The Margin Math

The Margin Math

GTRA's net margin rose from 9.2% to 10.6% in FY2025 even as gross margin fell 8 points to 34.2% (related-party bought-in haulage reaching Rp249.6bn, 57.8% of cost of revenue) only because financing cost fell to a 9.9%-of-revenue trough, yet Rp240.8bn of new debt-funded fixed assets added late in the year carry ~Rp25-29bn of annual interest (28-32% of pre-tax profit) and a 10% affiliate-haulage repricing would cost Rp25.0bn (27.5%) — so the offset that holds the margin up is itself about to erode. [1] [2]

The reported improvement reads like a business getting stronger; the mechanics say otherwise. A full year of interest on the Rp240.8bn of trucks added late in FY2025 runs to roughly Rp25bn to Rp29bn — 28-32% of the Rp90.9bn pre-tax profit — and the 9.9%-of-revenue financing ratio that flattered net margin is a trough against 16.3% a year earlier. Against that, the administrative operating leverage behind part of the improvement is genuine and recurring, so a high-single-digit net margin may prove sustainable rather than a one-year peak. What follows separates the durable supports from the borrowed ones: gross margin fell from 42.3% to 34.2% as a rising share of the work shifted to a related affiliate, while net margin rose only because financing cost and tax fell as a share of revenue — tailwinds that look temporary.

A divergence that points the wrong way

The headline is a widening gap between two margins that normally move together. Over FY2023–FY2025 revenue grew 90%, gross margin fell more than eleven points, and net margin drifted up.

Gross and net margin computed from consolidated revenue, cost of revenues and net profit: FY2025 and FY2024 from the FY2025 statement of profit or loss [3]; FY2023 from the FY2024 statement [4].

The obvious suspect — a shift in mix toward the low-margin Karoseri body-building unit — is not the answer. Karoseri revenue did grow faster than the group, from Rp22.7bn to Rp71.3bn, but its own gross margin improved sharply, from 8.2% to 28.3%, so it lifted the blend rather than dragging it [5]. The compression sits entirely in the Land Transportation segment, which is 89% of revenue: its gross margin fell from 44.2% to 34.9% in a single year [6].

The gross margin went into bought-in haulage

The cost-of-revenues note splits the trucking cost base into two kinds of cost: haulage bought in from other operators, and the running cost of the fleet Graha Trans owns — depreciation, tyres, spare parts, insurance. The two moved in opposite directions relative to revenue.

Bought-in haulage ("Transportation expense") and owned-fleet running cost (depreciation, spare parts, insurance, tyres, other) from Note 25, divided by Land Transportation segment revenue of Rp585.0bn (2025) and Rp406.2bn (2024) [7].

Bought-in haulage rose from Rp163.8bn to Rp296.9bn — up 81% against 44% revenue growth — so it climbed from 40% to 51% of segment revenue [8]. The owned-fleet running cost — Rp62.8bn rising to Rp84.0bn — grew 34%, slower than revenue, and held near 15% of segment sales in both years [9]. The whole 9.3-point fall in Land Transportation gross margin is the bought-in line: the company's own trucks remained as efficient as before, but a growing share of the freight is being fulfilled with hired-in capacity carried at a much thinner spread.

The capacity is hired from an affiliate

That hired-in capacity is not an arm's-length spot market. Purchases from related parties reached Rp249.6bn in FY2025 — 38.0% of consolidated revenue and 57.8% of the total cost of revenues, against just Rp36.6bn (8.5% of revenue) a year earlier [10]. That figure is about 84% of the Rp296.9bn bought-in haulage line, so the growth in subcontracted freight is overwhelmingly one affiliated supplier — the common-control counterparty examined in Related-Party Flows. In two years Graha Trans has moved from running the trucks that carry its freight to routing the majority of new volume through a related operator and booking the margin that is left.

