14 of 14 defence-PSUs have repriced. The capex cycle, not the war drum, did it

Three hundred and twenty percent in eighteen months. All fourteen names in the cohort positive. No stragglers. The screen looks like it was drawn by someone who had already read the order books.

It wasn't. It was drawn by a market catching up to information disclosed, filed, and available in exchange submissions for quarters before the price moved.

The popular explanation across brokerage notes and television segments was geopolitics: border tensions, a rearmed neighbourhood, a government that suddenly discovered procurement urgency. Convenient. Also analytically insufficient. Geopolitics explains a sentiment spike. It does not explain a sustained eighteen-month re-rating that lifted every name in a fourteen-stock cohort without exception. For that, you need ₹2.56 lakh crore of disclosed order book and a structural procurement floor that is not going anywhere in this decade.

The 14, and the ₹2.56 Lakh Crore They Converted

HAL, BEL, BDL, BEML, Mazagon Dock Shipbuilders, Garden Reach Shipbuilders, Cochin Shipyard. That is the listed-PSU spine. Layer on the private-defence names riding the same procurement tailwind: Astra Microwave, Apollo Micro Systems, Solar Industries, Premier Explosives, Data Patterns, MTAR Technologies, Paras Defence. Fourteen stocks, approximately 320 percent over eighteen months ending early 2026, cohort-weighted. Every name positive.

The breadth is the signal. A single headline drives two or three directly exposed names. A capex cycle drives the whole supply stack: airframes, electronics, propulsion, munitions, shipbuilding, support. When all fourteen names reprice simultaneously, the driver is structural, not situational.

Now look at the order books. At end-FY25: HAL disclosed ₹1.25 lakh crore. BEL disclosed ₹74,000 crore. Mazagon Dock disclosed ₹38,000 crore. BDL disclosed ₹19,000 crore. Combined: ₹2.56 lakh crore across four names alone. Filed, public, accessible to any analyst with a Bloomberg terminal and the patience to read exchange disclosures. These were not pipeline estimates or back-of-envelope projections. These were signed contracts, sitting in filings, waiting on conversion.

The market spent eighteen months deciding whether to price those order books at face value. When execution became undeniable, it did. That transition is the 320 percent.

The Procurement Floor: ₹6 Lakh Crore and a 75 Percent Rule

The order books are the near-term story. The structural story runs farther out.

The FY24-FY29 defence procurement plan, documented in Ministry of Defence planning material, commits approximately ₹6 lakh crore over five years. Not a wish list. Defence capital expenditure has expanded consistently across the past five Union Budgets, and the indigenisation mandate converts that budget into a directed demand pipeline for domestic manufacturers rather than an import window dressed up as procurement.

The 75 percent indigenisation target, to be met by FY27, is the mechanism that closes the import leak. Prior to the indigenisation push, a defence budget expansion could partially route to foreign platforms, imported sub-systems, and offset arrangements that provided headline compliance without genuine domestic manufacturing depth. The 75 percent floor redirects that spending. A rupee of defence capital allocation is now worth materially more to the cohort than it was five years ago.

The iDEX scheme, launched in 2018 and scaled materially since, created the private-sector channel. Before iDEX, the procurement route for a startup developing drone payloads or field communication systems was indirect at best. iDEX created a formal MoD contracting path for smaller private-defence firms. That is why Data Patterns, MTAR Technologies, Astra Microwave, and Paras Defence appear in the same cohort sentence as HAL and BEL. They are not the same kind of company. They are plugged into the same demand architecture.

₹6 lakh crore. A hard floor on domestic sourcing. A scheme routing demand into both large PSUs and the private-defence tier. That is the architecture behind the re-rating, not a geopolitical event with no budget line.

The Lag: Why the Rally Arrived Late

Defence-PSU price action historically trails disclosed order book by two to three quarters. This is structural, not an inefficiency to be arbitraged away.

The mechanics are simple. Order intake is disclosed when contracts are signed. Revenue recognition happens when deliveries occur. For large platforms (fighters, submarines, frigates), the gap between contract signing and delivery is measured in years. A market that prices on near-term earnings will, by construction, lag a market that prices on order-book coverage ratios and conversion visibility. Making that analytical shift requires accepting that a ₹1.25 lakh crore order book is not speculative inventory; it is a multi-year revenue commitment from a counterparty with a Constitutional obligation to pay.

Most of the market did not make that shift until earnings prints began confirming what the order books had telegraphed. By then, the price had already moved. Investors who treated the FY24 order-book disclosures as the signal, rather than the subsequent earnings beat, captured the bulk of the 320 percent.

The current position of the cohort: the lag has largely been consumed. The price now reflects approximately what the disclosed order books support. The next re-rating leg requires fresh order intake at scale, improving conversion efficiency, or the export optionality closing into actual revenue. Absent those catalysts, the cohort trades on execution quality.

Exports: Call Options, Not Base Case

Four export programmes sit in disclosed fact. BrahMos missiles to the Philippines. Pinaka multi-barrel rocket systems to Armenia. ATAGS artillery gun systems to Bhutan. Tejas in active discussions with Latin American buyers.

