08.5 · Industrials, capital goods, defence, railways & EMS¶
Sector playbook. The capital-goods cycle is India's 2020s story: record government capex (₹12.2 lakh Cr proposed for FY27), a defence-indigenisation push with order books that run to a decade of revenue, railways capex of ₹2.9 lakh Cr, and a PLI-driven electronics manufacturing (EMS) boom. Order books are the vocabulary; execution, working capital and margins are the analysis. This is also Kaveri Pumps' home sector.
Time: ~80 min · Prerequisites: 05.4 The capital cycle, 04.4 Working capital
1. How the industry makes money¶
| Segment | Model | EBITDA margin | Working capital | Order visibility | Customer |
|---|---|---|---|---|---|
| Engineered products (pumps, motors, compressors, valves, bearings, abrasives, cables) | Standard products via dealers/OEMs; some project sales | 12–20% | 60–100 days of NWC | Short cycle; dealer restocking | Fragmented industrial/agri; some government |
| Capital equipment (turbines, transformers, switchgear, machine tools, cranes) | Engineered-to-order; project-linked | 10–18% | High (advances offset) | Order book 1–3 years | Utilities, industry, government |
| EPC / infrastructure (L&T-type; roads, power T&D, water, buildings) | Fixed-price/lump-sum contracts; execution risk | 8–12% | Very high (retention, unbilled, BGs) | Order book 2–4 years; book-to-bill | Government, PSUs, private |
| Defence (aircraft, ships, missiles, electronics, ammunition) | Cost-plus/nominated or competitive; long-gestation; DPSUs dominate | 20–30% (DPSUs) / 10–20% (private) | Customer advances make NWC low or negative for DPSUs; receivables from MoD can lag | Order book 3–10 years of revenue (HAL ~₹2.5 lakh Cr — Defence Standard) | Ministry of Defence; exports |
| Railways (wagons, coaches, locomotives, signalling, track) | Tender-based; Indian Railways as near-monopsony | 8–15% | Moderate; payment discipline good | Order book 2–4 years; budget-driven | Indian Railways, metros |
| EMS (electronics manufacturing services — phones, TVs, appliances, wearables, PCBs) | Contract manufacturing for brands; PLI-linked | 3–5% (assembly) to 8–12% (components, design) | Moderate; customer-funded | Customer programmes; PLI targets | Global and Indian brands |
Government is the customer in defence, railways, T&D and most EPC — payment cycles, budget allocations and policy (indigenisation lists, PLI, ECMS) drive the sector as much as demand does (05.2 India notes).
2. The cycle as of September 2026¶
- Public capex: FY27 budget capex ₹12.2 lakh Cr (IBEF); railways capex ₹2.93 lakh Cr, ~39% spent by July (Business Standard). The question for the cycle is private capex: has it followed public capex (power, cement, steel, EMS say yes; broad-based manufacturing less clearly)?
- Defence: capital outlay ₹2.2 lakh Cr for FY27 (+22%), ₹1.39 lakh Cr for domestic procurement (KPMG); HAL's 97 additional Tejas Mk1A (₹62,000 Cr) and 156 Prachand helicopters (₹62,800 Cr); execution — engine supply from GE, delivery slippage — is the constraint, not orders.
- EMS: Dixon and peers reached ~28% of India's smartphone assembly in FY26; the ₹40,000 Cr Electronics Components Manufacturing Scheme (ECMS) shifts the incentive from assembly to components (Techovedas). Assembly margins are thin and PLI-dependent; the value question is whether component/design capabilities arrive before PLI expires.
- Capital-cycle warning (05.4): defence and railway stocks re-rated 3–10x in 2022–24 on order-book narratives, corrected in 2024–25, and remain priced for execution that has historically slipped. Solar-equipment capacity (modules, cells) has been added far faster than demand — the classic overshoot.
