Discussion of Dedman, Mouselli, Shen and Stark
Michael Bradbury ([email protected]) is a Professor in the School of Accountancy at Massey University.
My thanks go to Jo Greenland and Janet Ritchie at the New Zealand Institute of Chartered Accountants library for searching and providing articles. I also thank Robert Keys, Majella Percy and especially Wayne Lonergan for their comments on an earlier draft of this paper.
‘Accounting for Intangible Assets – Views from Australia’
Dedman et al. (2009), along with other contributors to the forum, provide evidence (and views) for standard setters across a range of intangibles (organizational capital, research and development, and goodwill). In this commentary I summarize the Australian research on the reporting of intangibles.1 Evidence from the Australian institutional setting complements the evidence from other jurisdictions (e.g., U.S. and U.K.) reported at the forum. For example, Australian firms adopted the practice of capitalizing internally generated identifiable intangible assets; a practice not available in most other jurisdictions.
BACKGROUND
In this section I provide a brief background to the development of financial reporting in Australia and in the following section I relate this to the reporting practices of Australian firms. Table 1 describes the development of accounting standards in Australia relating to intangibles.
| March 1983 | AAS 13, Accounting for Research and Development Costs |
| • R&D may be capitalized if net benefits are expected ‘beyond reasonable doubt’• No maximum amortization period• No reinstatement of expensed R&D | |
| May 1983 | ED 23, Accounting for Goodwill |
| Mar 1984 | AAS 18, Accounting for Goodwill |
| • Internally generated goodwill not to be recognized• Purchased goodwill to be recognized• Required amortization over benefits (maximum 20 years) | |
| December 1985 | AGR5, Accounting for Intangible Assets |
| • Intangibles required to be amortized | |
| July 1986 | NCSC Policy Statement (Release 135), ‘Revaluation of Intangible Assets’ |
| May 1987 | AASB 1011, Accounting for Research and Development |
| April 1988 | ASRB 1013, Accounting for Goodwill |
| • Excess of cost over fair value of identifiable net assets• Systematic amortization over expected benefits (maximum 20 years) | |
| August 1989 | ED 49, Accounting for Identifiable Intangible Assets |
| • IIA amortized over finite period• Few assets over 20 years | |
| 1991 | AASB 1010, Accounting for Revaluation of Non-Current Assets |
| • Requires revalued assets to be depreciated | |
| March 1992 | AARF Media Release |
| • ED 49 withdrawn ‘in view of the lack of consensus’ | |
| November 1993 | ASC Practice Note 39 |
| • ISOYD justified in ‘rare and exceptional circumstances’ | |
| December 1995 | ED 68, Amortization of Purchased Goodwill |
| • Amendments to AAS 18/AASB 1013 | |
| June 1995 | Media Release 95-6 (UIG Meeting #5) |
| • Discussion commenced on the proposal that straight line method to be the only method of goodwill amortization | |
| August 1995 | Media Release 95-8 (UIG Meeting # 6) |
| • Moratorium on the use of ISOYD for new acquisitions made on or after 25 August 1995 to 1 July 1997 | |
| April 1996 | AASB 1013, Accounting for Goodwill |
| • Goodwill to be amortized on SL basis | |
| 1996 | AASB Policy Statement 6, International Harmonization Policy |
| 1998 | IASC IAS 38, Accounting for Intangible Assets |
| August 1998 | ASIC Media Release. ‘Corporate Financial Reporting Surveillance Initiative’ |
| June 1999 | Accounting Interpretation 1, Amortization of Identifiable Intangible Assets |
| September 1999 | ASIC Media Release, ‘ASIC Reviews IIA Accounting Problem’ |
| 2000 | AASB Harmonization policy with IASC |
| • IAS 38• Intangibles priority project | |
| 2005 | Move to AIFRS; AASB138 |
| • Goodwill impairment rather than amortization• Inability to capitalize research• Derecognition of IIA• Inability to revalue IIA (no active market) | |
| 2008 | AASB Staff Discussion Paper (Keys and Ardern, 2008) |
- a GW = goodwill, IIA = identifiable intangible assets (excluding R&D), R&D = research and development.
