An investigation of goodwill impairment testing practices in Australia public companies

Transcription

1 Swinburne University of Technology Faculty Business and Enterprise An investigation of goodwill impairment testing practices in Australia public companies Merina Farida Bachelor of Business (Honours) Supervisor: Dr Geoff Speight Submitted: November 2008

2 DECLARATION I assert, to best of my knowledge, that this thesis does not contain any work that has been previously submitted for a degree at any educational institution or for publication without due reference made. Merina Farida 7 November 2008 i

3 ACKNOWLEDGEMENTS First of all, I want to thank my beloved family, even though you all were not here during preparation of my thesis. By knowing that you will always be there for me, helped me went through this one year. I would like to thank my supervisor, Geoff Speight. for his invaluable guidance. Thank you very much for guiding me and for making me think harder. I would also like to thank my Research Methodology lecturer, Sharon Grant, for preparing me in writing of my thesis. Thanks a lot for your time in assisting me to write better thesis. Last but not least, I would like to thank my flatmate, Angeline Elias, for being there. I really appreciate your cheering attitude and for making this year easier and more fun for me. ii

4 Table of Contents DECLARATION i ACKNOWLEDGEMENTS ii Table of Contents iii List of Tables...v ABSTRACT vi 1.0 INTRODUCTION Statement of Study Focus Definition of Key Terms Background 4 1.4Review and Synthesis of Relevant Literature Introduction to Business Combinations and Goodwill A Brief History and Advantages or Disadvantages of Goodwill Impairment Overview of Goodwill Impairment Method for Calculating Recoverable Amount: FVLCS Method for Calculating Recoverable Amount: VIU Formulation of Study Objectives and Hypotheses Study Objectives Hypotheses RESEARCH METHODOLOGY Research Design Description/Rationale for Research Methodology Sampling Procedure Data Analysis FVLCS Method VIU Method...18 iii

5 2.4 Summary of Research Methods Limitations and Constraints of Research Method DATA ANALYSIS AND DISCUSSION Preferred Method Adherence to AASB Guidelines Consistency of Impairment Testing Practices Pre-tax Discount Rates Growth Rate DISCUSSION AND CONCLUSION Impairment Testing Method Rate of Risk Capital Goods Industry Commercial Supplies and Services Information Technology (IT) Industry Growth Rates Capital Goods Industry Commercial Supplies and Services Information Technology (IT) Industry..42 References and Bibliography..43 Appendix.46 iv

6 List of Tables Table 1: Definition of Key Terms...2 Table 2: Capital Goods Industry Data..20 Table 3: Commercial Services and Supplies Industry Data...22 Table 4: Information Technology (IT) Industry Data..25 Table 5: Rates of Risk Used by Companies in Capital Goods Industry...27 Table 6: Rates of Risk Used by Commercial Services and Supplies Industry.29 Table 7: Rates of Risk Used by Companies in IT Industry..30 v

7 ABSTRACT In 2005, Australian Accounting Standard Board (AASB) adopted International Financial Reporting Standards (IFRS) to apply impairment testing on acquired goodwill, instead of amortising it on straight line basis. The Australian companies can choose either fair value less cost to sell (FVLCS) method and value in use (VIU) method. All companies under this study (consisted of 18 companies) choose to use VIU method to test whether their goodwill is impaired. The 18 companies were divided into three industries, i.e. capital goods industry, commercial services and supplies industry and information technology (IT) industry. In order to investigate whether the companies impairment testing practices are consistent with the standard and with each other, comparison between the companies (using pretax discount rates) and in the company itself (using growth rate) were undertaken. For consistency, only companies in same industry were compared with each others. Overall, based on information provided in the companies annual reports, impairment testing practice (VIU method) in Australian companies is consistent with the standard guidelines. Furthermore, the companies practices in capital goods industry and commercial services and supplies industry are consistent with each other. However, companies practices in IT industry are not consistent with each other. This may due to the fact that companies in IT industry have their own competitive advantage which differ them from each other. vi

8 1.0 INTRODUCTION 1.1 Statement of Study Focus Acquired goodwill, as defined in Australian Accounting Standards, is the excess of consideration paid over the fair value of company acquired. Acquired goodwill represents the economic benefits expected to arise from unidentifiable assets of company acquired. Acquired goodwill is subject to impairment testing which defined as process of identifying whether the value of an asset or a group of assets has been impaired. Australian Accounting Standards provide guidelines regarding how to conduct impairment testing for an asset and/or a group of assets. However, since the guidelines are merely guidelines, how the testing is done may depend on management discretion. The guidelines identify two different methods, namely fair value less cost to sell (FVLCS) method and value in use (VIU) method. Companies with acquired goodwill in Australia are required, by the Australian Accounting Standard Board standard no. 136 (AASB 136) - Impairment of Assets, to choose one of these methods in conducting impairment testing of acquired goodwill. This thesis aims to investigate differences in (acquired) goodwill impairment testing in publicly listed Australian companies, in particular, differences in the extent to which companies adhere to Australian Accounting Standard guidelines. This study will compare companies impairment testing practices against the standard guidelines and determine whether the companies practices are consistent with the guidelines and with each other. Furthermore, the thesis examines which method is mostly used by the companies. Another purpose of this research is to investigate whether it is necessary for the Australian Accounting Standard Board (AASB) should consider providing more detailed guidelines for VIU method and whether to abolish FVLCS method, or vice versa. Page 1 of 47

