Full Terms & Conditions of access and use can be found athttps://www.tandfonline.com/action/journalInformation?journalCode=rabr20

Accounting and Business Research

ISSN: (Print) (Online) Journal homepage: https://www.tandfonline.com/loi/rabr20

The real effects of a new accounting standard:the case of IFRS 15 Revenue from Contracts withCustomers

Christopher J. Napier & Christian Stadler

To cite this article: Christopher J. Napier & Christian Stadler (2020) The real effects of a newaccounting standard: the case of IFRS 15 Revenue�from�Contracts�with�Customers , Accountingand Business Research, 50:5, 474-503, DOI: 10.1080/00014788.2020.1770933

To link to this article: https://doi.org/10.1080/00014788.2020.1770933

© 2020 The Author(s). Published by InformaUK Limited, trading as Taylor & FrancisGroup

Published online: 26 Jun 2020.

Submit your article to this journal

Article views: 4199

View related articles

View Crossmark data

Citing articles: 3 View citing articles

The real effects of a new accountingstandard: the case of IFRS 15 Revenue from

Contracts with Customers

CHRISTOPHER J. NAPIER * and CHRISTIAN STADLER

School of Business and Management, Royal Holloway University of London, Egham, UK

International Financial Reporting Standard 15 (IFRS 15) Revenue from Contracts withCustomers has significantly changed the philosophy of revenue recognition, not only toprovide a fairer representation of corporate revenues, but also to inhibit the use of revenuesfor ‘earnings management’ purposes. We provide a framework to analyse the various effectsof new and amended accounting standards. Changes in how companies recognise, measure,present and disclose their revenues (accounting effects) can affect how companies and theirtransactions are understood, both internally and externally (information effects), can changesecurity prices (capital market effects) and can change how companies operate, and theircosts and cash flows (real effects). We provide empirical evidence, based on a review ofcorporate annual reports, comment letters and interviews, on the effects of IFRS 15. We findevidence of accounting, information and, to a lesser extent, real effects, although, outside afew industries, IFRS 15 has had relatively little impact on the recognition and measurementof revenue.

Keywords: financial reporting; IFRS 15; revenue; accounting standards; real effectsJEL Classification: M41

1. Introduction

The introduction or amendment of accounting regulations, including financial reporting stan-dards, may lead to changes in how entities carry on their activities. A new standard may revisethe calculation of accounting numbers through amending the ways in which assets, liabilities,income and expenses are recognised and measured. As accounting numbers are used in contracts,and form the basis for determining tax liabilities, different numbers are likely to lead to differentcash flows, unless contracts are written in such a way as to ignore subsequent accounting changes(this is sometimes referred to as ‘frozen GAAP’ – see, for example, Leftwich 1981; Christensenand Nikolaev 2017).

© 2020 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis GroupThis is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives License(http://creativecommons.org/licenses/by-nc-nd/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium,provided the original work is properly cited, and is not altered, transformed, or built upon in any way.

*Corresponding author. Email: [email protected]

Accounting and Business Research, 2020Vol. 50, No. 5, 474–503, https://doi.org/10.1080/00014788.2020.1770933

However, changes in accounting regulations may also lead entities to revise the ways in whichthey operate. Changing how certain transactions are accounted for may make these transactionsappear more, or less, attractive to those outside the entity. This is particularly the case where thetransactions were originally selected and structured so that they would be accounted for in waysthat were, at the time, seen to be favourable to the entity involved. The impact of a revised stan-dard may also be experienced at a more ‘micro’ level, as entities change the detailed structure oftheir transactions and contracts both to comply more easily with the newly introduced provisionsand to maintain an attractive appearance under the new rules. The information required by the newform of accounting may even change howmanagement regard the operations of the entity, leadingthem to realise that past economic decisions were sub-optimal. Where a change in an accountingrule or standard gives rise to changes in how an entity operates or affects its cash flows, we can saythat the accounting change has ‘real effects’. Although we consider that real effects arise at thelevel of the individual entity, the combined impact of such real effects across all entities,which may affect the economy as a whole, could be described as the ‘economic consequences’(Zeff 1978) of an accounting change.

In May 2014, the International Accounting Standards Board (IASB) published InternationalFinancial Reporting Standard 15 Revenue from Contracts with Customers (IFRS 15 – IASB2014). At the same time, the Financial Accounting Standards Board (FASB) published AccountingStandards Classification Topic 606 (ASC 606 – FASB 2014), with the same title. These standardswere the result of a development process extending over twelve years, intended to replace, in theIASB’s case, a brief and outdated revenue recognition standard, and in the FASB’s case, over100 separate guidelines for recognising revenue. The revenue number in the income statement rep-resents the ‘top line’ in the same way that profit after tax (earnings) is the ‘bottom line’. The revenuenumber is an overall indication of what an entity has achieved in a period, in terms of selling goodsand providing services. Because recognition of revenue is usually associated with recognition ofexpenses, such as cost of sales, revenue and earnings are intimately linked. If an entity seeks toreport higher earnings, one way of achieving this is to increase reported revenue. As we discussin section 3.1 below, the call for a new revenue standard was stimulated not only by a sense thatexisting standards provided inadequate guidance for entities entering the twenty-first century, butalso by fears, evidenced in some cases by financial scandals, that entities were able to engage in‘earnings management’ through recognising revenues earlier (or in exceptional cases later) thanmore conventional accounting practice would accept. A new revenue accounting standard, there-fore, would at the same time give clearer guidance on revenue recognition for all entities with con-tracts with customers and reduce the potential for earnings management.

Entities have been required to apply IFRS 15 for accounting periods beginning on or after 1January 2018 (although earlier adoption was possible). The standard replaced InternationalAccounting Standard 18 Revenue (IAS 18 – IASC 1993b), which had first been issued in1981, as well as International Accounting Standard 11 Construction Contracts (IAS11 – IASC1993a), which had first been issued in 1979. Both these standards were last substantiallyrevised in 1993. While IAS 18 required revenues to be recognised when the risks and rewardsof ownership of goods had been substantially transferred from seller to buyer, IFRS 15 hasadopted a ‘performance obligation’ approach, where revenue is recognised as and when anentity performs obligations included in a contract with a customer. Entities should thereforereview their contracts to identify the performance obligations that the contracts impose on theentity. Such a review provides entities with the opportunity of changing how they structure con-tracts, and in extreme cases to make substantial modifications to their business models. In otherwords, the new revenue accounting standard may lead to real effects.

The aim of this paper is to provide a framework for understanding the effects of a change inaccounting regulations, which would generally take the form of new or amended legislation, or

Accounting and Business Research 475

new or amended standards, and to apply this framework to analyse a specific example of a newaccounting standard. The framework includes accounting, information, capital market and realeffects, but in the present paper we do not address whether the adoption of IFRS 15 has hadany effects on capital markets, for example changes to security prices, concentrating instead onaccounting, information and real effects. We chose IFRS 15 because it affects all active entitiesthat sell goods or provide services, and so the implementation of IFRS 15 is likely to reveal awide range of effects. Our contribution is twofold. First, our framework could be used in otherstudies of the effects of accounting change, ensuring that future researchers consider the widerange of potential effects. Second, through our example study of IFRS 15, we show how entitieshave addressed the impact of a new financial reporting standard. We conclude that IFRS 15 hasled to much effort in implementing the standard, but that, in terms of the impact on reportedrevenue in the income statement, the new standard has not generated significant changes in theaccounting numbers for the majority of companies.

We begin by briefly discussing the existing literature on real effects of accounting, and weprovide a framework setting out the different types of effect that a new accounting requirementmay induce. We show how direct accounting effects of applying a new standard can induceadditional effects, which we classify as information effects, capital market effects and realeffects, and that these additional effects can themselves induce indirect accounting effects. Wethen consider the history of the concept of ‘revenue’ in financial reporting, with a focus on theUK, the USA and the international standard-setting context. This is followed by a discussionof our empirical findings, based on an examination of annual reports prepared by the largest Euro-pean companies, comment letters submitted to the IASB during the development process leadingto IFRS 15, and a small number of interviews. The final section sets out some conclusions.

2. Accounting changes and real effects

2.1. The real effects concept

It has long been recognised that entities pay regard to accounting requirements when structuringtheir transactions, and even when deciding which transactions to undertake. Accounting numbersmay be used for regulatory and contractual purposes, and changes in how accounting numbers aredetermined will have an impact on cash flows that depend on particular laws and other regulationsusing specific accounting numbers, as well as cash flows that are set by contracts. As early as1972, Ball (1972, p. 1) noted that:

The distinction between real and accounting effects on income is far from being clear, since changes inaccounting techniques can be responses to real variables (such as changes in expected future inventoryprices, or the firm moving into a new industry), and they can also induce real effects (such as changesin taxable income).

Ball uses the term ‘real effects’ in a restricted sense, the implication of the phrase ‘changes intaxable income’ being that an accounting change can alter the amount of profit subject to taxand therefore can lead to the entity paying a different amount of tax from its liability under theold accounting rules. However, the notion of ‘real effects’ has gradually expanded to embracechanges that an entity makes to how it carries out its activities.

Zeff (1978, p. 56) was among the first to point out that the standard-setting process in the USAin the 1960s and 1970s was increasingly being influenced by ‘individuals and groups that hadrarely shown any interest in the setting of accounting standards’, who ‘began to invoke argumentsother than those which have traditionally been employed in accounting discussions’. These argu-ments were not related to the technical characteristics of a proposed accounting standard, but

476 C.J. Napier and C. Stadler

rather the possibility that the standard would have a detrimental impact on both individualbusinesses and the wider economy and society. Zeff (1978, p. 59) observed that such externalpressure arose in particular in the context of accounting for marketable equity securities,leases, and oil and gas exploration and drilling costs.

More recently, Kanodia and Sapra (2016, p. 624, emphasis in original) have set out what theydescribe as the ‘real effects hypothesis’. This states that:

[T]he measurement and disclosure rules that govern the functioning of accounting systems – whicheconomic transactions are measured and which are not measured, how they are measured and aggre-gated, what is disclosed to capital markets and how frequently such disclosures are made – have sig-nificant effects on the real decisions that firms make.

These authors argue that the notion of providing decision-useful information to investors and theprinciple of ‘representational faithfulness’ may be ‘insufficient guides to standard setting’(Kanodia and Sapra 2016, p. 624), because they do not necessarily reflect the extent to whichaccounting requirements may lead entities to change what they do. To more critical accountingscholars, the recognition that accounting changes can have real effects is hardly a novelty:classic discussions such as that of Ruth Hines (1988) already pointed out how financial reportingis not a simple mirror of an externally given ‘reality’, but helps to construct the very ‘reality’ that itclaims to represent.

