Accounting for goodwill:

The transition from

amortisation to impairment –
an impact assessment
A Wiese
Department of Accountancy
University of Stellenbosch

Abstract
When the FASB adopted an impairment test approach in 2001, rather than
amortisation, the accounting for goodwill arising from an acquisition took a step in a
new direction. The IASB, seeking international convergence and global
harmonisation, also implemented this change when it issued IFRS 3 in 2004.
Moving away from amortisation towards an impairment test involves a radical
change. The research on which this paper is based was undertaken to examine these
two very different accounting practices for the treatment of goodwill and to assess the
possible impact that a transition from the one to the other may have on financial
reporting.
Key words
Accounting practices Acquisition
Amortisation Business combinations
Earnings Financial reporting
Goodwill Impairment

1 Introduction
Some debate regarding the most appropriate method of accounting for goodwill that arises
from an acquisition (commonly referred to as “purchased” or “acquired” goodwill) raged
during the early 1990s and again during the early 2000s.
During 1993, the International Accounting Standards Board (IASB) amended an
International Accounting Standard (IAS), namely IAS 22, Business Combinations (IASB
1993), by removing the option of writing off purchased goodwill (hereafter referred to
merely as goodwill) immediately on acquisition. The accounting treatment of goodwill that
arises from an acquisition took a significant step in a new direction with this amendment,
whereby goodwill now has to be recognised as an asset and amortised over its useful life.
A number of alternative treatments were used in South Africa at that time, as there was
no definitive accounting standard for the accounting treatment of goodwill before the

Meditari Accountancy Research Vol. 13 No. 1 2005 : 105-120 105

Accounting for goodwill: The transition from amortisation to impairment – an impact assessment

introduction of the new accounting statement, AC 131, Business Combinations (SAICA

1999). AC 131 was based on IAS 22 and covered all periods commencing on or after 1
January 2000.
The Financial Accounting Standards Board (FASB) in the United States of America
(USA) sparked an international debate during 2001 when it adopted a Statement of
Financial Accounting Standards (SFAS), SFAS 142, Goodwill and other intangible assets
(FASB 2001a), thereby replacing the existing requirement to amortise goodwill with an
impairment testing approach.
The IASB, seeking international convergence and global harmonisation, followed the
FASB, and issued a new International Financial Reporting Standard (IFRS), namely IFRS
3, Business Combinations (IASB 2004a) in March 2004. The South African Institute of
Chartered Accountants (SAICA) adopted IFRS 3 at the same time. According to IFRS 3,
from the beginning of the first annual period beginning on or after 31 March 2004, all
entities must discontinue amortising goodwill and must test the goodwill for impairment
annually instead.
The debate in the early 1990’s resulted in the amortisation of goodwill. The current
change is to an impairment test rather than amortisation. This is a radical change, and it
involves a very different accounting process (Moehrle 2001).

2 Aim and research method

Accounting for goodwill is one of the most difficult aspects of financial reporting. The
difficulty arises from the nature of goodwill, which is almost impossible to quantify. It is
therefore also very difficult to find an accurate method for measuring the consumption of
goodwill.
The aim of this study is to document the current amendments of national and
international accounting and financial reporting practices in respect of the treatment of
goodwill, and to compare them with previous practices. An example will serve to illustrate
and compare these practices. Against this background, the study examines how goodwill
originates. It then explores the effect that a change between these two very different
accounting practices regarding the treatment of goodwill may have on companies in future.
IFRS 3 requires all business combinations within its scope to be accounted for using the
purchase method, whereas IAS 22 permitted the use of the pooling of interests method, as
well as the purchase method. This choice allowed companies to avoid the recognition of
goodwill by structuring a business combination so that it could be classified as a uniting of
interests. (The possible linkage between the abandonment of the pooling of interests
method and the resulting change in the treatment of goodwill is not explored in this article.)
This article only discusses positive goodwill, because the accounting treatment of
negative goodwill has already been discussed in a previous article (Wiese 2003).

3 Accounting practices for the treatment of goodwill

Goodwill arising on acquisition refers to the excess of the cost of the acquisition over the
acquirer’s interest in the fair value of the identifiable assets and liabilities acquired on the
date of the exchange transaction.

106 Meditari Accountancy Research Vol. 13 No. 1 2005 : 105-120

 Wiese

3.1 IAS 22 / AC 131

According to IAS 22 and AC 131, goodwill should be recognised as an asset and carried at
cost, less any accumulated amortisation and any accumulated impairment losses. Goodwill
amortised over a period exceeding 20 years has to be tested for impairment annually.
Otherwise, the carrying amount of goodwill must only be reduced if there is an indication
that it is impaired.

