Insured Asset Exposure for Solar Portfolios: Why Maximum Probable Loss Models Are Incomplete

Insured asset exposure valuation for solar portfolios has a structural problem that the industry is only beginning to confront directly. Maximum probable loss (MPL) models tell you what the worst realistic loss event looks like under a defined set of peril scenarios. They model the hazard. They estimate the damage. They produce a loss figure that drives reinsurance purchasing, capacity deployment, and portfolio risk decisions.

What most MPL models do not adequately capture is what the assets are actually worth at the time of that loss, what they would cost to replace under current market conditions, and what recoverable value remains in damaged or destroyed components. That gap, between modeled loss and actual financial exposure, is widening as the market conditions underlying solar asset valuation move faster than the static inputs most models rely on.

Severe convective storms drove $60 billion in insured losses in 2025, the third-costliest SCS year on record for insurers. Hail accounted for just 6% of loss incidents but 73% of financial losses at solar farms. (PV Magazine, May 2026 / Renew Risk, May 2026) When the primary peril driving financial losses in the US solar market operates at that ratio of severity to frequency, the accuracy of the asset valuation inputs underlying the loss model matters enormously. A modeled loss built on stale replacement cost assumptions is not a loss estimate. It is a historical projection dressed up as a current risk assessment.

This blog examines what insured asset exposure analysis for solar portfolios typically includes, where the gaps are, why those gaps are growing, and what complete insured asset exposure valuation requires.

What Maximum Probable Loss Modeling Is Built to Do

Probable maximum loss is a term used across the insurance industry to describe the largest loss that could result from a single event, assuming the normal functioning of passive protective features and the proper functioning of most active suppression systems. PML represents the worst-case scenario for an insurer, calculating the maximum loss expected on a policy at any given time. Underwriting decisions and the amount of reinsurance ceded on a risk can be predicated on the PML valuation. (Arlington/Roe, January 2025)

For solar portfolios, MPL modeling typically involves three interconnected components.
The hazard module estimates the probability and severity of specific peril events affecting the project locations. For US solar, the primary perils are severe convective storms including hail, wildfire, flood, and hurricane. Catastrophe models exist for natural catastrophes including hurricanes, earthquakes, floods, and convective storms. The hazard module generates various event scenarios, determines the path associated with each scenario, and assesses the local impact as the event progresses in both time and space. (NAIC Catastrophe Modeling Primer, March 2025)

The vulnerability module estimates the damage ratio, the proportion of asset value that would be destroyed or impaired, for a given level of hazard intensity at a specific asset type. For solar modules, this translates to a relationship between hail stone size, impact energy, and the proportion of modules that would require replacement.

The financial module converts the damage ratio applied to the declared asset value into a modeled loss figure. This is where the insured asset exposure valuation inputs enter the model. The hazard and vulnerability components of modern solar cat models are becoming increasingly sophisticated. For the solar market, traditional models fail to capture how this complex peril interacts with varied engineering systems. The development of next-generation US solar catastrophe models is driven by the need to incorporate detailed insights into asset value distribution, total insured value, and business interruption calculations. (Renew Risk via Solar Power World, May 2026)

But the financial module, the component that translates physical damage into financial loss, is only as accurate as the asset valuation data it operates on. And that data, in most solar portfolio MPL models, is where the significant gap lies.

The Four Ways MPL Models Understate Solar Portfolio Exposure

1. Replacement cost inputs are static while the market is dynamic

• The financial module of a solar portfolio MPL model applies a declared asset value to the damage ratio to produce a loss estimate. That declared value is typically sourced from the portfolio's Statements of Values, which are prepared at policy inception and updated infrequently.
• US solar module prices rose 14% between January and November 2025. Antidumping and countervailing duties of up to 3,521% on imports from Cambodia, Vietnam, Malaysia, and Thailand, countries that represented 77% of total US module imports, have materially changed the replacement cost economics for the majority of modules currently in the field. (US Department of Commerce, April 2025) A loss model built on pre-tariff module pricing systematically understates the financial loss from a hail event that destroys a significant share of the module fleet.
• Hailstorms alone caused an estimated $300 to $400 million in losses to solar installations in 2022, a peril category that most cat models treat as secondary. Secondary perils such as hail receive less modeling attention than headline perils. (Repath.earth, April 2026) When hail is both the dominant financial loss driver and the most poorly modeled peril for solar assets, and when the replacement cost inputs underlying those models have not been updated to reflect current tariff conditions, the compound error in modeled loss estimates is significant.