The notion of studying real effects of financial reporting provisions has been increasinglyadvocated, particularly over the past 10–15 years. For example, Leuz and Wysocki (2016,p. 530, emphasis in original) call for ‘researchers to examine non-traditional disclosure andreporting settings, especially to learn about the real effects of disclosure mandates’. Theydefine real effects as ‘situations in which the disclosing person or reporting entity changes its be-havior in the real economy (e.g. investment, use of resources, consumption) as a result of the dis-closure mandate’. An indicator of how little attention was given to real effects in past accountingresearch is the review of empirical research on accounting choice undertaken by Fields et al.(2001). This paper does not mention the expression ‘real effects’, and, in her comment on thepaper, Jennifer Francis (2001, p. 311) observes that ‘the motivation for a real decision [onewith cash flow implications] may be unrelated to the accounting outcome’, suggesting that explor-ing for real effects may be unproductive. This negativity did not stop scholars such as ChandraKanodia from discussing, over many decades, how disclosure of accounting information mayaffect not just security prices but also corporate production-investment decisions (Kanodia1980; see also Kanodia and Mukherji 1996; Kanodia 2006), but perhaps the difficulty of observ-ing corporate decisions and actions, as compared to variables such as security prices, acted as adeterrent to extensive work on real effects.

Nonetheless, a growing literature is addressing the impact not just of disclosures but also ofchanges in accounting regulations and practices on corporate decisions. For example, Barth et al.(2017) have looked at how the quality of integrated reporting in South African companies mayenhance the companies’ stock market value not only through providing better information forcapital markets but also by improving investment efficiency – which they see as a ‘real effectschannel’ for enhancing corporate value. Ernstberger et al. (2017) suggest that the introductionof a European Union requirement for listed companies to provide quarterly ‘interim managementstatements’ has led to an increase in what they call (following Roychowdhury 2006) ‘real activi-ties manipulations’, such as changing production levels and varying expenditure on research anddevelopment or on selling, general and administrative expenses. These commercial effects havebeen identified as methods by which companies can engage in ‘real earnings management’ (see,for example, Healy and Wahlen 1999), and Bereskin et al. (2018) have suggested that real

Accounting and Business Research 477

earnings management can have measurable real effects on the level of innovation: companies thatmanaged reported earnings by spending less on research and development ultimately filed fewerpatents. Although in one sense the result was unsurprising, the research demonstrated that redu-cing research and development spending was more likely to be driven by the earnings manage-ment motive than by other economic and commercial considerations. Additionally, there isevidence of real effects that benefit employees: Christensen et al. (2017) find that mine-relatedinjuries decreased after the introduction of mandatory disclosure about mine safety records infinancial reports.

Studies of real effects rarely address specific changes in accounting regulations and standards.An exception to the general lack of work in this field is the study by Dou et al. (2018), who inves-tigate the real effects of a change in US accounting standards effective in 2010 relating to account-ing by banks for securitisations and the consolidation of variable interest entities. They found thatthe requirement to recognise substantial securitised assets (about US$800 billion) on balancesheets affected banks’ mortgage approval rates: banks recognising greater amounts of ‘new’securitised assets tended to show larger decreases in mortgage approvals. Arguably, the account-ing change in question was a requirement for additional disclosure, rather than one that affectedthe measurement of assets, liabilities, income and expense, and much ongoing work in the area ofreal effects still focuses on disclosure rather than measurement: for example, Dou and Zou (2019)examine whether US banks’ disclosure of the geographical distribution of small business lendinghas an impact on lending policies (as measured by the proportion of non-performing loans tosmall businesses). This means that there is a need for studies that examine how accounting regu-lation changes that require different measurement methods to be adopted, rather than simply man-dating additional disclosures, can have real effects. The existence and nature of such effects couldbe examined at a ‘macro’ level, by identifying variables that could be used as evidence for realeffects (this is the approach of Barth et al. 2017), but also at the ‘micro’ level, by attemptingto identify real effects in specific entities. In this paper, we adopt a mainly ‘micro’ approach,using a range of evidence to probe for real effects. However, before turning to the subjectmatter of this paper, the new revenue accounting standard, we set out a framework for understand-ing the various effects of a change in an accounting regulation.

2.2. Effects of a new or amended accounting standard

Although we refer to accounting standards in this section, our analysis would apply to any changein accounting regulation, for example, changes in legislation relating to corporate reporting. Anew or amended accounting standard (hereafter ‘new standard’ for simplicity) can havevarious effects. Figure 1 shows how such effects may be classified and how the different cat-egories are related to each other. We will use the classification throughout this paper in orderto structure our discussion of the various effects of a specific new accounting standard.

A new accounting standard necessarily leads to accounting effects [A]. These include changesin recognition [A.1], measurement [A.2], presentation [A.3] and disclosures [A.4]. A new stan-dard may require items previously not included in the financial statements to be recognised[A.1], for example a leasing standard may require certain leases previously kept off-balancesheet to be accounted for as assets and liabilities. In some situations, a new standard may deter-mine that certain items, such as deferred costs, are no longer to be recognised. In itself, recog-nition of a new item will require that item to be measured, but a new standard may change thebasis on which particular items already included in the financial statements are measured[A.2]. For example, a standard might require entities to measure a specific asset at fair value,where previously the required measurement basis was historical cost. A new standard maychange the way in which certain items are presented [A.3], for example, items previously included

478 C.J. Napier and C. Stadler

in other comprehensive income may be moved into profit or loss. Many changes in recognitionand measurement are associated with additional or changed disclosure requirements, and new dis-closures are often the result of legislation rather than accounting standards [A.4]. Disclosure couldrelate to new statements (for example, a cash flow statement) or to specific items. A new standardmay not lead to all of these changes. For example, an increase in disclosures need not be associ-ated with recognition, measurement or presentation changes. These accounting effects areregarded as direct effects of the standard. In Figure 1, we represent the direct accountingeffects of a new standard using a double arrow.

The direct accounting effects of a new standard can induce various additional effects, which werefer to as primary additional effects. We classify these as information effects [I], capital marketeffects [C] and real effects [R]. The induction of these various effects is represented as solidsingle arrows in Figure 1. Regarding information effects [I], first, the new standard might leadinternal users of accounting information to a better understanding of transactions [I.1]. This canhappen because the commercial implications of a transaction are more appropriately reflected inthe accounting numbers. For example, managers may come to realise that the rights and obligationsarising from a particular transaction are more extensive and complex than they had previouslythought. Changes in financial reporting as a consequence of adopting a new standard may alsoenhance (but could impair) how external users understand transactions [I.2]. If the accountingeffects [A] are significant, management might communicate the effects to stakeholders [I.3], poss-ibly before the publication of financial information under the new standard.

Regarding capital market effects [C], first, the accounting effects [A] may impact both equitymarkets [C.1] and debt markets [C.2]. In equity markets, share prices may change because thedisclosure of additional information leads investors to revise their expectations of the amounts,timing and uncertainty of future cash flows, or because new numbers suggest that previous esti-mates of corporate value are no longer tenable. New accounting information following adoption

Figure 1. Effects of a new or amended accounting standard. Note: This figure shows the effects of a new oramended accounting standard. The double arrow represents direct accounting effects. The solid single arrowsrepresent primary additional effects. The dashed single arrows represent secondary additional effects. Thedashed single arrows going into [A] represent indirect accounting effects.

Accounting and Business Research 479

of a new standard may also affect bid-ask spreads and trading volume. In debt markets, newaccounting information may change perceptions of risk, and therefore borrowing costs and thepricing of credit default swaps may change. Managers face pressures to maintain or improve cor-porate performance and financial position [C.3], and changes in accounting numbers that are usedby investors and lenders to assess managerial performance may induce further effects (see ‘sec-ondary additional effects’ below).

Real effects [R] are those effects that change how an entity undertakes its operations or thataffect its cash flows. First, a new standard is likely to result in both implementation and ongoingapplication costs [R.1], and for some new standards these costs may be substantial for specificentities. On the other hand, a new standard could reduce costs through simplifying recognitionand measurement or reducing disclosure. We consider that the need to use resources forimplementation is a real effect because it involves the allocation of resources to a use thatmany managers and stakeholders would consider to be unproductive. This is consistent withthe definition of real effects provided by Leuz and Wysocki (2016) quoted above. Second, ifthe new standard results in accounting effects that managers consider to be undesirablebecause of how accounting numbers are used in contracts, the entity may amend its contracts[R.2]. For example, a new standard might affect accounting-related debt covenants and any result-ing problem can be solved by amending the debt contract. Third, if a new standard results in unde-sirable accounting outcomes because of certain business practices, which cannot be solved bychanging contracts, a company may change its behaviour [R.3]. This may mean withdrawingfrom certain activities or, less dramatically, modifying how the business operates.

Changes in accounting numbers and disclosures could have regulatory effects [R.4], forexample, an entity may come under fresh scrutiny from regulators, or avoid previous regulation,because its reported revenues or profits are higher (or lower) as a result of an accounting change,and this brings the entity inside (or takes it outside) the domain of particular regulators. Wheretax liabilities and the scope for dividend payments depend on reported profits or on other accountingnumbers, changes in these figures may lead to a different tax liability and may restrict, or enable,dividend payments [R.5]. Finally, there may be other real effects [R.6]. For example, where account-ing numbers are used in contracts, such as remuneration contracts, that have cash flow implications,changes in entity cash flows may also be considered as real effects of the new accounting standard.

In addition to the primary additional effects, there are likely to be secondary additional effects.These are represented as dashed single arrows in Figure 1. First, information effects [I], capitalmarket effects [C] and real effects [R] may feed back into accounting effects [A]. For example,fears that external users such as competitors may gain advantage from more detailed knowledge ofa company’s operations [I.2] may influence the accounting disclosures ultimately included in anentity’s financial statements [A.4]. Pressures to improve performance [C.3] may result in changesin the accounting numbers [A.1 and A.2]. Behavioural changes, such as discontinuing certain activi-ties [R.3] will result in changes in the accounting numbers [A.1 and A.2]. Therefore, a new standardcan induce both direct and indirect accounting effects [A]. Moreover, there may be indirect effectsbetween [I], [C] and [R]. For example, a better understanding of transactions on the part of investors[I.2] may cause share prices to change [C.1] and a better understanding of transactions on the part ofentity [I.1] may lead the entity to revise contracts [R.2] and change its behaviour [R.3].

3. The new revenue accounting standard

3.1. The road to the new standard

3.1.1. The UK’s road

To older British accountants, the word ‘revenue’, to refer to sales or turnover, is an innovation.The Oxford English Dictionary (OED 2011) notes that the word ‘revenue’ has ‘senses relating

480 C.J. Napier and C. Stadler

to income’, and defines revenue as ‘the return, yield or profit of any land, property or other impor-tant source of income’, as ‘income, specifically that deriving from property, possessions, trade orinvestment’, as ‘an amount of money regularly accruing to a person’, and as ‘the annual income ofa government, state, etc., from which public expenses are met’. The word ‘revenue’ was oftenfound in the expression ‘revenue account’, which was distinguished from the capital accountin a double-account system of financial reporting.