This is the mechanical reconciliation of the divergence the report has carried since Growth on Borrowed Money: the business is not losing pricing power on its own fleet so much as re-routing growth through a thin-spread, asset-light, related-party channel. It has a defensible logic — it lets the company serve a surging e-commerce customer without buying and financing ever more trucks — but it means reported gross profit increasingly reflects a markup on an affiliate's invoice rather than the economics of an owned asset.

Why net margin rose anyway

If gross margin fell eight points at the group level, net margin could only rise because everything below gross profit fell faster still. It did.

Each line as a share of consolidated revenue; effective tax rate is tax expense over pre-tax profit. Administrative overhead from Note 26, financing cost is finance charges plus musyarakah financing cost, tax and pre-tax profit from the statement of profit or loss [11] [12]; FY2023 from the FY2024 statement [13].

Two of the three helpers are ordinary. Administrative overhead fell from 12.9% to 10.4% of revenue — genuine operating leverage on a fixed cost base [14]. The effective tax rate dropped from 30.1% to 23.8% [15]. The third is the anomaly. Total financing cost — finance charges plus musyarakah cost — fell from Rp69.9bn to Rp65.1bn even though interest-bearing debt rose roughly Rp150bn over the year, to about Rp728bn, as traced in Financing the Fleet. Financing dropped from 16.3% of revenue to 9.9% while the debt it services grew by a quarter.

The reconciling item is timing. Note 33 records Rp240.8bn of trucks acquired during FY2025 through new consumer-financing (Rp193.3bn) and bank debt (Rp47.5bn) [16], while consumer-financing interest itself fell from Rp41.6bn to Rp29.8bn as older, higher-cost leases ran down [17]. Much of the new stack was added late enough that only a partial year of its interest reached the FY2025 income statement. Even at a low-double-digit borrowing rate, a full year on Rp240.8bn of fresh debt is on the order of Rp25bn to Rp29bn of financing cost — a quarter or more of the Rp90.9bn pre-tax profit — before any offset from continued amortisation of the older leases. The FY2025 financing ratio is more likely a trough than a new normal.

Durability of the ten-percent margin

The net margin therefore leans on two supports the gross-margin trend does not: a financing ratio that looks set to rise, and an affiliate haulage bill whose pricing the company itself flags as not independently established. Both make the ~10% margin sensitive in the direction of lower, not higher.

The affiliate exposure is straightforward arithmetic. On Rp249.6bn of related-party purchases, a 5% increase in the rate the affiliate charges would cost Rp12.5bn — 13.7% of FY2025 pre-tax profit; a 10% increase would cost Rp25.0bn, or 27.5% [18]. Because the counterparty is under common control, the split of margin between Graha Trans and the affiliate is a decision made inside the ownership group rather than by a market, and the earnings base is only as durable as that decision.

FY2024–FY2025 growth in bought-in haulage and owned-fleet running cost, affiliate purchases as a share of the bought-in line, and the pre-tax profit impact of a 10% rise in affiliate haulage rates, all from Note 25 [19] [20].

The fair reading of the other side: none of this has yet reduced profit in absolute terms. Gross profit still grew 23.6% in FY2025, to Rp224.3bn, and net profit grew 75% [21]. An asset-light model that hires capacity instead of buying trucks is genuinely capital-efficient: it is the company's most capital-efficient route to keep serving its largest customer without the additional lease debt the balance sheet can least afford. If the affiliate haulage is priced fairly and the financing normalisation is modest, a high-single-digit net margin on fast-growing revenue is a perfectly reasonable base case — and the owned fleet, the part the company controls, is running at a stable cost ratio.

What would settle it is close at hand. The FY2025 pattern — gross margin down, net margin up on falling financing cost — should reverse first at the gross line and then at the net line as the new debt seasons. The Q1 FY2026 run-rate discussed in Valuation and Estimates already shows net margin easing toward 10% and gross margin lower year on year, the first data point in that direction. A H1 FY2026 print on 4 August 2026 that holds net margin near 10.5% with gross margin stabilising above 34% would say the bought-in model is accretive and the earnings more durable than this chapter treats them; a gross margin below 33% with net margin under 9% would confirm that FY2025 borrowed its net margin from below the operating line.