How to Audit Impairment of Assets Under IAS 36

Why impairment is the estimate that draws the most scrutiny

Impairment testing under IAS 36 sits at the intersection of two things regulators distrust: management optimism and accounting estimates. A value-in-use calculation is built entirely from management’s assumptions about the future. Revenue growth, margin trajectories, capital expenditure, working capital, discount rates, terminal growth. Every input is a judgment. And the judgment is being made by the same management team whose bonus may depend on the carrying amount of those assets.

The FRC’s annual inspection results consistently identify impairment as a top-five area of deficiency. In its 2023 thematic review on accounting estimates, the AFM devoted an entire section to goodwill impairment testing, noting that in a significant proportion of reviewed files, the auditor accepted management’s model without independently assessing the reasonableness of key assumptions. PCAOB inspection observations in 2023 raised similar concerns about auditors failing to test the completeness of impairment indicators.

This isn’t a niche topic. For any client that has made an acquisition in the past decade and carries goodwill on the balance sheet, impairment testing is likely the highest-risk estimate in the financial statements. It deserves more than a recalculation of management’s spreadsheet.

Cash-generating unit identification under IAS 36.66

What a CGU is and why it matters

IAS 36.6 defines a cash-generating unit as the smallest identifiable group of assets that generates cash inflows largely independent of the cash inflows from other assets or groups of assets. Getting the CGU identification wrong cascades through every subsequent step of the impairment test. A CGU that’s too broad masks impairment in a profitable unit. A CGU that’s too narrow creates artificial impairment triggers.

IAS 36.68 provides the key criterion: if an active market exists for the output produced by an asset or group of assets, that asset or group is a CGU even if some or all of the output is used internally. In practice, on mid-market engagements, CGU identification often aligns with operating segments or business units.

The problem arises when management defines CGUs at a level that’s convenient rather than correct.

A client with four retail locations that share purchasing, branding, and marketing but generate independent revenue streams has four CGUs under IAS 36.68, not one. If management treats all four as a single CGU, a loss-making location that should trigger an impairment charge gets absorbed by the profitable locations. The carrying amount of the individual location’s assets never gets tested against its own recoverable amount. This is the most common CGU-identification deficiency on retail and multi-site engagements, and it’s entirely preventable if you ask one question at planning: do the individual locations generate independent cash inflows?

Consistency of CGU identification year over year

IAS 36.72 requires that if an asset or CGU has changed from the prior period, the client discloses the change and the reasons for it. From an audit perspective, changes in CGU identification should be rare and well justified. If your client restructured its operations (merged two divisions, sold a product line, opened a new market), a CGU change may be appropriate. If nothing operationally changed but the CGU definition shifted, that’s a red flag.

Request the prior year CGU identification and compare it to the current year. Where differences exist, obtain management’s explanation, assess it against the operational facts, and document your conclusion. The FRC noted in its 2022 inspection results that CGU changes without adequate disclosure were flagged in multiple files.

Value in use: the assumptions that drive the number

The discount rate under IAS 36.55

IAS 36.55 requires a pre-tax discount rate that reflects current market assessments of the time value of money and the risks specific to the asset for which the cash flow estimates have not been adjusted. In practice, most clients use a post-tax WACC and then gross it up to a pre-tax rate, which IAS 36.BCZ85 acknowledges as an acceptable approach provided the result approximates the pre-tax rate.

Your audit work on the discount rate covers four elements. First, verify the risk-free rate against observable government bond yields at the reporting date. Second, check the equity risk premium against published sources (Damodaran’s annual data is the most commonly cited). Third, assess the beta used. If the client is unlisted, they’ll typically use a peer group of listed comparables. Confirm the peer group is reasonable (same industry, similar size, same geography where possible) and that the beta has been unlevered and relevered for the client’s capital structure. Fourth, verify the cost of debt against the client’s actual borrowing rate or, if using a theoretical rate, against the rate for similarly rated debt.

A 100 basis point change in the discount rate on a €15M value-in-use calculation can move the result by €1M to €3M depending on the cash flow profile and terminal value weight.

That sensitivity makes the discount rate the single most material input in most impairment models. It also makes it the assumption management is most likely to optimise. When you see a discount rate at the bottom of any reasonable range, combined with cash flow projections at the top, document the compound effect explicitly.

Cash flow projections under IAS 36.33

IAS 36.33 limits the projection period to a maximum of five years unless a longer period can be justified. Management’s cash flow projections should be based on the most recent budgets and forecasts approved by the board. IAS 36.34 is explicit: projections beyond the budget period should be based on extrapolations using a steady or declining growth rate, unless an increasing rate can be justified.