None of these belongs in a base-case revenue model. They are call options, each carrying real programme risk: timeline slippage, domestic capacity constraints, counterparty stability. The correct analytical treatment is that these programmes, if they close, add a revenue layer not priced into the domestic-procurement base case. They do not break the thesis if they stall.

What the export pipeline does prove is that Indian defence product has reached a quality threshold at which it competes internationally, not merely complies domestically. That transition matters for long-run valuation framing. An export-capable industrial base supports a different multiple than a captive-market one. But at this stage, the option has not exercised. Model it as optionality, not scheduled revenue; the desk that front-runs export income on named programmes before contracts are signed is the desk that gets caught revising numbers.

Conversion Efficiency: Why the Cohort Dispersed Inside the Rally

All fourteen names are positive. They are not equally positive, and the dispersion is instructive without being a stock-picking exercise.

The efficient converters, BEL and HAL most visibly, re-rated cleanest. BEL's electronics portfolio (radars, electronic warfare suites, sonar systems, secure communications) converts order intake into revenue with relative predictability. These are serial production programmes, not bespoke one-off platforms. Once a factory line is qualified to a specification, units ship on a known schedule. HAL's mix includes long-cycle platform deliveries (LCA Tejas variants, ALH Dhruv, LCH Prachand) alongside a substantial MRO component covering Sukhoi overhaul and helicopter maintenance. The MRO segment provides more consistent near-term revenue recognition than platform delivery alone, giving the order book a dual-speed conversion structure.

BEML's order mix is more heterogeneous. Defence mobility sits alongside metro coach manufacturing and mining equipment. The civilian segments carry different cycle characteristics and margin profiles. A market trying to apply a defence-PSU re-rating to a three-business conglomerate will calibrate that multiple conservatively and inconsistently. The procurement tailwind is real for BEML; the translation into a clean valuation re-rating is noisier. Underperformance relative to BEL and HAL within the same cohort is not a broken thesis. It is a business-mix discount, applied predictably.

This dispersion is the granular read on the cycle. Not all order-book growth converts to earnings at the same pace. Conversion efficiency and revenue visibility determine how cleanly a procurement tailwind reaches the bottom line, and the market has begun pricing that distinction with some precision.

The Actual Risk Stack

The risks to this thesis are programme-level. They are not trivial, and they are not the border.

A budget recast is the highest-probability risk. Defence capital expenditure is visible, large, and one of the few discretionary line items a Finance Ministry under fiscal pressure can defer without immediately disrupting public services. If India's fiscal arithmetic tightens materially, the capital procurement line is not immune. Any meaningful reduction in the ₹6 lakh crore plan directly reduces order intake for the cohort, and the re-rating works in both directions.

AMCA delay is the most-watched programme-level risk for HAL specifically. The Advanced Medium Combat Aircraft is the platform that extends HAL's fighter manufacturing franchise beyond the current Tejas production run. Significant slippage in AMCA timelines does not impair the existing ₹1.25 lakh crore order book, but it compresses the long-dated optionality that sophisticated investors have begun incorporating into their framing.

Private-JV margin compression is an emerging concern for the smaller private-defence names in the cohort. As iDEX programmes scale and the qualified domestic supplier base grows, competition for sub-system and technology contracts intensifies. Margins that were wide in a low-competition environment will compress as more entrants qualify and price aggressively. PSU names carry some insulation through captive programme access. The private-defence tier does not have that floor.

Programme cancellation risk is real but low-probability for flagship platforms. Historically, cancellations are followed by retendering rather than abandonment, which makes a cancellation a delay and a re-rating event rather than a permanent order-book impairment. Track it; weight it appropriately.

The Geopolitics Frame Keeps Getting the Cycle Wrong

Border tensions rise. Defence stocks rally. The narrative writes itself in about forty seconds. The analysis is wrong, and it is wrong at turning points when the cost is highest.

The geopolitics frame fails because it cannot fund procurement. An incident creates urgency; urgency without a budget line produces press releases and accelerated timelines on paper. India's defence order books are funded through annual capital allocation, multi-year procurement plans, and disclosed contracts. The chain from a geopolitical headline to a signed contract is long enough that by the time the order book is large, the headline that supposedly drove it is usually old.

The frame fails at cycle turns as well. When tensions de-escalate, the geopolitics trader looks for the exit. The order book does not de-escalate. HAL's ₹1.25 lakh crore does not shrink because a diplomatic summit was held. The analyst who correctly held through a de-escalation headline understood that the procurement cycle runs on its own calendar, separate from the diplomatic one. That distinction, internalized early, is worth several hundred basis points annually in this sector.

Finally, the frame is unfalsifiable in a way that should concern rigorous investors. When stocks rally, geopolitics receives the credit. When the cycle turns and capex is unfunded, geopolitics is quietly dropped as the explanation, replaced by something about fiscal headwinds or execution risk. A thesis that only claims causality in one direction is not a thesis. Cycle the framing back to the order book every time the news pulls it toward the border.