In an early accounting textbook, Advanced Accounting, Lawrence Dicksee, the first account-ing professor at a British university, discusses revenue only in the context of the distinctionbetween capital and revenue: ‘[I]t is hardly overstating the case to say that most of the errorsof principle that are perpetrated in practice arise from the lack of ability, or lack of desire, tostrictly discriminate between Capital and Revenue items’ (Dicksee 1903, p. 3). He defines‘revenue receipts’ as:

[T]hose which properly arise out of the business operations of the undertaking – i.e. earnings.Unless the business is upon a cash basis there will, however, always be some discrepancybetween the actual earnings and the receipts in respect thereof, and the proper item to credit toRevenue Account will be the true earnings for the period, rather than the actual receipts incash. (Dicksee 1903, p. 7)

Dicksee provides no guidance, either in the chapter on capital and revenue, or elsewhere, on what‘actual’ or ‘true’ earnings might be, suggesting that British accountants at the beginning of thetwentieth century had an unproblematic notion of what constituted sales revenues. This maybe because the relatively recent Sale of Goods Act 1893 had codified English law on commercialtransactions, and provided a default set of conditions that specified the point in time when a salecould be deemed to take place. These conditions emphasised the transfer of ‘property’ in goods,usually evidenced by possession of the goods passing from the seller to the buyer. Although theSale of Goods Act has been amended over the years, its core legal principles still largely apply inthe UK.

This lack of interest continued until the 1960s, when financial scandals involving the rec-ognition and measurement of sales revenues drew attention to the issue. For example, theRolls Razor Company Limited, which imported cheap domestic appliances and sold them onextended credit terms, was criticised following its collapse in 1964 for recognising salesrevenue in full on delivery of appliances, where there was a substantial exposure to creditdefault on the part of customers, and not recognising liabilities for customer claims and rejec-tions because of the poor quality of the machines being sold. The company’s chairman, JohnBloom, was subsequently fined £30,000 under false accounting charges (Benson 1989,p. 143). The Rolls Razor scandal prompted the Institute of Chartered Accountants inEngland and Wales (ICAEW) to issue a Recommendation on Accounting Principles on HirePurchase, Credit Sale and Rental Transactions (N23 – ICAEW 1964). Suppliers of goodson hire purchase and instalment sale terms were recommended to split the sales proceedsbetween the sale of the item and interest income arising over the period of the transaction,and to make appropriate adjustments (as expenses rather than as deductions from the sales pro-ceeds) to take account of credit risks and bad debts.

British companies were required to disclose an amount for sales revenue, described as‘turnover’, for the first time by the Companies Act 1967, but no definition of turnoverwas provided in the Act. Companies were required to state how the amount of turnoverwas determined, but there was no consistency of practice. An example of a definition of turn-over appears in the financial statements of Rolls-Royce Limited for the year ended 31December 1969:

Accounting and Business Research 481

Sales at invoice value to external customers, and in the case of uncompleted contracts, the estimatedsales value of actual deliveries or work done on development contracts. (Quoted in ICAEW 1971,p. 23)

Interestingly, the turnover figure was included in a note headed ‘Revenue from all sources’, whichalso included royalties, management fees and income from unquoted investments.

Although the application of the Sale of Goods Act 1893 meant that determining when a saletook place, at which point revenue would be recognised, was normally uncontroversial, a law casein 1976 (Aluminium Industrie Vaassen B.V. v. Romalpa Aluminium Limited [1976] 1 WLR 676 –the ‘Romalpa case’) raised the issue of how to account for sales where physical possession ofgoods was transferred to the buyer, but the seller retained legal title to the goods until theywere paid for, and could recover possession of the goods if the buyer did not pay for them.The ICAEW issued an Accounting Recommendation in October 1976, stating that ‘the commer-cial substance of the transaction should take precedence over its legal form where they conflict’(ICAEW 1976, para. 4). The ICAEW noted that recognising revenue from transactions wherelegal title did not pass on transfer of the goods themselves was already standard practice inhire purchase transactions, where legal ownership was transferred only after all payments hadbeen made by the buyer. Subsequent reviews of the Romalpa case (for example, De Lacy1995) have concluded that the significance of the case both legally and for financial reportinghas been minimal, but the case illustrates how by the 1970s accountants in the UK werealready regarding transfer of legal title in goods as neither necessary or sufficient for recognisingrevenue from the sale of those goods.

The Companies Act 1981 brought into effect the requirements of the then European Commu-nity’s Fourth Directive on Company Law (EC 1978), which specified standard formats for thepresentation of company financial statements. The formats for the profit and loss account specifieddisclosure of ‘net turnover’, so British companies continued to use the turnover term in theirfinancial statements. Article 28 of the Fourth Directive defined ‘net turnover’ as comprising:

[T]he amounts derived from the sale of products and the provision of services falling within the com-pany’s ordinary activities, after deduction of sales rebates and of value added tax and other taxesdirectly linked to the turnover.

This definition was included, in modified form (‘trade discounts’ replacing ‘sales rebates’) in theCompanies Act 1981 and in subsequent UK companies legislation.

The Accounting Standards Board (ASB) commented on revenue in its Statement of Principlesfor Financial Reporting (ASB 1999). The Statement of Principles used the term ‘gains’ for ‘allforms of income and revenue as well as all recognised gains (realised and unrealised) on non-revenue items’ (ASB 1999, chapter 4). As ‘gains’ include ‘revenue’, it would seem natural toinclude a definition of ‘revenue’, but no such definition was provided. Gains were defined as‘increases in ownership interest not resulting from contributions from owners’ (ASB 1999,para. 4.39). Recognition of a gain, in particular revenue, was linked with the occurrence of the‘critical event’, which was defined as ‘the point in an operating cycle at which there willusually be sufficient evidence that the gain exists and it will usually be possible to measurethat gain with sufficient reliability’ (ASB 1999, para. 5.33). The definition merely restated thegeneric recognition criteria for any element of financial statements. Some further clarificationwas provided: carrying out all obligations under an agreement except for a few minor acts of per-formance would imply that the critical event had occurred; the critical event would not usuallyhave occurred if there was a significant likelihood that the buyer would reject the goods or

482 C.J. Napier and C. Stadler

services; and a contract performed in stages might involve a series of critical events, among whichthe expected gain would have to be allocated.

The British accounting standard setters did not produce a statement of standard accountingpractice or a financial reporting standard on revenue until 2003, and even then this was in theform of an application note to a more general standard (Financial Reporting Standard 5 Account-ing for the substance of transactions – ASB 1994). However, by 2003, countries in the EuropeanUnion, including the UK, were preparing to adopt IFRS for the consolidated financial statementsof public interest entities. Standard setters in the UK focused on companies that were not coveredby of the new IFRS requirement, producing the omnibus Financial Reporting Standard 102: TheFinancial Reporting Standard Applicable in the UK and Republic of Ireland. The most recentversion of this (FRC 2018) substantially incorporates IAS 18, which means that, as at early2020, UK and International GAAP differ in the area of accounting for revenue.

3.1.2. The USA’s road

North American accountants were more likely than their British counterparts to use the word‘revenue’ to refer to sales. William Paton, one of the earliest US academic accounting theorists,conceptualised the business entity as consisting of ‘properties’ and ‘equities’, and saw revenue asthe conversion of properties, represented by ‘the effective cost of purchased commodities and ser-vices’ (Paton 1922, p. 146), into equities. Paton distinguished between ‘gross revenue’, whichwas the amount of sales, and ‘net revenue’, which was the increase in equities that arosethrough a sales transaction, and was equal to the amount receivable from customers less thecost of the properties involved in the sale.

Paton devoted an entire chapter of his book Accounting Theory (Paton 1922) to discussingrevenue. He noted that, in most of his previous discussion, he had assumed that the mainproblem of corporate accounting was ‘expense determination and apportionment’, and that ‘theassignment of revenue credits to the particular accounting period can be readily accomplishedby means of ordinary bookkeeping’ (Paton 1922, p. 443). However, he demonstrated that thisview of revenue determination as straightforward was misconceived, by asking ‘What is a satis-factory test or evidence of revenue? When is revenue realized?’ (Paton 1922, p. 444). Patonrejected the view that the key event for recognising revenue is the securing of a customer’sorder as evidenced by a contract with the customer (Paton 1922, p. 453), but proposed thatrevenue may be recognised when there is an enforceable right to receive payment from thecustomer: this would normally arise when goods are delivered (Paton 1922, p. 455). Paton wasprepared to accept recognition of interim revenue in the case of long-term processes (theexample he gives is building a ship – Paton 1922, p. 461) extending over several accountingperiods, but adoption of a ‘percentage of completion’ approach was, to Paton, the exceptionrather than the rule.

Paton did not offer a clear definition of revenue, but he linked revenue recognition closely torealisation. He carried this view forward into An Introduction to Corporate Accounting Stan-dards, which he wrote with A.C. Littleton for the American Accounting Association (Patonand Littleton 1940). Zeff (2018) has suggested that Paton was the main author of the monograph’schapter on revenue. Revenue was defined as ‘the product of the enterprise, measured by theamount of new assets received from customers’, and a distinction was drawn between earnedrevenue, created ‘by the entire process of operation, by the totality of business effort’, and realisedrevenue, which results from ‘conversion of product into cash or other valid assets’ (Paton and Lit-tleton 1940, p. 46). There is still some equivocation about whether revenue is a gross or netconcept, although the authors tend to use the word ‘income’ to describe net revenues after deduc-tion of all expenses. The authors do not provide an explicit rationale for disclosing a revenue

Accounting and Business Research 483

figure in the first place, with their emphasis falling on cost determination. The monograph advo-cates a matching approach, with the aim of determining ‘income’ (profit) for the period. The ideasof Paton and Littleton were to be highly influential in providing a framework for subsequent con-ceptual thinking about financial reporting in the USA until the late 1960s, and even beyond.

In the USA, the late 1960s saw a series of scandals where companies would use generallyaccepted accounting principles (GAAP) creatively, and often inappropriately, to boost revenues.The staunch critic of US financial reporting Abraham Briloff (1972, 1976) discusses cases such asNational Student Marketing Corporation (which inflated revenues from marketing and marketresearch contracts by using a percentage of completion method that strictly applied only to gov-ernment-funded construction contracts), Telex and Memorex (which reported revenues in respectof assets that they had constructed and planned to sell to finance companies to be leased out, butwhere the transfer did not actually take place by the end of the financial year), Liberty Equities(which sold options to buy land to a company connected to the chief executive officer on thelast day of the financial year, and subsequently reacquired the options through a related party),and Stirling Homex (which recognised revenues when construction of modular housing unitswas completed and specific units assigned to particular contracts, even when payment was notdue until delivery and installation of the units). Scandals such as these formed part of the back-ground that led to the establishment of the FASB in 1973 (Previts and Merino 1998, pp. 365–366).