9 1.2 Definition of Key Terms This study contains a considerable amount of technical terms from the accounting area. Key terms used throughout the thesis include: acquired goodwill, cash generating unit (CGU), fair value less cost to sell (FVLCS), value in use (VIU), recoverable amount, and impairment testing. The terms are defined in Table 1 below. Table 1 Definition of Key Terms Term Definition Source of Definition Acquired Goodwill CGU FVLCS An asset representing the future economic benefits arising from other assets acquired in a business combination that are not individually identified and separately recognised. It is normally calculated as the excess of cost of business combination paid over the fair value of net assets acquired. The smallest identifiable group of assets that generates cash inflows that are largely independent of cash inflows from other assets or groups of assets Goodwill must be allocated to CGU for the purpose of impairment testing. The amount obtainable from the sale of an asset or CGU in an arm s length transaction between knowledgeable, willing parties, less the costs of disposal. AASB 3 Business Combinations AASB 136 Impairment of Assets AASB 136 Impairment of Assets Page 2 of 47

10 VIU Recoverable Amount Impairment Testing The present value of the future pretax cash flows expected to be derived from an asset or CGU. It s required by the standard to apply appropriate discount rate to the cash flows. The higher of an asset s or cashgenerating units (CGU) fair value less cost to sell (FVLCS) and its value in use (VIU). However, if either of these amounts exceeds the asset s carrying amount, the asset is not impaired and it is not necessary to estimate the other amount. An asset is impaired if the asset s carrying amount exceeds the recoverable amount of the asset, as the asset s value should be written down to its recoverable amount. Therefore, impairment testing is the process to identify whether the recoverable amount of an asset is less than its carrying amount. AASB 136 Impairment of Assets AASB 136 Impairment of Assets AASB 136 Impairment of Assets Page 3 of 47

11 1.3 Background In general, there are two methods of growth for a company. It can grow organically, i.e. growing by improving and/or developing itself internally, and/or it can grow by acquiring other companies, i.e. business combination, which eventually will result in the acquiring company expanding its business and market share. Companies nowadays are growing by a combination of both. Per AASB 3, when a company purchases another company, acquired goodwill (from this point onwards, referred as goodwill ) will arise if the purchase consideration given exceeds the fair value of net assets acquired. The amount of the excess is the value of the goodwill. Practically, the excess represents the willingness of the acquirer to pay a premium due to qualities of the acquired company, e.g. efficient management, competent staff, and considerable amount of loyal customers. Goodwill is recognised under the Australian Accounting Standards as an intangible asset (despite the fact it is not individually identified and separately recognised) with assumption that it has indefinite useful life. Companies are required by AASB 136 to test their intangible assets annually to see if the value has declined, irrespective of whether there is an indication of impairment. AASB 136 also requires companies to recognise impairment loss in their statement of comprehensive income (formerly known as profit and loss (P&L) statement) if the carrying amount of any asset exceeds the asset s recoverable amount. Therefore, after goodwill is allocated to cash generating unit, the company is required to do impairment testing on the goodwill allocated to cash generating unit on an annual basis. If there is any impairment loss, the amount of loss should be deducted against the value of the goodwill. Par. 104 of AASB 136 states that if the amount of the impairment loss exceeds the carrying amount of the goodwill, the excess is allocated to other assets of the unit on a pro-rata basis based on their carrying amount. As stated above, an entity can choose either FVLCS method or VIU method to calculate the recoverable amount of the cash generating unit. Page 4 of 47

12 1.4 Review and Synthesis of Relevant Literature Introduction to Business Combinations and Goodwill Per AASB 3, business combination is the event of combining two or more businesses where one party (the acquirer) obtains control of the combination. According to Seetharaman, Sreenivasan, Sudha and Yee (2005), in business combination, goodwill represents the willingness of acquirer company to pay in excess of the value of net assets stated in the balance sheet. Nethercott and Hanlon (2002) also state that goodwill is the remaining balance of purchase consideration (price paid) for a company after all its identifiable assets have been valued and deducted from the purchase consideration. For example, Company A acquires Company B. Company A buys Company B for the amount of A$1 million. However, the net assets value of B is A$900,000. Hence, A will report goodwill asset of A$100,000 in its financial statement. Seetharaman et al. (2005) state that, unlike other assets, goodwill is intangible and not separately measurable. This means it is not possible for an acquiring company to acquire goodwill without acquiring the whole or a considerable portion of the acquired company as goodwill cannot be separated from the company. According to Badawi and Dorata (2007), goodwill is a remaining value which is computed as the difference between the fair value of the acquired company as a whole and the fair value of the net assets actually acquired. Furthermore, Nethercott and Hanlon (2002) argue that goodwill represents the assets that are not capable of being individually identified and separately recognised. That is, goodwill represents the future benefits the acquirer company expects to derive, mainly through the efficiency and effectiveness of the acquired company which cannot be separately recognised as an asset in the balance sheet (generally future benefits would include superior operating teams and market penetration). Duangploy, Shelton and Omer (2005) state that research has shown that the market perceives goodwill as an asset. Page 5 of 47

