ICAEW.com works better with JavaScript enabled.
The third instalment in a series looking at the differences between International Financial Reporting Standards (IFRS) and International Public Sector Accounting Standards (IPSAS), and the suitability of each for public sector financial reporting.

Financial reporting is the process of recognising, measuring and disclosing information that enable users to get an informed view of the financial position and performance of an entity that should allow them to make useful decisions and hold the entity to account. Impairment is a permanent reduction in the value of an asset which will negatively affect the services that can be provided by that asset or the profits it can generate.


There are two impairment standards under IPSAS, one for cash generating assets (IPSAS 26) and one for non-cash generating assets (IPSAS 21), whilst the IFRS standard has only one impairment standard for cash generating assets (IAS 36). Since this article compares the two accounting frameworks, most differences will arise from the non-cash generating assets, since these are often specialist assets used in the provision of services.

Cash generating assets are held with a primary objective of generating a commercial return. In the public sector, most assets will be held for the delivery of services and will not generate commercial return. However, there will be some assets that generate cash flows yet are primarily held for service delivery purposes. Sometimes it may not be clear whether the primary objective of holding an asset is to generate a commercial return or not. Examples would include fee paying hospital patients being treated in the same hospital ward as non-fee-paying patients or rental properties that include social tenants.

When the primary objective of holding an asset is not clear, it will be necessary to evaluate the significance of the cash flows and judgement will be required to determine whether to apply either the IPSAS for cash generating or for non-cash generating assets.

What is impairment?

Accounting standards (both IFRS and IPSAS) describe an asset as impaired when the carrying amount of an asset exceeds its recoverable amount (or recoverable service amount in the case of the non-cash generating asset, IPSAS 21).

The recoverable amount is the higher of an asset's fair value less costs to sell and its value in use. Measuring the recoverable amount can be quite challenging but this only needs to happen if there are indicators of impairment. However, there are exceptions to this which are noted below. At each reporting date, an entity is obliged to assess whether there is any indication that an asset may be impaired and if there is such an indication then the entity shall estimate the recoverable amount of the asset or cash generating unit.

Some are of the view that fair value is not an appropriate measurement basis for the public sector assets due to the application of the notion of highest and best use and reference to market participants. However, even if historic cost is applied as a measurement basis, judgement will be required to estimate the recoverable amount if there are indications of impairment. Establishing the recoverable amount involves estimating the fair value of an asset and its value in use which can be complex. This point is often overlooked when advocating the benefits of historical cost accounting as it is assumed that only observable inputs are required to account for assets at historical cost. This is not true due to the impairment standard requiring account preparers to provide fair value and value in use estimates.

Indicators of impairment

Both IFRS and IPSAS state that an entity shall assess at each reporting date whether there is an indication that an asset might be impaired and if there is such an indication then the entity shall estimate the recoverable amount or recoverable service amount of the asset.

The accounting standards split the indicators of impairment between external and internal sources of information, and, whilst the list is not exhaustive, each standard lists out in bold which indicators need to be considered as a minimum.

The lists of external and internal indicators of impairment differ. For example, IAS 36 contains an indicator relating to dividends from a subsidiary, joint venture or associate; and in contrast IPSAS 21 and IPSAS 26 include a decision to halt the construction of an asset before it is complete or in a usable condition.

IPSAS 21 provides specific examples of indicators for impairment of non-cash generating assets, such as cessation of the demand or need for services provided by the asset. In contrast to IPSAS 26 and IAS 36, it does not include a change in the market value of the asset as part of the minimum set of indicators of impairment, instead it simply refers to it in a commentary to the standard.

