2% of Capex Halves the Exposure: Turning Resilience Spend Into a Priced Insurance Credit

A joint assessment by Zurich Kotak General Insurance and Zurich Resilience Solutions, published in June 2026, evaluated 871 planned renewable energy projects in India, representing roughly 90% of the country's active renewable development pipeline. Nearly 90% of those projects were found to face high or critical exposure to climate-related hazards.

Two things about that sample matter for anyone buying or placing cover on a wind, solar or hybrid project right now.

The first is coverage. This was not a sample of distressed sites or a case-study set assembled to make a point. At roughly nine in ten of the active pipeline, it is close to a census. When an underwriter sits down to price a new solar park in Rajasthan or a wind cluster in the Western Ghats, the base rate they are working from is now informed by a screening of almost every comparable project in the country, not by a handful of loss files.

The second is the distribution. If nearly 90% of screened projects carry high or critical hazard exposure, high exposure is the market default. A developer who walks into a renewal expecting a discount for being climate-aware is arguing against a benchmark in which almost everyone is exposed. The commercial question shifts from whether your project has hazard exposure to whether you can evidence that you have engineered it down relative to the peer set.

That reframing is the whole point of this post. Climate exposure on its own is not a negotiating position. Documented, quantified reduction of that exposure is.

The headline financial figure from the assessment is that about $55 billion of renewable energy assets could be exposed to climate risk by 2030. The more interesting figure is the counterfactual: allocating around 2% of project capital expenditure to climate adaptation and resilience is estimated to cut the value of assets at risk to roughly $27 billion.

Read that as a ratio before reading it as a target. Two per cent of capex against a halving of exposed asset value is an unusually steep return curve, and it tells you something about where the exposure sits. Losses concentrated in cheap-to-fix failure modes, drainage that cannot clear a cloudburst, module mounting that fails at wind speeds below the site's revised design basis, inverter and switchgear rooms placed at grade in a flood-prone plot, respond disproportionately to modest spend. Losses that require moving the project do not.

The practical failure at this point is treating the 2% as an ESG line item. Capex booked under a sustainability heading with no engineering register behind it produces nothing an underwriter can rate. The same rupees, booked against named hazard mitigations with design parameters and as-built evidence, are the raw material for every commercial argument in the rest of this post.

For the underlying method of putting numbers on physical hazard at asset level, see our walkthrough of physical climate risk quantification for Indian assets.

Underwriters are pricing this year, not in 2030, and the 2026 monsoon gave them fresh loss experience in exactly the terrain where a large share of India's hydro and hill-state renewable capacity sits.

Himachal Pradesh's 2026 monsoon losses crossed Rs 1,202.64 crore. The Public Works Department accounted for Rs 914.76 crore of that and Jal Shakti Vibhag for Rs 258.61 crore, with 259 to 261 deaths reported during the season.

Note where the damage landed. Roughly three-quarters of the recorded loss sat with the roads and works department, and another fifth with water infrastructure. That is an access-and-utilities loss profile, not a generation-asset loss profile. For a project owner it is the difference between two very different claim conversations:

• Damage to the insured asset itself, which the material damage section responds to.
• Loss of access, evacuation capacity or water supply because public infrastructure around the asset failed, where recovery depends entirely on how the policy defines the trigger.

The second category is where renewable projects in hill and coastal terrain get hurt without a corresponding recovery. A wind farm whose turbines are undamaged but whose only approach road is gone still stops earning. A solar plant intact behind a substation that flooded still exports nothing. Standard business interruption wording keyed to physical damage at the insured premises does not reach either situation unless a denial-of-access or contingent extension has been bought and sized.

Resilience capex that addresses site access, internal drainage and utility redundancy therefore does double work. It reduces the loss, and it gives you the factual basis to ask for the extension that would have paid if the loss happened anyway.

Indian property and engineering pricing is negotiated, which means a credit has to be argued rather than claimed. The argument that works follows the same structure a risk engineer uses.

Match each measure to the hazard the rate is loaded for

Start from the underwriter's own view of your site. Ask the broker for the hazard drivers behind the quoted rate: flood zone classification, wind zone, seismic zone, aggregation with other insured assets in the same postcode. Then map each resilience item to a specific driver. A rate loaded for flood does not come down because you upgraded lightning protection.

Then express the mitigation as a change in modelled loss rather than as a spend. The sentence that moves a rate reads: "finished floor level for the inverter and control rooms is 1.2 metres above the recorded 2023 high-water mark, site drainage is designed to clear the 100-year one-hour intensity, and the modelled 100-year loss falls from X to Y as a result". That is a revised probable maximum loss, which is the currency an underwriting file is built to handle. A rupee figure on its own tells the underwriter what you spent and nothing about what it bought.

Ask to be paid in the right currency

A premium credit is one of several ways to be paid for resilience, and often not the best one. The alternatives, in rough order of how often they are worth more than an equivalent rate cut:

1. A lower or scrapped percentage deductible on the named hazard you have engineered against.
2. Removal or lifting of a hazard sub-limit that sits below your modelled loss.
3. A wider delay in start-up structure during construction, covered in the next section.
4. Reinstatement on improved-standard terms, so that a post-loss rebuild locks in the resilience rather than restoring the original vulnerability.
5. The headline rate reduction.

