When a company sells a product with a warranty, it simultaneously creates a liability — the obligation to repair or replace defective products within the warranty period. This liability must be recognised on the balance sheet as a provision under Ind AS 37 (Provisions, Contingent Liabilities and Contingent Assets), measured as the best estimate of the expenditure required to settle the obligation.

The challenge is that warranty claims are paid in the future, develop over time, and depend on variables — claim rates, repair costs, product volumes, return rates — that require actuarial modelling to quantify accurately.

Why Simple Provision Methods Fail

Many manufacturers provision warranty costs as a fixed percentage of revenue — often based on the prior year's actual claims as a percentage of sales. This approach systematically fails because:

It ignores claims already incurred but not yet reported (IBNR) — claims from products sold in prior periods that will emerge in future periods
It ignores the development pattern of claims — products fail at different rates at different points in the warranty period
It uses current-year claims as a proxy for prior-year liabilities, confusing incurred-period with reported-period
It produces provisions that are systematically too low (if claim rates are rising) or too high (if claim rates are falling)

The Actuarial Approach — Loss Development Method

Kapadia & Kochrekar applies the Loss Development Factor (LDF) methodology — the same technique used in non-life insurance reserving — to warranty liability estimation:

Step 1 — Build the development triangle

Claims data is organised by the period in which the product was sold (accident year equivalent) and the period in which the claim was reported. This produces a triangle showing how claims develop from the point of sale through the warranty period.

Step 2 — Compute Loss Development Factors

LDFs measure the ratio of cumulative claims at each development period to cumulative claims at the prior period. They capture the development pattern — for example, a product warranty where 30% of lifetime claims are reported in the first 6 months and 70% develop over the next 18 months.

Step 3 — Project ultimate losses

The LDFs are applied to current incurred claims to project the ultimate lifetime claims for each cohort of products sold. The difference between projected ultimate losses and claims already paid is the outstanding reserve — the warranty liability.

Step 4 — Cashflow projection

The reserve is distributed across future periods based on the development pattern, producing a cashflow schedule. This enables present-value discounting (where material) and the current/non-current split of the provision under Ind AS 1.

Applications Beyond Annual Provisioning

Extended warranty pricing: Actuarial claim data is the basis for pricing extended warranty products offered to customers — ensuring the premium is adequate to cover the incremental liability
Warranty insurance: If the company purchases warranty insurance, the actuarial claim projections determine whether the coverage limit and premium are appropriate
M&A due diligence: Acquiring a company with warranty obligations requires independent actuarial assessment of the true liability — book provisions are frequently inadequate
Product recall planning: Actuarial models can estimate the total financial exposure from a potential recall event across different recall scenarios

Industries Served

Industry	Typical Warranty Period	Key Valuation Challenge
Consumer electronics	1–2 years	Short tail, high volume, rapid technology change affecting repair costs
White goods / appliances	1–5 years	Longer tail, spare parts cost inflation, brand-differentiated failure rates
Fans and HVAC	2–5 years	Seasonal failure patterns, installation-related claims
Automotive / auto components	2–5 years or km-based	Kilometre-correlated claim rates, high severity, safety recall risk
Industrial equipment	1–3 years	Low frequency, high severity, long claim resolution period
Solar and clean energy	5–25 years	Very long tail, performance guarantees, limited historical data

Applicable Standards

Standard	Application
Ind AS 37	Warranty provision as a constructive obligation — best estimate basis
Ind AS 115	Extended warranties sold as separate performance obligations — revenue deferral
Ind AS 113	Fair value measurement for warranty provisions where market-based inputs are available
IAS 37 / IFRS 15	For IFRS-reporting entities and group consolidations

Frequently Asked Questions

No — these are different. A standard product warranty is a provision under Ind AS 37 — an obligation that arises from past sales. An extended warranty sold as an add-on for an additional fee is a separate performance obligation under Ind AS 115, creating a contract liability (deferred revenue) that is recognised as revenue as the warranty period passes. We handle both correctly.

Ideally 3–5 years of claims data organised by the period of sale (not the period of claim). In the absence of sufficient historical data, we use industry benchmarks and analogous product experience, clearly disclosed in the valuation report.

Yes — frequently significantly different. Most simple provisions understate the liability because they miss IBNR claims. On occasion, provisions are overstated where claim rates are declining. The actuarial reserve is the correct measure under Ind AS 37's 'best estimate' standard.

Request a warranty liability assessment

Kapadia & Kochrekar applies actuarial claims development methodology to warranty provisioning — producing defensible, Ind AS 37-compliant reserve estimates.

Request an Assessment arrow_forward

Warranty Valuation — Actuarial Estimation of Warranty Claim Liabilities