3.2 IFRS 3
In terms of the new standard, goodwill acquired in a business combination is an asset and
must initially be measured at cost (IFRS 3 par.51). After initial recognition, the acquirer
must measure this goodwill at cost, less any accumulated impairment losses (I FRS 3
par.54). The acquirer must test goodwill for impairment annually, or more frequently, if
events or changes in circumstances indicate that it might be impaired, in accordance with
IAS 36, Impairment of Assets (IFRS 3 par.55).
IAS 36, Impairment of Assets (IASB 2004b) requires goodwill to be tested for
impairment as part of the impairment testing of the cash-generating unit to which it relates,
using a two-step approach. The carrying amount of goodwill must be allocated, from the
acquisition date, 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. A cash-generating
unit (hereafter referred to as a unit) is defined as the smallest identifiable group of assets
that generates cash inflows from continuing use that are largely independent of the cash
flows from other assets or groups of assets.
The two-step impairment test involves the following
Step 1: Compare the carrying amount of the unit, including the goodwill, with its
recoverable amount. The recoverable amount of such a unit should be measured,
consistent with the requirements in IAS 36, as the higher of value in use and net selling
price. If the recoverable amount of the unit exceeds its carrying amount, goodwill is not
impaired. If not, then follow Step 2.
Step 2: Compare the implied value of goodwill with its carrying amount. Implied
goodwill is the excess of the recoverable amount of the unit to which the goodwill has
been allocated over the fair value of the net identifiable assets that the entity would
recognise if it acquired that unit in a business combination on the date of the impairment
test. Any excess of the carrying amount of goodwill over its implied value is recognised
immediately, in profit or loss, as an impairment loss. Any remaining excess of the
carrying amount of the unit over its recoverable amount is recognised as an impairment
loss and allocated to the other assets of the unit on a pro rata basis, based on the
carrying amount of each asset in the unit.

3.3 SFAS 142

Goodwill must be recognised as an asset and tested for impairment annually. This will also
involve a two-step process that begins with an estimation of the fair value of a reporting
unit.
The main differences between IFRS 3 and SFAS 142 are that

Meditari Accountancy Research Vol. 13 No. 1 2005 : 105-120 107

Accounting for goodwill: The transition from amortisation to impairment – an impact assessment

SFAS 142 uses fair value (the amount for which the whole unit could be bought or sold
between willing parties) as the basis for the impairment testing of goodwill, whereas the
recoverable amount is used as basis for the impairment test in IAS 36; and
SFAS 142 refers to a reporting unit, whereas IAS 36 refers to a cash-generating unit. A
reporting unit is the level at which management reviews and assesses the operating
segment’s performance – in other words, units can be discrete business lines or grouped
by geography and can produce stand-alone financial statements. The level at which
goodwill is tested for impairment under SFAS 142 will thus often be higher than the
level at which it would be tested under IAS 36 (IASB 2003a).

3.4 Example
Company X acquires a 100% interest in Company Y for R1 600 000 on 1 January 20X3. At
this date, Y’s identifiable net assets have a fair value of R1 200 000. At the end of 20X3,
the following relates to Y (assume Y is a cash-generating unit):
Recoverable amount of Y R1 600 000
Carrying amount of the identifiable net assets of Y R1 350 000
Fair value of the identifiable net assets of Y which X would recognise if it acquired
Y at this date R1 500 000

The amounts relating to the investment in Y in the consolidated financial statements of X

would be as follows, according to the two methods (assuming that the goodwill is
amortised over five years):
Amortisation (IAS 22)
Net assets of Y R1 350 000
Goodwill ((1 600 000 – 1 200 000) – 80 000) 320 000
Net total R1 670 000
Impairment test (IFRS 3)
Net assets of Y R1 350 000
Goodwill (400 000 – 300 000) 100 000
Net total R1 450 000

Step 1 of the impairment test: The carrying amount of Y in the consolidated financial
statements is R1 750 000 (R1 350 000 + R400 000), which is higher than the recoverable
amount of R1 600 000.
Step 2 of the impairment test: Implied goodwill is calculated as the excess of the
recoverable amount of Y (R1 600 000) over the identifiable net assets of Y which X would
recognise if it acquired Y at this date (R1 500 000). Goodwill is therefore written off to this
implied value of R100 000.

4 How goodwill arises

4.1 Purchased or acquired goodwill
An existing business with a successful operating history normally has an infrastructure in
the form of a customer base, technical expertise, managerial expertise, and many other
intangible factors that create value over and above the amortised historical costs of its
identifiable assets. This goodwill can only be reported in the financial statements of a
business enterprise after a business combination transaction (Herring 2002). In the

108 Meditari Accountancy Research Vol. 13 No. 1 2005 : 105-120

 Wiese

accounting sense, goodwill can be thought of as a “premium” for buying a business. One
characteristic of goodwill that has emerged over the past century is that it cannot be
separated from the business. It cannot be sold without selling the associated business
(Fontanot 2003).

4.2 Inherent goodwill

According to Stickney and Weil (quoted by Cunningham, 2002), inherent goodwill is the
non-physical, non-current rights of an entity that give it a preferred or exclusive position in
the market. Unlike purchased goodwill, inherent goodwill is not subject to a market
transaction. It is therefore not capable of valuation in the same way as purchased goodwill,
and its value merely reflects the value to a particular party at a given point in time. In terms
of IAS 38, Intangible Assets (IASB 2004c), internally generated goodwill shall not be
recognised as an asset, because it is not an identifiable resource controlled by the entity and
it cannot be measured reliably. The recognition and capitalisation of inherent goodwill is
therefore largely conditional on the recognition of purchased goodwill (Tollington 1998).
This would tend to support Van Mesdag’s view of goodwill as an accountants’ acceptable
alternative in recognising inherent goodwill on the balance sheet (cited by Tollington
1998).

