Q4 Bunker: VLSFO Averages $876.50/mt at 20 Ports, Singapore $904, MGO $1,500

The Analysis

Ship & Bunker published its 6 October snapshot: a 20-port average of $876.50/mt for VLSFO, $751.50 for HSFO and $1,500 for MGO. One day later, ENGINE priced Singapore VLSFO at $904/mt. Rotterdam sits at $682.50, Fujairah at $979.50, Los Angeles-Long Beach at $1,055. July averaged $744/mt and September $865/mt, and the Q4 view has been lifted by more than $100/mt across two monthly revisions. That is what the bulletin says. Now we take the numbers apart, because three of them disagree with each other and one of them has not happened yet.

The bulletin puts three prices on the table and they are not interchangeable. VLSFO (very low sulphur fuel oil, the standard marine fuel since 2020) averages $876.50/mt across 20 ports. HSFO (high sulphur fuel oil, legal to burn only behind an exhaust scrubber) averages $751.50. MGO (marine gasoil, the distillate every ship without a scrubber must burn inside an emission control area) is $1,500. The gap between the first two is $125/mt. The gap between the first and the third is $623.50/mt. Every bunker clause in your file names one of those three as its reference grade, and the cost of naming the wrong one is bigger than the move the headline is describing.

Start with the spread, because the bulletin's own arithmetic is worth more than its headline. Rotterdam at $682.50 and Fujairah at $979.50 differ by exactly $297.00/mt. That is the near-$297 spread everyone repeats, and it is not Singapore against Los Angeles. Singapore at $890 against Rotterdam at $682.50 is $207.50. Singapore at $890 against Los Angeles-Long Beach at $1,055 is $165. Three pairs, three different numbers, and the widest one is the one that made the headline. The 20-port average of $876.50 sits $194.00 above Rotterdam and $178.50 below Los Angeles, which means the average describes a port your ship probably does not bunker in. Averages are fine for a forecast.

In a contract they are expensive. Work out which ports your own rotation actually calls at, because that list, not the 20-port basket, is what your fuel bill is built on. A service that stems in Rotterdam and Singapore is priced off $682.50 and $890, and the weighted cost of those two stems is nothing like $876.50. A service that stems in Fujairah and Los Angeles is priced off $979.50 and $1,055, and every tonne of it sits above the average. Ask your carrier for the stem list on your lane. It is one email, and it is the cheapest piece of information in this whole story.

Now the trend, and this is where the number gets shaky. July averaged $744/mt. September averaged $865/mt. The 6 October print is $876.50/mt. July to October is plus $132.50/mt, or 17.8%, which is real money. September to October is plus $11.50/mt, or 1.3%, which is noise. And the Q4 view has been lifted by more than $100/mt across two monthly revisions. Put those three moves side by side: $132.50 over three months, $11.50 over one month, and a forecast pushed up by more than $100 in two months. Most of what you are being asked to price for Q4 has not happened yet. It is a pencil line in a forecast table.

We calculate how much. Take the Q4 forecast as the September average of $865 plus the stated uplift of more than $100, so somewhere north of $965/mt. October spot is printing $876.50. If October ends up averaging $876.50 and the quarter is meant to average $965, then November and December together have to average $1,009.25/mt. That is the hidden number: the forecast is not saying fuel is expensive today, it is saying fuel has to gain another $132.75/mt before Christmas. At 0.9 mt per 40ft, which I will justify in a moment, that implied climb is about $119 a box that nobody has quoted you yet.

Singapore gets its own paragraph because most bunker clauses in this market reference it. Ship & Bunker had Singapore VLSFO at $890/mt on 6 October. ENGINE had it at $904/mt on 7 October. That is $14.00/mt, or 1.6%, in one day, from two desks pricing the same grade in the same port. Part of it is a genuine day-on-day move. Part of it is methodology: different contributors, different cut-offs, ex-barge versus ex-wharf, different credit terms. A clause that says Singapore VLSFO without naming the source and the window hands the counterparty a $14/mt option he can exercise every time he reads the tape. Name the source. Name the window. Name whether it is a spot print, a weekly average or a monthly average.

Right, we calculate properly, with the assumptions written down, because a number without its assumptions is a rumour. Assume a 14,000 teu ship, main engine burning 180 mt/day of VLSFO, 85% slot utilisation, so about 11,900 earning slots. Assume a 30-day Shanghai-Rotterdam passage. That is 5,400 mt across 11,900 slots, or 0.454 mt per teu, so a 40ft box counted as two teu carries about 0.91 mt. Call it 0.9 mt and keep it in your file. At 0.9 mt per 40ft: July's $744 against October's $876.50 is plus $132.50/mt, so about $119 a box. September's $865 against $876.50 is plus $11.50/mt, so about $10 a box.

