On August 18, President Irfaan Ali said Guyana is now entitled to 39.8% of crude production from the Stabroek block, up from 12.5%. Reuters carried it, the Guyanese press ran it, and the framing everywhere was a threefold jump in a country's take from the largest oil development of the last decade.
It is a real event and the mechanism is unusual enough to be worth explaining. But the number that matters to anyone holding Exxon was published three weeks earlier, by Exxon, in the second-quarter materials of July 31: Guyana net entitlement volumes fall by approximately 100,000 barrels a day in the third quarter, as the production-sharing agreement moves from cost recovery to profit sharing.
Those are the same event. Nobody has put them next to each other, and when you do, two things fall out — the president's percentages and the company's barrels reconcile almost exactly, and the size of the whole thing against Exxon is 2.2% of production.
This is our first article on Exxon, which we have tracked for forty-four quarters without writing a word. Starting on a contractual event rather than a routine quarter is the better place to begin, and the baseline is at the end.
First, the multiple is wrong
The 12.5% figure everyone is dividing into 39.8% is the profit-oil share alone. Guyana also takes a 2% royalty on gross production, so its prior total entitlement was 14.5%, not 12.5%.
That makes the increase 2.7×, not the 3.2× in circulation. It is a smaller headline and it is the correct one.
The mechanics behind it, from the 2016 agreement: up to 75% of monthly production goes to the contractor group as cost oil, and what remains is profit oil, split 50–50. At the 75% ceiling, Guyana receives 2% royalty plus half of the remaining 25% — 12.5% — for 14.5% in total, and the contractor group keeps 85.5%.
Reverse the announced figure and you get the state of the cost bank. For Guyana to take 39.8%, profit oil must be 75.6% of production, which means cost oil has fallen to about 24.4%. Ali described it as roughly twenty barrels in a hundred going to costs, which would produce 42%; the announced 39.8% implies a slightly fuller cost bank than his description. Either way the contractor group's share drops from 85.5% to 60.2% — a 25.3-point transfer.
Nobody renegotiated anything. The agreement did this by itself, on schedule, because the costs it was written to repay have been repaid.
The conversion: 25.3 points of a block is 102.5 thousand barrels of Exxon
The consortium's share is not Exxon's share, and this is where most of the coverage stops. Exxon holds 45% of Stabroek as operator; Chevron holds 30% (through its acquisition of Hess) and CNOOC 25%.
Exxon reported gross Stabroek production of about 900,000 barrels a day in the second quarter. Run that through both splits:
| Contractor share | Gross to contractors | Exxon at 45% | |
|---|---|---|---|
| At the 75% cost-oil ceiling | 85.5% | 769.5 kb/d | 346.3 kb/d |
| On the announced 39.8% | 60.2% | 541.8 kb/d | 243.8 kb/d |
| Change | −25.3 pts | −227.7 kb/d | −102.5 kb/d |
102.5 thousand barrels a day. Exxon's own third-quarter guidance, given on July 31, is a fall of approximately 100 thousand barrels a day.
Two statements made three weeks apart by parties with directly opposing incentives — a head of state describing what his country gained, and an oil major telling investors what it lost — agree to within about 2.5%. That is the strongest evidence available that the 39.8% is a real contractual figure and not a political one, and it is the reason this article can be written at all: the president supplied the percentage, and Exxon independently supplied the barrels.
The denominator, which is the whole point
Exxon produced 4,514 thousand barrels of oil equivalent a day in the second quarter of 2026. The entitlement loss is 2.2% of that.
Put the other way: Exxon's Guyana entitlement falls from about 7.7% of its total production to about 5.4%. The most important growth asset in the portfolio, the one the company has spent $55 billion on since 2014, just handed a quarter of its production share to the state — and the company-level effect is a rounding error and a half.
Priced at the $89.73 average Brent price for August 2026, 100 thousand barrels a day is about $3.3 billion of gross revenue a year — roughly 1% of the $332.2 billion of revenue in our stored series. At $70 Brent it is $2.6 billion; at $100, $3.7 billion.