13 1.4.2 A Brief History and Advantages or Disadvantages of Goodwill Impairment Accounting for goodwill changed in Australia for all reports produced on or after the 1 January 2005 when Australia adopted the International Financial Reporting Standards (IFRS) (Wines, Dagwell and Windsor 2007). Goodwill value was previously subject to mandatory amortisation over a maximum 20-year period which would be completely written off its value over a period of time. Lamond (1995, p.68) (cited in Wines et al. 2007, p.865) argues that there is no explanation for the magical 20 year selection of the maximum amortisation period for goodwill in Australia. However, after the adoption of the IFRS, goodwill acquired in a business combination is no longer amortised but shall be tested for impairment whenever there is an indication that its value may have been impaired or at least annually (Wines et al. 2007). Wines et al. (2007, p.868) argued that the overall advantages of the adoption, from financial statements perspective, are that the goodwill value will be more closely aligned to a real assessment of asset value and a real economic decline in value. Donnelly and Keys (2002) stated that, unlike goodwill amortisation, goodwill impairment testing can adequately reflect a decline in value of goodwill. The change provides for more valid financial statements, but also requires accountants to select proper methods to estimate whether there is any impairment losses ( Lander and Reinstein 2003). The change from amortisation to annual impairment testing improves a new and continuous responsibility on management to estimate the recoverable amount of CGU of which the goodwill belongs, as well as a new burden on auditors, regulatory bodies, and investors to evaluate management s estimation (Hayn and Hughes 2006). Wines et al. (2007) suggested the requirement for more estimation of goodwill value may introduce increased uncertainty and decreased transparency, as the new goodwill treatment relies on professional judgement by financial reports preparers and auditors. Page 6 of 47

14 1.4.3 Overview of Goodwill Impairment The purpose of testing goodwill for impairment annually is to determine whether it continues maintaining its value (Massoud and Raiborn 2003). Impairment exists when the carrying amount of goodwill exceeds its implied recoverable amount, and if there is any impairment loss, it is stated as an expense item in the income statement (Duangploy et al. 2005). The recoverable amount of CGU is its higher value calculated under FVLCS and VIU methods (Romano 2008). If the value calculated under one method is higher than the CGU s carrying amount, then the other value does not have to be calculated (Shoaf and Zaldivar 2005). Romano (2008) suggested that as goodwill does not generate cash inflows independently of other assets, it cannot be tested for impairment independently. Furthermore, for the purpose of impairment testing, goodwill acquired in business combination should, from business combination date, be allocated by the acquirer company to its CGU (Shoaf and Zaldivar 2005). Hayn and Hughes (2006) argued that identifying CGU and allocating goodwill to them is one of the most difficult tasks. This is being so, as discussed by Shoaf and Zaldivar (2005), the acquirer should allocate the goodwill to CGU expected to benefit from the business combination from which the goodwill is derived. Shoaf and Zaldivar (2005) stated that, in conducting goodwill impairment testing, the carrying amount of CGU is compared with its recoverable amount. An impairment loss is recognised for goodwill when its CGU recoverable amount is less than its CGU carrying amount (Wines et al. 2007). Accordingly, goodwill value must be written down by an amount equivalent to amount of which the CGU s carrying amount exceeds its recoverable amount (Shoaf and Zaldivar 2005). However, if the carrying amount of goodwill allocated to the CGU does not cover the impairment loss, the excess of impairment loss is allocated to other assets in the CGU on a pro rata basis (Wines et al. 2007). Page 7 of 47

15 1.4.4 Method for Calculating Recoverable Amount: FVLCS According to Romano (2008), the best evidence in determining FVLCS of a CGU is the CGU s binding sale agreement less any related costs to sell the CGU (excluding finance cost, income tax expense, and costs already recognised as liability). The FVLCS method is the only appropriate valuation of recoverable amount of a CGU held for sale under binding sale agreement (Olde 2007). However, if there is an active market for CGUs instead of a binding sale agreement, then the recent transaction or market price (less costs to sell) for similar CGUs within the same industry might be the best evidence in determining the recoverable amount of CGUs under FVLCS method. Barth and Landsman (1995) argued that there is lack of availability of market prices for all assets (including CGU) since not all assets are actively traded. Furthermore, if both binding sale agreement and transaction price for similar CGUs within the same industry are not available, companies are allowed to make an estimation (based on best available information) for the CGU s selling price less costs to sell (Ivory 2006). However, this information might not be readily available and its estimation might be costly (Shoaf and Zaldivar 2005). Nevertheless, the AASB 136 does not prescribe how the estimation should be calculated (Ivory 2006). It is suggested that more specific information on the performance of the CGU is needed to assess the fair value of goodwill and its change over time (Hayn and Hughes 2006). Ivory (2006) argued that FVLCS can be divided into its component parts, namely fair value and costs to sell. Fair value is determined from an estimation of price obtainable from the sale of asset(s) in an arm s length transaction between willing parties, i.e. the seller and the buyer (Shoaf and Zaldivar 2005). The fair value computed for CGU in the current year is allowed to be carried forward to next years if no significant change occurs (Duangploy et al. 2006). According to Barth and Landsman (1995), when the market is not perfect or complete, fair values are ambiguously unique as, alternatively, it may be calculated using three concepts; being entry value, exit value and value in use which are not likely to be equal. Measurement of fair value involves using models, estimations and assumptions (Schipper 2003). This statement is also supported by Wines et al. (2007), who argued that estimation of fair value is often based on subjective assumptions and judgement Page 8 of 47

