ARSENALIND
Supply chain
04

Subsystems

BEL's kingdom — the profit pool of Indian defence lives at the subsystem layer

CompetitiveImport dependence 35%Margins: PremiumOligopoly

Radars, EW suites, sonars, fire-control computers, communication sets, simulators. The subsystem layer combines electronics margins (25-40% OPM) with DPSU-scale order books, and it is the layer most protected by security-clearance moats: a private newcomer can machine a bracket, but shipping a certified naval EW suite requires years of DRDO co-development and classified lab access. BEL converts that moat into a ₹75,000 Cr order book at 29% margins with zero debt.

Structure & moats

  • ▸DRDO designs, BEL/Data Patterns/Astra productionise: the Indian model splits IP (state labs) from manufacturing profit (listed players) — understanding which programs carry production-linked royalties is key to margin forecasting.
  • ▸Security clearances + classified test infrastructure form a 5-10 year moat; only ~8 private firms hold system-level EW/radar clearances.
  • ▸QRSAM (₹30K Cr+ program), Akash-NG, and naval EW (Shakti follow-ons) are the three near-term production waves — BEL is prime on all three.
  • ▸Simulators (Zen) monetise training-hour mandates: DGMO's simulator-hours-per-soldier policy converts directly into Zen's ₹36-40% OPM annuity — but ordering is lumpy (FY26 proved it).

The numbers experts anchor on

  • ▸BEL: book-to-bill sustained >2x since FY22; EBITDA margin walked 22%→29% on indigenised BOM — every 5pt of import substitution ≈ 1pt of margin.
  • ▸Data Patterns: 40% OPM, 90x P/E — the market prices design-IP scarcity; the risk is order lumpiness against a ₹1,300 Cr book (14 months of revenue).
  • ▸Astra Microwave: OPM 12%→29% in five years as mix shifted from build-to-print to build-to-spec TR modules for AESA programs (Uttam, Virupaksha for Su-30 upgrade).
  • ▸Op Sindoor (May 2025) validated Akash + EW stack in combat — export enquiries (Brazil, Philippines, Armenia) follow proven-in-battle systems.

Watch items

  • ▸QRSAM production order split (BEL vs BDL share) — announced tranches decide FY28-30 revenue mix.
  • ▸Uttam AESA integration on Tejas Mk1A tranche-2 — pulls Astra/Data Patterns TR-module volumes.
  • ▸GaN foundry access (AGNIT + SCL) — GaN TR modules double radar power-efficiency; whoever locks domestic GaN supply owns the next radar generation.

Players at this stage

Private / PSU-unlisted
DRDO labsTata AdvancedAlpha Design (Adani)
Global benchmark
RaytheonThalesIAI/ELTAHensoldtSafran Electronics