Petrol Up Rs2.02, Diesel Down Rs3.59: What Geopolitics Actually Costs in Pakistan's Fuel Formula

September 26, 2026, early morning. Before the new rate board went up at a pump in Islamabad, three numbers had already landed in my notebook — 391.30, 408.53 and 2.02. The first is the new ex-depot price per litre of petrol, the second that of high-speed diesel (HSD), and the third is the size of the petrol increase. What never made the notebook was the headline itself: diesel fell by 3.59 rupees. Two products moved in opposite directions in the same notification, and that opposition is the real story.

A reader who saw only the headline two days ago thinks prices are coming down. A reader who buys five litres of petrol two days from now will feel the bill rise. In Pakistan's fuel pricing, the coexistence of both truths is the rule, because the market does not set the price — a formula does. And inside that formula sit three layers of arithmetic: the global benchmark, the dollar exchange rate, and taxes and levies.

The arithmetic you must understand first

Petrol and diesel prices in Pakistan are not a single company's decision. The Oil and Gas Regulatory Authority (OGRA) and the Petroleum Division jointly revise ex-depot prices on a fortnightly cycle. 'Ex-depot' means the price at the moment the product leaves the depot; dealer margins, the inland freight equalisation margin, and federal and provincial taxes are added on top. The retail pump price therefore never equals the ex-depot price.

The first layer of the formula is import parity. The calculation runs roughly like this: the international benchmark price, plus a premium, plus incidental costs, multiplied by the interbank dollar rate, plus local freight-equivalent costs. On top of that sit the petroleum development levy, sales tax and customs duty. A move in any of these layers changes the pump price — even when the price of the product itself does not move at all.

So the first question when reading this news should be: what changed in the international market, and what changed in the domestic formula? The two must be separated. In this review window, the international benchmark was mixed. Brent crude stood at 105.26 dollars and WTI at 92.78 dollars — a gap of more than 12 dollars between the two benchmarks, which is unusual when markets are moving in the same direction. That gap is itself the clue inside the decision.

Why petrol rose and diesel fell — because they are not the same product

Motor spirit (MS) and high-speed diesel (HSD) are two separate markets in refined fuel. Both come from crude, but the crack spread — the gap between refined product and crude — does not widen in both at the same time. Diesel demand comes from industry, freight transport, agricultural pumping and power generation; petrol demand comes from private cars and two-wheelers. Diesel demand rises in winter, petrol in summer.

In September 2026 the domestic demand picture in Pakistan was mixed as well. Under high interest rates and a contractionary fiscal stance, diesel-heavy freight and construction slowed, while the number of passenger vehicles did not fall — only the reach of fuel subsidies did. When diesel demand weakens and petrol demand stays rigid, the international spread shifts in exactly the direction today's notification indicates.

One more calculation belongs in the notebook: after today, the gap between the two products stands at 408.53 against 391.30 — that is 17.23 rupees a litre. A week earlier the diesel-petrol gap was wider. A narrowing gap is relief for agriculture and freight, but not for the urban middle class that runs on petrol. The same government notification is therefore two different truths for two different classes.

The part of geopolitics that gets priced in

This is where it gets opaque. The background to the review carried two geopolitical signals — Houthi attacks on Saudi supply infrastructure, and talk of a possible US-Iran truce. The first pushes prices up, the second pulls them down. In formula-based pricing, the net effect of the two does not appear directly, because OGRA does not buy the product — it takes market settlements and calculates from them.

In other words, geopolitics enters fuel prices one step late — not through rumour, but through the premium. Truce talk creates a 'risk discount' in the market; Houthi attacks create a 'risk premium'. The price that emerges from that tug-of-war is what shows up in the international benchmark, and that benchmark reaches a Pakistani pump two weeks later.

In an import-parity formula, that lag is not only of time but of direction. If international prices fall today, the effect may reach a Pakistani pump two or three review cycles later — unless the government spreads the adjustment across phases. And when international prices rise, the pass-through comes faster, because there is often a backlog of deferred adjustment. In administered pricing, this asymmetry is close to a rule.

One thing needs stating plainly: the link between the international benchmark and your local pump price is not linear. The dollar rate, the premium, incidentals and taxes — four layers can amplify or dampen a benchmark move before it reaches the consumer. In nine years of keeping market notebooks, this is the least discussed truth I have recorded: global oil prices and your pump bill are two separate lines, and tax policy is the bridge between them.

The part being sold as 'relief'

Now the other side. Calling the 3.59-rupee diesel cut 'relief' is easy, but when the number itself speaks, a different picture appears. First, the validity window is only three days — September 26 to 28, 2026. Notifications on such a short cycle are usually not a final policy decision; they are an interim adjustment, with the next step arriving in the following review. Today's 'cut' can therefore become part of tomorrow's 'correction'.

Second, the petrol increase is 2.02 rupees — a number that looks small only because the previous ex-depot price moved from 389.28 to 391.30. In percentage terms that is roughly 0.52 percent, seemingly marginal. But month on month, for a user consuming five litres a day, that marginality compounds into a distinctly larger burden — and it lands precisely on the class that has no alternative transport.

Third, and this is the biggest point of contention: the entire 'geopolitical price' shown in today's news is expectation-based. A US-Iran truce is still an estimate, not a decision. And Houthi attacks on Saudi supply are still a weekly event, not a spike. If the risk premium now sitting in the market for these two narratives does not materialise, the next one or two cycles will build downward adjustment pressure. If the opposite happens — if supply risk grows — then today's cut will prove temporary, and this notification's three-day window will remain a small event in the past.

Fourth, the notification says nothing about the budget impact. Pakistan habitually uses adjustments to the petroleum development levy to keep fuel prices stable. Without that mention, we cannot know whether this price came out of the consumer's tunnel or out of the government's revenue adjustment. These are policy-wise two completely different events, but in a headline they look identical.

What the notebook says overall

First, today's figures are internally consistent if we assume the petrol spread widened and the diesel spread narrowed during the review window. The size of the diesel cut is nearly one and a half times the petrol increase — an asymmetry that signals the correction pressure is not equal across the two products.

Second, the mixed benchmarks (Brent at 105.26, WTI at 92.78) and the roughly 12-dollar gap tell us the global market currently lacks direction. In a directionless market, administered prices always lag, and closing that distance requires yet another adjustment next cycle — which may be upward or downward.

Third, what we read as 'fuel news' is really three separate stories published together: the commodity market story, the administrative decision story out of the Prime Minister's Office, and the consumer's daily cost story. One set of numbers cannot prove the other's truth — yet in a headline, it very easily does.

What to watch now

My notebook's next entry will track two things. First, whether the petrol-diesel directions stay the same in the notification after September 29, 2026 — if diesel rises again, today's cut was a market adjustment, not a policy one. Second, whether the petroleum development levy figure moves in the next notification; if it does, today's numbers should be read through an entirely different lens.

For the reader who thinks the bill went down, I leave one question: when two fuels move in opposite directions on the same day, is that a market signal or an accountant's reconciliation? The answer is written in the next two review notifications — you just have to know how to read the numbers.