One of the early projects of the FASB was the development of a conceptual framework forfinancial reporting, and in 1980, a statement on the ‘elements’ of financial statements of businessenterprises (SFAC 3 – FASB 1980) was published. One of the elements, the building blocks of thebalance sheet and income statement, was revenues, defined as:

[I]nflows or other enhancements of assets of an entity or settlements of its liabilities (or a combinationof both) during a period from delivering or producing goods, rendering services, or other activities thatconstitute the entity’s ongoing major or central operations. (FASB 1980, para. 63).

This concepts statement had little to say about revenue, beyond noting that revenue is usuallyrecognised when goods are delivered or services performed, but may also be recognised whencash is received, when production is completed, or as production progresses. Further consider-ation of revenue recognition was deferred to the statement on recognition and measurement(SFAC 5 – FASB 1984).

That statement drew a distinction between ‘earnings’ (which the statement suggested wereequivalent to the previously used term ‘net income’), and ‘comprehensive income’. Revenuewas regarded as a component of earnings, from which expenses were deducted. The statementprovided criteria for recognising revenues: they needed to be realised or realisable, and theyneeded to be earned. Realisation involved exchange of goods or services for cash or for claimsto cash, while revenue was realisable if goods or services were exchanged for assets that couldbe readily convertible into cash or claims to cash. Revenues were earned if ‘the entity has sub-stantially accomplished what it must do to be entitled to the benefits represented by the revenues’(FASB 1980, para. 83). A few points of guidance on revenue recognition were included, coveringissues such as the percentage of completion method, recognition of services or rights to use assetscontinuously over time, and recognition of revenue in exceptional cases when production is com-plete even though the customer has not taken possession of the goods (FASB 1980, para. 84).

However, this conceptual guidance was not enough for US companies, and further pronounce-ments on recognising revenue for specific transactions or for particular industries were made bythe FASB, the American Institute of Certified Public Accountants (AICPA) and industry bodies. Itwas estimated that there were over 100 different pronouncements on revenue recognition in USGAAP by the time the new revenue accounting standard was published (Trainer 2019). Even

484 C.J. Napier and C. Stadler

before this, US regulators were concerned that there were significant problems with revenue rec-ognition. For example, Arthur Levitt, the Chairman of the Securities and Exchange Commission(SEC), expressed concern over earnings management, specifically referring to ‘premature recog-nition of revenue’ as an ‘accounting gimmick’ and an ‘illusion’ (Levitt 1998). Levitt’s criticismwas supported by Lynn E. Turner, the SEC’s Chief Accountant. Turner (1998) noted that ‘thenumber of reported instances of companies’ improperly booking revenues has become an embar-rassment to all of the accounting profession’. He gave a list of situations where he thought thatrevenues had been recognised prematurely:

Delivery of the product to the end user’s site has not occurred.Agreements have not yet been accepted and executed by the customer.The seller has to complete remaining obligations, such as installation or training.The customer unilaterally can terminate or cancel the agreement.‘Just in time’ arrangements exist, with [free on board] delivery terms, and revenue is recognized priorto arrival at the delivery destination.Upfront fees are recognized immediately upon receipt notwithstanding an agreement to provide ser-vices, discounts or products during an ensuing membership period. (Turner 1998)

The concern of Levitt and Turner was based on recurrent interventions by the SEC, whichrequired many US companies to restate their financial statements. Dechow et al. (2011, p. 29) ana-lysed the 2,261 Accounting and Auditing Enforcement Releases issued by the SEC from May1982 to June 2005. The SEC had identified accounting misstatements in the case of 676 compa-nies, of which 54% related to misstated revenue. Concern over the volume of detailed, and some-times contradictory, guidance on revenue recognition prompted the FASB to consider developinga comprehensive revenue recognition standard, firmly grounded in sound principles. However, asnoted below, the FASB would not develop a new revenue standard on its own.

3.1.3. The international road

The International Accounting Standards Committee (IASC) initially considered accounting forconstruction contracts, issuing IAS 11 in 1979. This standard allowed, and from 1993 required,the use of the percentage of completion method for such contracts where certain conditions weremet. The IASC was at this time engaged in an active programme of standard-setting, addressingvarious complex accounting problems, and the revenue standard IAS 18, issued in 1982, was rela-tively straightforward, taking up only 12 pages of text (not including illustrative examples). IAS11 and IAS 18 were updated as part of the Comparability Project undertaken by the IASC in thelate 1980s, to make its standards more attractive to international security market regulators (Camf-ferman and Zeff 2015, pp. 11–12; Roberts et al. 1996). As amended by the Comparability Projectand by subsequent small changes, IAS 18’s definition of revenue was:

[T]he gross inflow of economic benefits during the period arising in the course of the ordinary activi-ties of an entity when those inflows result in increases in equity, other than increases relating to con-tributions from equity participants. (IASC 1993b, para. 7).

This definition is an adaptation of the definition of ‘income’ in the IASC’s Framework for thePreparation and Presentation of Financial Statements (IASC 1989), with revenue being con-sidered as income arising in the course of the entity’s ‘ordinary activities’. Barker (2010) hasquestioned whether the definition of income provided by the IASC (and subsequently adoptedby the IASB in its Conceptual Framework for Financial Reporting – IASB 2018) is appropriate,and Nobes (2012) has extended this critique to definitions of revenue. Barker’s view is that

Accounting and Business Research 485

income should be regarded simply as an increase in equity (excluding contributions from equityparticipants) rather than as an increase in assets or a decrease in liabilities that results in anincrease in equity, while Nobes suggests that a literal application of the definitions of incomeand revenue would lead to double-counting of revenue in some cases and omissions ofrevenue in other cases.

IAS 18 set out two sets of conditions, one for the sale of goods and the other for the renderingof services, for revenue to be recognised. The key condition for recognising revenue from the saleof goods was that ‘the entity has transferred to the buyer the significant risks and rewards of own-ership of the goods’, reinforced by a requirement that ‘the entity retains neither continuing man-agerial involvement to the degree usually associated with ownership nor effective control over thegoods sold’ (IASC 1993b, para. 14 (a) and (b)). For services, the wording was largely the same asfor construction contracts under IAS 11, referring to the percentage of completion method. Thisapproach was justified as providing ‘useful information on the extent of service activity and per-formance during a period’ (IASC 1993b, para. 21). The standard also covered interest, royaltiesand dividends as types of revenue, and required disclosure of ‘the amount of each significant cat-egory of revenue recognised during the period’ (IASC 1993b, para. 35 (b)).

IAS 18 was a ‘principles-based’ standard rather than one setting out a collection of detailedrules, and the IASC (and subsequently the IASB) largely resisted demands to provide moredetailed guidance, beyond a few illustrative examples. None of these addressed clearly theissue of how to recognise revenue from complex contracts in which goods and services werebundled together. It was not unusual for companies attempting to apply IAS 18 to look for gui-dance to the increasing number of ‘rules-based’ provisions for revenue recognition in the USA(Camfferman and Zeff 2015, p. 355). As already noted, by the beginning of the 2000s, theFASB was minded to undertake a major review of revenue recognition, which was announcedin 2002 (Camfferman and Zeff 2015, p. 133). The IASC had been restructured as the IASBwith effect from January 2001 (Camfferman and Zeff 2015, p. 18), and it also decided toadopt a project on accounting for revenue, announced in July 2002 (Camfferman and Zeff2015, p. 133). Against the background of the Sarbanes-Oxley Act of 2002, a US legislative reac-tion to the Enron scandal (see, for example, Hart 2009), the FASB and the IASB decided to workmore closely together, with the aspiration of converging the different sets of financial reportingrequirements in the USA and internationally. This was formalised in the ‘Norwalk Agreement’,a memorandum of understanding between the FASB and IASB to work towards ‘bringing theirrespective standards into compatibility with each other’ (Camfferman and Zeff 2015, p. 76).

One of the joint projects specifically endorsed by the Norwalk Agreement was the project onrevenue. This continued to be endorsed in a further memorandum of understanding issued in 2006(Camfferman and Zeff 2015, p. 338). However, a major source of concern was the underlyingphilosophy to be adopted. A standard that was consistent with the ‘asset-liability’ basis of finan-cial reporting implied by both FASB’s Conceptual Framework (FASB 1980) and IASB’s (adoptedfrom IASC) Framework for the Preparation and Presentation of Financial Statements (IASC1989), and that used fair value measurement as far as possible, appeared to lead entities to recog-nise profits on contracts as soon as they were agreed, irrespective of notions of ‘realisation’ and‘earning’ of revenues (Camfferman and Zeff 2015, pp. 356–357; see also Wüstemann andKierzek 2005). Two groups, each advocating a different solution for revenue recognition, cameto be known as the ‘space cadets’ (who supported the pervasive use of fair values in revenue rec-ognition) and the ‘dinosaurs’ (who supported an approach involving allocation of the total con-sideration to different aspects of a contract). A compromise involving the latter approach, withsome fair value remeasurement in the case of onerous contracts, was ultimately to form thebasis of a joint discussion paper Preliminary Views on Revenue Recognition in Contracts withCustomers (FASB 2008).

486 C.J. Napier and C. Stadler

This discussion paper was revolutionary from the US perspective, because it attempted toprovide a single model for the recognition of revenue, which would apply across the wholerange of entities. The broad principles set out in the discussion paper were to be modified bytwo subsequent exposure drafts, published in June 2010 (IASB 2010) and November 2011(IASB 2011), with the final standards IFRS 15 (IASB 2014) and ASC 606 (FASB 2014) beingissued in May 2014 (Camfferman and Zeff 2015, p. 576). Following comments from companies,a long transition period was allowed, with IFRS 15 originally being mandatory for accountingperiods beginning on or after 1 January 2017, but subsequent pressure led the IASB to extendthe implementation date to 1 January 2018. The broad requirements of IFRS 15 (there aresome variations between this document and the text of ASC 606, particularly as regards the defi-nition of revenue) are set out in the next section.

3.2. IFRS 15 and its potential real effects

IFRS 15 is a much longer document than IAS 11 and IAS 18 combined, extending over some 80pages (not including the basis for conclusions and illustrative examples). Revenue is defined as‘income arising in the course of an entity’s ordinary activities’ (IASB 2014, Appendix), withincome being defined in the same terms as the IASB’s then conceptual framework (based onthe 1989 Framework – IASC 1989). The broad principle is that:

[A]n entity recognises revenue to depict the transfer of promised goods or services to customers in anamount that reflects the consideration to which the entity expects to be entitled in exchange for thosegoods or services. (IASB 2014, para IN7).

This involves a five-step approach to revenue recognition, and each of the five steps has the poten-tial to generate real effects.

First, it is necessary to identify the contract with a customer, and this may involve combiningseveral separate legal agreements into one contract or splitting a single agreement into severalcontracts for recognition purposes. Business contracts are likely to be written to achieve commer-cial goals, rather than to fit an accounting standard. However, at the margin, businesses may seekto align the form of their legal agreements to the way in which these agreements are identified forthe purposes of IFRS 15, by rewriting standard terms and conditions of operation.