2. Technology-specific replacement cost variation is not captured

• Solar portfolio risk assessment models typically apply a single per-watt or per-MW replacement cost assumption across the entire portfolio. This assumption smooths over component-level variation that is materially significant for accurate loss estimation.
• Mono PERC modules hit $0.33 per watt in February 2026. TOPCon modules were at $0.285 per watt in the same period, a pricing inversion driven by FEOC compliance demand for domestically produced cells. (PV Tech, April 2026) A portfolio that contains both technology types across projects at different vintages carries different replacement cost exposure than a system-level per-watt estimate reflects. The difference between applying a blended rate and applying technology-specific rates compounds across a large portfolio into a material gap in modeled versus actual exposure.
• For BESS components, the variation is more pronounced. NMC system replacement cost is sensitive to nickel, manganese, and cobalt commodity movements. LFP replacement cost is sensitive to lithium and iron phosphate pricing. 75% of BESS sites show early HVAC-related thermal risk signals. (PowerUp via kWh Analytics, May 2026) A BESS system that has experienced HVAC-related degradation carries different replacement cost exposure than a new system specification. Models that treat BESS replacement cost as equivalent to original procurement cost are not modeling actual exposure.

3. Residual and salvage value offsets are absent from most loss models

• This is the most significant and least discussed gap in solar portfolio MPL modeling.
  MPL models estimate gross loss. They estimate what the replacement cost of the destroyed or impaired assets would be. What most models do not estimate is the net loss after accounting for the salvage value and residual recovery value of the damaged assets. When hail damages a solar panel, it does not reduce the panel's copper, silver, and aluminum content to zero. It does not eliminate the glass, the aluminum frame, or the silicon cells as recyclable materials.
• Depending on the severity of the damage, the panel may retain reuse value, recycling value, or scrap value that partially offsets the gross replacement cost.
  Copper hit a record $14,527 per metric ton in January 2026. Silver reached an all-time high above $121 per ounce in the same period after gaining more than 130% during 2025. (Crux Investor, March 2026 / Investing News Network, April 2026) The metallic content of a utility-scale solar asset is worth materially more today than any pre-2025 benchmark would reflect.
• A gross loss model that does not account for the recoverable material value in damaged assets overstates the true net financial exposure. For a portfolio with $500 million in declared module value and a significant hail event affecting 40% of the fleet, the difference between gross replacement cost and net replacement cost after salvage recovery can represent tens of millions of dollars in misstated exposure.
  From the other direction, an underwriter or lender whose MPL model does not capture residual value is also likely overpaying for reinsurance coverage sized against a gross loss figure that overstates actual net exposure.

4. Business interruption exposure is modeled against revenue assumptions that predate current grid conditions

• Business interruption and delay in start-up exposures are expected to increase in 2026 as aging transmission equipment and increased congestion create longer and more frequent outages, and data centers amplify local grid stress. In solar, grid-originating outages were one of the biggest BI cost drivers of 2024 and 2025. (NARDAC, December 2025)
• MPL models typically estimate BI exposure as a function of replacement timeline and declared revenue. Both inputs are dynamic. Transformer lead times extended to 18 months or longer during 2023 and 2024, and while they have partially recovered, the structural demand from grid modernization and data center buildout continues to create supply pressure. Inverter availability varies by manufacturer. Module delivery timelines vary with tariff conditions and supply chain dynamics.
• A BI estimate built on standard replacement timelines without accounting for current supply chain conditions systematically understates the duration of income loss following a significant loss event.

What Solar Catastrophe Modeling Limitations Mean for Underwriters and Lenders

The practical consequences of these gaps flow differently for insurers and lenders but converge on the same underlying data problem.

For insurers:

75% of renewable energy tax insurance underwriters will not cover valuation step-ups above 25%. (kWh Analytics via CAC, May 2026) This constraint exists precisely because of the valuation accuracy problem. When the declared asset values underlying a portfolio's MPL model are materially below actual current replacement cost, the modeled loss is understated, the premium is calculated against an understated TIV, and the reinsurance is structured against an understated exposure. A large loss event then reveals the full gap simultaneously.

99.27% of photovoltaic plants have a 10% annual chance of experiencing hail over two inches in diameter in nearby locations. (Central Michigan University via Insurance Business America) The probability of a hail event sufficient to trigger significant module replacement at virtually any US solar site within any given decade is not a tail risk. It is a near-certainty over the life of the project. An MPL model built on stale replacement cost inputs is not modeling an edge case incorrectly. It is modeling the primary loss scenario incorrectly.