The forecast premium of roughly $88.50/mt is about $80 a box, prepaid. The 20-port average against an actual Rotterdam stem is $194/mt, so about $175 a box billed on fuel nobody burned. Los Angeles at $1,055 against the $876.50 average is $178.50/mt, so about $161 a box the carrier absorbs today and recovers at the next GRI. The scrubber spread of $125/mt is about $113 a box, which is why a scrubber-fitted service can underbid a non-scrubber service by three figures and still make money.

MGO at $1,500/mt is the line people skip, and it is the most expensive tonne in the bulletin. On a Transpacific northbound sailing the ship spends roughly the last day inside the North American emission control area, 200 nautical miles off the coast. Assume 150 mt/day on that leg across 11,900 slots and a 40ft box carries about 0.025 mt of MGO. At $1,500 that is $37.50 a box. The same volume at the VLSFO average of $876.50 is $21.90. The ECA premium is about $15.60 a box. Small, and it lands on your invoice only if your clause lets the carrier switch grade for the ECA leg and bill at distillate. Most clauses do. Find out whether yours does before you sign the tender.

Next question, the one I ask every time: split the freight into base rate and surcharge and work out who is doing the passing through. Ocean freight is a base plus a bunker adjustment factor, the BAF, and the two move independently. When VLSFO fell back towards $600 a few years ago carriers cut BAF and rebuilt base rates quietly. Now that fuel is climbing, base rates are being defended and the adjustment does the work again. You cannot argue about the total. You can argue about the split, because that is where the margin sits.

Pull the last three BAF notices from each carrier and look for four fields: the grade, the source, the window, and the base price below which no BAF applies. That fourth field decides whether you pay anything at all. A BAF built on a $600/mt base charges you on $276.50/mt of today's spot. One built on an $800/mt base charges you on $76.50/mt. Same fuel, same ship, same day, a $200/mt difference, which at 0.9 mt per 40ft is $180 a box. No headline will ever tell you which base your carrier used. The notice will.

Timing, because a number without a date is also a rumour. A spot booking made today carries today's surcharge from the booking date. A contract BAF usually resets on a lag: the October-December figure is often computed from a trailing window that closed in September, and then sits unchanged for the quarter. So the $876.50 print of 6 October is very unlikely to be the number on your December invoice. The December number was fixed before 6 October happened.

That lag cuts both ways. If the forecast is right, you are billed below the market until the next reset and then you take the whole step at once. If the forecast is wrong, you prepay a step that never arrives and nobody refunds it. Establish which lag your own contracts use before you argue about the level. Write it down as three dates: the date the window opens, the date it closes, and the date the resulting figure goes live on your invoice.

If those three dates are not in the clause, they are whatever the carrier says they are, and a carrier who chooses his own averaging window is choosing his own price. Then count how many bookings fall inside each reset. A shipper booking weekly across a quarterly reset is riding one price for thirteen weeks of cargo, which is fine while fuel is falling and expensive while it is climbing.

Who eats what, in the bulletin's numbers rather than adjectives. A shipper on a yearly contract with a formula BAF inherits whatever the formula says and has almost no room to argue inside the quarter, so his job this month is the clause, not the rate. A shipper buying spot is re-priced on every booking, pays the full $119-a-box move against July immediately, and also gets the relief immediately if fuel retreats. A reefer or out-of-gauge shipper carries more fuel per box than our 0.9 mt, because reefer plugs draw power and heavy lifts slow the ship, so multiply the per-box figures by 1.5 to 2 and stop arguing about the last $10. A shipper on scrubber-fitted tonnage is implicitly short the $125/mt spread and should be asking his carrier to share it.

A shipper into Los Angeles-Long Beach is exposed to the most expensive bunker in the bulletin at $1,055 plus the MGO leg, and should expect the loudest GRI language on that lane. Scale it up: at 200 forty-foot boxes a month on Asia-Europe, the July-to-October move is about $23,800 a month, the forecast premium is about $16,000 a month, and the 20-port-average wedge is about $35,000 a month. Over a quarter that is $71,400, $48,000 and $105,000. Those three numbers belong in front of your finance team before the tender closes, not after the first invoice.

