Business Economics

Business Economics — Bajaj Mobility AG

Figures converted from EUR at historical FX rates — see data/company.json.fx_rates. Ratios, margins, and multiples are unitless and unchanged.

Bottom line. Bajaj Mobility is an annuity-tailed cyclical premium-OEM with an India royalty bridge — about 68 % of revenue is wholesale premium motorcycle hardware sold one cycle at a time, 20.5 % is a high-margin PowerParts/PowerWear aftersales annuity attached to a 3 m+ bike installed base, and a quietly growing slice is a license/royalty stream from Bajaj-built 125–390 cc KTM/Husqvarna sold through the Indian parent's own network (78,906 units in 2025, +27 % YoY against group volume down 28 %) (AR_2025 p.L-12). The pre-crisis economic baseline is the FY2016–FY2022 average EBITDA margin of 14.7 % on ~$1.9 bn average revenue, with ROE in the high-teens (income.json; ratios). The Mattighofen plant needs roughly 140 k units a year to absorb fixed overhead; it produced 48,377 in 2025 (AR_2025 p.L-12) — which is why a ~25 % revenue drop in 2023→2024 turned into an EBITDA swing from +$358 m to −$500 m, ~4–5× operating leverage in reverse (income.json). The single underwriting debate is whether mid-cycle EBITDA margin returns to 14–16 % on a rebuilt $1.8–2.4 bn revenue base — that is the math that decides whether this is a Harley template (premium-OEM with annuity tail), a Royal Enfield template (emerging-premium volume play) or a Piaggio template (structurally compressed European motorcycle franchise).

1. Revenue Mix & Where the Profit Pool Actually Sits

The headline number is one P&L; underneath it sit four economically distinct slices, each with a different margin shape and a different cycle exposure. The aftersales line is small in revenue but disproportionately large in gross-profit dollars; the India royalty stream is small in revenue (booked inside Group revenue, no plant capex) but the only growing volume pool inside the consolidated business.

Source for all six rows: AR_2025 p.L-12 segment table.

Profit pool, qualified. Bajaj Mobility does not publish segment EBIT. What it does publish — that 2025 gross margin was 1.8 % versus −2.7 % in 2024 (AR_2025 p.L-13), held positive by aftersales — tells you the profit pool must be skewed toward parts and apparel. With new-bike gross margins compressed to near zero by destocking discounts and US tariff drag, and an industry-standard 40–60 % gross margin on premium aftersales, the majority of FY2025 gross-profit dollars came from aftersales, despite aftersales being only 20.5 % of revenue. A "normalized" cycle gross margin (FY2016–FY2022 average ~20–22 % of revenue, reverse-engineered from EBITDA margin 14.7 % plus normalized opex run-rates) would still show aftersales pulling above its weight: at ~20 % of revenue and 2× the gross margin of new bikes, it likely throws off ~30–35 % of cycle gross-profit pool.

The India royalty/license stream is not separately disclosed on the FY2025 P&L. The volume disclosure is unambiguous — 78,906 Bajaj-direct units in 2025 against 62,084 in 2024 (+27.1 % YoY) (AR_2025 p.L-11) — but the dollars booked back into Bajaj Mobility revenue per unit is a related-party transfer price that the company has not published. This is a material gap (see Section 8).

Source: AR_2025 p.L-12 geographical revenue table. Europe revenue down −54.0 % YoY (largest absolute hit); North America still 33.4 % of revenue but down −38.2 % YoY.

2. Unit Economics

The right unit for premium motorcycles is revenue per unit by product family, not the blended number. In 2025 BMAG sold 132,580 units inside the consolidated motorcycle segment (KTM/Husq/GASGAS + CFMOTO + MV Agusta residual + others), plus 78,906 Bajaj-direct units, for 211,486 total motorcycles (AR_2025 p.L-11). The implied ASPs differ by an order of magnitude across the slices.

Sources: AR_2025 p.L-11 (units), AR_2025 p.L-12 (revenue). Note that the consolidated ASP including aftersales/Bajaj/bicycles falls to ~$5,653 ($1,186 m ÷ 209,704 motorcycles sold (operating_kpis.json approximation), 28 % below the FY2023 ASP of ~$7,894 ($2,940 m ÷ 372,511 units).

The peer table — pulled from competition-claude.md and the underlying competitor 10-Ks — anchors what these ASPs mean.

Sources: BMAG income.json + operating_kpis.json; HOG 10-K FY2025; EICHERMOT AR FY2025; PIAGGIO AR 2024 (per competition-claude.md).