A 5% rate credit on a project whose flood sub-limit is capped at a fraction of the modelled event loss is a poor trade. Fix the limit structure first.

For projects still in the pipeline, and the screened 871 are planned projects, the cover that most deserves the resilience argument is delay in start-up, written alongside erection all risks or contractors all risks.

DSU pays the debt service and lost gross profit caused by a delay to commercial operation date, where that delay follows an indemnifiable physical loss. Three terms decide whether it is useful:

• The waiting period. Days of delay absorbed before the policy responds. A monsoon-season interruption that runs to three weeks is worth very little under a 30-day waiting period.
• The indemnity period. The maximum period of delay compensated. Grid connection queues and long-lead equipment replacement can push a real delay well past a short indemnity period.
• The hazard sub-limit. Flood, storm, tempest and inundation are frequently sub-limited within DSU at a level that has nothing to do with the site's modelled exposure.

Resilience evidence bites harder on these three terms than on the property rate. Each of them is a judgement the underwriter makes about how long your site will be out of action, and that is precisely what drainage, spares and access engineering change. Site drainage that clears in hours rather than days, a stockholding policy for long-lead items with a documented replacement pathway, and an access road engineered for the design event are direct arguments for a shorter waiting period and a longer indemnity period.

Bring the construction programme to the DSU negotiation alongside the risk report. The underwriter is pricing a delay to a date. Showing which activities sit on the critical path during the monsoon window, and what resilience spend protects them, converts an abstract hazard discussion into a schedule discussion the underwriter can rate.

Our detailed treatment of how these structures work on Indian renewable and industrial projects sits in DSU and ALOP insurance for renewable and industrial projects, and the sector's broader placement difficulties are covered in renewable energy insurance challenges in India.

Project finance and insurance are reading the same risk, and increasingly from the same documents. A lender's technical adviser reviews the insurance programme as part of the conditions precedent, and the debt sizing depends on the reliability of the cash flow that services it.

Resilience evidence reaches the credit decision through three channels.

The first is insurance adequacy. Lenders set minimum insurance requirements, and a project whose flood sub-limit falls short of the modelled event loss creates an uninsured gap the technical adviser has to flag. Closing that gap through better limit structure, itself obtained through resilience evidence, removes a condition rather than adding a cost.

The second is availability assumptions. The base-case generation forecast embeds assumed downtime. Documented mitigation of the outage drivers, siltation, access loss, substation flooding, supports a tighter downtime assumption, which flows into the debt service coverage ratio.

The third is the resilience of the DSU structure itself during construction, because the lender's exposure through to commercial operation date is protected by that cover and nothing else.

None of this is automatic. A developer who wants the pricing benefit has to put the resilience file in front of the lender's adviser in the same form the insurer's risk engineer accepted it, and ask explicitly for the sizing assumption to move. Absent that request, resilience spend is treated as ordinary capex and quietly worsens the return. The financing side of adaptation spend is covered further in climate adaptation and risk financing for Indian corporates.

Ask a risk engineer what would change their view of a site and the answer is consistently specific. The following pack is what survives a technical review.

1. A site-level hazard assessment naming each peril, its return period basis and the data source used. Screening scores from a hazard platform are a starting point. A study that states the design event and its recurrence interval is what gets rated.
2. A mitigation register listing each measure, the hazard it addresses, the design parameter it meets, the capex booked against it, and the completion date. One row per measure.
3. As-built confirmation rather than design intent: elevation certificates for critical equipment, drainage commissioning results, anchoring and foundation test records, cable route and switchgear layout drawings.
4. The revised loss estimate, ideally a modelled probable maximum loss before and after mitigation, with the methodology stated. This is the number the credit is argued against.
5. An operational plan for the design event, covering monsoon-season standing instructions, shutdown and isolation protocol, spares held on site, and contracted access to heavy equipment for post-event recovery.
6. Loss history with root causes, including near misses that produced no claim. Suppressing a near miss buys nothing, because the underwriter's benchmark already assumes events at your hazard level.

The order in which a developer does this determines whether the 2% ever converts into anything.

Twelve to nine months before financial close or renewal. Commission the site hazard study. Get the underwriter's current view of your hazard drivers through the broker before the study scope is fixed, so the study answers the questions the pricing actually turns on.

Nine to six months. Fix the mitigation scope and book it as identified engineering scope with its own cost codes. Capex that cannot be traced to a hazard measure at renewal time cannot be argued at renewal time.

Six to three months. Run a pre-renewal risk engineering visit with the incumbent insurer or a lead market, on the mitigation package as designed. Get their written view of what would earn a credit and in what form. This is where the negotiation is actually won, and it is the step most often skipped.

At renewal or close. Present the pack, and ask for the concessions in the priority order set out earlier: deductible and sub-limit structure on the engineered hazard first, DSU waiting and indemnity periods second, rate last.

After the first event. Whether or not it produces a claim, document how the mitigation performed. A measure that demonstrably worked during a real monsoon event is worth more at the following renewal than any pre-loss study, and it is the evidence that carries into the next project in the pipeline.

Against a pipeline where nearly 90% of projects sit at high or critical hazard exposure and $55 billion of assets is projected to be at risk by 2030, the developers who get paid for resilience will be the ones who can show, on paper and per site, exactly what they built and what it changed.