For the purpose of this review, the development of Australian generally accepted accounting principles can be divided into five distinct phases. The first phase represents pre-accounting standard (when firms had considerable freedom with regard to accounting policy choice). The second phase represents the development of accounting standards by the professional accounting bodies, which were enforceable only under their code of ethics. From 1966 the professional bodies jointly operated the Australian Accounting Research Foundation (AARF). This phase resulted in AAS 13, Accounting for Research and Development Costs, and AAS 18, Accounting for Goodwill. A third phase provided accounting standards with legal backing. In 1984 accounting standards produced by the profession were to be approved by the Accounting Standards Review Board (ASRB). Standards approved by the ASRB had the force of company law under the Companies Act 1981.2 In 1991, major changes to the legislative system governing financial reporting were introduced. The Australian Accounting Standards Board (AASB) replaced the ASRB and in addition to the approval process, it also took over the role of producing accounting standards. The AASB's accounting standards achieve legal status under the Corporations Law. The third phase resulted in the production of AASB 1011, Accounting for Research and Development, and AASB 1013, Accounting for Goodwill. A fourth phase began in 1996, when the AASB announced its intention to harmonize with International Accounting Standards Committee's standards. The fifth phase was the adoption of the International Accounting Standard Board's standards in 2005, which led to the issuance of Australian equivalents of International Financial Reporting Standards.
RESEARCH AND DEVELOPMENT
AAS 13 was introduced in March 1983. It required research and development (R&D) costs to be expensed as incurred, except to the extent that where net future benefits are expected beyond reasonable doubt they must be carried forward. Capitalized expenditure was required to be amortized ‘to match costs with related benefits’ (with no maximum amortization period specified). The unamortized balance of R&D was required to be reviewed annually for impairment. The reinstatement of previously written-off R&D was prohibited. AAS 13 obtained statutory backing through the issue of ASRB 1011. In 1995 the Australian Securities Commission (ASC) undertook a review of R&D reporting practice and concluded that R&D disclosures were often incomplete and arbitrary (Schmidt, 1996, p. 24). At least nine companies changed their R&D policy from capitalizing to expensing all R&D costs following the ASC statement (Tutticci et al., 2007, p. 88).
The majority of Australian firms have no accounting policy on R&D (presumably because it is not material). McGregor (1980) reports that 80 per cent of firms do not have an accounting policy. Wyatt et al. (2001) report 64 per cent of their 1993 to 1997 sample have no accounting policy. Similarly, few firms report R&D expenditure. Abrahams and Sidhu (1998) find that 48 per cent of their 1975 sample report R&D expenditure. Godfrey and Koh (2001) find only 18 per cent report R&D.
Of the firms that report R&D, most firms expense it (Godfrey and Koh, 2001; Wyatt, 2005; Tutticci et al., 2007). This was also the case pre-AAS13 (Carnegie and Turner, 1987; McGregor, 1980). Firms capitalizing R&D are small (Godfrey and Koh, 2001, p. 43; Wyatt, 2005, p. 995); younger (Tutticci et al., 2007); R&D intensive (Percy, 2000; Wyatt, 2005; Tutticci et al., 2007); not generating sales and have short business cycles (Wyatt, 2005); and have lower return on assets, losses, tax losses (Percy, 2000); lower growth potential and higher leverage (Tuttici et al., 2007).3
The Australian evidence on the value relevance of R&D is mixed. Abrahams and Sidhu (1998) find that R&D capitalization is value relevant. They also find weak evidence that the amount of R&D capitalized improves earnings as a measure of firm performance. In contrast Godfrey and Koh (2001) and Wyatt (2006) find the coefficient on the R&D asset is not significant. Smith et al. (2001) find that capitalized R&D expenditure results in balance sheet and income statement amounts that are more highly associated with market value than ‘as if’ expensing. They also show that a dollar of capitalized R&D is ‘worth’ more than a dollar of expensed R&D. Tutticci et al. (2007) find the reliability of capitalized R&D investment is less than conclusive. The most persistent finding is that the market positively values R&D expenditure. However, the market places greater weight on potential growth from R&D when it is expensed than when it is capitalized.