Bajaj Mobility's pre-crisis revenue-per-unit (~$7,900) sat firmly between Royal Enfield (~$2,300, lower-tech bikes at India unit cost) and Harley-Davidson (~$28,700, mostly cruiser ticket + finance revenue). The 2025 ASP compression to ~$5,653 is partly mix (more Bajaj-built small-displacement, fewer high-end adventure bikes from a halted Mattighofen) and partly forced clearance discounting (AR_2025 p.L-12: "die angestrebte Reduktion der Lagerbestände negativ beeinflusst, da diese Zielsetzung mit zusätzlichen Rabattmaßnahmen verbunden war"). Investors should expect this to recover only modestly because the next leg of unit growth is small-displacement India.

Aftersales per bike-in-the-wild. BMAG does not disclose installed-base count. Cumulative KTM/Husq/GASGAS sales over the last decade (~3 m units, derived from annual unit data) imply aftersales of $243 m FY2025 corresponds to roughly $70–82 per active bike per year in PowerParts/PowerWear. That is well below Harley HDFS-adjusted aftersales (HOG 10-K: parts & accessories + general merchandise ≈ $1,074 m / ~5 m+ active US bikes = ~$200 per bike), suggesting attachment headroom.

R&D per unit. FY2025 R&D expense was $165.7 m, 14.0 % of revenue (AR_2025 p.L-13) — versus a normalized 8–10 % on healthy revenue. Spread over 211,486 motorcycles sold (consolidated incl. Bajaj-direct), that is roughly $784 of R&D per unit, versus Royal Enfield's ~$13 per unit (Eicher FY2025 R&D ~2.4 % of revenue) and Harley's ~$1,000 per unit. BMAG's R&D-per-unit sits above premium pure-play Royal Enfield by ~50× and approaches HOG's — a signal that brand-IP investment is being preserved through the crisis, but also that fixed-cost gearing is unsustainable at low volume.

Factory motorsport spend. AR_2024 disclosed combined sales/race-program opex of $423 m, of which industry estimates place ~$94 m on factory motorsport (race operations + sponsorship + riders). In 2025 the combined Vertriebs- und Rennsportaufwendungen line fell −31.6 % alongside the overall opex cut (AR_2025 p.L-14), implying motorsport was held roughly intact in absolute terms while distribution opex absorbed most of the cut — which is consistent with KTM winning 29 world titles in 2025 (its best season ever) and the 2026 Dakar (21st KTM bikes Dakar title) during the production stop (Q1_2026 interim p.4–5). Per world title won, that is ~$3.1 m of factory race spend — a marketing channel competitors cannot replicate without similar multi-year commitment.

3. Cost Structure & Operating Leverage

The economic distinctiveness of Bajaj Mobility's cost base is the Mattighofen single-plant fixed-cost stack plus structurally elevated R&D intensity. Together these set the break-even volume and explain the violent operating leverage when demand swings.

Sources: income.json (FY2022–FY2025), AR_2025 p.L-13 to L-14 for clean EBIT and 2025 expense mix.

The cost block that didn't flex with revenue is what bit. Between FY2022 and FY2025, revenue fell −54.4 % ($2,599 m → $1,186 m), but R&D expense fell only ~24 % (~$198 m → $166 m gross; AR_2025 p.L-13 reports total R&D incl. capitalisation effects "weitgehend unverändert", −1.8 %). Admin + sales/racing opex flexed harder (sales/race −31.6 %, R&D gross −42.7 % when measuring expense before capitalisation/grants), but the capitalisation rate of development cost dropped from 60.1 % to 36.3 % (AR_2025 p.L-14) — cleaner accounting that pushes more R&D through the P&L on a smaller denominator and lifts R&D-as-percent-of-revenue optically.

Mattighofen break-even math. The plant is the high-margin end of the franchise. Empirically (operating_kpis.json + AR cover Konzern-Kennzahlen):

The 2018 EBITDA-margin peak (16.2 %) came at ~145 k Austrian units (estimated; 2018 not separately disclosed but consistent with 2020's 140 k baseline). The 2021–2022 EBITDA-margin run-rate of ~16 % came at 179–222 k Austrian units. The math:

• Below 100 k Austrian units → fixed-cost absorption broken (EBITDA margin under 10 %, falling fast).
• 140–180 k Austrian units → mid-teens EBITDA margin (the franchise's natural state).
• 220 k+ Austrian units → 16 % EBITDA margin ceiling, set by R&D intensity not by COGS.