Examples of indicators, list is not exhaustive:

External sources of information IAS 36 IPSAS 26 - cash-generating assets IPSAS 21 - non-cash generating assets
Market value declines significantly more than would be expected Yes Yes No
Cessation or near cessation of the demand or need for services provided by the asset No No Yes
Significant adverse changes – technological, market, economic, legal (& gov policy for IPSAS 21 only) Yes Yes Yes
Internal sources of information IAS 36 IPSAS 26 - cash-generating assets IPSAS 21 - non-cash generating assets
Physical damage to the asset (obsolescence addition for IAS 36 and IPSAS 26 only) Yes Yes Yes
Decision to halt the construction of the asset before it is complete or in a usable condition No Yes Yes
Dividend from a subsidiary, joint venture or associate exceeds total comprehensive income Yes No No

Note that intangible assets with an indefinite useful life and goodwill have to be tested for impairment irrespective of whether there is any indication of impairment. These can be tested at any time during the reporting period as long as it is at the same time every year.

Historical differences in scope

When IPSAS 21 and 26 were first endorsed in 2004 and 2008 respectively, neither standard required an impairment test for assets that were carried at revalued amounts or for goodwill, since these were out of scope. This was a major departure from IAS 36 which mandates impairment testing for both revalued assets and goodwill.

Regarding property, plant and equipment held at revalued amounts, IPSASB argued that the revaluation already incorporates impairment and that in cases where the fair value is the market value, the maximum amount of impairment loss is the disposal cost, which they deemed to be immaterial. Hence these assets were out of scope for practical reasons.

However, IPSASB acknowledged that it was ambiguous whether impairment losses (and their reversal) were revaluations, given that they are accounted for in the same way. If impairment losses are treated as revaluations, then the standard requires the entire class of the assets to be revalued – an onerous requirement. Impairments are limited to specific assets, not entire asset classes. To fix this ambiguity, revalued assets are required to be tested for impairment as well as going through a regular revaluation exercise. The practical expedient thus no longer exists.

Revaluation frequency and impairment

IPSAS 17 'Property Plant and Equipment' states that the carrying value of an asset should not be materially different to that asset's current value (such as fair value or current operational value) at the reporting date. If it is then the asset will need to be revalued.

Much like the impairment indicators above, IPSAS 17 also provides a list of indicators to help with the assessment of whether the revalued asset's carrying amount differs materially from that which would be determined if the asset were revalued at the reporting date.

The indicators relating to significant (long-term) changes with an adverse effect on the entity in the technological, legal, or government policy environment in which the entity operates, and evidence of physical damage are examples of indicators that are applied to both revaluations and impairments.

The IPSAS require account preparers to differentiate between revaluation and impairment but in practice this will require judgement. Whilst it would be helpful to have more tailored indicators, the potential for an overlap of triggers could result in an entire asset class having to be revalued on the one hand versus only a single asset being impaired on the other hand. The accounting impact is quite significant with impairments affecting the performance statement immediately whereas a downward revaluation could go against the revaluation reserve (in equity) if there has been a previous upward evaluation (ie, there is headroom to be used up).

In the UK, the accounting manual for central government states that revaluations only need to be carried out every five years. There are ongoing discussions taking place how an asset should be treated between valuations to ensure the carrying value is not materially different to the fair value. The proposals centre around some form of indexation based on values provided centrally by HM Treasury.

Regarding goodwill, it was originally excluded from IPSAS's impairment standards because there was no public sector combinations standard (that would recognise goodwill in the first place) and there was some concern about public sector specific differences such as always having to be able to identify the acquirer in a combination. IPSASB issued IPSAS 40 'Public Sector Combinations' in 2017 and hence no longer exclude goodwill from impairment.

Goodwill can only be measured by reference to cash flows and because goodwill can't generate economic benefits independently of other assets, it is assessed for impairment as part of a group of assets. IPSAS 21 only deals with the impairment of individual assets and therefore this standard is not appropriate to apply to the impairment of goodwill. Goodwill should be treated as a cash-generating asset for the purposes of impairment. Consequently, IPSAS 26 includes the extensive requirements and guidance on a) the impairment of goodwill, b) the allocation of goodwill to cash-generating units, and c) testing for impairment cash-generating units with goodwill.

Measurement differences

An asset is impaired if the recoverable amount is less than the carrying amount. The recoverable amount is defined in both IAS 36 and IPSAS 26 as the higher of an asset's or cash generating unit's fair value less costs of disposal and its value in use.