Second, businesses must identify the performance obligations – the promises to transfer dis-tinct goods or services to a customer – that are undertaken because of the contract. In some cases,a business may take on obligations to a customer associated with a contract that in the past wouldhave been regarded as promotional or marketing activities. For example, a salesperson may havehad the discretion to offer incentives, such as a ‘free’maintenance agreement, to encourage a cus-tomer to buy goods. In the past, the costs resulting from this incentive might have been recognisedonly as incurred, or at best as a provision for the expected costs of fulfilling the maintenanceagreement. The implementation of IFRS 15 may encourage businesses to review what theyprovide customers so that distinct performance obligations may be more easily identified andtaken into consideration when contracts are negotiated and accounted for.

Third, it is necessary to determine the transaction price, and this causes particular difficultieswhere part of the consideration involved in a contract is variable. Variable consideration caninclude standard commercial practices such as ‘discounts, rebates, refunds, credits, price conces-sions, incentives, performance bonuses, penalties or other similar items’ (IASB 2014, para. 51),and reflecting such items in the determination of revenue may involve making estimates and jud-gements. For example, a company with a policy of giving performance rebates to customerswhose purchases exceed a given target may have to reduce the amount of revenue for the expected

Accounting and Business Research 487

rebates rather than including the rebates in cost of sales. Businesses may decide to amend con-tracts to reduce the measurement uncertainty involved in estimating variable consideration.The transaction price is net of amounts collected on behalf of third parties. The example providedfor this is ‘some sales taxes’ (IASB 2014, para. 47), but there will be situations, particularly wherethe business is an intermediary, where the amounts collected on behalf of third parties could beconsiderable. It may be more transparent to restructure transactions so that the customer pays theultimate supplier of a good or service directly, so that the revenue of the business does not need tobe adjusted. A contract where some or all of the customer’s payments are deferred may imply thatthere is a significant financing component in the contract, requiring deferred payments to beincluded at their discounted present value, and the discount treated as finance income ratherthan revenue (IASB 2014, paras. 60–65). Companies that, in the past, offered deferredpayment terms as a contract incentive may seek to replace such terms by another structure thatavoids splitting the transaction price between revenue and financial income.

Fourth, the transaction price must be allocated to each performance obligation on the basis ofthe relative ‘stand-alone selling prices’ of each distinct good or service promised in the contract.In some cases, determination of stand-alone selling prices may be a complex process, andbusinesses may decide no longer to include obligations in contracts where they need to estimatestand-alone selling prices, or to change their operations so that stand-alone selling prices can beobserved rather than estimated.

Finally, revenue is recognised when an obligation is satisfied. In many cases, decidingwhether an obligation has been satisfied will be straightforward, but there will be transactionsat the margin where the precise point of satisfaction is not clear (this was a problem with previousrevenue recognition standards). Businesses may decide to reduce any degree of uncertainty byamending contracts (or their commercial practices) to include a definite point at which an obli-gation may be considered to be satisfied. A particular issue relates to whether obligations are sat-isfied over time, rather than at a specific point of time. IFRS 15’s criteria for determining whetherperformance obligations are satisfied over time are rather arcane (IASB 2014, para. 35), but theyappear to reduce the ability of companies to use the percentage of completion method.

As well as the five-step process, IFRS 15 has brought in criteria limiting companies’ ability totreat costs incurred in fulfilling contracts as assets (IASB 2014, para. 95), and requires companiesto recognise the incremental costs of obtaining a contract as an asset if the company expects torecover the costs (IASB 2014, para. 91). It may be easier for a company to measure such costsif obtaining and fulfilling contracts are undertaken for a fee by an external party rather than intern-ally. Finally, IFRS 15 has greatly extended the scope of disclosure (IASB 2018, paras. 110–129),and complying with the increased disclosure will require companies to establish, or enhance, theiraccounting information systems to provide the necessary information. At the margin, companiesmay decide that particularly complex contracts with customers create more problems in terms ofaccounting than they provide commercial benefits, encouraging greater simplicity in how thebusiness operates. We therefore expected that the introduction of IFRS 15 would not only havedirect accounting effects, but also real effects in terms of how the business operates, cash flowchanges, and implementation and application costs. Evidence as to the actual impact of IFRS15 is set out in the next section.

4. Empirical evidence

In this section we provide empirical evidence on accounting effects [A], information effects [I]and real effects [R] following the introduction of IFRS 15. Capital market effects [C] of account-ing are beyond the scope of the present study. Our evidence is mainly based on corporate annualreports, comment letters and interviews. We discuss only those effects shown in Figure 1 for

488 C.J. Napier and C. Stadler

which we have found empirical evidence. Our analysis of all of the effects is in the followingsequence: we provide any evidence first from annual reports, then from comment letters andfinally from our interviews.

4.1. Data sources

4.1.1. Annual reports

Annual reports include information on the effects of new accounting standards and they are argu-ably the most objective source of information. We analyse the effects of IFRS 15 on a sample ofthe largest European companies. Specifically, we use the STOXX Europe 50 companies as shownon the index’s component factsheet as at 31 December 2018. The list of constituent companiesalso includes country and industry data. Table 1 shows the country and industry distribution.Our sample includes nine countries: Belgium (BE), Switzerland (CH), Germany (DE),Denmark (DK), Spain (ES), France (FR), United Kingdom (GB), Italy (IT) and the Netherlands(NL). Industry is according to the ‘supersector’ as reported by STOXX. Our sample includescompanies from 14 different industries. Our analysis includes data for 48 instead of 50 companiesbecause we excluded two constituents of the index: ABB (Switzerland) because it used US GAAPand Unilever NV (Netherlands) because its consolidated financial statements are the same as thosefor Unilever plc (UK), also a constituent of the index.

Our sample companies adopted IFRS 15 in the accounting period ending on 31 December2018 or nearest after, except for Siemens, which adopted IFRS 15 in the period ending on 30 Sep-tember 2018. For simplicity, we refer to all these annual reports as ‘2018’ reports. Any ‘prior year’data relate to the period before the implementation of IFRS 15. We downloaded the annual reportsfrom the corporate websites. Specifically, we used the English language reports that include thefinancial statements and notes. For most companies, this is described as the ‘annual report’, or the

Table 1. Country and industry distribution of STOXX Europe 50 companies in 2018.

BE CH DE DK ES FR GB IT NL TOTAL

Automobiles & parts – – 1 – – – – – – 1Banks – 1 – – 2 1 3 1 1 9Basic resources – – – – – – 2 – – 2Chemicals – – 1 – – 1 – – – 2Construction & materials – – – – – 1 – – – 1Food & beverage 1 1 – – – – 1 – – 3Health care – 2 1 1 – 1 2 – – 7Industrial goods & services – – 1 – – 3 – – – 4Insurance – 1 1 – – 1 1 – – 4Oil & gas – – – – – 1 2 1 – 4Personal & household goods – – – – – 2 3 – – 5Technology – – 1 – – – – – 1 2Telecommunications – – 1 – 1 – 1 – – 3Utilities – – – – – – 1 – – 1TOTAL 1 5 7 1 3 11 16 2 2 48

Note: This table shows the country and industry distribution of STOXX Europe 50 companies at 31 December 2018. Thecountries are Belgium (BE), Switzerland (CH), Germany (DE), Denmark (DK), Spain (ES), France (FR), United Kingdom(GB), Italy (IT) and the Netherlands (NL). Industry is according to the ‘supersector’ as reported by STOXX. The tableincludes data for 48 instead of 50 companies because we excluded two companies: ABB (Switzerland) because it usedUS GAAP and Unilever NV (Netherlands) because there is also Unilever plc (UK). All data are from the STOXXEurope 50 ‘Component Information’ factsheet (https://www.stoxx.com/document/Indices/Factsheets_Components/2018/December/SX5GR.pdf, Accessed on 14 October 2019).

Accounting and Business Research 489

‘financial report’. For a few companies, the only English language report containing the financialstatements and notes is a Form 20-F prepared for filing with the Securities and Exchange Com-mission in the USA. For most French companies, we use the ‘registration document’ (or ‘refer-ence document’). All accounting data are hand-collected from these reports.

4.1.2. Comment letters

Comment letters on Exposure Drafts (ED) are documents which are likely to include infor-mation on anticipated effects. However, some of the information in comment letters may notbe objective because the main purpose of these documents is to influence the standard settingprocess. We analyse those comment letters on the IASB’s revised ED/2011/6 (published on14 November 2011 – IASB 2011) which were submitted by our STOXX Europe 50 com-panies. Specifically, we analyse 12 comment letters from the following companies: EADS(now Airbus), Barclays, BP, Daimler, Deutsche Telekom, HSBC, Nestlé, Roche, SAP,Siemens, Telefónica and Vodafone.

The IASB states that it is ‘not seeking specific comments on all matters in this exposure draft’(IASB 2011, p. 14) but all companies except one (Barclays) include additional comments beyondthe specific questions asked in the ED. Some of these additional comments are about anticipatedeffects. In our analysis regarding disclosure changes [A.4], we only consider general commentsand not the answers to Question 5 of the ED about disclosures in interim financial reports.

4.1.3. Interviews

Interviews let us ask experts about the effects of implementing IFRS 15. This allows us to gaininsights beyond what is available in publicly available documents such as annual reports andcomment letters. We conducted three semi-structured telephone interviews, each lastingbetween 30 and 50 min. We interviewed a preparer (the chief accounting officer of a FTSE250 company), an advisor (a partner at a Big-4 firm) and an auditor (a partner at another Big-4firm). While the preparer’s current employer was not heavily impacted by IFRS 15, the previousemployer (a FTSE 100 company) was.

4.2. Direct accounting effects

Our first empirical analysis is on the direct accounting effects [A] of implementing IFRS 15. Asdescribed in Section 3.2, the new standard resulted in significant changes in revenue accounting.We jointly analyse recognition and measurement changes [A.1 and A.2] via changes in reportednumbers. Additionally, we analyse disclosure changes [A.4].

Before analysing the accounting effects [A], we provide some evidence on the importance ofrevenue. We do this by investigating the extent to which revenue is a key audit matter (KAM).The audit reports of all of our sample companies include details on KAM. Table 2, Panel Areports our findings. First, there are 4.35 KAM on average, ranging from 2 to 7. Second,revenue accounting is an important KAM. Roughly half (48%) of our sample companies haveat least one KAM related to revenue. However, it does not appear that this is driven by theimplementation of IFRS 15 because all companies that have a KAM related to revenue in2018 also have one in the prior year. Third, the implementation of IFRS 15 is only importantin the audit of some companies. Just 17% of companies have a KAM which specifically mentionsthe implementation of IFRS 15 and only 8% of companies have a KAMwhich includes ‘IFRS 15’in the heading. The analysis of KAM suggests that the importance of IFRS 15 varies by industry:all telecommunication companies, all technology companies and most (3 out of 4) of the industrial

490 C.J. Napier and C. Stadler

goods and services companies have a KAM that specifically mentions the implementation ofIFRS 15; in contrast, none of the companies from other industries have such a KAM.

