charters, and put two existing VLCCs onto two- and three-year cover in August.

GEORGE: Four different effects on supply. A sale changes ownership. Recycling removes a hull. A delivery adds one. A time charter changes commercial availability. Only one of those actually takes a ship out of the market.

MARIA: So what would tell us this is changing? George — what are you watching?

GEORGE: Three things. Whether direct Gulf liftings resume in a sustained way, not just for one programme. Whether West African stems stay split once VLCC availability improves. And charter redeliveries — because existing ships coming off multi-year cover can add capacity far faster than any newbuilding.

MARIA: I'd add a fourth, and I think it's the one that gets underweighted — transport efficiency. Fewer transfer legs, shorter waiting time, a wider acceptable fleet. Any of those releases capacity without a single ship being delivered.

GEORGE: That's the real forward risk. Efficiency returning at the same moment the newbuildings arrive.

MARIA: And one long-range flag before we close. A Sinopec researcher estimated EVs would displace 56 million tonnes of oil this year — about 1.2 million barrels a day. That's a counterfactual estimate, not an observed fall in Chinese crude imports.

GEORGE: Which is the right way to hold it. A demand-growth question for later in the decade — not something you'd trade off this quarter.

MARIA: So here's the takeaway. The tonnage story in crude right now isn't mainly about how many barrels move. It's about how many times each barrel gets handled, and how far it travels once it is. A shuttle leg through Hormuz. A VLCC stem split into two Suezmaxes. A Brazilian cargo going to Rotterdam instead of Ningbo. None of those change the barrel count. All of them change the ship count. Bauxite, grain, coal — and now crude. Four cargoes, one thread: when the energy map gets redrawn, the ships move first and the statistics catch up later. George, thank you.

GEORGE: Thank you, Maria.

MARIA: And thank you for watching Current Intelligence. Follow the signal — we'll see you in the next episode.