BC 128 Integral to the Board's decision that indefinite-lived intangibles should be tested for impairment annually was the view that many entities should be able to conclude that the recoverable amount of such an asset is greater than its carrying amount without actually recomputing recoverable amount. However, the Board concluded that this would be the case only if the last recoverable amount determination exceeded the carrying amount by a substantial margin, and nothing had happened since then to make the likelihood of an impairment loss other than remote. The Board concluded that, in such circumstances, permitting a detailed calculation of the recoverable amount of an indefinite-lived intangible to be carried forward from the preceding period for use in the current period's impairment test would significantly reduce the costs of applying the impairment test, without compromising its integrity.
Measuring recoverable amount and accounting for impairment losses and reversals of impairment losses

BC 129 The Board could see no compelling reason why the measurement basis adopted for determining recoverable amount and the treatment of impairment losses and reversals of impairment losses for one group of identifiable assets should differ from those applying to other identifiable assets. Adopting different methods would impair the usefulness of the information provided to users about an entity's identifiable assets, because both comparability and reliability, which rest on the notion that similar transactions are accounted for in the same way, would be diminished. Therefore, the Board concluded that the recoverable amounts of indefinite-lived intangibles should be measured, and impairment losses and reversals of impairment losses in respect of those assets should be accounted for, consistently with other identifiable assets covered by the Standard.
BC 130 The Board expressed some concern over the measurement basis adopted in the previous version of IAS 36 for determining recoverable amount (ie higher of value in use and net selling price) and its treatment of impairment losses and reversals of impairment losses for assets other than goodwill. However, the Board's intention in revising IAS 36 was not to reconsider the general approach to impairment testing. Accordingly, the Board decided that it should address concerns over that general approach as part of its future re-examination of IAS 36 in its entirety, rather than as part of its Business Combinations project.
Testing goodwill for impairment
(paragraphs 80-99)

BC 131 The Board concluded that if a rigorous and operational impairment test could be devised, more useful information would be provided to users of an entity's financial statements under an approach in which goodwill is not amortised, but is instead tested for impairment annually or more frequently if events or changes in circumstances indicate that the goodwill might be impaired. An outline of the Board's deliberations in reaching this conclusion is provided in the Basis for Conclusions on IFRS 3 Business Combinations.
BC 132 Paragraphs BC133-BC177 outline the Board's deliberations on the form that the impairment test for goodwill should take:
(a) paragraphs BC137-BC159 discuss the requirements relating to the allocation of goodwill to cash-generating units and the level at which goodwill is tested for impairment.
(b) paragraphs BC160-BC170 discuss the requirements relating to the recognition and measurement of impairment losses for goodwill, including the frequency of impairment testing.
(c) paragraphs BC171-BC177 discuss the requirements relating to the timing of goodwill impairment tests.

BC 133 As a first step in its deliberations, the Board considered the objective of the goodwill impairment test and the measure of recoverable amount that should be adopted for such a test. The Board observed that recent North American standards use fair value as the basis for impairment testing goodwill, whereas the previous version of IAS 36 and the United Kingdom standard are based on an approach under which recoverable amount is measured as the higher of value in use and net selling price.
BC 134 The Board also observed that goodwill acquired in a business combination represents a payment made by an acquirer in anticipation of future economic benefits from assets that are not capable of being individually identified and separately recognised. Goodwill does not generate cash flows independently of other assets or groups of assets and therefore cannot be measured directly. Instead, it is measured as a residual amount, being the excess of the cost of a business combination over the acquirer's interest in the net fair value of the acquiree's identifiable assets, liabilities and contingent liabilities. Moreover, goodwill acquired in a business combination and goodwill generated after that business combination cannot be separately identified, because they contribute jointly to the same cash flows.

BC 135 The Board concluded that because it is not possible to measure separately goodwill generated internally after a business combination and to factor that measure into the impairment test for acquired goodwill, the carrying amount of goodwill will always be shielded from impairment by that internally generated goodwill. Therefore, the Board took the view that the objective of the goodwill impairment test could at best be to ensure that the carrying amount of goodwill is recoverable from future cash flows expected to be generated by both acquired goodwill and goodwill generated internally after the business combination.
BC 136 The Board noted that because goodwill is measured as a residual amount, the starting point in any goodwill impairment test would have to be the recoverable amount of the operation or unit to which the goodwill relates, regardless of the measurement basis adopted for determining recoverable amount. The Board decided that until it considers and resolves the broader question of the appropriate measurement objective(s) in accounting, identifying the appropriate measure of recoverable amount for that unit would be problematic. Therefore, although the Board expressed concern over the measurement basis adopted in IAS 36 for determining recoverable amount, it decided that it should not depart from that basis when measuring the recoverable amount of a unit whose carrying amount includes acquired goodwill. The Board noted that this would have the added advantage of allowing the impairment test for goodwill to be integrated with the impairment test in IAS 36 for other assets and cash-generating units that include goodwill.
Allocating goodwill to cash-generating units