Possible explanations for the mixed results are that smaller and more research intensive firms are more dependant on the success of R&D projects. For example, the size of firms in Godfrey and Koh (2001, p. 43) is much larger than that of Abrahams and Sidhu (1998). This is consistent with the findings that small firms are more likely to capitalize expenditure in the U.S. (Daley and Vigeland, 1983) and the U.K. (Oswald, 2008). Furthermore, the results in Smith et al. (2001) only hold for a sample of firms with material R&D (based on an above median ratio of capitalized R&D to market capitalization).4
PURCHASED GOODWILL AND IDENTIFIABLE INTANGIBLE ASSETS
Pre-Regulation
In 1974, pre-regulation, the most common method (64 per cent) of accounting for goodwill on consolidation was to write down the investment against equity in the parent's books (Gibson and Francis, 1975, p. 170). Immediately prior to the release of AAS 18 (in 1983), 33 per cent reported goodwill as an asset, 26 per cent chose an immediate write-off and 41 per cent left goodwill as a ‘dangling debit’ in equity. Of those reporting goodwill as an asset 59 per cent retained it as an asset without amortization or write-off (Goodwin, 1986b, pp. 21–7). Similar evidence is provided by Anderson and Zimmer (1992).
AAS 18 was released in March 1984. Its most controversial provision was the requirement to amortize goodwill over the period of expected benefits (Wines and Ferguson, 1993, p. 91). The focus on amortization of goodwill is a direct result of the requirements to recognize goodwill under the purchase method of accounting for acquisitions (AASB 1024 and 1015), the prohibition on revaluing purchased goodwill (AASB 1013), and the inability to recognize internally generated goodwill.5
To minimize the impact of amortized goodwill on operating profit, Australian firms adopted at least four approaches: (a) non-compliance, (b) pre-standard ‘clearing the decks’, (c) recognizing identifiable intangibles, and (d) using the inverse-sum-of-the-years-digits to amortize goodwill.
Non-Compliance
While AAS 18 increased the number of firms systematically amortizing goodwill, there was also a significant degree of non-compliance (Carnegie and Gibson, 1987; Kirkness, 1987; Williams and Carnegie, 1989; Dunstan et al., 1993; Wines and Ferguson, 1993).
Wines and Ferguson (1993, p. 97) report that four firms in 1988 (5 per cent) received statutory relief to permit transitional exemption from the provisions of ASRB 1013. In the same year, a further twenty-one firms (25 per cent) applied complete or partial ‘extraordinary’ write down of goodwill against income. Apart from transitional issues, the statutory-backed AASB 1013 brought about significant changes and consistency in the reporting of goodwill (Williams and Carnegie, 1989; Carnegie and Gibson, 1991; Goodwin and Harris, 1991; Dunstan et al., 1993; Wines and Ferguson, 1993).
Clearing the Decks
Many firms eliminated or reduced their goodwill balances before AAS 18 became operative (Kirkness, 1987, p. 51). The immediate write-off of goodwill, either direct through to reserves or through income (typically as an extraordinary or abnormal item), continued until ASRB 1013 became effective (Goodwin and Harris, 1991; Dunstan et al., 1993; Wines and Ferguson, 1993).
Recognizing IIA (and the Regulators' Response)
Another response to AAS 18 was to recognize identifiable intangible assets (IIA), on the basis that they are not amortized, thereby reducing the amount that would otherwise be recorded as goodwill. Goodwin and Harris (1991, p. 24) report that over the 1987 to 1989 period the number of firms recording goodwill increased by 68 per cent, while those recording IIA increased by 27 per cent. In 80 per cent of these cases the recorded amount for IIA was less than 5 per cent of non-current assets.