5 Amortisation
Conceptually, amortisation is a method to allocate the cost of goodwill over the periods it is
consumed. This is consistent with the approach taken with regard to other fixed assets that
do not have indefinite useful lives (IASB 2004d). Overpayment for the assets of an
acquired company generally reflects an expectation of high future earnings. Amortisation of
this overpayment ensures that the overpayment is matched with the expected future
earnings (Fontanot 2003).
Systematic amortisation with additional impairment testing acknowledges that the factors
that constitute acquired goodwill generally diminish in value over time, and that the related
costs are systematically charged to income over the useful life of the goodwill (EFRAG
2003). Although the useful life of goodwill cannot be predicted with a satisfactory level of
reliability, systematic amortisation provides an appropriate balance between conceptual
soundness and operationality at an acceptable cost. According to the respondents to ED 3,
Business Combinations (IASB 2002), this is the only practical solution to an intractable
problem (IASB 2004d).

5.1 Arguments against amortisation

Amortisation of goodwill is based on the assumption that goodwill is a wasting asset (that
is, finite), and thus ignores the fact that some kinds of goodwill can have an indefinite
useful life. The value of a business, and consequently of its goodwill, does not necessarily
wear out. It can be maintained or even improved by careful management and by cash
expenditure charged against the income stream. Such management action does not only
replace purchased goodwill with inherent goodwill, but maintains the value measured at the
date of acquisition. Reducing profits further through the application of regular amortisation
does not lead to a meaningful measurement of earnings (Njeke 1991).

Meditari Accountancy Research Vol. 13 No. 1 2005 : 105-120 109

Accounting for goodwill: The transition from amortisation to impairment – an impact assessment

In considering the comments on ED 3, the IASB observed that it is generally not possible
to predict the useful life of goodwill and the pattern according to which it diminishes.
Hence, the amount amortised in any given period can at best be described as an arbitrary
estimate of the consumption of goodwill during that period (IASB 2004d). The FASB
(2001b) argued that it is virtually impossible to predict accurately what the useful life of
goodwill would be and that amortisation is therefore not a faithful representation of the
pattern of decline (FASB 2001b).
Moehrle (2001) argues that goodwill is a wasting asset because it has a definite life, but
does not decline in value over a straight line for an observable period. The imposition of a
maximum life and a systematic write-down, irrespective of the efforts of management and
the ongoing business value, seems to be an overly mechanistic approach to judging the
possible erosion of goodwill value.
Most of the controversy about goodwill over the last four decades has centred on the
earnings drag of goodwill, rather than its usefulness (Waxman 2001). Financial statement
users in the USA have indicated that they do not regard goodwill amortisation as useful
information in investment analyses (FASB 2001b). For this reason, most analysts ignore
goodwill amortisation when they calculate their ratios (Hattingh 2002).

6 Impairment
One of the main arguments of the FASB in proposing the impairment approach was that it
would lead to improved financial reporting, because the financial statements of entities that
acquire goodwill would reflect the underlying economics of those assets better. As a result,
financial statement users would be better able to understand the investments made in those
assets and the subsequent performance of those investments (FASB 2001b).
The IASB (2004d), in considering the comments on ED 3, agreed with the FASB that
achieving an acceptable level of reliability in the form of representational faithfulness,
while at the same time striking some balance with what is practicable, was the primary
challenge faced in deliberating accounting for goodwill. They acknowledged that, if
goodwill is an asset, it must be true that goodwill acquired in a business combination is
being consumed and replaced by internally generated goodwill, provided the entity is able
to maintain the overall value of goodwill. They were, however, doubtful about the
usefulness of an amortisation charge that reflects the consumption of goodwill, whilst the
internally recognised goodwill replacing it is not recognised. They argued that a rigorous
and operational impairment test would provide more useful information to users of an
entity’s financial statements.
According to Moehrle (2001), a good impairment test promotes transparency, because
the trigger is a change in underlying economic or business conditions, not an arbitrary
period. As a result, reporting is based on current events that affect the business. If it is
properly managed, goodwill is an appreciating asset, and if it is not properly managed, the
impairment test will recognise any reduction in value.

6.1 Arguments against impairment

6.1.1 Capitalisation of internally generated goodwill
Acquired goodwill is generally an asset that is consumed over time and replaced by
internally generated goodwill. One of the main arguments of the respondents to ED 3 in

110 Meditari Accountancy Research Vol. 13 No. 1 2005 : 105-120

 Wiese

support of amortisation was that it prohibits the recognition of internally generated

goodwill, which is consistent with the general prohibition in IAS 38 on the recognition of
internally generated goodwill (IASB 2004d).
In some cases, the value of goodwill may appear not to decrease over time. This is
because the potential for economic benefits purchased initially is progressively replaced by
the potential for economic benefits resulting from subsequent enhancement of goodwill,
which is brought about by the ongoing expenditure of the business (Moodie 2000). The
operating policies, focus and levels of ongoing expenditure, which are, in turn, determined
by the new controllers, are factors that establish or increase the inherent goodwill of the
acquired company. Over time, these factors will have an impact on the value of the
business, and then the purchased goodwill will be replaced by inherent goodwill (Njeke
1991).
Implied goodwill is calculated as the excess of the recoverable amount of the unit to
which the goodwill has been allocated, over the fair value of the identifiable net assets that
the entity would recognise if it acquired that unit in a business combination on the date of
the impairment test. What is absurd about this algorithm, according to Ketz (2001b), is that
it assumes that the residual amounts relate solely to the goodwill purchased as a result of a
previous business combination. This computed goodwill, however, could easily be tainted
by internally generated goodwill generated by other activities, events, resources and social
arrangements. The calculated number may thus have little to do with the purchased
goodwill.
Internally generated goodwill is not recognised on balance sheets , since there is no
reliable measure of cost or value (Nobes & Norton 1996). It is therefore likely that many
companies have unrecorded goodwill associated with the reputation of their products,
service(s) and managerial expertise (Herring 2002). The impairment test does not
distinguish between acquired goodwill and this pre-existing goodwill of the company that is
being acquired, nor between acquired goodwill and the goodwill internally generated after
the combination. The result is a cushion that would shield many large acquirers from ever
having to recognise an impairment of goodwill. According to Moehrle (2001), analysts will
now have to face the challenge of recognising when the goodwill asset is wasting, and is
being replaced by internally generated goodwill, compared to instances in which value
remains related to past business acquisitions.