4.2.1. Recognition and measurement changes [A.1 and A.2]

Table 2, Panel B reports our findings on recognition and measurement changes [A.1 and A.2] inthe annual reports of our STOXX Europe 50 companies. We do this by analysing changes in threereported numbers: the effect on the opening balance of retained earnings, the difference in revenueunder IFRS 15 compared to IAS 11/18, and the difference in profit under IFRS 15 compared toIAS 11/18. Most numbers relate to the year 2018 but when a company chose retrospective appli-cation of IFRS 15 then the numbers relate to either 2016 or 2017. Untabulated findings show thatonly 25% of companies chose retrospective application. The effects are minor for most compa-nies, but significant for a few companies from particular industries. Almost half of the samplecompanies (48%) state that the effect of IFRS 15 is overall not material. Additionally, the analysesbelow imply that there are further companies where the change in numbers is not material,although they do not provide a statement confirming this.

Details of the three numbers analysed are: First, ΔRE is the change of the opening balance ofretained earnings divided by the corresponding opening balance of shareholders’ equity. Second,ΔR is the difference in revenue between IFRS 15 and IAS 11/18 divided by the IFRS 15 revenue.Third, ΔP is the difference in profit between IFRS 15 and IAS 11/18 divided by the IFRS 15revenue. We do not use profit as a deflator because it has undesirable properties (e.g. it can besmall and therefore it may be unclear whether a high ΔP is because of a large difference inprofit between IFRS 15 and IAS 11/18 or a small IFRS 15 profit). For ΔRE, most companies dis-close the required numbers. For 14 companies, we set ΔRE to zero because they do not disclose anumber but state that the effect of IFRS 15 is not material. For ΔR and ΔP, most companies do notdisclose the required numbers. For companies without disclosure, we assume that the effect is notmaterial and set ΔR and ΔP to zero. When we calculate the mean, median, minimum andmaximum, we use absolute values for the change/difference, that is, we do not consider thesign of the effect (though this is done in the last three columns of Table 2).

Regarding the effect on retained earnings as a percentage of shareholders’ equity (ΔRE), theaverage absolute change is 1.88% but the median change is only 0.05%. The three telecommunicationcompanies have large positive adjustments (5.79%, 3.63% and 3.03% for Deutsche Telekom, Vodafoneand Telefónica, respectively) and Airbus and Safran have large negative adjustments (57.31% and11.55%, respectively). When we exclude Airbus, the average ΔRE is 0.70%. For Airbus, a significantreason for the reduction is that certain ‘revenue and production costs relative to the manufacture of air-craft are recognised at a point in time (e.g. upon delivery of the aircraft to the customer)’, instead ofbeing recognised in stages over the period of manufacture (Financial Statements 2018, p. 18). Addition-ally, the relatively high ΔRE for Airbus is because its opening balance of shareholders’ equity is rela-tively low. Only 13% of our sample companies experience a ΔRE adjustment above 1%. The numberof companies with a negative adjustment to retained earnings (20) is larger than the number of com-panies with a positive adjustment (11) – see columns ‘#Δ < 0’ and ‘#Δ > 0’ in Table 2. Our sampleincludes 13 financial companies (banks and insurance, see Table 1), where the revenue standardmay be expected to have a more limited impact, because most of the income of financial companiesdoes not fall within the scope of IFRS 15. If we exclude these companies, we find that only 17% ofour non-financial companies experience a ΔRE adjustment above 1%.

Regarding the difference in revenue between IFRS 15 and IAS 11/18 as a percentage of IFRS15 revenue (ΔR), the average absolute difference is 0.72%. Only 13% of our sample companiesand 17% of our non-financial companies experience a ΔR adjustment above 1%. Ten companiesshow a lower revenue figure under IFRS 15 than under IAS 11/18, against five companies with a

Accounting and Business Research 491

Table 2. Accounting effects of implementing IFRS 15 in STOXX Europe 50 companies in 2018.

Mean Median Min Max #Δ > 0 #Δ = 0 #Δ < 0

Panel A: Key audit matters (KAM)# KAM 4.35 4 2 7KAM about revenue 48%KAM about revenue prior year 50%KAM includes implementation of IFRS 15 17%KAM with ‘IFRS 15’ in the heading 8%Panel B: Recognition and measurement changes [A.1 and A.2]Company reports that effect is not material 48%Δ retained earnings / shareholders’ equity [ΔRE] 1.88% 0.05% 0% 57.31% 11 17 20ΔRE excluding the company Airbus 0.70% 0.05% 0% 11.55% 11 17 19Effect (ΔRE) >1% 13%Effect (ΔRE) >1% excluding financial companies 17%Δ revenue / revenue [ΔR] 0.72% 0% 0% 13.12% 5 33 10Effect (ΔR) >1% 13%Effect (ΔR) >1% excluding financial companies 17%Δ profit / revenue [ΔP] 0.16% 0% 0% 1.61% 4 38 6Effect (ΔP) >1% 8%Effect (ΔP) >1% excluding financial companies 11%Panel C: Disclosure changes [A.4]# pages revenue note 0.89 0.75 0 6# pages revenue note prior year 0.46 0.25 0 1.75Δ pages revenue note 0.43 0 −0.25 5 21 26 1*Separate revenue note 38%Separate revenue note prior year 25%

Note: This table reports the accounting effects of implementing IFRS 15 in STOXX Europe 50 companies in 2018. Panel Areports information on key audit matters (KAM). Panel B reports information on recognition and measurement changes[A.1 and A.2] via changes in reported numbers. Panel C reports information on disclosure changes [A.4]. All data arehand-collected from annual reports. The results are based on 48 companies, as described in the note to Table 1. Oursample companies adopted IFRS 15 in the accounting period ending on 31 December 2018 or nearest after, except forSiemens (which adopted IFRS 15 in the period ending on 30 September 2018). For simplicity, we refer to all theseannual reports as ‘2018’ reports. Any ‘prior year’ data relate to the period before the implementation of IFRS 15. Mostrows report dichotomous data, except those starting with ‘#’ (denoting ‘number of’) or ‘Δ’ (denoting ‘change in’).Dichotomous data are coded as 1 (yes) or 0 (no) and we report the mean (as a percentage). For rows starting with ‘#’,we report the mean, median, minimum (Min) and maximum (Max). For rows starting with ‘Δ’, we report the mean,median, minimum (Min) and maximum (Max), which are presented as a percentage for ratios. Additionally, we reportthe number of companies for which the data increases (#Δ > 0), remains unchanged (#Δ = 0) or decreases (#Δ < 0). InPanel B, ‘Δ retained earnings’ is the change of the opening balance of retained earnings, ‘shareholders’ equity’ is thecorresponding opening balance of shareholders’ equity, ‘Δ revenue’ (‘Δ profit’) is the difference in revenue (profit)between IFRS 15 and IAS 11/18, and ‘revenue’ is the IFRS 15 revenue. Most numbers relate to the year 2018 butwhen a company chose retrospective application of IFRS 15 then the numbers relate to either 2016 or 2017. For ΔRE,most companies disclose the required numbers. For 14 companies, we set ΔRE to zero because they do not disclosebut state that the effect of IFRS 15 is not material. For ΔR and ΔP, most companies do not disclose the requirednumbers. For companies without disclosure, we assume that the effect is not material and set ΔR and ΔP to zero.When we calculate the mean, median, min and max we use absolute values for the change, i.e. we do not consider thesign of the effect (this is done in the last three columns). Additionally, when we calculate the mean, median, min andmax we set ΔRE of GlaxoSmithKline to zero. This is because the company has Δ retained earnings of £4 m, negativeshareholders’ equity of £68 m and states that ‘IFRS 15 did not have a material impact’ (Annual Report 2018, p. 145).Financial companies are banks and insurance companies (see Table 1). In Panel C, we count the number of pages inthe notes which are predominantly about revenue. Specifically, anything up to a quarter of a page is counted as 0.25,anything between a quarter and half of a page is counted as 0.5 and anything between half and three quarters of a pageis counted as 0.75. In our count, we do not include the ‘revenue’ part of the company’s general note on accountingpolicies. The revenue disclosure may be part of the note on segment reporting, but we only count it if it ispredominantly about revenue, i.e. we do not count disclosure (e.g. in the form of a table) that reports revenue togetherwith other numbers (such as profits or assets). Regarding the separate revenue note, banks and insurance companiesusually have a note on ‘fee and commission income’ and such income is within the scope of IFRS 15. However, wedo not score a note on ‘fee and commission income’ as a separate revenue note.*This relates to Reckitt Benckiser and can be explained by a change in operating segments. The previous segments werebased on geographical areas and the company disclosed additional revenue data for its business lines (which we scored as0.25 in the ‘prior year’).

492 C.J. Napier and C. Stadler

higher figure – see columns ‘#Δ < 0’ and ‘#Δ > 0’ in Table 2. The effect is largest for Airbus andNational Grid, which have reductions in revenue of 13.12% and 8.22%, respectively. Airbus hasalready been discussed above. For National Grid, this is mainly because ‘[c]ertain pass-throughrevenues […] will be recorded net of operating costs, whereas previously they were recognisedgross of operating costs’ (Annual Report and Accounts 2018/19, p. 188). The effect of this isthat revenues and operating costs are reduced by the same amount.

Regarding the difference in profit between IFRS 15 and IAS 11/18 as a percentage of IFRS 15revenue (ΔP), the average absolute difference is 0.16% and the largest difference is 1.61%. Only8% of our sample companies and 11% of our non-financial companies experience a ΔP adjustmentabove 1%. Six companies show a lower profit figure under IFRS 15 than under IAS 11/18 andfour companies show a higher figure – see columns ‘#Δ < 0’ and ‘#Δ > 0’ in Table 2. This doesnot suggest a clear trend. Therefore, the effect of IFRS 15 on profit is overall insignificant.

Our results show that the effect on reported numbers is minor for most companies. Forexample, Nestlé’s 2018 opening balance of shareholders’ equity was 82,870 million SwissFrancs and its 2018 IFRS 15 revenue was 89,590 million Swiss Francs. The effects of implement-ing IFRS 15 were a reduction in the opening balance of returned earnings of 268 million, areduction in revenue of 169 million and a reduction in profit of 25 million Swiss Francs.These relatively small changes are understandable when the company discloses that a maineffect was that ‘a small proportion of sales (less than 0.5% of annual sales) is recognized onaverage 2 days later under the new standard’ (Financial Statements 2018, p. 75).

Our interviews suggest that our findings based on STOXX Europe 50 companies apply morebroadly. The auditor said:

And one of the things which struck me quite a lot was actually how little impact the standard had hadfor many of them because actually they have quite simple businesses […]. There are obviously a lot ofexceptions to that but I would say that when you look across the board you’ve got quite a lot ofbusinesses where there is relatively little impact […]. And then there is relatively few where thereis a really dramatic change.

The advisor also confirmed this, and additionally observed that the lack of more change wassomewhat surprising:

I find it personally a little bit surprising then that we didn’t have more changes to revenue recognitionwhen we moved to IFRS 15. […] And I haven’t seen as many companies which accelerated revenuethan I might have expected.