This is where an article would normally convert that into earnings, and we are not going to. Exxon does not disclose Guyana unit costs, realisations or country-level earnings, so the margin on those specific barrels is not a number anyone outside the company has. The revenue figure above is a ceiling on the cash effect and should be read as one — the barrels carry operating cost and tax, and under Article 15.4 of the agreement Guyana pays the consortium's Guyanese income tax out of its own share. What can be said with the disclosures available is the volume, and the volume is 2.2%.
Exxon's own read: the entitlement falls and the cash goes up
The company was not defensive about this on the call, which is itself informative. CFO Neil Hansen confirmed the $55 billion invested since 2014 had been recovered, roughly two years ahead of the original schedule, on faster execution, better reliability and higher prices — and framed the crossover as a positive:
"This is about value, not volume. We're going to see two times the level of free cash flow in 2030 than we saw in 2025."
The logic is straightforward once the cost bank is out of the way. Cost oil is a repayment mechanism, not a profit: barrels taken as cost oil were reimbursing capital already spent. What replaces them is a smaller number of profit-oil barrels arriving alongside a capex line that stops climbing — and gross volumes that keep growing, with the fifth FPSO, Errea Wittu, sailed in June and on plan for fourth-quarter startup at 250 thousand barrels a day of added capacity.
So there are two true sentences here and they point in opposite directions. Guyana's share of every barrel roughly tripled — 2.7×, properly measured. Exxon's cash from Guyana is guided to double. Both can happen because the barrel count is rising underneath them and the spending that was consuming the cash has been repaid.
The genuine irony is in the trigger. Exxon paid itself back two years early because the project went well, and the reward for executing ahead of schedule is a permanently smaller share of every future barrel, arriving two years sooner than it otherwise would have.
The qualification that decides whether 39.8% survives
The 39.8% is a high-water mark, not a floor, and the reason is a feature of the contract that has been argued about in Guyana for years: the Stabroek agreement is not ring-fenced.
Costs are recovered block-wide rather than project by project. There is one cost bank for all of Stabroek, so capital spent on the next development is recoverable out of production from the last one. Three developments are already sanctioned behind the four FPSOs producing today — Uaru/Errea Wittu at about $12.7 billion, Whiptail at about $12.7 billion, and Hammerhead at about $6.8 billion, roughly $32 billion in all, with Longtail proposed and unsanctioned behind them.
Every dollar of that flows into the same cost bank the block just drained. Cost oil is calculated monthly against the balance, so the contractor group's share does not stay at 60.2% by right — it stays there only while new recoverable spending runs below the rate at which production repays it. The ceiling is still 75%, and nothing in the agreement prevents a return toward it.
That cuts both ways for a reader. It is the largest risk to "Guyana now gets 39.8%" as a forward statement, and it is the largest support for Exxon's claim that Guyana free cash flow doubles by 2030, because it is the same mechanism seen from the other side. It is also the live political question: the opposition has spent August demanding a full 50%, and Guyanese analysts put the cost of the missing ring-fence at roughly $4 to $4.9 billion of foregone profit oil in 2025 alone. Those are advocacy figures, not audited ones, and we have not verified them; the absence of ring-fencing in the agreement itself is not in dispute.
The Exxon baseline, since this is the first time
For context rather than argument, from our stored series, which is why the coverage gap was worth closing:
- Revenue of $332.2 billion over the four quarters through December 2025, down 5.0% year over year.
- Free cash flow of $29.1 billion over the four quarters through September 2025 — an 8.7% margin on the matching revenue window — for a Rule of 40 score of 3.8.
- Capital expenditure of $27.7 billion over that same window, against $55 billion committed to one block over eleven years.