6.1.2 Effect on earnings

One of the most important intangible assets on a company’s balance sheet that can have a
material effect on earnings is goodwill. The capitalisation of goodwill without amortisation
may allow the most advantageous financial reporting figures due to net income not to be
reduced periodically, but could also result in more abuse than any other method (Fontanot
2003).
Annual systematic charges to goodwill are more objective than periodic reviews for
impairment. The latter would allow firms greater opportunities to manage their earnings
(Schoderbek & Slaubaugh 2001). The change could generate a boost to earnings per share
that could deceive the market into thinking that a company is doing better than anticipated,
causing the stock market to move higher (Basi & Penning 2002).
Companies generally do not want a goodwill write-off to distort income and cause a
negative earnings hit (Moehrle 2001). Many commentators strongly favour the impairment

Meditari Accountancy Research Vol. 13 No. 1 2005 : 105-120 111

Accounting for goodwill: The transition from amortisation to impairment – an impact assessment

model, simply because it will increase earnings and earnings per share (Waxman 2001). It
should be noted that this would not apply to headline earnings per share, as Circular 7/2002
(SAICA 2002) specifically excludes goodwill from the calculation of headline earnings.
Headline earnings per share has become the accepted earnings per share measure of many
South African listed companies, but is unique to South Africa and is not an accepted
earnings per share measure in global markets (Moodie 2000).
There may also be more volatility in reported income than under previous standards,
because impairment losses are likely to occur irregularly and in varying amounts (FASB
2001b).

6.1.3 Complexity
IFRS 3 puts its faith in a potentially unreliable and very complex impairment test. This was
the opinion of the two IASB members who dissented from the issue of IFRS 3 (IASB
2004e). The projection of future cash flows is difficult, especially in developing and
volatile industries (such as the “high tech” and telecommunications industries). The
question is whether fair presentation will be achieved in the recognising and measuring of
goodwill in view of the high dependency on the ability of an entity’s financial personnel to
predict future cash flows accurately (White 2003).
In its analysis of the main issues raised during field visits and roundtable discussions, the
IASB noted that its decision to adopt a non-amortisation approach for goodwill was
contingent on its ability to devise a sufficiently rigorous yet operational impairment test.
The Board noted that it was the second step of the proposed impairment test and the method
for measuring any impairment loss for the goodwill that caused the greatest concern for
both respondents and field visit participants (IASB 2003a).
In the USA, businesses indicated that they do not like the reporting unit requirement
because of the difficulty they have in determining the reporting unit level on which
goodwill is to be tested, as well as what specific methodology to use to calculate the fair
value of the reporting unit. Will the fair value be derived from an independent third party
valuation, an earnings model, a multiple of book value, or some other model? Stock prices
represent an ideal estimate, for accounting purposes, of fair value, because they are
objective and verifiable (Moehrle 2001). However, many finance managers believe that the
current price at which their company stock trades does not reflect its fair value and may
also not be representative of the fair value of the reporting unit as a whole (SmartPros
2001).
Non-amortisation of goodwill increases the reliance that must be placed on impairment
tests of those assets. Therefore IAS 36 requires additional disclosure to provide users with
information needed to evaluate the reliability of the estimates used by management.
Although the IASB has relaxed the disclosure requirements in IAS 36, based on the field
visit participants’ and respondents’ concerns that the proposed disclosures go beyond their
intended objective of providing users with relevant information for evaluating the reliability
of the impairment tests (IASB 2003b), IAS 36 still contains very rigorous disclosure
requirements.

6.1.4 Cost
The costs of the impairment tests are likely to be high and the benefits may be diminished
by their potential unreliability. For smaller companies, both quoted and unquoted, the costs

112 Meditari Accountancy Research Vol. 13 No. 1 2005 : 105-120

 Wiese

may outweigh any possible benefit. The proposed impairment test requires the fair
valuation of all assets and liabilities in a cash-generating unit that are potentially impaired.
Such a valuation will be expensive and time -consuming (White 2003).
To ensure compliance with SFAS 142 and to avoid unexpected charges, many companies
in the USA are paying more for professional valuation services to value goodwill and other
intangibles. George D. Shaw, the Boston-based managing director of Grant Thornton
Corporate Finance LLC, the accounting firm’s M&A advisory subsidiary (quoted by
Reason 2003), has commented that the adoption of SFAS 142 in the USA had been good for
the valuation business and that this trend would continue because ongoing testing is
required. Valuation firms offer a structured process and a paper trail, which may be useful
if a company’s valuation practices are challenged. The US Securities Commission (SEC)
released a review of the 2002 filings by Fortune 500 companies during February 2003
(cited by Reason 2003), noting that goodwill impairment was among the critical disclosures
that often seemed to be materially deficient in explanation or clarity. Among the additional
information that the SEC demanded were clearer descriptions of accounting policies for
measuring impairment, as well as better information on how reporting units are determined
and how goodwill is allocated to those units (Reason 2003).