That ceiling is what differentiates this from Royal Enfield (~24 % EBITDA margin at $2,300 ASP and minimal R&D). KTM's product cadence and 29-world-titles-a-year race program is paid for here; the brand gets the R&D, but the margin never catches Eicher.

Why a ~25 % revenue drop in 2023→2024 produced an −$858 m EBITDA swing. Revenue −$988 m × incremental gross margin (~30 % on the lost units, premium-mix) = ~−$296 m of mechanical gross-profit loss. Add ~−$520 m of one-time impairments (Husqvarna E-Bicycles wind-down, dealer-receivables provisions, restructuring charges in COGS/sales) and ~−$73 m of fixed-overhead unabsorption at half-utilised Mattighofen, and you reproduce the full −$858 m swing. The operating-leverage ratio of ~4–5× quoted in business-claude.md is the right shorthand, but the majority of the swing was impairments, not pure operating leverage — that distinction matters when you forecast FY2026 because the impairment block does not repeat.

4. Capital Intensity & Returns

Premium motorcycles look "capital-light" from a distance — single plant, no aluminum smelter — but BMAG's reinvestment intensity through the cycle has been 9–14 % of revenue in capex plus large swings in working capital. The result is a business that converts P&L profit to FCF poorly except in the back half of the cycle, and that ran a working-capital build straight off the road in 2022–23.

Sources: cash_flow.json (capex, CFO, FCF), balance_sheet.json (working capital), income.json (revenue).

Working capital was the leading indicator. It ran 12–18 % of revenue from FY2016–FY2022, peaked at 19.9 % in FY2023 as dealer-channel stuffing broke, and was forcibly cut to $403 m FY2025 (still 34 % of a collapsed revenue base). The $407 m WC build in 2023 alone ($222 m → $587 m) is what tipped the parent into the November 2024 insolvency filing — see numbers-claude.md §3 for the full forensic. In Q1 2026, WC ticked up again to $442 m on $43 m of trade-receivables expansion (Q1_2026 interim p.5) — the first place a second cycle would show up.

Returns on capital, pre-crisis only. FY2024–25 ROE and ROIC are not meaningful: equity went through zero in 2024 and the 2025 print embeds the $1.4 bn creditor haircut. The pre-crisis cycle:

Source: ratios.json (via numbers-claude.md §5).

The pre-crisis 7-year ROE average is 18.2 %, pre-2020 5-year ROIC average is 10.8 %, with FY2018 the strongest year (ROE 21.2 %, ROIC 13.6 %). Both numbers are above any reasonable cost of equity/capital for a European premium-OEM cyclical — and below Royal Enfield's ~20–25 % ROE on a smaller R&D burden. The franchise earned its capital cost in the pre-crisis cycle; it did not over-earn it. That is the right anchor for a normalized-returns view.

5. Cash Conversion Through the Cycle

The cleanest single discipline on this company is EBITDA → OCF → FCF. The series tells you when reported profit was paper and when it was real.

Sources: income.json (EBITDA), cash_flow.json (CFO, FCF).

FY2023 vs FY2025 — the most informative comparison. FY2023 was the last "normal-looking" P&L ($2,940 m revenue, $358 m EBITDA, $84 m net income reported) but the cash statement showed cash conversion broken: CFO −$123 m, FCF −$456 m. FY2025 reverses the optics — $1,027 m reported EBITDA on $1,186 m revenue — but cash conversion is, again, negative: CFO −$26 m, FCF −$40 m. The conclusion is unambiguous: neither year is a usable read on cash earnings power. The $1.4 bn Sanierungsgewinn is explicitly excluded from operating cash flow (AR_2025 Konzernanhang p.K-25: "die Auszahlungen zur Realisierung des Sanierungsgewinns wurden vollständig im Cashflow aus der Finanzierungstätigkeit als eigenständige…" — i.e., the creditor cash quota of $616 m flowed through financing, not operating).

Normalized mid-cycle FCF estimate. Anchor on the FY2019–FY2021 average (the cleanest pre-crisis window): CFO $363 m, capex $184 m, FCF $167 m on average revenue of $1,966 m → a mid-cycle FCF margin of ~8 %. Applied to a recovered $2,000–$2,235 m revenue base, that is $160–$180 m of normalized FCF — call it $165 m a year mid-cycle. FY2022 (peak revenue) shows you why you cannot use the cycle peak: working-capital build at the top consumes incremental FCF. The FY2019–FY2021 window is the right anchor, and even that is generous because it precedes the Bajaj-channel ramp that should compress capex/revenue going forward (royalty model = asset-light).