While the recognition and revaluation of IIA had not reached the ‘plague proportions’ predicted by Curran (1989), it became a major concern to standards setters and regulators. The AARF issued Accounting Guidance Release (AGR) 5 (December 1985), pointing out that IIA should be accounted for in accordance with AAS 4, Depreciation of Non-Current Assets, and accordingly should be amortized. In July 1986 the National Companies Securities Commission issued policy statement, ‘Revaluation of Intangible Assets’ (Release 135). This statement was primarily concerned with revalued IIA in prospectuses, but also covered disclosures in subsequent financial statements.
Carnegie and Kallio (1988) examined the effect of reporting practice immediately after AGR 5 and found that 35.4 per cent capitalized IIA without amortization. By 1989, Wines and Ferguson (1993, p. 99) report that 54.5 per cent of firms recognizing IIA failed to follow the guidance in ASR 5. The unamortized IIA were mostly brands and business names. The reasons provided for non-amortization were these assets did not diminish in value and/or they were subject to periodic revaluation (Carnegie and Kallio, 1988). In contrast, patents and rights (which usually have limited legal lives) were usually amortized and the treatment of trademarks and licences was mixed.6
In August 1989 the AARF issued ED 49, Accounting for Identifiable Intangible Assets. This exposure draft permitted the revaluation of purchased intangibles and the recording of internally developed intangibles. While there was support for issuing a standard prescribing the accounting for IIA, there was a great diversity of views on some issues and strong opposition on others (Alfredson, 2001, p. 13). Content analysis on the submissions made to ED 49 indicates that the requirement to amortize IIA was the single most substantive issue, with 83 per cent of respondents disagreeing with the requirement to amortize (Tutticci et al., 1994, p. 96).
An AARF media release, issued August 1990, stated that the Board was ‘paying close attention to overseas developments on this topic [IIA accounting]’. A second paragraph contained a reminder of the accounting requirements for, and in particular the depreciation of, non-current assets. In March 1992, ED 49 was withdrawn. The media release further advised that the board would continue to monitor overseas developments and might issue a revised exposure draft (Alfredson, 2001, p. 15).
In April 1996 the AASB issued a policy statement giving its intention to harmonize with international standards (AASB 1996). This had the effect of implementing a program to make international accounting standards issued by the International Accounting Standards Committee a basis for (but not necessarily identical to) Australian standards. Wyatt et al. (2001) conclude that during the 1993–97 period, the discretion available to managers resulted in considerable diversity of recognized IIA and reporting practices. However, despite the variety in accounting practice, the AASB's ‘resolve was not sufficient at that time to use its International Harmonization Policy to as a lever to finalize a standard in harmony with IAS 38’ (Alfredson, 2001, p. 17).
Late 1998 the Australian Securities Investment Commission (ASIC), the corporate regulator, entered into the debate by announcing that it was undertaking a surveillance program, of which one topic was ‘amortization of intangibles’. The results of this review was reported in ASIC's Media Release 99/219. ASIC reviewed 111 listed companies' financial statements and reported a number of instances where intangibles were not amortized. The ASIC release also mentioned a recent AARF Accounting Interpretation, Amortization of Identifiable Intangible Assets (issued June 1999) that clarified that AASB1021, Depreciation of Non-Current Assets, applies to intangible assets in the same way that it applies to physical assets. ASIC Media Release 00/264 reported the surveillance results of a further hundred financial statements, stating that some entities still refused to amortize because some intangibles had indefinite or infinite lives or that the amortization expense was immaterial because of high residual values.
Using the Inverse-Sum-of-the-Years Digits
A particular concern of Australian regulators was the development of the inverse-sum-of-the-years digits method (ISOYD) to amortize goodwill, which deferred the bulk of the amortization expense to the latter periods. After several years debate with the corporate sector, the ASC issued Practice Note 39, which formally documented the ASC's negative view of ISOYD. The issue was then considered by the Urgent Issues Group (UIG) and more decisively by the AASB, in October 1995, when it announced its intention to issue a revised standard on accounting for goodwill that would mandate the use of straight-line amortization. Despite strong opposition, AASB 1013 was reissued in April 1996 containing amendments specifically requiring goodwill be amortized on a straight-line basis.