Two things about those figures. The series is stale — it ends at the December 2025 quarter, and Exxon has since reported a June quarter with $14.5 billion of earnings, $23.6 billion of operating cash flow and $17.2 billion of free cash flow. And the Rule of 40 is close to meaningless for this business: a score of 3.8 on a company that generated $17.2 billion of free cash flow in three months is measuring a growth framework against something that is not trying to grow. We keep it here because it is the site's common denominator, not because it settles anything about Exxon.
There is no valuation model for Exxon on this site, so this piece makes no fair-value claim. Where a model exists we reconcile against it; here there is nothing to reconcile, and inventing a discounted cash flow to fill the section would be exactly the kind of number this article has spent its length refusing to print.
What to watch
- Exxon's third-quarter net production, due late October. The guided ~100 kb/d entitlement fall is the first checkable number in this whole story. If reported Guyana volumes drop by materially less, the 39.8% was optimistic.
- Monthly Stabroek gross output, from the Guyanese government's own data. Production is not monotonic — output fell in May and June — and every percentage here is applied to a moving 900 kb/d.
- Whether the cost-oil share starts climbing again once Errea Wittu's capital enters the cost bank at fourth-quarter startup. The first monthly figure showing cost oil back above 30% would make 39.8% a peak rather than a level.
- Any Exxon disclosure of Guyana unit economics. Country-level realisations or costs would convert the 2.2% volume effect into an earnings effect, which is currently not computable from outside.
- Chevron's version of the same event. It holds 30% of the block through Hess and takes the identical hit, proportionally larger against a smaller production base. It has said far less about it than Exxon has, and it is tracked here.
Guyana's 39.8% entitlement and the 12.5% prior figure are President Irfaan Ali's statements of August 18, 2026, reported by Reuters via OE Digital and the Guyanese press; Exxon has confirmed the cost recovery but has not itself published the 39.8%, which is a claimed figure and treated as one throughout. The contract terms — the 75% cost-oil ceiling, the 2% royalty and the 50–50 profit-oil split — are the 2016 production-sharing agreement's, as reported consistently across Reuters, OilNOW and Exxon's own second-quarter presentation. Exxon's disclosed figures — the approximately 100 kb/d third-quarter entitlement decline, 4,514 kboe/d of net production, ~900 kb/d of gross Stabroek output, $14,525M of earnings, $23,555M of operating cash flow, $17,236M of free cash flow, $6,787M of capex, and the fifth FPSO's fourth-quarter startup and 250 kb/d of capacity — are from its second-quarter 2026 results of July 31, 2026 and the accompanying materials. The $55 billion recovery, the roughly two-year acceleration and the "value, not volume" and 2030 free-cash-flow quotations are CFO Neil Hansen's on the July 31 call, quoted from published transcripts of that call rather than from a filing we have read. Derived by us: Guyana's 14.5% prior total take, the implied 24.4% cost-oil share, the 25.3-point transfer, the 2.7× multiple, the 346.3 and 243.8 kb/d Exxon entitlements and the 102.5 kb/d change, the 7.7%-to-5.4% and 2.2% shares of total production, and the $2.6–3.7 billion annual revenue range. Working interests of Exxon 45%, Chevron 30% and CNOOC 25% are long-standing public terms of the block. The $89.73 August 2026 Brent average is press-reported and is used as a stand-in because Exxon does not disclose Guyana realisations; no Guyanese crude differential is assumed. Project capital costs of about $12.7bn each for Uaru and Whiptail and about $6.8bn for Hammerhead are press-reported sanction figures. The absence of ring-fencing in the agreement, and the Article 15.4 provision under which Guyana settles the contractor group's income tax from its share, are well documented; the $4–4.9 billion estimates of its cost to Guyana in 2025 are advocacy-group figures we have not verified. Exxon's trailing revenue, free cash flow, capex and Rule of 40 are our stored series, which ends at the quarter ended 2025-12-31 for revenue and 2025-09-30 for cash flow, and the free-cash-flow margin is computed against the matching four revenue quarters. No share price, market capitalisation or valuation figure appears here, and this site holds no model for Exxon.