6.1.5 Subjectivity
The impairment test is subject to a high degree of subjectivity and uncertainty, which may
make it no less arbitrary than amortisation.
Annual impairment testing is an onerous process, which requires companies to make
subjective decisions about whether goodwill has declined in value. The corporate executive
now has to absorb more questions because the recoverable amount of goodwill is one more
item that needs to be audited (Basi & Penning 2002).
The determination of the fair value of a unit and the detailed measurement of the implied
fair value of goodwill may be so subjective that the timing and amount of write-downs may
not always be independently verifiable (Waxman 2001). Ketz (2001b) comments that
managers will have the opportunity to deny impairments in many situations. He suggests
that executives will simply find those appraisers and consultants who will provide the
desired numbers leaving auditors with no benchmark by which to evaluate the treatments.
The difficulty is that no one can value goodwill with precision. Ketz (2001a) is convinced
that investment bankers, accountants and others will, for an appropriate fee, measure
goodwill, but compares this to gazing into crystal balls. This is because goodwill cannot be
sold separately and because there is no market for goodwill.

6.1.6 Different accounting treatments for goodwill and other assets

IFRS 3 does not differentiate goodwill in the same manner as IAS 38 differentiates other
intangible assets. In terms of IAS 38, intangible assets with finite useful lives should be
accounted for at cost, less any accumulated amortisation and any accumulated impairment
losses. Intangible assets with indefinite useful lives should not be amortised, but should be
tested for impairment annually.
Goodwill and other intangible assets that are similar in nature will thus be subject to
different accounting treatments, which will diminish comparability and reliability. All other
assets are subjected to an impairment test only if there is an indication that they are

Meditari Accountancy Research Vol. 13 No. 1 2005 : 105-120 113

Accounting for goodwill: The transition from amortisation to impairment – an impact assessment

impaired, according to IAS 36, whereas IFRS 3 requires goodwill to be tested for
impairment annually.
Goodwill is also subject to a two-step impairment test, whereas the other assets in IAS 36
that are written down to recoverable amounts are subjected to a one-step impairment test
only. According to the Institute of Chartered Accountants in England & Wales (ICAEW
2003), the screening test is consistent with the impairment test for the other assets in IAS 36
that are written down to recoverable amounts and no further testing is required.
In order to test goodwill for impairment, the recoverable amount of the investment in the
subsidiary has to be determined according to IAS 36. This implies that the investment in the
subsidiary must also be tested for impairment annually, in comparison with other
investments, which only have to be tested for impairment when there is an indication that
they are impaired.

7 Effect of the transition from amortisation to impairment

7.1 Effect of the change to a compulsory annual impairment test
for all goodwill
The change to an impairment test was, inter alia, intended to address the criticism against
previous rules, which required amortisation and an impairment test only if there was an
indication that goodwill may be impaired. These rules provided too much flexibility in the
measurement of goodwill and gave firms too much discretion in timing the write-off. A
study by Henning, Shaw and Stock (2004) suggests that in the USA firms delayed goodwill
write-offs before the enactment of SFAS 142, since a disproportionately large number of the
firms in their sample recognised impairments upon adopting SFAS 142. The transition to
the new standard in the USA might have triggered the recognition of impairments by
reducing the flexibility of previous Generally Accepted Accounting Practice.
In addition, transition period write-offs significantly exceeded predicted write-offs,
suggesting that firms might have used the transition period to minimize future write-offs.
This could be based on the belief that the market would view such impairments as relating
to a past problem, and as having no impact on the current firm value (Henning et al. 2004).
The analysis set out in Table 1 of companies in the USA with high impairments in 2002,
the year of adoption of SFAS 142, in comparison with significantly lower amortisation in
2001 and followed by lower impairments in 2003, may provide proof of these statements.
Table 1 Impairment versus the amortisation charges of USA companies
2003 2002 2001
Company Impairment Impairment Amortisation
US $ m US $ m US $ m
Time Warner Inc (2004) - 100 5
Comfort Systems Inc (2004) 3 229 11
Instinet Group Inc (2004) 22 571 8
Boeing Co (2004) 931 2 410 158
Omnova Solutions Inc (2004) 41 104 9
Storage Computer Corp (2004) - 14 3
Exide Technologies (2004) 37 105 15

114 Meditari Accountancy Research Vol. 13 No. 1 2005 : 105-120

 Wiese

7.2 Effect on earnings

Basi and Penning (2002) note that the one-time charge-offs that may be made after a
change to impairment could further depress already weak earnings in the financial records
of some companies. In the example in Section 3.4 of this article, Company X would have to
record an impairment charge of R300 000 at the end of 20x3 under IFRS 3, in comparison
with an annual amortisation charge of R80 000 under IAS 22.
One estimate in the USA in 2001 was that nearly two-thirds of major companies would
have to record some impairment of goodwill on adoption of SFAS 142 (Investor Relations
Business 2001). According to Harrington (2003), losses due to the impairment of goodwill
and other tangible assets decreased the combined reported profits of the Fortune 500
companies from $260 billion to $69.6 billion in the USA in 2002, due to notable charges by
some firms in the changeover year of 2002. Henning et al. (2004) also predicted a
significant increase in the number of recognised impairments in the transition year. In their
examination, 205 of the 681 sample firms in the USA that announced the results of their
transition impairment tests in the first quarter of fiscal 2002 reported an impairment of
goodwill on adoption of SFAS 142.
The significant effect of the high impairment charges on the profits of certain companies
in the USA in 2002, the year of adoption of SFAS 142, is set out below.
Table 2 Impairment charges and net (loss)/profit of USA companies
Impairment Net (loss)/profit
Company US $ m US $ m
2002 2002 2001
Time Warner Inc (2004) 100 (99) (5)
Comfort Systems Inc (2004) 229 (209) 13
Instinet Group (2004) 571 (735) 145
Boeing Co (2004) 2 410 492 2 827
Omnova Solutions Inc (2004) 104 (136) (7)
Storage Computer Corp (2004) 14 (28) (13)
Exide Technologies (2004) 105 (304) (165)