Day and Harnett (1999/2000) find that firms using ISOYD had higher purchased goodwill (and therefore higher goodwill expense) but were indistinguishable in terms of firm size and profitability. Interest cover was marginally lower for ISOYD firms. Day and Harnett also used event methodology to examine the share price reaction of ISOYD firms to seven key event dates towards the outlawing of ISOYD. They find no evidence that the stock market reacted to the events that mandated changes to the amortization of goodwill.
The Move to AIFRS
In July 2004 the AASB decided that Australian firms, from 1 January 2005, would be required to follow Australian equivalents of International Financial Reporting Standards (AIFRS). Adopting IAS 38 meant the replacement of amortization with an impairment regime for indefinite life intangibles; the inability to capitalize research expenditure associated with the research phase of an internally generated IIA; and the inability to revalue an IIA for which there is no active and liquid market (as was the existing practice).
Chalmers and Godfrey (2006, p. 64) find that in 2002 (pre-adoption of international accounting standards) 67 per cent of their sample of 476 listed firms had intangible assets with median intangible asset to total asset ratio of 11.5 per cent. They conclude the adoption of AIFRS will result in significant changes to the diversity of intangible accounting policies. Post-AIFRS adoption, Carlin et al. (2007) provide some descriptive data of accounting disclosures in relation to impairment testing of goodwill for a small sample of large Australian firms.
The 2006 Memorandum of Understanding between the IASB and the Financial Accounting Standards Board (FASB) notes that intangible assets is a topic ‘already being researched, but not yet on an active agenda’. In October 2006 the IASB discussed the scope, approach and draft plan of an intangible assets project, to be led by the AASB.7 At their December 2007 meetings the IASB and FASB considered the proposal but decided not to take the project on to their active agenda, because of competing agenda priorities. The AASB continued to work on the project, with the encouragement and support of the national standards setters, and the resulting AASB Staff Discussion Paper was issued (Keys and Arden, 2008).8
Value Relevance
The Australian evidence on the value relevance of goodwill and IIA provided by Barth and Clinch (1998), Godfrey and Koh (2001), Wyatt (2005), Ritter and Wells (2006), and Bugeja and Gallery (2006) is consistent with overseas evidence.9 The conclusions reported in Clinch (1995, p. 28) still hold:
- •
Goodwill and IIA are positively associated with share market value;
- •
The association between equity values and goodwill appears weaker than association with IIA; and
- •
There is no consistent evidence of an association between reported goodwill amortization and share prices.10
Ritter and Wells (2006, p. 861) also provide evidence of a positive association between IIA and realized future periodic income.
Accounting Policy Choice
One area in which Australasian research has a strong tradition is the documentation of corporate characteristics relating to accounting policy choice. Anderson and Zimmer (1992) report that in 1980 (pre-AAS 18) the most popular accounting methods for goodwill were dangling debit in equity (55 per cent) and capitalization without amortization (39 per cent). Only five firms (1.8 per cent) capitalized and amortized and two firms (3.6 per cent) wrote off goodwill. In contrast, by 1985, 56 per cent of firms wrote-off goodwill, 39 per cent amortized, one firm capitalized without amortization and two reported goodwill as a dangling debit. Anderson and Zimmer point out that the dangling debit policy is consistent with the way goodwill was treated in many debt agreements at that time.