4.2.2. Disclosure changes [A.4]

Table 2, Panel C reports our findings on changes in disclosures [A.4] in the annual reports of ourSTOXX Europe 50 companies. There is clear evidence of increased disclosure following theimplementation of IFRS 15.

An analysis of the quantity of disclosures requires judgement by the researcher and there aremany possible ways of doing this. We adopt the following approach: we count the number ofpages in the notes which are predominantly about revenue. Specifically, anything up to aquarter of a page is counted as 0.25, anything between a quarter and half of a page is countedas 0.5 and anything between half and three quarters of a page is counted as 0.75. In our count,we do not include the ‘revenue’ part of the company’s general note on accounting policies.The revenue disclosure may be part of the note on segment reporting, but we only count it if itis predominantly about revenue. That is, we do not count disclosure (for example, in the formof a table) that reports revenue together with other numbers (such as profits or assets).

Accounting and Business Research 493

For 2018, the first year under IFRS 15, the average number of pages of revenue disclosures inthe notes is 0.89, ranging from 0 to 6. In the prior year, the average number of pages is 0.45,ranging from 0 to 1.75. Therefore, the amount of disclosures almost doubles. The increase in dis-closure is statistically significant, based on a paired sample t-test (t-statistic = 3.49, two-tailed p-value = 0.001). Out of the 48 sample companies, 21 increased their disclosures, 26 are unchangedand only one has a reduction. The latter relates to Reckitt Benckiser and can be explained by achange in their operating segments following a new business unit structure. The previous seg-ments were based on geographical areas and the company disclosed additional revenue datafor its business lines (which we scored as 0.25 in the ‘prior year’). We also counted which com-panies have a separate revenue note. In 2018, 38% of the sample companies had a separaterevenue note compared to 25% in 2017.

Qualitative disclosures by companies confirm our quantitative analysis. For example, Rochestates that ‘[t]he new standard results in an increased volume of disclosure information in theAnnual Financial Statements’ and that it has ‘[a]s a result of implementing IFRS 15, […]created a new note for “Revenue” as Note 3’ (Finance Report 2018, p. 139).

In their comment letters, 8 out of the 12 companies address proposed disclosure changes[A.4]. All companies think that they are extensive and many companies use the term ‘excessive’.For example, Daimler states (Comment Letter No. 80, p. 1):

The proposed disclosure requirements are substantially more extensive and detailed than the existingrequirements. The inclusion of the detailed qualitative disclosure requirements will likely increase thesize of disclosures.

In our discussion above of how we have conducted our page count of disclosures, we have men-tioned that revenue information is often part of the segment reporting disclosures. This issue ishighlighted in the comment letter of BP, which states that they ‘are concerned that the proposedrequirement to disaggregate revenue into categories will duplicate disclosures already required byIFRS 8, and so request that the Board clarify how the requirements interact with those in IFRS 8’(Comment Letter No. 279, p. 11).

Our interviews confirm that there has been an increase in disclosure, but there is a mixed reac-tion about whether this represents an improvement. The auditor said:

I think if you ask most people they would say that revenue disclosures were generally poor. […] Theywere very boilerplate and didn’t tell you very much. And so the intention was that IFRS 15 would giveusers a lot more information about the approach to recognising revenue, which I think has happened. Ithink we’re probably only about 60% of the way there in year one and I think year two should showsome further improvement.

In contrast, the preparer thinks that ‘some of the additional disclosures […] are a waste of time’and providing them ‘doesn’t really add anything’.

4.3. Information effects

4.3.1. Internal users’ understanding of transactions [I.1]

Our evidence on internal users’ understanding of transactions [I.1] is based on our interviews.We asked whether the implementation of IFRS 15 was used as an opportunity by companies tofundamentally rethink their business model or was just seen as a compliance exercise. Takingthe diverse evidence of our three interviewees together, we conclude that the implementationof IFRS 15 was seen by many companies as a compliance exercise but for some companies

494 C.J. Napier and C. Stadler

with complicated businesses it was used to better understand their transactions. The advisorsaid:

I’d love to tell you I have clients that you know welcomed me and sort of see that opportunity toenhance and improve their business but that wasn’t the case. Most people treated it as a pure compli-ance exercise and left it relatively late in the process.

The auditor’s experience, focusing more on the technical side of accounting for revenue, was‘much more’ of a positive approach by companies than treating it as a compliance exercise.However, the auditor, who is a revenue expert, said that this ‘might reflect where I spend mytime’. The preparer provided an example where the implementation of IFRS 15 has improvedthe understanding of transactions related to the preparer’s former company which is ‘quite com-plicated from a business perspective’:

To start with it was seen to be a compliance exercise, especially by the CFO. […] And so as we were goingthrough, and understanding how the contracts are written, it was taken as an opportunity, especially in onedivision, to actually formalise what the various salespeople around the world were allowed to agree and notto agree […] because we found that that has quite dire consequences on the accounting […].

4.3.2. External users’ understanding of transactions [I.2]

All three telecommunication companies argued in their comment letters that the new standard wouldnegatively affect external users’ understanding of transactions [I.2]. For example, under the heading‘Inconsistent accounting for similar transactions’, Vodafone argues that IFRS 15 reduces the useful-ness and comparability of accounting information (Comment Letter No. 273, p. 3):

In common with other telecom operators, Vodafone sells airtime contracts to customers through itsown shops and distribution channels (the ‘direct’ channel’) and via third party dealers (the ‘indirect’channel).Key performance indicators for users of our accounts, particularly service revenue and EBITDA, willbe materially different for the same customer airtime tariffs, depending on whether customer contractsare acquired through the direct channel or the indirect channel.Since the proportion of direct and indirect channel sales can vary significantly between operators,countries and reporting periods, we believe this will materially undermine the usefulness and compar-ability of telecom operators’ financial statements.

Additionally, when commenting on disclosure changes [A.4], six companies link it to externalusers’ understanding of transactions [I.2]. While the argument in most cases is that the additionaldisclosures do not provide useful information (that is, there is no information effect), there are alsocases where companies argue that the implementation of IFRS 15 will have an impact on thewider reporting environment. For example, EADS states (Comment Letter No. 335):

We note that the collection and preparation of the data required will be an ‘accounting-only’ exercisethat will divert our limited internal resources away from other financial reporting areas that we con-sider to be of more interest to users.

Regarding our interviews, the above quote of the auditor regarding the additional disclosures indi-cates that IFRS 15 has improved the information available to users. The auditor expects that thiswill be particularly relevant for regulators:

I think that IFRS 15 will give regulators a deeper understanding of companies and with that deeperunderstanding they will be looking at what’s happening and challenging whether the accounting iscorrect and whether the disclosures are correct.

Accounting and Business Research 495

4.3.3. Communication of effects to stakeholders [I.3]

Some of our sample companies have hosted presentations about the impact of IFRS 15 in order tocommunicate the effects to stakeholders [I.3]. We adopted a standardised approach to identifysuch companies, by performing the following Google search for each of our 48 sample compa-nies: company name ‘IFRS 15’ filetype:pdf. This search identified six companies with relevant‘communication’ documents/presentations. Four companies (Airbus, Daimler, Safran and SAP)have documents that are exclusively about IFRS 15 (for example, Safran published a documententitled ‘IFRS 15 Workshop’ on 12 March 2018). Two further companies (Deutsche Telekom andTelefónica) have documents that include a detailed discussion of IFRS 15 but do not refer to thestandard in the document’s title.

4.4. Real effects

4.4.1. Implementation and application costs [R.1]

There is indirect evidence in annual reports that the implementation and application of IFRS 15was costly [R.1] because it required investments in information systems and related processes.For example, as part of its description of key audit matters, Deutsche Telekom’s auditor PwCstates that ‘[t]he initial application of IFRS 15 required group-wide adjustments or implemen-tation of systems and processes regarding the new requirements for revenue recognition’(Deutsche Telekom Annual Report 2018, p. 266).

Additionally, we analysed audit fee data from 2016 to 2018 and found one specific example ofimplementation costs. Vodafone discloses that the audit fees of €21 million in the year prior to theimplementation of IFRS 15 (2018) include €5 m related to ‘[f]ees during the implementationphase of new accounting standards, notably preparations for IFRS 15 “Revenue fromContracts with Customers” in the year ended 31 March 2018’ (Vodafone Annual Report 2019,p. 129).

In its comment letter, Deutsche Telekom provides details on the expected implementation andapplication costs [R.1] and gives an estimate of the expected costs (Comment Letter No. 215,p. 15):

Implementing the ED would require us to build a new, highly complex IT system connected to severaldata basis (performance obligations data base, stand-alone selling prices data base, tool for determin-ing the transaction price etc.). Furthermore, operating this IT system would require a huge manpowerwith regard to permanent monitoring, data update, making assessments and estimates etc. This con-nection between a complex IT solution and permanent manual assistance will inevitably cause tremen-dous costs which we estimate to amount at least hundreds of millions of Euros.

Additionally, when commenting on disclosure changes [A.4], seven companies link it toimplementation and application costs [R.1]. For example, Roche states that ‘[c]ollecting thisdata is not free; it requires internal resources to implement, and costs to maintain and audit’(Comment Letter No. 242, p. 3).

Our interviews revealed some further insights. The advisor highlights that ‘changing youraccounting around revenue involves a lot of people across the business in many differentareas, many of whom are not accountants’. Companies would spend money on an ‘impact assess-ment’ either internally or using an advisor, ‘updating their policies and the related processes andthe controls’ and ‘really big money […] where they identified a major change to the debits andcredits’ which would require them to build or buy an IT system (‘accounting engine’). RegardingDeutsche Telekom’s estimated cost (see above), the advisor suggested that telecommunicationcompanies did not spend that much money: they initially worried that they would have to

496 C.J. Napier and C. Stadler

‘upgrade their billing systems’ but ‘in reality, what most telcos did was they didn’t change billingsystems, they left them untouched because they are so expensive, and built an accounting enginethat takes information from the billing system’. Overall, the advisor’s view on implementationcosts is: ‘I think a lot of companies didn’t spend that much money, but I question whetherthey’ve done enough work’.

The preparer mentioned that the auditors increased their fees. Additionally, in the preparer’sprevious company, which was heavily impacted, the implementation of IFRS 15 constituted a‘significant cost’ and it was ‘way more expensive than we thought’. Regarding audit, theauditor said that ‘the extra work now is much more around challenging whether the policyadopted is an appropriate policy, and certainly in year one that was a lot of extra work forpeople and reasonably difficult work’.

4.4.2. Contractual changes [R.2]

In comment letters, we found one reference to potential contractual changes [R.2], in a commentfrom the German software company SAP (Comment Letter No. 307, p. 21):

Changes in business practice in response to changes in accounting guidance are common in ourindustry as evidenced by the changes SAP experienced when it moved from German GAAP toUS GAAP. One potential example for such a change in business practice is the requirementunder para 35(b) (iii) of the ED that the entity must have a right to payment for performancecompleted to date. Current contract exit clauses may not be worded in a manner that meetsthe requirement of para 35(b) (iii) but an entity may be confident to have appropriate clausesin its future contracts.