6. Post-Restructuring Capital Stack & What It Means for Economics

The new balance sheet, capital allocation, and divestments together change the underlying unit economics in three measurable ways. (1) Net debt is $938 m at year-end 2025 ($1,075 m long-term debt + $300 m short-term + $300 m other; offset by $161 m cash); the $529 m Bajaj-parent restructuring loan was repaid in Q1 2026 and replaced by the $646 m five-year unsecured facility from a JPM/HSBC/DBS/MUFG consortium at low-to-mid single-digit margin, plus a separate $411 m shareholder loan from Bajaj Auto International Holdings (Q1_2026 interim p.6; AR_2025 Konzernanhang p.K-30). (2) The exited businesses (E-Bicycles, MV Agusta, X-BOW, FELT, CERO Design Studio) removed loss-making revenue and management bandwidth — the bicycle segment alone was $85 m of revenue and a documented strategic-mistake margin drag (numbers-claude.md §6). (3) The Bajaj-channel India volume is now the only structurally growing volume pool, +27 % YoY in 2025 against group volume −28 %, +62.2 % in Q1 2026 against group motorcycle volume +93 % (Q1_2026 interim p.3–4).

What this changes for unit economics going forward:

• Interest expense. FY2024 reported interest expense of $128 m; FY2025 $99 m (lower because insolvency-period interest accrual was paused; AR_2025 p.L-14). On the new capital stack ($646 m senior at ~EURIBOR + ~3 %, $411 m shareholder loan at related-party terms, $32 m investment loans, $49 m leases), normalized full-year interest expense should run $65–$88 m — well above pre-crisis ($23–34 m) but materially below FY2024. In Q1 2026 interest expense was already $29 m, annualizing to ~$115 m, indicating the run-rate is still settling (Q1_2026 interim p.5).
• Aftersales economics under Bajaj scale. Bajaj Auto's procurement scale on PowerParts/PowerWear (especially apparel and accessories) should compress landed cost over 24–36 months. There is no public commitment or disclosed savings target — this is a thesis, not a fact.
• Bajaj-channel royalty. Not disclosed. The growth in volumes (+27 % YoY, +62 % Q1) is real; the dollars that flow back through Group revenue per unit are not. This is the single biggest data gap in the post-restructuring economic model.
• Capex profile. FY2025 capex of $86 m was survival-mode (−68.7 % YoY). Steady-state capex for premium-OEM at restored volume should run $153–$212 m a year — i.e., FY2026 capex will step up materially, which will compress FCF on the way to volume recovery.
• No captive finance. Unchanged. Bajaj does not have a captive finance arm in EU/NA, and building one is a 3–5 year regulatory project. This remains the structural disadvantage versus Harley-Davidson Financial Services (~19 % of HOG consolidated revenue and the bulk of HOG's economic profit; see moat-claude.md §4).

Q1 2026 — first clean quarter post-insolvency. Revenue +70.2 % YoY to $381 m. KTM/Husq/GASGAS motorcycles revenue +151.6 % to $313 m. Aftersales (PG&A) +19.7 % to $67 m, now 17.5 % of segment revenue (down from 20.5 % FY2025 because new-bike revenue rebounded faster). EBITDA back in the black at $6.3 m (1.7 % margin). Clean EBIT improved $75 m YoY to −$30 m (Q1_2026 interim p.2–5). The annualized run-rate is ~$1.5 bn revenue — half the FY2023 peak, two-thirds of an FY2026 plausible base. This validates that demand exists when supply returns; it does not validate that mid-cycle margins return on schedule.

7. Verdict — Which Business-Model Template?

The economics of Bajaj Mobility are not those of any single peer. The discipline is to ask which template best predicts steady-state ROIC, not which marketing narrative sounds most ambitious.

Sources: BMAG income.json + AR_2025; HOG 10-K FY2025; EICHERMOT AR FY2025 (per competition-claude.md).

8. What's Not Disclosed (and Would Move the Verdict)

Three data gaps materially constrain the analysis above. Naming them is part of the answer.

The investor framing: if all three gaps (segment EBIT, Bajaj royalty per unit, aftersales gross margin) were disclosed honestly, the Harley-template verdict above could move in either direction by 200–400 bps of mid-cycle EBITDA margin. That is the scale of the disclosure-driven uncertainty in normalized earnings power. Until then, anchor on the pre-crisis 14.7 % EBITDA-margin average, the 8 % normalized FCF margin, and the 18.2 % ROE average — and treat anything tighter as false precision the data does not support.