Compare management’s projections to the approved budget. If the impairment model shows year-one revenue 12% above the board-approved budget, ask why. Compare projected margins to historical achieved margins over the past four years. If the model assumes a margin expansion from 8% to 14% over five years but the client has never achieved a margin above 10%, the projections need justification that goes beyond aspiration. This comparison is the most productive five minutes you’ll spend on the impairment workpaper.

Nominal versus real is a persistent trap. IAS 36.40 requires consistency: nominal cash flows with a nominal discount rate, or real cash flows with a real discount rate. If the client’s model includes 2.5% annual price inflation in revenue but the discount rate doesn’t include an inflation component, the mismatch systematically overstates value in use. On Dutch mid-market engagements, this is the most common projection error, and it can shift the value-in-use result by 10% to 20% over a five-year forecast horizon.

Terminal value

The terminal value typically represents 60% to 80% of the total value in use. More than half the recoverable amount rests on a single growth rate applied in perpetuity.

IAS 36.33(c) requires the terminal growth rate not to exceed the long-term average growth rate for the products, industries, or country in which the client operates. For a Dutch client, the reference point is the ECB’s long-term inflation target (2%) plus long-term real GDP growth for the Netherlands (typically 1% to 1.5%). A terminal growth rate above 3% requires specific justification, and justification means evidence that the CGU’s growth will permanently outpace the broader economy. For most mid-market industrial or services businesses, it won’t. When a client uses 4% and can’t explain why their CGU is structurally different from the economy it sits in, you have an unsupported assumption driving the majority of the recoverable amount.

The ciferi Financial Ratio Calculator can cross-check whether the implied valuation from management’s model is consistent with the client’s observable financial performance metrics.

Goodwill allocation and the annual test

Allocation under IAS 36.80

IAS 36.80 requires goodwill acquired in a business combination to be allocated to each CGU (or group of CGUs) expected to benefit from the synergies of the combination. The allocation must be completed by the end of the first annual period beginning after the acquisition.

This deadline is not flexible. IAS 36.84 gives the client until the end of that first annual period, not longer. If the client acquired a business in February 2024 and the reporting date is 31 December, the goodwill allocation must be finalised in the financial statements for the year ending 31 December 2025 at the latest. Unallocated goodwill beyond this window means the impairment test cannot be performed correctly, because there’s no CGU to test it against.

When CGUs change after acquisition

When a client reorganises and the CGU to which goodwill was originally allocated changes, IAS 36.87 requires the goodwill to be reallocated using the relative values of the reorganised parts. Not optional.

Many mid-market clients ignore this. They reorganise operations, redefine segments, but leave the goodwill sitting on the same line it was originally allocated to. The impairment test then uses cash flows from the new CGU structure with goodwill allocated to the old structure. The test is internally inconsistent. If you rely on it, the audit conclusion on goodwill is not adequately supported, and the working paper contains a logical contradiction that a quality reviewer will identify.

The interaction between CGU-level and entity-level testing

IAS 36.90 addresses the situation where goodwill cannot be allocated to individual CGUs on a non-arbitrary basis. In that case, goodwill is tested at the level of the smallest group of CGUs to which it can be allocated. On a mid-market engagement with a single operating segment and one pool of goodwill, this often defaults to entity-level testing.

Entity-level testing creates built-in headroom. A profitable client’s total recoverable amount (including self-generated intangibles and assembled workforce that aren’t on the balance sheet) will almost always exceed the carrying amount of recognised net assets. Individual asset groups can be impaired in economic substance but never recognised, because this surplus at group level absorbs them.

Your file should acknowledge this limitation explicitly. Document why entity-level allocation is appropriate under IAS 36.90, not just default to it because it’s convenient.

Worked example: auditing Hendriksen Groep B.V.

Client: Hendriksen Groep B.V., a mid-market industrial services group based in Arnhem. Revenue: €86M. Two operating divisions: Maintenance Services (€54M revenue, 11% EBITDA margin) and Technical Staffing (€32M revenue, 6% EBITDA margin). Goodwill on balance sheet: €7.8M (€5.2M allocated to Maintenance Services, €2.6M allocated to Technical Staffing, from a 2021 acquisition). Performance materiality: €430K.

1. Confirm CGU identification

Hendriksen operates two divisions with separate management, separate customer bases, and separate P&L reporting to the board. Technical Staffing places temporary workers. Maintenance Services provides contracted industrial maintenance. Revenue streams are independent. Under IAS 36.68, these are two CGUs. You confirm the identification is unchanged from the prior year and consistent with the operating segment disclosure under IFRS 8.

2. Challenge the discount rate for Technical Staffing

Management uses a post-tax WACC of 9.2% for Technical Staffing, grossed up to a pre-tax rate of 11.8%. You decompose the WACC. Risk-free rate: 2.6% (10-year Dutch government bond at reporting date). Equity risk premium: 5.5% (Damodaran 2024 data for the Netherlands). Beta: 0.95, derived from a peer group of four listed European staffing companies. Cost of debt: 4.8% (the client’s actual borrowing rate on its revolving credit facility). Debt-to-equity ratio: 35/65.