7.3 Effect of previous amortisation period on transitional

impairment cost
Amortisation results in a very small effect on the profitability of the acquiring company,
especially where it is written off over a long period (Basi & Penning 2002).
This was confirmed by the significant effect of the changeover to an impairment test in
2002 on companies in the USA where goodwill had been amortised over long periods
before. The companies recorded in Table 2 all amortised goodwill over periods longer than
15 years prior to 2002. In the case of companies that previously amortised goodwill over
periods shorter than five years, the effect was insignificant. The amortisation cost for the Zi
Corporation (2004), for example, was Canadian $1.08 million in 2001, compared to an
impairment charge of Canadian $1.98 million in 2002. For Cognos Inc (2004), the
amortisation cost was US $2.5 million in 2001 and US $4.4 million in 2002, compared to
no impairment charge for the year ended 28 February 2003, when SFAS 142 was
implemented.

Meditari Accountancy Research Vol. 13 No. 1 2005 : 105-120 115

Accounting for goodwill: The transition from amortisation to impairment – an impact assessment

8 Characteristics of companies likely to be affected most by

the transition to impairment
8.1 Significant abnormal returns
The study by Henning et al. (2004) suggests that when they adopted SFAS 142, firms with
significant abnormal returns impaired a greater proportion of their goodwill. Of the
transition firms with significant negative abnormal returns in their sample, 93% recognised
goodwill impairments. The following cases also suggest evidence of this proposition.
Table 3 Goodwill assets and net (loss)/profit of US companies
Net loss after Net loss after
Goodwill asset impairment amortisation
Company US $ m charge charge
US $ m US $ m
2002 2001 2002 2001
Time Warner Inc (2004) 37 127 (99) (5)
Omnova Solutions Inc (2004) 41 143 (136) (7)
Storage Computer Corp (2004) 3 16 (28) (13)

8.2 Significant goodwill assets

A balance sheet that consists substantially of intangible assets holds high risks for investors
and bankers, as was demonstrated in the examples of W&A and Tollgate (Finansies &
Tegniek 1995).
The study by Henning et al. (2004) suggests that goodwill is an important component of
firm value and confirms that, on adopting SFAS 142, firms with older goodwill impaired a
greater proportion of their goodwill. The average age of the impaired goodwill of the firms
with significant negative abnormal returns (see Section 8.1) was 10.1 years.
In the case of Comfort Systems Inc, a leading provider of heating, ventilation and air
conditioning services in the USA, goodwill as a percentage of total assets dropped from
50% in 2001 to 31% in 2002. The 2002 results included a charge of US $229 million
relating to the adoption of SFAS 142. A significant part of the company’s growth from 1997
to 1999 resulted from acquisitions leading to significant goodwill asset amounts that
required reduction under the new, more stringent standards of SFAS 142 (Business Wire
2002).
Other companies which confirmed this trend on adoption of SFAS 142 were Time
Warner Inc (2004) (goodwill as a percentage of the total assets dropped from 61% in 2001
to 32% in 2002), the Instinet Group (2004) (goodwill as a percentage of ot tal assets
dropped from 5% in 2001 to 0% in 2002), the Boeing Co (2004) (goodwill as a percentage
of total assets dropped from 10% in 2001 to 5% in 2002), the Storage Computer Corp
(2004) (goodwill as a percentage of total assets dropped from 47% in 2001 to 30% in 2002)
and Omnova Solutions Inc (2004) (goodwill as a percentage of total assets dropped from
22% in 2001 to 9% in 2002).

8.3 Number of cash-generating / reporting units

The market value of a company with only one reporting unit that acquires another company
would have to fall below the company’s book value before the goodwill acquired would be

116 Meditari Accountancy Research Vol. 13 No. 1 2005 : 105-120

 Wiese

subject to further impairment testing (for example, Cognos Inc acquired Adaytum in 2003
for US $157 million in cash, resulting in goodwill of US $154 million, but has only one
reporting unit – the goodwill will therefore only be impaired if the book value of Cognos
falls below the market value). In a case study done by Frucot, Jordan and Lebow (2004),
the acquiring company also consisted of only one reporting unit. Here, a drop in the share
price of the acquiring company’s shares due to market fluctuations resulted in an
impairment of 25% of the goodwill, arising from the acquisition of another company six
years earlier, although the combined company was very profitable and paid very high
dividends.
Large companies with multiple units would thus seem more vulnerable, although those
units can often provide a substantial cushion against impairments (Reason 2003).