Using New Zealand data Wong and Wong (2001) and Bradbury et al. (2003) examine alternative explanations for purchased goodwill on acquisition accounting and goodwill amortization, respectively. Wong and Wong show that both the level of debt and the allocation that takes place apportioning the acquisition price between net tangible assets and goodwill are driven by the firm's investment opportunities. Bradbury et al. examine the marginal contributions of alternative explanations for managers' choice of goodwill amortization. They find that asset structure (i.e., the investment opportunities) makes a greater contribution than traditional contracting or opportunism variables. Using an extensive Australian data set and an array of variables representing underlying economics of intangible assets (e.g., technology cycles and property rights), Wyatt (2005) compares traditional contracting and signalling explanations with asset-based explanations for intangible asset accounting. She concludes that limiting management's choice to record intangible assets tends to reduce, rather than improve, the quality of the balance sheet.
Analysts, Preparers and Auditors
Matolcsy and Wyatt (2006) report that firms with higher underlying intangibles (i.e., the difference between market equity and net tangible assets) have lower analyst following. To attract higher analyst following, firms with more certain intangible assets will signal this by capitalizing IIA. Without the capitalization signal, firms have lower analyst following and higher analyst forecast errors.
In contrast, direct evidence from analysts indicates that they ignore this signal. For example, based on a 56 per cent response rate of 200 of its members, the Securities Institute of Australia (1995) found that 86 per cent adjusted income for goodwill amortization, regardless of the method of amortization (Securities Institute of Australia, 1995). This suggests that goodwill amortization does not have a significant impact on pricing. Furthermore, Smith et al. (2001) show that while managers' capitalized R&D is associated with firm value it has only a modest advantage over a simple algorithm that might be used by analysts.11
Keys and Arden (2008) include in their Staff Discussion Paper the results of fourteen interviews with Australian preparers, auditors and users of financial statements. They report that users expressed a preference for cash expenditure over asset valuation information and did not express a preference for any particular method of accounting but were more interested in consistency across all intangible assets and all reporting entities. These results are similar to those of analysts' surveys elsewhere (e.g., AIMR, 1993; PricewaterhouseCoopers, 2007).
A PricewaterhouseCoopers (2007) survey with over fifty buy-side analysts showed that 70 per cent described the reported goodwill balance as ‘not useful’ in their analysis and most were ambivalent about its measurement. With regard to IIA, 74 per cent described the item as ‘not useful’. Many respondents described the current model, whereby some expenditure is capitalized and some expensed, as ‘confusing’.
The AIMR (1993, p. 49) position paper suggests that once the amount has been established for how much was paid to acquire goodwill, it should ‘be removed from the list of assets forthwith’. Furthermore, in a subsequent publication there is an absence of goodwill and intangibles from the model balance sheet (CFA Institute, 2007, p. 23).12 The AMIR paper also rejects the idea of internally generated intangibles being recorded without authentication of a transaction.
A Caution on Interpretation
In my view, the interpretation that voluntary reporting of IIA and goodwill is a signal to the market of the value of intangibles is somewhat naive. The history documented above demonstrates that the reporting of IIA resulted from an accounting arbitrage or the apparent irrational flight from the requirement to recognize and amortized purchased goodwill.13 At best we can say that is that these firms did not abuse the reporting of IIA (as the reported intangible assets are value relevant and reflect the investment opportunities). However, under a voluntary regime (i.e., pre-accounting standards) most firms wrote-off goodwill or left it as a dangling debit in equity (see Anderson and Zimmer 1992).
COMMENTARY ON DEDMAN ET AL. (2009)
The Dedman et al. (2009) paper (DMSS) is difficult to comment on, as it is an extension of prior work that has already been published in good journals. Quibbling over minor technical details does not seem a worthy pursuit.
One feature of DMSS that I liked was that it was a response to calls for academic research to be more relevant (e.g., Leisenring and Johnson, 1994; Schipper, 1994; Brown and Howieson, 1998).
Another positive feature of DMSS is that it does not simply summarize the existing U.K. evidence on R&D (to a large extent this has been done by Clinch, 1995; Wyatt, 2008; and for the U.K. in particular by Stark, 2008). Rather it extends the value relevance evidence by including more recent years' data and adds further sensitivity analysis. The results suggest that R&D expenditures are impounded into share prices and there is little to suggest that R&D activities are systematically misunderstood by the market.
