The Government Will Subsidise the Launch Bill. It Will Not Subsidise the Liability.

On 5 August 2026, following Skyroot's Vikram-1 becoming India's first privately built vehicle to reach orbit, the Centre opened an incentive scheme for private space companies. The Times of India reported it as covering 30 to 100 per cent of costs across four heads: launch, data, technology and the use of ISRO facilities. India Today described the same scheme as offering up to 100 per cent incentives to Indian space companies.

For a founder rebuilding a mission cost model, the effect is straightforward on the capital side. A line item that was priced at full commercial cost in the Series A model now carries a government share. Range-and-facility charges for using an ISRO launch complex, data purchase costs, and technology transfer or development costs all move.

The scheme is aimed outward, not only downward. IN-SPACe chairman Pawan K Goenka told Moneycontrol on 7 August 2026 that the launch subsidies are meant to win global customers rather than merely cut costs for domestic operators. That framing matters for insurance, because a foreign customer's satellite riding an Indian vehicle brings its own contractual insurance requirements into the programme, and those requirements do not scale down when the launch price does.

The money is arriving from private sources at the same time. NDTV reported on 3 August 2026 that Indian space startups had raised USD 871 million following Skyroot's success. Combined with the subsidy, the sector's cost of building and flying has moved sharply. The cost of being liable has not.

Read the four heads again: launch, data, technology, ISRO facility use. Every one of them is a cost of producing a mission. None of them is a cost of what the mission might do to someone else.

Third-party liability and in-orbit cover sit outside all four. They are not procurement costs for a service the government supplies; they are the financial security the state requires before it will let the mission fly at all. That obligation attaches to the operator through the IN-SPACe authorisation, and it attaches regardless of who paid for the rocket.

The reason sits in the treaty chain rather than in the incentive scheme. Under Article VI of the Outer Space Treaty 1967, India bears international responsibility for national space activities carried out by non-governmental entities, and must authorise and continuously supervise them. Article VII makes the launching state internationally liable for damage caused by its space objects. The Liability Convention 1972 then makes that liability absolute for damage caused on the surface of the Earth or to aircraft in flight. A claim under that chain is presented against the Republic of India, and India then looks to the operator through the domestic authorisation and indemnity. We covered that chain in detail in our launch vehicle third-party liability guide.

The subsidy sits on one side of that structure. The liability sits on the other, and the two do not talk to each other.

IN-SPACe had issued 130 authorisations by 22 July 2026, against a private sector that ANI reported as having crossed 450 startups. That is a regulator processing real volume, which means the authorisation conditions are becoming standardised rather than negotiated case by case.

For a founder, the practical sequence is what matters:

1. Application. The operator files mission details: vehicle, trajectory, payload, orbit, re-entry profile, launch site.
2. Risk assessment. The regulator forms a view of maximum probable loss to third parties from that specific mission profile. Overflight of populated areas, launch azimuth, stage impact zones and debris footprint drive this.
3. Insurance condition. The authorisation carries a financial security requirement sized to that assessment, usually satisfied by a third-party liability policy naming the government among the insured parties.
4. Evidence before flight. The certificate of insurance and the underlying policy wording go to the regulator before launch clearance, not after.
5. Continuing supervision. For an orbital mission the obligation does not end at separation. In-orbit third-party exposure continues through operations and, where applicable, disposal.

The point founders miss is at step 2. The limit is an output of the regulator's loss model. The operator is a price taker on limit and a price maker only on how efficiently that limit is bought.

Maximum probable loss for a launch is driven by physics and geography. A vehicle's mass, propellant load, trajectory and the population under the flight corridor determine what a failure could destroy. None of those change because the Centre paid part of the range fee.

There is a second-order effect that runs the wrong way for the operator. A subsidy that raises flight cadence raises aggregate exposure. Ten subsidised launches in a year generate ten separate liability attachments, and an annual programme placed on an aggregate limit will exhaust faster than a single-launch policy would suggest. If the cost model assumes the same insurance spend per launch as cadence triples, the programme is under-bought in the aggregate even while it looks correctly sized per mission.

The third-party liability layer is priced as a percentage of limit, so the operator's lever is the structure of the placement rather than the size of the requirement. A single reinsurance-backed tower covering a full year's manifest, with defined reinstatement, is generally more efficient than repeated single-shot placements, and it gives the regulator a cleaner evidence trail across launches.

Other launching states cap the operator and stand behind the tail. The United States, France, the United Kingdom and Japan each require the operator to insure up to a set limit and then have the state absorb claims above it, at least in part. India has not enacted that structure. The draft Space Activities framework has been expected to introduce a comparable cap, but until it is law the position above the insured layer is set by the authorisation rather than by statute.