(paragraphs 80-87)
BC 137 The previous version of IAS 36 required goodwill to be tested for impairment as part of impairment testing the cash-generating units to which it relates. It employed a 'bottom-up/top-down' approach under which the goodwill was in effect tested for impairment by allocating its carrying amount to each of the smallest cash-generating units to which a portion of that carrying amount could be allocated on a reasonable and consistent basis.
BC 138 Consistently with the previous version of IAS 36, the Exposure Draft proposed that:
(a) goodwill should be tested for impairment as part of impairment testing the cash-generating units to which it relates; and
(b) the carrying amount of goodwill should be allocated to each of the smallest cash-generating units to which a portion of that carrying amount can be allocated on a reasonable and consistent basis.
However, the Exposure Draft proposed additional guidance clarifying that a portion of the carrying amount of goodwill should be regarded as capable of being allocated to a cash-generating unit on a reasonable and consistent basis only when that unit represents the lowest level at which management monitors the return on investment in assets that include the goodwill. That cash-generating unit could not, however, be larger than a segment based on the entity's primary reporting format determined in accordance with IAS 14 Segment Reporting.

BC 139 In developing this proposal, the Board noted that because acquired goodwill does not generate cash flows independently of other assets or groups of assets, it can be tested for impairment only as part of impairment testing the cash-generating units to which it relates. However, the Board was concerned that in the absence of any guidance on the precise meaning of 'allocated on a reasonable and consistent basis', some might conclude that when a business combination enhances the value of all of the acquirer's pre-existing cash-generating units, any goodwill acquired in that business combination should be tested for impairment only at the level of the entity itself. The Board concluded that this should not be the case. Rather, there should be a link between the level at which goodwill is tested for impairment and the level of internal reporting that reflects the way an entity manages its operations and with which the goodwill naturally would be associated. Therefore, it was important to the Board that goodwill should be tested for impairment at a level at which information about the operations of an entity and the assets that support them is provided for internal reporting purposes.

BC 140 In redeliberating this issue, the Board noted that respondents' and field visit participants' comments indicated that the Board's intention relating to the allocation of goodwill had been widely misunderstood, with many concluding that goodwill would need to be allocated to a much lower level than that intended by the Board. For example, some respondents and field visit participants were concerned that the proposal to allocate goodwill to such a low level would force entities to allocate goodwill arbitrarily to cash-generating units, and therefore to develop new or additional reporting systems to perform the test. The Board confirmed that its intention was that there should be a link between the level at which goodwill is tested for impairment and the level of internal reporting that reflects the way an entity manages its operations. Therefore, except for entities that do not monitor goodwill at or below the segment level, the proposals relating to the level of the goodwill impairment test should not cause entities to allocate goodwill arbitrarily to cash-generating units. Nor should they create the need for entities to develop new or additional reporting systems.

BC 141 The Board observed from its discussions with field visit participants that much of the confusion stemmed from the definition of a 'cash-generating unit', when coupled with the proposal in paragraph 73 of the Exposure Draft for goodwill to be allocated to each "smallest cash-generating unit to which a portion of the carrying amount of the goodwill can be allocated on a reasonable and consistent basis". Additionally, field visit participants and respondents were unclear about the reference in paragraph 74 of the Exposure Draft to "the lowest level at which management monitors the return on investments in assets that include goodwill", the most frequent question being "what level of management?" (eg board of directors, chief executive officer, or segment management).
BC 142 The Board noted that once its intention on this issue was clarified for field visit participants, they all, with the exception of one company that believes goodwill should be tested for impairment at the entity level, supported the level at which the Board believes goodwill should be tested for impairment.
BC 143 The Board also noted the comment from a number of respondents and field visit participants that for some organisations, particularly those managed on a matrix basis, the proposal for cash-generating units to which the goodwill is allocated to be no larger than a segment based on the entity's primary reporting format could result in an outcome that is inconsistent with the Board's intention, ie that there should be a link between the level at which goodwill is tested for impairment and the level of internal reporting that reflects the way an entity manages its operations. The following example illustrates this point:
A company managed on a matrix basis is organised primarily on a geographical basis, with product groups providing the secondary basis of segmentation. Goodwill is acquired as part of an acquisition of a product group that is present in several geographical regions, and is then monitored on an ongoing basis for internal reporting purposes as part of the product group/secondary segment. It is feasible that the secondary segment might, depending on the definition of 'larger', be 'larger' than a primary segment.

BC 144 Therefore, the Board decided:
(a) that the Standard should require each unit or group of units to which goodwill is allocated to represent the lowest level within the entity at which the goodwill is monitored for internal management purposes.
(b) to clarify in the Standard that acquired goodwill should, from the acquisition date, be allocated to each of the acquirer's cashgenerating units, or groups of cash-generating units, that are expected to benefit from the combination, irrespective of whether other assets or liabilities of the acquiree are assigned to those units or groups of units.
(c) to replace the proposal for cash-generating units or groups of units to which goodwill is allocated to be no larger than a segment based on the entity's primary reporting format, with the requirement that they be no larger than a segment based on either the entity's primary or the entity's secondary reporting format. The Board concluded that this amendment is necessary to ensure that entities managed on a matrix basis are able to test goodwill for impairment at the level of internal reporting that reflects the way they manage their operations.