3. KPIs¶
| KPI | Definition | Notes |
|---|---|---|
| Order book / order inflow / book-to-bill | Unexecuted orders; new orders in the period; inflow ÷ revenue | Book-to-bill > 1 = growing backlog; check the executable order book (some orders are framework/optional) |
| Order book / revenue (years of visibility) | 2–4x normal for EPC; 5–10x for DPSUs — but conversion is slow | |
| Execution / conversion rate | Revenue ÷ opening order book | Historical rate predicts revenue better than management guidance |
| Margin by contract type | Fixed-price vs cost-plus; segment margins | Fixed-price EPC margins fall when input costs rise mid-contract (2021–22 steel) |
| Working capital: NWC days, unbilled revenue, retention money, customer advances, mobilisation advances, BGs outstanding | Notes; presentations | The sector's cash-flow killer; Kaveri's solar receivables in miniature |
| CFO/EBITDA over 3 years | EPC companies often show < 50% | |
| Capacity utilisation | For product companies | Kaveri: pumps 78%, motors 57% |
| PLI / incentive income | Other operating income | Policy-dependent; separate from core margins |
| Export share | Defence exports policy target; engineering exports to US/EU |
4. What drives earnings and multiples¶
Government budget and payment discipline; private capex revival; commodity input costs (steel, copper — Kaveri's gross-margin story); policy (PLI, ECMS, indigenisation lists, Make-in-India procurement preference, "Atmanirbhar"); interest rates (financing for buyers); execution capability (skilled labour, engines, components); exports. Multiples: product companies with strong franchises (Cummins, ABB, Siemens, Polycab, Havells) 40–60x at the cycle peak; EPC 20–30x; DPSUs re-rated from 10–15x to 30–50x in 2022–24, then partially corrected; EMS 60–100x at peak on growth narratives. The sector's multiples embed cycle-peak margins and order-book conversion that the capital-cycle framework says to fade.
5. Accounting quirks¶
Percentage-of-completion revenue on contracts (Ind AS 115 over-time recognition): unbilled revenue/contract assets, cost-to-complete estimates that can be optimistic; retention money in receivables; customer advances as contract liabilities (negative working capital for DPSUs is customer money, not efficiency); liquidated damages provisions; warranty provisions; PLI income recognition timing; joint ventures and consortium accounting in EPC; CWIP ageing for capacity builds (Kaveri's Hosur); export incentives.
6. Red flags¶
Order book growing while execution rate falls; unbilled revenue and retention rising faster than revenue; margins held by capitalising costs or under-provisioning liquidated damages; working-capital borrowings funding receivables from state entities (Kaveri); order announcements that never appear in revenue; related-party subcontracting; a company "diversifying" into defence/EV/solar to catch a narrative; PLI-dependent margins presented as structural; promoter pledges rising with the stock (the 2023–24 small-cap defence/railway rally saw many).
7. How to value¶
Product companies: DCF with capacity/utilisation and margin drivers, cross-checked with P/E vs growth and ROCE (Kaveri's reference valuation is the template). EPC/defence: order-book conversion DCF (revenue = executable backlog × conversion + new orders × win rate), margin by contract type, explicit working-capital build, and a scenario for payment delays; P/E only with the cycle stage stated. EMS: DCF with PLI expiry and margin-mix drivers; EV/Sales sanity check against terminal margins (07.4 logic applies).
8. Representative companies (for study; verify)¶
Larsen & Toubro (EPC + everything), Siemens, ABB, Cummins India, Thermax, Bharat Forge, Polycab, KEI, Havells, Kirloskar Brothers, KSB (pumps — Kaveri's real peers), Grindwell, Carborundum, Timken, SKF, Schaeffler; HAL, BEL, Bharat Dynamics, Mazagon Dock, GRSE, Cochin Shipyard, Data Patterns, Solar Industries, Astra Microwave (defence); RVNL, IRCON, Titagarh Rail, Jupiter Wagons, Texmaco (railways); Dixon, Kaynes, Syrma SGS, Amber, PG Electroplast (EMS); Waaree, Premier Energies (solar equipment — capital-cycle case).
9. The 10-question sector checklist¶
- Order book: size, executable portion, ageing, and historical conversion rate?
- Margin by contract type, and exposure to input-cost inflation on fixed-price work?
- Working capital: unbilled, retention, advances, BGs, receivables from government entities — trend?
- CFO/EBITDA over three years?
- Customer: which budget line funds this, and what is the payment record?
- Policy dependence: PLI/ECMS/indigenisation — what happens when it changes?
- Capacity: utilisation, capex plans, and industry-wide capacity announcements (the capital cycle)?
- Execution constraints: components, engines, skilled labour, land?
- Governance: promoter pledges, related-party subcontracting, announcements vs delivery?
- What conversion rate and margin does the multiple imply, and how does that compare with history?
← Previous: 08.4 Pharma & healthcare · Module index · Next: 08.6 Autos & ancillaries →