Work the increment, not the total. Revenue of 1.5 trillion yuan after 5.3 per cent growth [3] implies a prior-year half of about 1.4245 trillion [3], so the top line added roughly 75.5bn yuan [6]; net profit of 103.94bn after a 22 per cent rise [2][1] implies a prior-year 85.2bn [2], an increase of about 18.7bn [7]. Divide one by the other and something close to 24.8 per cent of each additional yuan of revenue arrived as net income [8], against an average net margin of 6.9 per cent for the half [1]. The average margin moved from about 6.0 to 6.9 per cent, call it nine tenths of a point [4][5], which sounds modest until you notice that profit grew 4.2 times as fast as the revenue underneath it [9].
The reported 1.5 trillion is two significant figures, so treat the incremental figure as a band rather than a point: flex the base between 1.45 and 1.54 trillion and the drop-through sits somewhere between about 24 and 26 per cent [11]. Loose, but not loose enough to be confused with seven.
Drop-through at that level does not come from selling more barrels at an unchanged spread. It comes from the barrel repricing, which is what PetroChina itself points to in crediting elevated global energy prices [1], and what the report generalises when it says the market continues to support upstream earnings for major producers despite varied downstream conditions [5]. Note the provenance, though: this account comes from cryptobriefing.com, which closes by selling prediction-market analysis [10], and its own takeaway that market pricing implies better odds of crude reaching new highs [7] arrives without a single number attached, so there is nothing there to check.
The counter-thesis sits in the same sentence as the thesis. The report also credits enhanced fuel sales [4], which is a volume claim about the downstream half of an integrated company, and it supplies no segment breakdown at all [9]. The gap admits more than one reading. Price-led drop-through, in which the 18.7bn is mostly upstream realisation and repeats for as long as crude holds. A two-legged version, or rather the more interesting version, in which refining and marketing spreads widened at the same time, making the increment sturdier than a pure price call. And a mix in which inventory or currency effects did real work, invisible at this level of disclosure. This is probably wrong in one particular, but the first reading is where I would put the weight, because 4.2 times revenue growth is very hard to manufacture out of throughput.
What would kill it: a segment table showing refining and marketing profit, not exploration and production, contributing the bulk of that 18.7bn. That single disclosure settles the argument, and nothing in the supplied account does.
One discipline on reading it as a shareholder-return story. Nothing here says what happens to the 18.7bn, because the material carries no capex line, no dividend, no buyback, and the company is described only as China's largest oil and gas producer with a bigger profit than last year [6][2]. Price drove this result, and price will decide what stays.