For lenders:

Collateral assessments built on MPL models share the same valuation data dependency. If the replacement cost inputs are stale and the residual value offsets are absent, the collateral assessment overstates gross exposure and understates net recovery. A lender sizing a debt service reserve or evaluating a distress scenario against an MPL model that does not account for current module pricing, current commodity values, and transaction-backed residual value benchmarks is working from an incomplete picture of actual collateral exposure.

The SEIA hosted a webinar in April 2026 specifically covering the latest tools and methodologies for assessing solar project residual value, the impact of decommissioning requirements, and emerging trends in the solar secondary market. (SEIA, April 2026) The fact that SEIA dedicated a formal event to residual value methodology signals that the industry has recognized this as a mainstream financial variable, not a decommissioning planning concern.

What Complete Insured Asset Exposure Valuation Requires

Closing the gap in solar portfolio MPL models requires addressing the four weaknesses identified above with specific data inputs that are not currently standard in most portfolio-level insurance underwriting data frameworks.

Current replacement cost by component type

Replacement cost inputs must reflect current tariff-inclusive market pricing by technology type, not project cost at financial close. That means module pricing by technology generation, inverter pricing by manufacturer and capacity range, BESS pricing by chemistry, and racking and balance of plant pricing by system type at current labor rates in the relevant geographies.

US solar is on track to add a record-breaking 86 GW of new utility-scale capacity in 2026. (Renew Risk via Insurance Edge, May 2026) At that scale of deployment, even a modest per-watt error in replacement cost assumptions across the portfolio represents hundreds of millions of dollars in misstated modeled exposure at the portfolio level.

Transaction-backed salvage value benchmarks

Net exposure requires net inputs. The recoverable material value in damaged or destroyed assets, across resale, refurbishment, recycling, and scrap pathways, must be incorporated into the financial module of any MPL model that purports to estimate actual financial loss rather than gross replacement cost.

This requires transaction-backed benchmarks built from real observed outcomes rather than static percentage assumptions. Buckstop's residual value indexes are built on observed market data tied to actual recovery outcomes across all relevant pathways, incorporating manufacturer, age, wattage, condition, configuration, and geographic and regulatory factors that influence recovery value. Buckstop platform methodology When commodity prices reprice sharply, these benchmarks update to reflect actual market conditions rather than holding a historical assumption in place.

Supply chain-aware BI assumptions

Business interruption inputs must account for current supply chain conditions rather than standard industry lead times. That means manufacturer-specific inverter availability, current transformer lead times, tariff-related module procurement timelines, and current grid interconnection queue conditions in the relevant ISO or RTO.

Regular update cadence

84% of PV fire loss events are equipment-driven brushfires originating from within the plant itself, meaning the source of ignition is the solar equipment rather than external wildfire. (kWh Analytics, May 2026) When the primary fire risk is internal rather than perimeter-based, the condition and maintenance status of the specific equipment in the field matters for both the hazard assessment and the replacement cost estimate. Static inputs updated at policy inception cannot track equipment condition changes that affect both loss probability and loss severity.

Annual updates tied to commodity price indices, module price benchmarks by technology type, and secondary market transaction data are the minimum appropriate cadence for MPL model inputs that are connected to live debt facilities or active reinsurance structures.

The Residual Value Gap in Catastrophe Models

The most consequential and least discussed gap in renewable energy asset valuation gaps within cat models is not on the hazard side. The models are becoming increasingly sophisticated in their ability to estimate where hail will fall, how large it will be, and what proportion of modules will sustain damage at a given impact energy.

The gap is on the financial side. Specifically, the failure to account for what damaged assets recover. When hail destroys 40% of the modules at a 200 MW utility-scale site, the gross replacement cost is the insurer's liability. But the damaged modules still contain copper wiring, aluminum frames, silver contacts, and silicon cells. Some proportion of those panels may be repairable or resalable in secondary markets. The remainder has recycling or scrap value that partially offsets the gross replacement cost.

The proportion of that offset that is real and recoverable depends on commodity prices, secondary market demand, recycling infrastructure accessibility, and the specific condition of the damaged assets. None of those variables are static. And none of them are standard inputs in current solar utility scale loss exposure models.

At current copper and silver prices, the metallic content offset on a large hail loss event is not a rounding error. It is a financially material component of the net loss calculation that the industry has been systematically ignoring because the data infrastructure to calculate it accurately did not exist.

That is changing. Transaction-backed residual value intelligence built from real secondary market data provides the inputs needed to close this gap at both the individual project level and the portfolio level.

The Bottom Line

Maximum probable loss modeling for solar portfolios is only as accurate as the valuation data it operates on. The hazard models are improving. The vulnerability functions are becoming more sophisticated. High-frequency, highly localized severe convective storms are now the primary loss driver in the US, accounting for 51% of natural catastrophe-related losses in 2025, totaling $46 billion. (Renew Risk via Solar Power World, May 2026) The physical risk side of the model is receiving significant investment and attention.