By contrast, the IPSAS standard on non-cash generating assets, IPSAS 21, only applies to individual assets. Under both IAS 36 and IPSAS 26, where it is not possible to determine the recoverable amount of an individual asset, then the recoverable amount of the asset's cash-generating unit will be determined. IPSASB considered the concept of a service-generating unit in a non-cash-generating context but noted that this was not necessary because it is possible to identify the service potential of individual assets and it would introduce undue complexity. Therefore, the recoverable service amount is defined as the higher of an asset's fair value less costs to sell and its value in use.

This leads us into the first difference between IFRS and IPSAS and between the two IPSAS impairment standards.

In IPSAS 26 and IAS 36, value in use is the present value of the estimated future cash flows expected from using the asset plus the disposal at the end of its useful life. By contrast, the impairment standard on non-cash generating assets defines value in use as the present value of the asset's remaining service potential.

There are several methods of estimating the value in use (present value of remaining service potential) for non-cash generating assets:

  1. Depreciated replacement cost – the replacement cost is the asset's gross service potential and is depreciated to reflect the asset in its used condition. The cost is measured as the reproduction or replacement cost of the asset (whichever is lower).
  2. Restoration cost approach – the cost of restoring the service potential of an asset to its pre-impaired level. Under this approach, the present value of the remaining service potential is determined by subtracting the estimated restoration cost from the current cost of replacing the remaining service potential of the asset before impairment, which is usually the depreciated reproduction or replacement cost of the asset (see above).
  3. Service units approach – the present value of the remaining service potential of the asset is determined by reducing the current cost of the remaining service potential of the asset before impairment to conform with the reduced number of service units expected from the asset in its impaired state.

For cash-generating assets, IAS 36 and IPSAS 26 are both identical in that the value in use is determined by estimating the future cash inflows and outflows from continuing use of the asset and from its ultimate disposal and applying the appropriate discount rate to those future cash flows.

Practical implications of measurement differences

Determining the value in use for non-cash generating assets can be approached in a number of ways. IPSASB did consider replicating the IAS 36 and IPSAS 26 approach which is to estimate discounted cash inflows that would have arisen had the entity sold its services or outputs to the market. But this would not really work in practice since it would be extremely difficult to determine the appropriate prices at which to value the service or output units and to also estimate the discount rate.

To recap, a major difference is the application of different methods to determine the value in use of non-cash generating assets which is to determine the present value of service potential as opposed to present value of cash flows.

Before looking into some more specific scenarios, it should be highlighted that if either the fair value less costs to sell or the value in use of an asset is greater than the asset's carrying amount then the asset is not impaired, and it is then not necessary to estimate the other amount. So, if fair value less costs to sell is higher than the carrying amount of an asset then it is not necessary to also estimate the value in use.

Australia case study

The Australian Accounting Standards Board's (AASB) impairment standard (AASB 136) applies to assets that are carried at revalued amounts in accordance with both the property, plant and equipment and the intangibles standards. This is no different to the impairment standard(s) under IFRS or IPSASB.

If the assets are held at fair value, then the only difference between the carrying amount and fair value less costs to sell is the direct incremental costs attributable to the disposal of the asset. If the disposal costs are negligible, the recoverable amount of the revalued asset is necessarily close to, or greater than, its revalued amount. In this case, after the revaluation requirements have been applied, it is unlikely that the revalued asset is impaired and the recoverable amount need not be estimated. This is not the case if the disposal costs are not negligible.

Neither IPSAS impairment standards explicitly discuss the relationship between fair value and fair value less costs to sell and therefore does not explicitly state that the recoverable service amount or recoverable amount need not be estimated where costs of sale are negligible. The fact that revalued amounts are no longer out of scope (see above for the reasons) would seem to suggest that an entity would still need to demonstrate that the recoverable amount is higher than the carrying amount.