Here is the part the headline does not carry, and it is the reason to read this far. The bulletin gives you one average. Two desks give you two different Singapore prices one day apart. Your ship lifts fuel in one specific port at one specific price. Reconcile those three and the biggest number in this story is not the $132.50 rise since July. It is the $194/mt wedge between the 20-port average of $876.50 and the actual Rotterdam stem of $682.50, about $175 a box on our assumption.

Every carrier that indexes a BAF to a broad average and bunkers in Rotterdam collects that wedge as margin, and every shipper who signs 20-port average without asking where the ship actually stems has agreed to pay it. Add the $88.50/mt forecast premium that has not happened and the $14/mt source spread, and the same physical fuel can be billed anywhere between $682.50 and $965/mt on nothing but drafting. That is a $282.50/mt range, about $254 on a 40ft box, manufactured entirely by paperwork.

Now take the other side, because I do not sign anything I have not tried to break. The Q4 view has been lifted by more than $100/mt in two monthly revisions while spot moved $11.50 in a month. If the forecast is running ahead of the market and crude softens into November, the forecast never lands, and a Q4 bunker fixed at $965/mt leaves you about $80 a box worse off than floating. If the revisions are early rather than wrong and the 20-port average clears $900 before the end of October, floating costs you roughly $21 a box for every further $23.50/mt of climb.

Both are live. The condition that flips the call is the print, not the forecast, so draw the line and let the tape cross it. If ENGINE's Singapore VLSFO holds above $904/mt for five consecutive business days, or the Ship & Bunker 20-port average clears $900/mt, fix the rest of Q4. If the 20-port average falls back through $820/mt, unlock and go back to floating, and ask for the base to be reset. Below $820 you would be paying for a forecast that has already missed twice.

In practice, for anyone tendering in October, begin by pulling the bunker clause out of every contract and filling the four fields: grade, source, window, base. A blank field is a negotiation, not a formality. Next, run the 0.9 mt arithmetic on your own volume and put the sheet in front of finance. Then ask each carrier for the actual stem port and grade on your lane and for the logic behind the surcharge. A carrier who will not name the port is telling you the average is worth money to him. On structure, a fixed BAF removes volatility and costs you roughly $80 a box if the forecast fails.

A floating BAF keeps you at the market and exposes you to the $119 a box that has already happened plus whatever comes. A collared BAF, fixed below $820 and above $950 with floating between, is what I would push for in a market this wide, and it is the structure most carriers resist, which tells you exactly why it is worth asking for. Slow steaming is the other lever and it is real: every knot off service speed trims daily consumption by roughly 8-10%, which is 14-18 mt/day on our 180 mt/day assumption, but the voyage lengthens by about a day and a half on a 30-day rotation, so you are buying fuel with transit time.

Decide which one your cargo can afford. And do the same exercise on the buying side: if you have 300 boxes a month riding one BAF formula, a $10/mt drafting error is $2,700 a month, which is $32,400 a year, which is roughly the cost of the person you have not hired to read these clauses. Ask for a quarterly reconciliation statement showing the index value used, the base applied and the resulting per-box figure. Carriers produce it when asked. Silence on that request tells you more than any forecast table will.

The traps are boring and they are expensive. A BAF quoted per teu and applied per 40ft doubles the bill. A surcharge with a carrier's discretion line is not a surcharge, it is a free option you wrote for somebody else. A quarterly reset averaging a window that ends on the 15th means your October booking is priced off September fuel. And the classic: a clause referencing VLSFO on a service your carrier runs with scrubber-fitted tonnage burning HSFO at $751.50/mt. You pay $876.50. He burns $751.50. That $125/mt is about $113 a box and it is entirely legal, because you signed it.

All told: fuel is up $132.50/mt since July, up $11.50/mt since September, and forecast up more than $100/mt across two revisions. Two of those three have happened. One of them is a pencil line, and the forecasting desk used two months and a pencil to move VLSFO by more than $100 while the market moved eleven dollars and fifty cents, which is either bold analysis or an expensive pencil. Price the two that happened, argue about the one that has not, and do not let anyone bill you for a number that lives in a forecast table. My line is ENGINE Singapore above $904 for five straight business days or the 20-port average through $900; above that, lock the rest of Q4 and stop reading forecasts. Below $820, unlock and refloat. And check the base before you check the rate, because a $200/mt difference in the base is $180 a box, which is more than every dollar of the move the bulletin is shouting about.

↑ Back to the key points

— By Vivian Zhao