16 which can lead to the determination of goodwill value (or CGU s recoverable amount) resulting in wide variations Method for Calculating Recoverable Amount: VIU Donnelly and Keys (2002) define VIU as present value of estimated future cash flows expected to arise from continuous use of an asset and from its disposal at the end of its useful life. Hence, the calculation of VIU should reflect an estimation of future cash flows based on reasonable and supportable assumptions (for example, recently approved budgets and forecasts), and VIU should also reflect the pre-tax discount rate which is consistent with current market assessments of the time value of money and the (rate of) risks specific to the asset or CGU (Ivory 2006). Companies may use a steady or declining growth rate that is consistent with that of the product, industry, or country in calculating the future cash flows projection of the CGU (Romano 2008). The projected future cash flows must exclude any inflows and outflows expected to arise from future asset improvements (Ivory 2006). Furthermore, according to Ivory (2006), if market pre-tax discount rate plus asset-specific risk rate is not available, a surrogate must be used based on other rates such as the company s own weighted average cost of capital (WACC), the company s incremental borrowing rate, or other market borrowing rates. Even though WACC fluctuates with the entity s capital structure, companies often assume a constant WACC based on the firm s target debt to equity ratio (Lander and Reinstein 2003). According to Barth and Landsman (1995), the VIU concept has a long history in the accounting literature and its estimation is often difficult because it involves prediction of future cash flows, selection of appropriate discount rate, and knowledge of asset synergy. An estimation of recoverable amount under VIU method can be maximised by estimating annual cash flows and salvage value (selling price of an asset at the end of its useful life) of the CGU at the maximum amount possible, and adopting the lowest possible discount rate (Wines et al. 2007). Page 9 of 47

17 As the future cash flows can be maximised at the maximum amount possible, the estimation of future cash flows might not be verifiable and valuation based on them are likely to be manipulated (Watts 2003). According to Massoud and Raiborn (2003), these incidents happen as companies might choose to recognise impairment losses in the manner that selectively fits their operating results. Besides that, the amount of goodwill impairment loss has a relatively large impact on total assets of a company as it can reduce the company assets book value. This might increase the debt-to-equity ratio, signalling the rising possible insolvency of the company which consequently will depress the value of the company (Duangploy et al. 2006). Lander and Reinstein (2003) argued that the discount rate is estimated with error. This is due to measurement errors. Barth and Landsman (1995) argued there are two types of measurement error. The first is unsystematic error arising from general uncertainty, and the second is systematic error arising from management s discretion in making the estimate. The error caused by the management would reduce investors perception of the company value (Barth and Landsman 1995). However, manipulation might not always be the case for company s management. As described by Wines et al. (2007), it s important for corporate governance mechanisms to result in effective and independent oversight of management s knowledge and economic power regarding the goodwill valuation process. Page 10 of 47

18 1.5 Formulation of Study Objectives and Hypotheses Study Objectives The objectives of this study are: to understand the practical implementation of goodwill impairment testing in Australian publicly listed companies; to investigate whether the companies practices are adhering to AASB guidelines; to examine the consistency of companies practices, by comparing the practices of companies within the same industry; and to investigate which method (for calculating the recoverable amount of CGU) is most commonly used by the companies under review. The study has implications for AASB impairment testing guidelines. If the commonly used method has been implemented according to AASB guidelines, perhaps AASB should consider developing more guidelines for the method. If all companies use same method, AASB should consider discarding other method Hypotheses There are no existing studies that are directly related to objectives of this study. Thus, this research will make a contribution to the literature by investigating impairment testing practices in Australian companies. However, even without explicit reason, Olde (2007) believed it is more logical and appropriate to calculate the recoverable amount of CGU under VIU method rather than under FVLCS method. Barth and Landsman (1995) argued that estimating value-in-use involves integrating company specific and potentially private information. Thus, it is hypothesised that most of Australian publicly listed companies will adopt VIU method. The rationale is that internally available information will be more relevant in estimating the CGU s recoverable amount (bearing in mind assets or CGU of each company have their own uniqueness), and it will be less costly. Page 11 of 47

19 2.0 RESEARCH METHODOLOGY 2.1 Research Design This research began with the question of whether Australian public companies showing goodwill on their balance sheet prefer to apply the value in use (VIU) method or the fair value less cost to sell (FVLCS) method, in conducting impairment testing on their cash generating units (CGU). This thesis aimed to investigate differences in the impairment testing methods used by Australian companies, and the extent to which companies adhere to Australian Accounting Standard Board (AASB) guidelines. Thus, this study would compare companies impairment testing with recommended practices of the AASB guidelines (under both VIU and FVLCS method) and/or would investigate whether the companies practices were consistent with the guidelines as well as with other companies using the same methodology (i.e. companies using VIU method would be compared with each other and companies using FVLCS will be compared with each other). This thesis investigated the companies impairment testing practices through their annual reports (year ) and any related data and/or information available in the market including the market interest rate for 2005 until The rationale for examining the annual reports from year 2005 until 2007 was to identify differences and changes in reporting practices over the time, as the goodwill impairment testing concept was first applied in This time frame was chosen as it spans across the final year of reporting under the old Australian developed standards and the introduction and use of the new international financial reporting standards (IFRS). During this transition, the AASB ceased to implement the Australian standard requiring goodwill amortization and replaced it with goodwill impairment testing as required under Australian International Financial Reporting Standards (AIFRS). Page 12 of 47