A new measurement basis: Current operational value

IPSASB has issued a new measurement standard (IPSAS 46) which includes a new current value measurement basis called current operational value. There is more detail on this in part II of our IFRS vs IPSAS series but this measurement basis is defined as the estimated amount required to replace the service potential of an asset at the measurement date.

The reason for creating this new measurement basis is to provide an alternative to fair value, which is not always applicable in a public sector context as explained below.

The objective of a fair value measurement is to estimate the price at which an orderly transaction to sell the asset or to transfer the liability would take place between market participants at the measurement date under current market conditions.

A common concern with fair value is that when an asset is held for its operational capacity, fair value is difficult and inappropriate to apply because the following concepts are generally not applicable:

  1. Highest and best use; and
  2. Maximising the use of market participant data.

In the public sector, many assets are used for the delivery of services which can result in them not being put to their highest and best financial use. A typical example would be operating a school in a city centre when the most financially lucrative use for the building might be office space. Whilst the standard says that highest and best use of non-financial assets takes into account the use of the asset that is physically possible, legally permissible and financially feasible, the highest and best use is nevertheless determined from the perspective of market participants, even if the entity intends a different use (of the asset).

This may make the application of fair value difficult in a public sector context, not only to calculate the value but also to interpret the result. Many believe that such information is not helpful to users of the financial statements since if an asset will never change its use or is unlikely to ever be sold, then why apply a hypothetical sales value to an asset that is being valued on a different basis than it is actually being used for (school vs office as per the example above)?

The current operational value was introduced by IPSASB as an alternative to fair value to cover public sector specific circumstances when fair value is not an appropriate measurement basis. But does it make sense to have impairment standards that require an estimation of fair value when at standards level fair value was purposefully avoided as a measurement basis? Furthermore, IPSAS 21 and IPSAS 26 don't make it clear which definition of fair value is being applied, does the fair value less costs to sell include highest and best use for example?

Another potential problem is the overlap of different measurement techniques available for current operational value and the impairment standards. Current operational value allows different techniques, market value (observable market data) and cost approach (current replacement cost). In cases where the measurement technique applied for current operational value doesn't align with the method applied in IPSAS 21 or IPSAS 26, there could still be a need for further impairment if the recoverable amount is lower than the current operational value.


There are no differences regarding the recognition of impairments between IPSAS and IFRS. Impairment losses are recognised straight away in surplus or deficit unless the asset is carried at revalued amount in accordance with another standard (ie, property, plant and equipment).

Any impairment loss of a revalued asset shall be treated as a revaluation decrease in revaluation surplus (reserves) to the extent that the impairment loss does not exceed the amount in revaluation surplus for that individual asset.

UK adaption to IAS 36

The UK's central government financial reporting manual (FReM) has made an adaptation to the impairment standard by stipulating that only impairment losses that result from a clear consumption of economic benefit (including as a result of loss or damage from normal business operations) should be taken to surplus or deficit.

The FReM, makes it very clear that a reduction in an asset's value due to fall in its market price should first be offset against a revaluation reserve if there is one, and once that element of the reserve is exhausted the fall in value should be taken to surplus or deficit.

Examples of impairments resulting from a consumption of economic benefit or service potential include impairments as a result of loss or damage, abandonment of projects, gold-plating and use of the asset for a lower specification purpose.

The reason for this is to maintain alignment with their own budgeting guidance which says that a fall in value relating to a consumption of economic benefit is an impairment.


The differences between IPSAS 21, IPSAS 26 and IAS 36 are not pervasive, especially since IPSASB updated their standards to remove the scope exclusion of goodwill and assets held at revalued amounts.

There are some differences in indicators of impairment, which reflect the two different environments in which public sector and private sector entities operate . The biggest difference is that IPSASB has created an impairment standard for non-cash generating assets which then has an impact on how to determine the recoverable amount (both fair value and value in use).

We would like to see IPSASB clarify the interaction between the new measurement basis – current operational value – and the impairment standards. There are a number of cross-cutting issues around the overlap of measurement basis and the definition of fair value.