The financial inputs side is not keeping pace. Replacement cost assumptions that have not been updated for a market where module prices moved 14% in a single year and tariffs have fundamentally changed replacement economics. Salvage value offsets that are absent from most models despite copper and silver prices at or near all-time highs. BI assumptions that do not reflect current supply chain realities.

The gap between modeled loss and actual net financial exposure is not a rounding error for a portfolio with hundreds of millions of dollars in declared asset values. It is a material discrepancy that affects premium adequacy, reinsurance sizing, collateral assessment, and claims outcomes simultaneously.

Closing that gap requires treating insured asset exposure valuation as a live data problem, not a one-time calculation. Transaction-backed replacement cost benchmarks by component type. Real-time commodity-linked salvage value offsets. Supply chain-aware BI inputs. Annual updates that reflect what the market is actually doing.

That is the standard that accurate solar portfolio MPL modeling requires. And the gap between that standard and what most models currently incorporate is the risk nobody is talking about. Buckstop builds transaction-backed residual value intelligence for solar and BESS assets across resale, refurbishment, recycling, liquidation, and scrap pathways. Explore Buckstop Residual Value Intelligence.

Frequently Asked Questions

What is insured asset exposure valuation for solar portfolios?

Insured asset exposure valuation for solar portfolios is the process of determining the financial value at risk across a portfolio of solar assets for insurance and risk modeling purposes. It involves establishing accurate replacement cost values by component type, calculating residual and salvage value offsets for net exposure determination, estimating business interruption exposure based on current supply chain conditions, and incorporating geographic and technology-specific factors that affect loss severity. It is the foundational input into maximum probable loss modeling, Statement of Values preparation, and reinsurance purchasing decisions.

Why are maximum probable loss models for solar portfolios incomplete?

Maximum probable loss models for solar portfolios are incomplete primarily because their financial module inputs, the replacement cost values and salvage offsets that translate physical damage into financial loss, are typically based on static data from the time of policy inception rather than current market conditions. US module prices moved 14% in a single year. Copper rose 16% in a single month in early 2026. Tariff-driven changes to replacement economics have materially altered the cost of replacing the majority of modules currently in the field. MPL models that do not update their financial inputs with current market data are systematically overstating or understating net exposure.

What is the difference between gross loss and net loss in solar MPL modeling?

Gross loss is the replacement cost of assets destroyed or impaired in a loss event, without accounting for any recovery of value from the damaged assets. Net loss is gross loss minus the recoverable value from those damaged assets across resale, refurbishment, recycling, and scrap pathways. For solar assets at current copper and silver prices, the difference between gross and net loss on a significant hail event can be material. Most solar MPL models estimate gross loss. Very few incorporate transaction-backed salvage and residual value benchmarks to produce net loss estimates.

How does hail risk affect insured asset exposure for solar portfolios?

Hail is the primary financial loss driver for US solar portfolios, accounting for 73% of financial losses at solar farms despite representing only 6% of loss incidents. 99.27% of PV plants have a 10% annual chance of experiencing hail over two inches in diameter in nearby locations. Hail events that damage a significant proportion of the module fleet at a large project trigger both property replacement cost exposure and business interruption exposure from extended replacement timelines. The accuracy of the replacement cost inputs and salvage value offsets underlying the MPL model determines whether modeled hail exposure reflects actual financial exposure.

What solar insurance underwriting data gaps are most significant in 2026?

The most significant solar insurance underwriting data gaps in 2026 are: stale replacement cost inputs that do not reflect tariff-driven module price increases, absent or understated salvage value offsets that overstate net replacement cost exposure, BI assumptions that do not account for current supply chain lead times, and BESS-specific valuation inputs that treat battery chemistry as homogeneous. 75% of renewable energy tax insurance underwriters will not cover valuation step-ups above 25%, meaning these data gaps have become access constraints rather than pricing adjustments.

How does residual value affect reinsurance purchasing for solar portfolios?

Reinsurance structures for solar portfolios are typically sized against the gross MPL estimate. If that estimate is built on stale replacement cost inputs and does not account for salvage value offsets, the reinsurance purchase may be oversized relative to actual net exposure. Conversely, if replacement cost inputs are understated relative to current tariff-inclusive market pricing, the reinsurance may be undersized. Transaction-backed residual value intelligence that provides accurate replacement cost and salvage offset inputs allows reinsurance structures to be sized against actual net financial exposure rather than a modeled gross loss that may diverge significantly from the true risk.