20 2.2 Description /Rationale for Research Methodology Sampling The population for this study was Australian publicly listed companies reporting goodwill on their balance sheet. This study used a purposive sampling technique. This means that companies are selected because they have been practising goodwill impairment testing since year This research focused on a total of 18 Australian publicly listed companies (17 of the companies are S&P/ASX300 companies). There are sets of six companies chosen from three different industries. S&P/ASX300 companies are the top 300 public companies by market capitalisation currently listed in ASX (Australian Securities Exchange). The companies were divided, based on industry, in order to make the comparison more relevant, as companies in the same industry are more likely to have similar assets (CGU) and/or business structure, as well as using same market or industry data (Deegan, 2007). Identification of the industry to which a company belongs was extracted from the Fin Analysis (Aspect Huntley) database. However, all industries in ASX300 which does not have companies reporting goodwill since year 2005 and/or companies have never reported goodwill, were excluded from this study. The first industry examined was the Capital Goods industry, represented by Alesco, Hastie Group, Leighton Holdings, MacMahon Holdings, Monadelphous Group, and United Group. The second industry was the Commercial Services and Supplies industry, represented by Coffey International, Corporate Express Australia, Downer EDI, Programmed Maintenance Services, Salmat, and SEEK. The last industry was the Information Technology (IT) industry, represented by Computershare, Iress Market Technology, MYOB, Oakton, Reckon, and UXC. Seventeen companies of these are listed in ASX300. Reckon is a listed company but is not in the ASX300. However, Reckon have had goodwill in their accounts since Since there is no information how many Australian publicly listed companies have been implementing goodwill impairment testing since 2005, the appropriate number (or percentage) of companies required for sample representatives for this study could not be determined. Considering the time limitation imposed, 18 companies seemed reasonable. Page 13 of 47

21 The 18 companies were also chosen by considering the fact that they are market leaders in their respective industries. As market leaders, they have responsibility to more stakeholders who are affected by the companies performances. In the Fin Analysis database, there are more than 10 industries to of which the ASX300 companies belong. After investigating annual reports of companies in the industries, three industries were considered relevant for this study. The three industries included in this study are the ones whose companies have been reporting goodwill in their balance sheet since Most of the industries have less than 10 companies with goodwill. Thus, choosing six companies from each of the three industries seemed to be an appropriate method in selecting the sample Procedure The data collection for this study was based on publicly available information. The sources for collecting the data would be the Fin Analysis website, companies official websites, the ASX website, and government s websites (e.g. ABS website and RBA website). Fin Analysis database would also be used to obtain forecast data regarding the growth of industries and companies in this study. The data for this research was mainly collected from the companies annual reports. The companies annual reports were obtained from Fin Analysis database. Data was also be collected from available market information, data and/or analysis from Fin Analysis, as well as any market and/or company announcement related to this research. The following are the data which were collected from the annual reports: For companies using FVLCS method, this study would identify which market data the company used to estimate the fair value of the CGU to which the goodwill was allocated. For companies using VIU method, first of all, a search for the growth rate and pretax discount rate used by the companies in projecting the net cash flows of the CGU would be undertaken. Page 14 of 47

22 Annual reports also provide the net profit of the company which may be used to justify the company s chosen growth rate. For example, if the company stated in the 2005 annual report that the projected annual growth rate for every year over the next five years will be 9%, but, the net profit in year 2006 and 2007 only grew 1% and 2% respectively from net profit in Then, the growth rate of 9% would be questionable. Bear in mind, growth rate may not be growth in net profit but growth in revenues or market share. For consistency, a company s growth rate would be compared with other growth rates in the company itself, including growth rates of operating revenue and earning after tax. The growth rates were also collected through the company s annual reports. It would also be compared against other companies (operating in the same industry) growth rates, bearing in mind that those companies might have their own average growth rate based on past experiences. The CGU s pre-tax discount rate would be compared with the market (interest) rate for time value of money calculations, which could be collected from market available information, i.e. Reserve Bank of Australia (RBA) website. The pre-tax discount rate used in 2005 would be compared with market (interest) rate in However, every company s assets have specific risks attached to those assets. Therefore, when estimating CGU s pre-tax discount rate, the company must have added the rate on the CGU s risk into the market risk-free interest rate. If there was no market data of pre-tax discount rate for similar CGU, comparison would be made with the companies within the same industry. Page 15 of 47