In contrast to the extensive evidence presented by DMSS, the Australian evidence on value relevance of R&D is relatively thin. Furthermore, the standards dealing with R&D impact only a few small firms with high research intensity. There is nothing in the Australian evidence on R&D that contradicts the conclusion in DMSS.
DMSS also report two case studies. These case studies suggest: (a) that non-accounting information (e.g., the results of clinical trials) is perhaps more important than balance sheet information on capitalized R&D expenditure; and (b) that price-sensitive disclosure rules and enforcement are important parts of the process (i.e., that market forces cannot be relied upon alone to produce honest and timely disclosures).
Unfortunately, these case studies provide only weak evidence for standard setters and regulators. The case studies are not supported by any theoretical propositions (e.g., with regard to regulation, enforcement, agency costs) and therefore have limited external validity. Hence, they are merely two very interesting, but anecdotal, stories that apply to the U.K. The Australian literature provides more systematic evidence on the role of monitoring and regulatory intervention. Tutticci et al. (2007) find that external monitoring by a Big 5 auditor and the regulator (ASC) has a dampening effect on R&D capitalization. Furthermore, the accounting policy choice literature (e.g., Panel D of Table 2) indicates that capitalization, amortization and disclosure practices are associated to the firm's investment opportunities.
| Authors | Scopea | Date, sample size | Description |
|---|---|---|---|
| Panel A: Descriptions of accounting practice with regard to intangibles | |||
| Gibson and Francis (1975) | GW | 1974, 196 | Pre-standard |
| McGregor (1980) | R&D | ?, 200 | Pre-standard |
| Goodwin (1986b) | GW | 1980–83, 101 | Pre-standard |
| Carnegie and Gibson (1987) | GW | 1985, 189 | AAS 18 |
| Carnegie and Turner (1987) | R&D | 1986, 141 | AAS 13 |
| Kirkness (1987) | GW | 1980–85, 100 | AAS 18 |
| Carnegie and Kallio (1988) | IIA | 1986/87, 100 | AGR 5 |
| Williams and Carnegie (1989) | GW | 1985–87, 75 | AAS 18 |
| Carnegie and Gibson (1991) | GW | 1988, 200 | ASRB 1013 |
| Goodwin and Harris (1991) | GW, IIA | 1987–89, 90 | ASRB 1013 |
| Wines and Ferguson (1993) | GW, IIA | 1985–89, 150, 750 firm-years | AAS 18 and ASRB 1013 |
| Dunstan et al. (1993) | GW | 1983–89, 84, 581 firm-years | AAS 18 and ASRB 1013 |
| Wyatt et al. (2001) | GW, IIA, R&D, | 1993–97, 1366 firm-years | International harmonization policy |
| Chalmers and Godfrey (2006) | GW, IIA | 2002, 476 | Move to AIFRS |
| Carlin et al. (2007) | GW | 2006, 50 | GW impairment disclosures |
| Panel B: Consequences of accounting rules on intangibles | |||
| Anderson and Zimmer (1992) | GW | 1985–89, 78 | Change in practice to AAS 18 |
| Tutticci et al. (1994) | IIA | 113 submissions | Review of submissions on ED 49 |
| Day and Harnett (1999) | GW | 1993–95, 7 event dates | Share price impact of events surrounding goodwill amortization (ISOYD) controversy |
| Panel C: Value relevance of intangibles | |||
| Abrahams and Sidhu (1998) | R&D | 1994–95, 78 167 firm-years | |
| Barth and Clinch (1998) | IIA | 1991–95, 250 | Also revalued intangible assets |
| Godfrey and Koh (2001) | IIA, GW, R&D | 1999, 172 | Hierarchical multiple regression |
| Smith et al. (2001) | R&D | 1992–97, 108 497 firm-years | Capitalized vs expensed R&DCapitalized vs simple analysts' algorithm |
| Wyatt (2005) | GW, IIA,R&D | 1993–97, 1366 firm-years | Supplementary analysis to accounting policy choice study |
| Ritter and Wells (2006) | IIA, GW | 1979–97, 1078 firm-years | Also examined realized future income |
| Bugeja and Gallery (2006) | GW | 1995–2001, 475 firm-years | |