What the authorisation does contain runs in the opposite direction. As a condition of authorisation the operator gives an indemnity in favour of the government, agreeing to reimburse the state for third-party claims arising from the operator's activity. The third-party liability policy exists to fund that indemnity, so that the promise to the state is backed rather than merely made.

Three things follow for a founder.

First, the insured layer is a floor on exposure, not a ceiling. Without a statutory cap, the operator's indemnity to the state can in principle run above the limit the authorisation required it to insure. Whether the specific authorisation caps that recourse is a question answered by the document rather than by general principle.

Second, whatever protection sits above the layer is conditional. Arrangements of this kind ordinarily fall away where the operator has breached the authorisation, misrepresented mission parameters, or acted with gross negligence or wilful misconduct. An operator that flies outside its approved trajectory has not just committed a regulatory breach; it has potentially stepped outside whatever backstop it thought it had.

Third, the layer that matters day to day is the insured one. Claims inside it come out of the policy and, where the policy responds imperfectly, out of the company.

Ask for the indemnity wording as a document, in writing, at authorisation stage. Read it against the exclusions in the third-party liability policy. Any gap between where the policy stops responding and where the operator's obligation to the state continues is uninsured exposure, and it is invisible unless the two documents are read side by side.

A launch programme is four distinct covers with four different triggers. Buy them in this order, and price each against what the subsidy did to the underlying value.

1. Third-party liability. First, because it is a condition of authorisation and nothing else can proceed without it. The limit is set by the regulator. Bind this before committing to a launch date.
2. Pre-launch cover. Physical damage to the vehicle and payload during manufacture, integration, transport to the range and pre-launch processing. This is the phase where value is highest relative to attention paid to it, and it is closest to conventional property and transit underwriting.
3. Launch cover. Attaches at intentional ignition and runs to separation or a defined post-separation checkout window. This is the most expensive layer per rupee of sum insured, because it prices the single riskiest minutes of the mission.
4. In-orbit cover. First-year in-orbit is the market standard, covering total and partial loss of satellite function. It is optional in commercial terms but frequently required by a paying customer or a lender.

Third-party liability attaches to the mission whether or not the operator buys anything else. The other three attach to assets, and assets are exactly what the subsidy repriced.

For an operator that has just moved to a subsidised cost base, the temptation is to redirect the freed capital into cadence. The disciplined alternative is to fix the liability layer for the full manifest first, then spend what is left on flying more often.

This is the question that determines whether a subsidised programme is correctly insured or quietly over-insured on one line and under-insured on another.

Sum insured on a payload is normally set at replacement or rebuild value: the cost to build and fly the satellite again. If the government has covered a share of the launch cost and a share of the technology cost, then the cost of replacing that mission is genuinely lower for the operator, but only to the extent the subsidy would be available again for the replacement flight.

That conditional is the whole issue. If the incentive is a one-time award tied to a specific mission, the replacement flight will be paid at full commercial cost, and insuring at the subsidised figure leaves a shortfall exactly when it is needed. If the scheme is a standing entitlement the operator can draw on again, the subsidised figure is the honest number.

The test is simple: insure the cost of the replacement mission the operator would actually have to fund, which is not always the cost of the mission that was flown.

Two related traps:

• Customer-owned payloads. Where a foreign customer's satellite is aboard, its insured value has nothing to do with what India subsidised. It is the customer's build cost, and the launch services agreement will usually settle it.
• Overstated value. Declaring a value the operator cannot substantiate does not buy extra recovery. Insurers pay indemnity, and a claim settles on demonstrable loss. Inflated declarations buy premium, not protection.

Cross-waivers of liability in the launch services agreement handle the first-party side between operator and customer, so each party carries its own damage. They do not touch the third-party layer, which exists for people who never signed anything.

The sector's economics changed in the space of a fortnight in August 2026: a subsidy scheme on 5 August, USD 871 million of private funding reported on 3 August, and Skyroot signalling on 2 August that it plans to build its own launch pad after the Vikram-1 flight. An operator building its own pad shifts risk it previously borrowed from ISRO onto its own balance sheet, including ground property, construction-phase exposure and public liability around the site.

Concrete steps:

• Model the liability layer as a fixed cost, not a variable one. It scales with cadence and mission profile, not with what the launch cost.
• Re-run the cost model with and without the subsidy on replacement value. Confirm with IN-SPACe whether the incentive would be available for a re-flight before lowering any sum insured.
• Place the liability tower across the manifest, not per launch. Confirm reinstatement terms so a first-quarter loss does not strand a fourth-quarter launch.
• Read the indemnity wording against the policy exclusions. Close the gap or price it consciously.
• Treat a new launch pad as a separate risk. Construction, ground property and public liability around a private range are conventional Indian commercial covers, and they are not part of the launch programme.

The general framing of the space startup insurance programme, including pricing anchors by phase, is in our space launch startup insurance guide and the smallsat and in-orbit cover post.