23 2.3 Data Analysis FVLCS Method The 18 companies would be divided into two groups based on their impairment testing methodology. For companies using FVLCS method, how the company determined the amount of fair value of the CGU would be analysed. If the recoverable amount of the CGU was based on the price under a binding sale agreement (for the CGU) less costs of disposal, then the selling price (without deducting the costs of disposal) recorded in the annual report of company under review (seller) would be compared to amount recorded in the buyer company s annual report. Australian accounting standards describe a binding sale agreement as an agreement between both parties to sell and to buy certain assets in the future. According to AASB 136, the costs of disposal are the amount (to be) estimated and/or actually paid by the seller company. Further investigation of the costs of sale would not be necessary for the purpose of this study. Hence, if the price recorded in both parties (of a binding sale agreement of particular CGU) annual reports are the same (ignoring the costs of disposal), it could be deduced that the selling company s practice in impairment testing has conformed to AASB guidelines. Furthermore, a company also can use the binding sale agreement of similar CGU owned by other companies in the same industry as the basis in estimating the recoverable amount of its own CGU under FVLCS method. If the assets within the CGU are actively traded, the assets market (current bid) price less costs of disposal will be the recoverable amount of the CGU. Therefore, the analysis would focus on whether the company (whose CGU s assets are actively traded) quoted the market price less costs of disposal for that period as the recoverable amount of the CGU in the same period. If the CGU s assets are actively traded, then the company would publish the costs of disposal as well. A company can choose to determine the recoverable amount of the CGU under FVLCS even in the absence of a binding sale agreement or transaction for similar CGU, and even when there is no active market for assets in the CGU. In this case, the Australian standard allows the use of a best estimate of the CGU s market selling price less costs of disposal, based on the best information the company can obtain. However, the standard does not explain or provide any guidelines of how the estimation process can Page 16 of 47

24 be implemented. For companies using this method, the data would be analysed by tracking how the companies obtain the best information to estimate the selling price of CGU VIU Method For companies using VIU method, investigation of whether they are consistent with standard s guidelines as well comparable with the fellow companies in the same industry, the pre-tax discount rate and growth rate would be analysed. Based on the standard guidelines from AASB, CGU s pre-tax discount rate should reflect the current market risk-free interest rate plus risks specific to the CGU. The applied risk-free interest rate would be deducted from the pre-tax discount rates of CGU s companies in the same industry which are using VIU method. The deduction would show how much (rate of) risk each company allocates to their CGU. Then, the rate of risk would be analysed and compared among the companies within the same industry. This analysis showed the consistency of the companies rate of risk, which eventually may summarise the consistency of the companies impairment testing practices with standard guidelines as well as with other companies within the same industry. Unlike the pre-tax discount rate, the method to generate growth rate for projecting the future cash flows is not guided by the standard. However, this study would analyse the growth rate used by the companies within the same industry, to test their consistency with each other. As stated by the previous example, the company s other growth rates would also be used to test the accuracy of the company s CGU predicted growth rates. It is necessary to analyse the pre-tax discount rate and growth rate of the companies. Inappropriate values allow companies which may manipulate the rates for their own benefit. For example, a company may choose a lower pre-tax discount rate to discount back the projected cash flows (in calculating the value in use of the CGU) to present value, which will lead to higher recoverable amount. Alternately, the company may choose higher growth rate when calculating the percentage growth in cash flows from year to year. This manipulation will also lead to higher recoverable amount. Page 17 of 47

25 2.4 Summary of Research Methods For data collection, the annual reports from 2005 to 2007 of the 18 companies were collected from the companies websites. The market risk-free interest rates will be collected from the RBA website. The industries growth rates will be collected from ASX and ABS websites. For data processing, the carrying amount of goodwill, the method for calculating the recoverable amount, recoverable amount of CGU, the pre-tax discount rate and growth rate of the companies under review were obtained from the companies annual reports. The annual average of market risk-free interest rates from 2005 to 2007 might need to be calculated for data collected from RBA website. For data analysis, companies using FVLCS method would be directly compared using any related market available information. For companies using VIU method, the pre-tax discount rates and the companies growth rates needed to be calculated in order to make meaningful comparison. This study also analysed whether FVLCS or VIU method is mostly used by the companies under review in calculating the recoverable amount of the CGU. Page 18 of 47

26 2.5 Limitations and Constraints of Research Method There are three limitations of this study. Firstly, as the research of this study is based on secondary data, there may be limited information. Thus, in-depth analysis, such as examining projected cash flows the company used in determining the recoverable amount of its CGU, could not be conducted for this study. Ideally, it should be done by examining the company s budget reports which are not available to public. Secondly, the CGU of each company has its own uniqueness which might differentiate it from each other. This means that the comparison (directly or indirectly) may not be that appropriate. Thirdly, there is no information regarding the population of Australian publicly listed companies which have been reporting goodwill in their balance sheet since The sampling selection may be biased as the population may be under-represented. Finally, the allocation of goodwill to the CGU is at the discretion of the company and this data is not available publicly and can be used to manipulate the calculation system. Page 19 of 47