| Tutticci et al. (2007) | R&D | 1992–2002, 789 firm-years | Considered the impact of higher-quality auditor, increased regulatory monitoring |
| Panel D: Accounting policy choices on intangibles and firm characteristics | |||
| Percy (2000) | R&D | 1993, 152 firms | R&D capitalizationVoluntary disclosures |
| Smith et al. (2001) | R&D | 1992–97 ? | R&D capitalization |
| Tutticci et al. (2007) | R&D | 1992–2002, 789 firm-years | R&D capitalization |
| Day and Harnett (1999) | GW | 1996, ? | ISOYD amortization |
| Wong and Wong (2001) | GW | 1989–93, 113 firm-years | Acquired goodwill: IOS vs contracting explanations |
| Bradbury et al. (2003) | GW | 1993, 41 firms | Goodwill amortization: IOS vs contracting and opportunism explanations |
| Wyatt (2005) | IIA, GW, R&D | 1993–97, 1366 firm-years | Proportion of intangible assets: technology and property rights vs contracting |
| Panel E: Financial analysts, preparers and auditors | |||
| Smith et al. (2001) | R&D | 1992–97, 108 497 firm-years | Capitalized vs simple analysts' algorithm |
| Securities Institute of Australia (1995) | GW | 1995 survey, 56% response rate | Analysts views on goodwill |
| Matolcsy and Wyatt (2006) | IIA | 1990–97, 291 firms | Analysts following and forecast accuracy |
| Keys and Ardern (2008) | IIA | 2008 14 interviews | Users, preparers and auditors views on GW |
- a GW = goodwill, IIA = identifiable intangible asset (excluding R&D), R&D = research and development.
One weakness of DMSS, from the view of standard setters, is that it focuses solely on R&D. The Australian history suggests that goodwill and IIA are so related that a piecemeal approach led to accounting arbitrage. Furthermore, an implication of Bugeja and Gallery (2006) is that an impairment testing regime for goodwill implies internally generated goodwill is being substituted for acquired goodwill. It is likely that this conclusion also applies to R&D. Hence, it is difficult to simply deal with purchased intangibles without implicitly dealing internally generated intangibles.
CONCLUDING REMARKS
My interpretation of the evidence for standard setters is that:
- •
Capitalization of historical cost amounts has short term (value) relevance.14 For example, the evidence suggests that goodwill declines with age and there is no significant value relevance after two years (Bugeja and Gallery, 2006, p. 533).
- •
Amortization of intangibles is irrelevant. At best, if a purchased intangible is regarded as an asset the amortization period should be very small.
- •
In the absence of transaction data, the revaluation of intangibles is fraught with valuation difficulties (i.e., mostly reliability) and provides managers with considerable discretion.15
- •
An impairment regime, which is simply a downward revaluation regime, is fraught with the same difficulties as a full valuation regime. The current AIFRS regime has particular difficulties relating to the discount rate applied and the concept of value-in-use (see Lonergan, 2007).
Finally, my own view on intangible accounting. If we define an asset in terms of property rights owned by the firm then most intangibles (e.g., R&D and goodwill in particular) are not existing property rights owned by the firm but anticipated property rights that might be owned in the future. Purchased goodwill and R&D are the disbursements of existing resources in anticipation of future earnings. Such expenditure should be expensed, not capitalized. This approach has long been advocated (see Catlett and Olson, 1968). It is notable that the dangling debit and immediate write-off are the most common forms of goodwill accounting practice pre-accounting standards (Anderson and Zimmer, 1992). It is also the way analysts treat goodwill (AIMR, 1993). Hence, the reporting of financial and non-financial information relating to intangibles is therefore a supplementary disclosure issue.