9 Conclusion
The accounting treatment of goodwill has been a matter of concern to accountants and
accounting standards committees for more than a decade. International standard setters
agreed in the early 1990s that goodwill should be recognised as an asset and amortised over
its useful life. The issue was re-introduced at the start of the 21st century, resulting in the
FASB replacing the amortisation requirement with an impairment testing approach in 2001.
The IASB also adopted this approach with the acceptance of IFRS 3 in 2004.
The examination of the nature of goodwill in Section 4 of this article has shown that
goodwill exists because the fair value of a business as a going concern exceeds the fair
value of its identifiable net assets. If this can be attributed to expectations of high future
earnings, amortisation would result in the matching of the cost of the goodwill with
expected future earnings from the acquisition. If the excess paid is due to internally
generated intangibles that are not recognised as assets, or expected synergies from
combining the businesses of the companies involved, the impairment approach would
permit the capitalisation of these internally generated intangible assets. This could result in
companies’ never having to recognise an impairment of the acquired goodwill, which
would give companies growing through acquisitions the advantage of recognising inherent
goodwill as an asset.
A comparison between amortisation and impairment revealed that amortisation is well
understood and that it is a well-established principle consistent with the approach taken to
other tangible and intangible assets with finite useful lives, but that it ignores the fact that
some forms of goodwill can have an indefinite useful life. The impairment test approach,
on the other hand, seems to involve a very different and much more complex, costly and
subjective accounting process. Fair value measurements can be difficult to do and may need
to be done by valuation experts. In cases where stock prices are used as estimates for fair
values, goodwill might have to be impaired, although the current price at which the
company’s stock trades might not reflect its fair value and thus not be representative of the
fair value of the reporting unit as a whole. It may therefore be more appropriate to treat
goodwill in the same way as other intangible assets: amortising finite life goodwill and
subjecting it to an impairment test only where there is an indication that it has been
impaired, while subjecting indefinite life goodwill, where the life of goodwill is difficult to
assess, to an annual impairment test. This will ensure that accounting for goodwill takes
into account the nature of goodwill as an intangible asset, as well as the reasons for its
existence, without unnecessarily subjecting it to a complex and costly annual impairment
test.

Meditari Accountancy Research Vol. 13 No. 1 2005 : 105-120 117

Accounting for goodwill: The transition from amortisation to impairment – an impact assessment

An examination of the effect that the transition from amortisation to impairment has had
on companies in the USA suggests that many South African companies may have to record
an impairment of goodwill when they adopt IFRS 3, especially those who were seen to be
“growing” because they were acquiring new businesses and merging them with their own
in the late 1990s and early 2000s.
Transition period write-offs significantly exceeded the predicted write-offs on adoption
of SFAS 142 in the USA in 2002, suggesting that firms might have used the transition
period to minimize future write-offs. It is especially where companies delayed goodwill
write-offs before the adoption of SFAS 142, that high impairment charges in the transitional
year had a significant impact on their earnings. Companies where goodwill was amortised
over long periods under the old accounting rule, as well as companies with significant and
older goodwill assets, had to impair a greater proportion of their goodwill on adoption of
SFAS 142.
Finally, it appears that amortisation provided the opportunity for goodwill to have a very
small and systematic effect on the profitability of the acquiring company. Impairment
losses, on the other hand, are likely to occur irregularly and in varying amounts, causing
volatility in reported earnings. Although impairments are a non-cash charge, they obviously
have a very big negative effect if a company is writing off a substantial amount of goodwill
because an acquisition does not live up to expectation. Overstating goodwill can make an
otherwise marginally lucrative deal look worthwhile, but companies will now have to be
careful with deals based on “back-of-the-envelope” calculations, as more rigorous
accounting now needs to be applied to the goodwill acquired.

Bibliography
Basi, B.A. & Penning, S.M. 2002. Let the investor beware. Supply House Times,
May:70-74.
Business Wire. 2002. Comfort systems USA reports first quarter results; sees improving
business trends; adopts new goodwill accounting standard. Business Wire, 14 May.
http://www.findarticles.com. Accessed: 9 September 2004.
Cunningham, P. 2002. A Framework for the Analysis of an Entity’s Market Value. Paper
presented at the Southern African Accounting Association Biennial International
Conference, Port Elizabeth, 26-28 June.
European Financial Reporting Advisory Group (EFRAG). 2003. Comment on IFRS 3
Business Combinations and IAS 36 and IAS 38, Accounting Education, April.
http://www.accountingeducation.com. Accessed: 10 September 2004.
Financial Accounting Standards Board (FASB). 2001. Goodwill and other intangible assets
– Statement No 142. June. Norwalk, Connecticut: FASB.
Financial Accounting Standards Board (FASB). 2001. Summary of Statement No 142.
http://www.fasb.org/st/summary/stsum142.shtml. Accessed: 20 August 2004.
Finansies & Tegniek . 1995. Regte plek op inkomstestaat? Finansies & Tegniek , 47,
Junie:15
Fontanot, P. 2003. How good is goodwill – you decide. Accountancy SA, May:9-11.