You note that the peer group beta of 0.95 seems low for a staffing business that lost two significant contracts in the past 18 months. Comparable companies with declining revenue typically show higher betas. You request management’s rationale for not applying a company-specific risk adjustment. The CFO argues that the contract losses are reflected in the cash flow projections, not in the discount rate. You assess whether this is reasonable under IAS 36.56(a), which states that risks already reflected in cash flows should not be double-counted in the discount rate. The argument has merit if the cash flows genuinely capture the downside.

3. Evaluate the cash flow projections for Technical Staffing

Management projects revenue growth of 8% per year for Technical Staffing over the five-year forecast period. Historical revenue growth over the past four years: 2021 (14%), 2022 (9%), 2023 (minus 3%), 2024 (minus 6%). The division has been declining for two consecutive years. You request the board-approved budget for year one. The budget shows 2% revenue growth, not 8%.

You raise the discrepancy with the CFO. She explains that the 8% reflects a new contract pipeline that is not yet in the budget because the contracts haven’t been signed. Under IAS 36.44, the cash flow projections must not include cash inflows from future restructurings or asset enhancements to which the client is not yet committed. Unsigned contracts fall into this category.

You conclude that the year-one projection should align with the approved budget (2% growth) and the outer years should use a growth rate consistent with historical performance and the approved budget, not the unsigned pipeline. You recalculate value in use with a 2% year-one growth rate and 3% for years two through five. The revised recoverable amount for Technical Staffing falls below the carrying amount (including goodwill) by approximately €1.1M.

4. Test the impairment allocation

The €1.1M impairment is allocated first to the goodwill allocated to Technical Staffing (€2.6M, so the goodwill can absorb the full charge) under IAS 36.104. No allocation to other assets in the CGU is necessary.

Your impairment audit checklist

1. At planning, identify all assets subject to annual impairment testing (goodwill, indefinite-life intangibles) and assess whether indicators exist for other assets under IAS 36.12. The indicator assessment must be documented even when the conclusion is that no indicators exist.
2. Confirm the CGU identification is consistent with the prior year. Where changes occurred, obtain management’s explanation, verify it against operational facts, and check whether the IAS 36.130 disclosure requirements for a change in CGU composition have been met.
3. For value-in-use models, compare year-one cash flow projections to the board-approved budget. If the model exceeds the budget, require a documented justification from management that addresses IAS 36.33(b) specifically. Document whether the explanation is persuasive.
4. Independently verify the discount rate by decomposing the WACC into its components and checking each against observable market data at the reporting date. Record the risk-free rate source, equity risk premium source, peer group for beta, cost of debt source, and the gross-up methodology from post-tax to pre-tax.
5. Assess the terminal growth rate against the long-term GDP growth rate for the relevant country and industry per IAS 36.33(c). Any rate exceeding the economy-wide average requires specific justification relating to the CGU’s competitive position and market dynamics, not generic industry growth claims.
6. Perform a sensitivity analysis on the two or four most material assumptions (typically discount rate and terminal growth rate, sometimes also year-one revenue and margin). Document the headroom or deficit under each scenario and identify which assumptions would need to change, and by how much, to trigger an impairment or reverse one.

Common mistakes

• Accepting management’s model without challenging assumptions: The AFM’s 2023 inspection findings on impairment testing identified this as the most frequent deficiency: auditors recalculating the model correctly but not challenging whether the inputs were reasonable. A model that is arithmetically correct but built on unsupported assumptions doesn’t meet ISA 540.
• Defaulting to entity-level goodwill testing: Testing goodwill impairment at entity level without documenting why CGU-level allocation under IAS 36.80 is not possible on a non-arbitrary basis. Entity-level testing creates built-in headroom from self-generated intangibles and assembled workforce that masks CGU-level impairment. The FRC flagged this in its 2022 thematic review, noting that several firms defaulted to entity-level testing without the IAS 36.90 justification required.

Related Ciferi content

Related guides:

Put audit concepts into practice with these free tools:

Frequently asked questions

Further reading and source references

• IAS 36, Impairment of Assets: the source standard governing impairment testing, CGU identification, and goodwill allocation.
• IFRS 3, Business Combinations: relevant to the initial goodwill recognition and purchase price allocation that feeds into subsequent impairment testing.
• ISA 540, Auditing Accounting Estimates and Related Disclosures: the audit standard governing the auditor’s approach to management’s impairment estimates.
• IAS 38, Intangible Assets: governs indefinite-life intangible assets that require annual impairment testing alongside goodwill.