27 3.0 DATA ANALYSIS AND DISCUSSION Table 2 Capital Goods Industry Data No. Company FV L CS VIU Growth Rate Pre tax Discount Rate Alesco - YES % 12.2% 11.3% 2 Hastie Group - YES % 13.5% 13.5% 3 Leighton - YES % 11.5% 11.5% 4 Macmahon - YES - 10% 10% % 15.4% 5 Monadelphous - YES % 10% 10% 6 United Group - YES Table 2 shows that: 1) Alesco financial year ends at 31 May. Alesco used WACC rate instead of pre tax discount rate. It did not provide its WACC rates in its annual reports. However, from information provided in Alesco s annual reports, WACC rates were calculated. For consistency with other companies, the WACC rates were converted to pre tax discount rates (please refer to Appendix A for the calculation). There was no goodwill impairment in 2006 and ) Hastie Group financial year ends at 30 June. There was no goodwill impairment in ) Leighton financial year ends at 30 June. The recoverable amount of its CGU is calculated using 5-year cash flow projections (there is no growth rate disclosed). Leighton had goodwill impairment in 2006, but there was no goodwill impairment in 2005 and Page 20 of 47

28 4) Macmahon financial year ends at 30 June. There was no goodwill impairment in 2005, 2006, and ) Monadelphous financial year ends at 30 June. The growth rate of its CGU is based on the entity s budgeted cash flows. There was no goodwill impairment in 2005, 2006, and ) United Group financial year ends at 30 June. There was no impairment loss on goodwill in 2005, 2006, and Page 21 of 47

29 Table 3 Commercial Services and Supplies Industry Data No. Company FV L CS VIU Growth Rate Pre tax Discount Rate Coffey International 2 Corporate Express Australia - YES 2% 2% 2% 17% 17% 17% - YES % 13% 3 Downer EDI - YES 3% 3% 3% 9.4% 9.5% 9.9% 4 Programmed Maintenance Services - YES - 20% 17.5% - 11% - 5 SALMAT - YES % 11.6% 11.6% 6 SEEK - YES 0% 0% 0% 11% 11% 11% Table 3 shows that: 1) Coffey International financial year ends at 30 June. In 2005, goodwill was not impaired. 2) Corporate Express Australia financial year ends at 31 December. In 2005, it decreased amortization of goodwill expense of $5,566,000. Instead of using pretax discount rate, it uses an after tax cost of capital. 3) Downer EDI financial year ends at 30 June. In 2005, the goodwill amortization expense was expected to decrease to nil (by 19,606,000), which means no goodwill impairment. 4) Programmed Maintenance Services financial year ends at 31 March. In 2005, the goodwill amortization expense was expected to decrease to nil (by $144,000). 5) SALMAT financial year ends at 30 June. Page 22 of 47

30 6) SEEK financial year ends at 30 June. There was no impairment in 2005, 2006 and Page 23 of 47

31 Table 4 Information Technology (IT) Industry Data No. Company FV L CS VIU Growth Rate Pre tax Discount Rate Computer share - YES - 1% 1% 15.4% 14.7% 15.7% 2 IRESS - YES MYOB - YES 4% 4% 4% 14.8% 15.3% 15.3% 4 Oakton - YES % 13% 5 Reckon - YES 10% 10% 10% 10.5% 13.7% 13.4% 6 UXC - YES - 2% 2.5% 13.25% 11.75% 11.2% Table 4 shows that: 1) Computershare financial year ends at 30 June. It did not have impairment loss in ) IRESS financial year ends at 31 December. It did not have any impairment loss in 2005, 2006, and ) MYOB financial year ends at 31 December. In 2005, for CGU in China, MYOB used 6% growth rate. In 2006, for CGU in China, MYOB used 5% growth rate. In 2005, pre-tax cash flow forecasts are derived from financial projections covering 5 years period. In 2006 and 2007, it covered respectively 14 year and 13 year period to ) Oakton financial year ends at 30 June. It did not have any impairment loss in 2005, 2006, and ) Reckon s financial year ends at 31 December. It uses cash flow projections based on annual financial budgets (5 years) for the coming year. 6) UXC financial year ends at 30 June. There was no goodwill impairment in Page 24 of 47

32 3.1 Preferred Method All of the 18 companies under review have been using VIU method since 2005 in conducting impairment test of their goodwill. Thus, VIU is the preferred method for goodwill impairment testing, across this sample. 3.2 Adherence to AASB Guidelines Per AASB 136 guidelines, companies in this study have allocated goodwill to their CGU. They have been using VIU method (one of two methods allowed by AASB) in conducting impairment test. As required under VIU method, the companies have been applying a pre-tax discount rate to their projected cash flows. Even though five of the companies under study stated in their annual reports that they applied pre-tax discount rates in calculating the value in use (recoverable amount) of their CGU, they did not state nominal values of those rates. Companies are allowed by the AASB to use growth rate in estimating projected (future) cash flows of their CGU to which the goodwill is allocated. Seventeen companies under review stated that they used growth rate in estimating future cash flows of their CGU. However, only eight companies stated nominal values of growth rates in their annual reports. Page 25 of 47