118 Meditari Accountancy Research Vol. 13 No. 1 2005 : 105-120

 Wiese

Frucot, V.G., Jordan, L.G. & Lebow, M.I. 2004. A & B Companies: Impairment of
Goodwill. Issues in Accounting Education, 19(3), August:369-376.
Harrington, A. 2003. Honey, I shrunk the profits: accounting made a bad year look a whole
lot worse. Fortune, 147, 14 April.
http://www.fortune.com/fortune/information/Magarchive. Accessed: 10 August 2004.
Hattingh, C. 2002. We’re old fashioned? Accountancy SA, July:27-28.
Henning, S.L., Shaw, W.H. & Stock, T. 2004. The amount and timing of goodwill write-
offs and revaluations: evidence from U.S. and U.K. firms. Review of Quantitative
Finance & Accounting, 23(2), September:99.
Herring, H.C. 2002. Business combinations & international accounting. 1st edition. Mason,
Ohio: Thomson (South-Western).
International Accounting Standards Board (IASB). 1993. Business Combinations, IAS 22.
London: IASB.
International Accounting Standards Board (IASB). 2002. Business Combinations, ED 3,
December. London: IASB.
International Accounting Standards Board (IASB). 2003a. IASB update on board decisions
on International Financial Reporting Standards. July. http://www.iasb.org.uk.
International Accounting Standards Board (IASB). 2003b. IASB update on board decisions
on International Financial Reporting Standards. October. http://www.iasb.org.uk.
International Accounting Standards Board (IASB). 2004a. Business Combinations, IFRS 3.
March. London: IASB.
International Accounting Standards Board (IASB). 2004b. Impairment of assets, IAS 36.
March. London: IASB.
International Accounting Standards Board (IASB). 2004c. Intangible assets, IAS 38.
March. London: IASB.
International Accounting Standards Board (IASB). 2004d. Basis for conclusions on IFRS 3
Business Combinations, IFRS 3 BC. March. London: IASB.
International Accounting Standards Board (IASB). 2004e. Dissenting opinions on IFRS 3,
IFRS 3 DO. March. London: IASB.
Institute of Chartered Accountants in England & Wales (ICAEW). 2003. Memorandum of
comment to the IASB concerning Exposure Draft 3, Business Combinations, Tech
12/03, April. http://www.icaew.co.uk/library. Accessed: 20 August 2004.
Investor Relations Business. 2001. Annual impairment debuts. 12 November. New York:
Securities Data.
Ketz, J.E. 2001a. Goodwill gunked. SmartPros, 22 January.
http://accounting.smartpros.com. Accessed: 9 September 2004.
Ketz, J.E. 2001b. The accounting cycle: wash, rinse and spin, FASB’s Exposure Draft on
accounting for goodwill. SmartPros, 12 March. http://accounting.smartpros.com.
Accessed: 9 September 2004.
Moehrle, S.R. 2001. Say good-bye to pooling and goodwill amortisation. Journal of
Accountancy, September:31.

Meditari Accountancy Research Vol. 13 No. 1 2005 : 105-120 119

Accounting for goodwill: The transition from amortisation to impairment – an impact assessment

Moodie, C. 2000. A growing pain in going global. Accountancy SA, November/

December:20-21.
Njeke, J.N. 1991. Accounting for goodwill revisited. Accountancy SA, July:188-192.
Nobes, C & Norton, J. 1996. International variations in the accounting and tax treatments of
goodwill and the implications for research. International Accounting, Auditing &
Taxation, 5(2):179-196.
Reason, T. 2003. Goodwill to all pieces: are companies properly valuing and assigning
acquired intangibles to business units? CFO, Magazine for Senior Financial Executives,
July. http//www.cfo.com/article.cfm. Accessed 20 August 2004.
South African Institute of Chartered Accountants (SAICA). 1999. Business combinations,
AC 131. June. Johannesburg: SAICA.
South African Institute of Chartered Accountants (SAICA). 2002. Headline earnings.
Circular 7/2002, December. Johannesburg: SAICA.
Schoderbek, M.P. & Slaubaugh, M.D. 2001. The FASB exposure draft on accounting for
business combinations and intangible assets: an instructional assignment. Journal of
Accounting Education, 19:265-281.
SmartPros Editorial Staff. 2001. FEI teleconference reveals goodwill confusion. SmartPros,
28 February. http://accounting.smartpros.com. Accessed: 9 September 2004.
Tollington, T. 1998. Separating the brand asset from the goodwill asset. Journal of Product
& Brand Management, 7(4):291-304.
Waxman, R.N. 2001. Goodwill convergence. CPA Journal, 71, October:18-25.
White, A. 2003. Proposals for business combinations. Accountancy SA, June:10-11.
Wiese, A. 2003. Treating the credit difference that arises from an acquisition as negative
goodwill. Meditari Accountancy Research, 10:295-309.

Annual financial reports and US Securities Commision (SEC) filings.

Boeing Co. http//www.boeing.com and http//www.sec.gov/cgi-bin/srch-edgar. Accessed:
20 November 2004.
Cognos Inc. http//www.cognos.com and http//www.sec.gov/cgi-bin/srch-edgar. Accessed:
20 November 2004.
Comfort Systems Inc. http//www.comfortsys.com and http//www.sec.gov/cgi-bin/srch-
edgar. Accessed: 20 November 2004.
Exide Technologies. http//www.exide.com and http//www.sec.gov/cgi-bin/srch-edgar.
Accessed: 22 November 2004.
Instinet Group Inc. http//www.instinet.com and http//www.sec.gov/cgi-bin/srch-edgar.
Accessed: 22 November 2004.
Omnova Solutions Inc. http//www.omnova.com and http//www.sec.gov/cgi-bin/srch-edgar.
Accessed: 22 November 2004.
Storage Computer Corp. http//www.storage.com and http//www.sec.gov/cgi-bin/srch-edgar.
Accessed: 23 November 2004.
Time Warner Inc. http//www.timewarner.com and http//www.sec.gov/cgi-bin/srch-edgar.
Accessed: 23 November 2004.
Zi Corporation. http//www.zicorp.com. Accessed 23 November 2004

120 Meditari Accountancy Research Vol. 13 No. 1 2005 : 105-120