33 3.3 The Consistency of Impairment Testing Practices Pre-tax Discount Rates Pre-tax discount rate will be used to examine the consistency of goodwill impairment testing practices among companies within the same industry. As stated in the research methods chapter, the pre-tax discount rate consists of risk-free market interest rate and specific risk rate of which companies allocated to their CGU. Based on Reserve Bank of Australia data, the risk-free market interest rate in 2005 was 5.5%, effective from 2 March 2005 until 3 May In 2006, there were 3 rates, i.e. 5.75% on 3 May 2006, 6% on 2 August 2006, and 6.25% on 8 November However, in early 2006, risk-free market interest rate was still 5.5% until 3 May Therefore, the average market interest rate for 2006 was around 5.9% (sum of 5.5%, 5.75%, 6%, and 6.25%, and then divided by 4). In 2007, there were 2 rates, i.e. 6.5% on 8 August 2007 and 6.75% on 7 November Hence, similar to calculation in 2006, the average interest rate for 2007 was 6.5% (sum of 6.25%, 6.5%, and 6.75%, and then divided by 3). The risk-free market interest rates are known, therefore the percentage of risk allocated by companies to their CGU can be calculated. Please note that, to make the comparison between companies more relevant, the average market interest rates for each year will be used. During the period under investigation, the Australia market interest rate was not volatile. Thus, using average market interest rate for this data analysis can be justified. Page 26 of 47

34 Table 5 Rates of Risk Used by Companies in Capital Goods Industry No. Company Rate of Risk Alesco 7.7% 6.3% 4.8% 2 Hastie Group 8% 7.6% 7% 3 Leighton 6.3% 5.6% 5% 4 Macmahon % 8.9% 5 Monadelphous 4.5% 4.1% 3.5% 6 United Group Average 6.6% 7.1% 5.8% In Table 5, Leighton s and Monadelphous rates of risk were under average. However, compared to Leighton s rates of risk, Monadelphous rates of risk were significantly lower than the average which might be affected by the fact that Macmahon had much higher rate of risk. The exact reason why Monadelphous had relatively lower rates of risk compared to the other companies cannot be investigated due to limited information. Macmahon had applied relatively much higher percentage of risk on its pre-tax discount rates compared to other companies in capital goods industry. There is no information to explain the higher percentage of risk. It might due to the uniqueness of its CGU. However, a company will not gain any benefit by applying higher percentage of risk on their CGU. Therefore it is not necessary to investigate why Macmahon had higher rate of risk assigned to its CGU. Page 27 of 47

35 Nevertheless, from 2005 to 2007, it can be assumed that goodwill impairment testing practices in capital goods industry is consistent. This being so because majority of the companies had rates of risk which is not much difference from the average and from each other. Page 28 of 47

36 Table 6 Rates of Risk Used by Companies in Commercial Supplies and Services Industry No. Company Percentage of Risk Rate Coffey International 11.5% 11.1% 10.5% 2 Corporate Express Australia - 7.1% 6.5% 3 Downer EDI 3.9% 3.6% 3.4% 4 Programmed Maintenance Services - 5.1% - 5 SALMAT 5.8% 5.7% 5.1% 6 SEEK 5.5% 5.1% 4.5% Average 6.7% 6.3% 6% Table 6 shows that majority of the companies had lower rates of risk than the average. This due to Coffey International who applied relatively high rates of risk compared to others. It was only Downer EDI who had significantly lower rates of risk than the average. Based on the industry s averages, it can be assumed that goodwill impairment testing practices in commercial services and supplies industry are consistent. It does not mean that Coffey International was not consistent with the industry. Coffey International might have relatively riskier CGU compared to others, which made it having much higher rates of risk. Page 29 of 47

37 Table 7 Rates of Risk Used by Companies in Information Technology Industry No. Company Rates of Risk Computershare 9.9% 8.8% 9.2% 2 IRESS MYOB 9.3% 9.4% 8.8% 4 Oakton % 6.5% 5 Reckon 5% 7.8% 6.9% 6 UXC 7.75% 5.85% 4.7% Average 8% 8.6% 7.2% In 2005, it was Reckon s rate of risk which is relatively much lower than average. Furthermore, compared to MYOB rate of risk, UXC rate of risk was much lower than Computershare s. In 2006 and 2007, compared to other companies, it was only UXC rate of risk which is relatively much lower than average. The variation between rates of risk is also high. It can be assumed that goodwill impairment testing practices in IT industry, in term of allocation of rate of risk to CGU, is less consistent compared to other industries under review. This may because of the fact that IT industry is a unique industry. Every IT company has its own competitive advantage and expertise which differentiate them from each other. Page 30 of 47

38 3.3.2 Growth Rate For testing consistency of growth rates inside the company itself, the growth rate used for projecting CGU cash flows will be compared with growths of companies operating revenues and earning after tax (EAT). The rationale of choosing to compare them with CGU growth rate is the relationship between the three growth rates. CGU projected cash flows are basically the net cash inflows from operating CGU. Operating revenues are gross cash inflows from operating companies business. EAT is net (after all expenses and tax) cash inflows that a company earns from any sources. However, compared to EAT, operating revenue growth would be a better number in comparing with the CGU growth rate. This due to the fact that EAT may include extraordinary items such as interest revenue. Most companies under review which have CGU more than one and the CGU might have different growth rates. Nevertheless, the companies only put one growth rate in projecting the future cash flows of their CGU. A company only have one growth rate in their annual report as the company might have total up the growth rates or have an average growth rates of all its CGU. Thus, the comparison with operating revenue and EAT growth rates was made possible by assuming of same growth rate of cash flows across all CGU in a company. This means that only CGU growth rates provided in the annual report were used for this study. Page 31 of 47