SAP was clearly hinting at a need to change certain aspects of contracts to allow it torecognise the revenue from such contracts over a period of time rather than only at the end ofthe contract.

Our interviews provide further evidence that IFRS 15 led to contractual changes [R.2].However, these changes were not fundamental but rather at the margin. The preparer said:

I think if you sell products, it does not really change very much. But as soon as you get into anythingthat’s long-term contracts or different performance obligations in one contract, you can get unintendedconsequences. And I know in [former company] when we were looking into our long-term contracts,[…] basically, we reviewed all the contracts and as we revised the contracts, we’re making sure thatthe wording was sufficient that it meant that we could still effectively account for it in the same way,even though the terminology changed.

Additionally, the advisor suggested that ‘there were some examples where, you know, this was anopportunity to put more control around contracting discipline and pricing discipline, and compa-nies did take advantage of that because it made it easier to apply the standard’.

4.4.3. Behavioural effects [R.3]

In annual reports, the disclosures of effects of IFRS 15 are usually about direct accounting effects[A] and not about behavioural effects [R.3]. However, SAP provides one example of an expectedchange in business practices in its Annual Report 2017, that is, in the year before the initial adop-tion of IFRS 15 (p. 160):

The most notable revenue impact relates to the accounting for options for the purchase ofadditional copies of already licensed on-premise software where those options provide a material

Accounting and Business Research 497

right to the customer (Material Right Additional Copy Options). Under our previous policies, weadhered to the guidance under previous U.S. GAAP to not account for these options. In contrast,IFRS 15 provides that such options are accounted for as a separate performance obligation. Thetransaction price portion allocated to a Material Right Additional Copy Option is recognized insoftware revenue upon exercise or forfeiture of the option, which will usually be later than thetiming under our previous policies. […] We currently estimate that our business practices willchange in the light of this new accounting policy in a way that Material Right AdditionalCopy Options are only provided rarely.

In our sample of comment letters, we have found one clear reference to potential behaviouraleffects [R.3]. This comes from Vodafone and relates to the quote about [I.2] above. Under theheading ‘A cost accounting model that is likely to drive changes to sales and remuneration struc-tures’ Vodafone states (Comment Letter No. 273, p. 7):

We do not believe that it is appropriate that the proposed standard should mandate the recognition ofthe incremental costs of acquiring a contract as an asset.

The proposals will incentivise entities to sell goods and services through third parties for the followingreasons:

. Fixed costs incurred by third parties, such as salaries and property-related overheads can be reim-bursed by the entity through the payment of incremental commissions that can be recognised as anasset, whereas such costs incurred by an entity in its own sales channels will be expensed as incurred;

. Assuming such assets are classified as intangible assets, the charge will ultimately be realised as amor-tisation and thus EBITDAwill depend on the mix of sales channel used and usefulness and compar-ability of this key measure will be significantly reduced; and

. Entities will have increased scope to manage reported profits through altering sales structures.

Entities will also be compelled to redesign remuneration structures.

Our interviews confirm that IFRS 15 led to behavioural effects [R.3] but these were considered tobe at the margin. There were no examples of fundamental changes to how businesses operate.However, in relation to the quotation from the preparer in sub-section 4.3.1, the preparer con-firmed that the work in relation to IFRS 15 did change what the salespeople were allowed todo, especially regarding add-ons to contracts. The advisor provided some further explanationabout this:

The accountants aren’t going to say you can’t do it, but they’ll say, well, if you do it we’re goingto have to do some accounting for it. Or there will be a debate around, well, could we achieve thesame aim without giving something away for free by providing some other attractive incentive tothe customer whatever that is that doesn’t have the same accounting impact. […] I’m giving a typeof discount that basically is a deduction of revenue and therefore doesn’t need to be trackedseparately.

4.4.4. Regulatory effects [R.4]

In its comment letter, Telefónica provides an example of potential regulatory effects [R.4](Comment Letter No. 159, p. 15):

[I]n many countries, the prices of different telecom services are regulated by establishing maximumprices. Any alteration of revenue allocated to different performance obligations could result in con-tradictions with applicable law and/or reporting requirements to regulators. (e.g. Universal ServiceObligation)

498 C.J. Napier and C. Stadler

4.4.5. Tax and dividend effects [R.5] and other real effects [R.6]

In its comment letter, Telefónica also provides an example of a potential tax effect [R.5](Comment Letter No. 159, p. 15):

In some countries, the different telecom services are taxed at different VAT rates. Any alteration ofrevenue allocated to different performance obligations could result in higher or lower taxes thancurrently.

In our interviews, the auditor made the following general comment about tax effects [R.5] andother real effects [R.6]:

It’s just the classic thing that accountants say to just warn people of traps they might otherwise fall into. Ifyou are a contract manufacturer and you’ve got quite a long lead time and you find that you previouslyrecognised revenue on delivery of goods and now you recognise some revenue as you make them, thenyou’re bringing revenue forward. And if you bring revenue forward, you typically bring profit forward. Asyou’re bringing profit forward, you typically pay tax earlier. And also if you’ve got bonus schemes that arepredicated on hitting similar profit targets you might find that you suddenly go from something that lookquite stretching to something that you’re bound to meet. So it’s rather less about should we do businessdifferently but rather more about are there other sort of knock-on effects that we might need to maketweaks to. Setting profit targets for bonus schemes and that sort of thing.

4.5. Indirect accounting effects

In our analysis of comment letters, we have found one example of a potential indirect accountingeffect. It is included in the above quote of Vodafone about behaviour changes [R.3], where thecompany suggests that ‘[e]ntities will have increased scope to manage reported profits throughaltering sales structures’ (Comment Letter No. 273, p. 7).

4.6. Other findings

Our interviews provided additional insights that are separate from the effects of IFRS 15 dis-cussed above. First, according to the auditor, IFRS 15 has been useful in dealing with difficultrevenue questions, not only in IFRS contexts but more generally:

[T]he standard has been a really helpful tool. And to give you an illustration of that, we, obviously,auditors, have a lot of companies that don’t apply IFRS, they apply UK GAAP, FRS 102, which ismore aligned to IAS 18 than to IFRS 15. So when we get revenue questions under FRS 102 weusually think about them through an IFRS 15 lens first. […] So I think amongst the sort ofrevenue experts, I would say that the standard’s been really helpful and is much appreciated.

Second, there are mixed opinions on whether the standard has improved revenue accounting. Theauditor has a very positive opinion about IFRS 15, as can be seen by the above quote. Addition-ally, the auditor said that IAS 18 ‘just left people inventing their own accounting policies’ and that‘we saw a lot of diversity as a result’. The advisor was more ambivalent. The advisor thought thatthe old IAS 18 was ‘too brief and too broad’. Regarding IFRS 15, the advisor thought that ‘themodel they put together works’ but ‘there were a number of key judgements in there whichcreate challenges […] in terms of practical implementation’. This meant that the advisor was‘not convinced yet […] that we got to that sort of consistency that was intended here’. Addition-ally, the advisor questioned whether ‘what was required for the amount of financial change’ was‘worth the effort’. The preparer saw both advantages and disadvantages of IFRS 15 because ‘thereare some areas where it’s improved because it’s more detailed and gives you more guidance tofollow, but it’s a lot more complicated’.

Accounting and Business Research 499

5. Conclusions

We have provided evidence on the accounting, information and real effects of IFRS 15 based oncorporate annual reports, comment letters and interviews. Our analysis of the largest Europeancompanies shows that the impact of the standard on accounting numbers was minimal for mostcompanies, except in a few industries (most notably telecommunications). On the other hand, dis-closures have generally increased. With respect to information effects, there was limited evidencethat companies would take the opportunity of the introduction of IFRS 15 to undertake an inten-sive review of their business.

Table 3. Summary of the empirical evidence on the effects of IFRS 15.

Effect Evidence Impact Section

[A] Accounting effects[A.1] Recognition changes AR + I Joint impact of [A.1] and [A.2]: minor for

most companies; significant for someSection4.2

[A.2] Measurement changes AR + I Joint impact of [A.1] and [A.2]: minor formost companies; significant for some

Section4.2

[A.3] Presentation changes N/A[A.4] Disclosure changes AR + CL + I Increase for many companies Section

4.2[I] Information effects[I.1] Internal users’ understanding

of transactionsI Yes for complicated contracts Section

4.3[I.2] External users’ understanding

of transactionsCL + I Possible impact Section

4.3[I.3] Communication of effects to

stakeholdersO Important for some companies Section

4.3[C] Capital market effects[C.1] Equity market effects Not

investigated[C.2] Debt market effects Not

investigated[C.3] Pressure to maintain/improve

performance and financialposition

Notinvestigated

[R] Real effects[R.1] Implementation and

application costsAR + CL + I Significant; large for telecommunication

companiesSection4.4

[R.2] Contractual changes CL + I At the margin for some companies Section4.4

[R.3] Behavioural effects AR + CL + I At the margin for some companies Section4.4

[R.4] Regulatory effects CL Possible impact for regulated companies Section4.4

[R.5] Tax and dividend effects CL + I Possible impact Section4.4

[R.6] Other real effects I Possible impact Section4.4

Note: This table shows a summary of the empirical evidence on the effects of IFRS 15. The column ‘Effect’ lists thepossible effects (see Figure 1). The column ‘Evidence’ reports where we have obtained evidence from: AR = annualreports, CL = comment letters, I = interviews, O = other. One effect does not apply to IFRS 15 [A.3] and therefore werecord ‘N/A’. Additionally, we record ‘Not investigated’ for [C] capital market effects because these are outside thescope of this paper. The column ‘Impact’ provides a summary of our findings. The column ‘Section’ shows whichsection of the paper provides details.

500 C.J. Napier and C. Stadler

Regarding real effects, in some cases, the introduction of IFRS 15 has involved substantialcosts relating to the development of additional computer software, to track transactions and ident-ify when obligations under contracts with customers have been performed. Some companieswould take IFRS 15 into account when drafting new contracts, but would not disturb existing con-tracts. Furthermore, some companies have chosen to make changes to their business operations sothat they map onto the requirements of IFRS 15 more closely. This has been a particular issue forindustries where contracts involve complex bundles of provisions, often including both goods andservices. IFRS 15 has certainly led to real effects, but for most companies, it appears that thesearise at the margin rather than representing significant restructuring of core operations. Table 3summarises our findings by reference to the various effects that we refer to in our frameworkin Figure 1.

We did not investigate how far the adoption of IFRS 15 has changed the ways in which userssuch as analysts interpret revenue information, and this would provide opportunities for futureresearch. A limitation of the present study is that only three interviews were undertaken, involvingpreparers and their advisors. It would be informative to undertake both surveys of and interviewswith financial analysts and other users, as well as value relevance and other quantitative studies toassess how far IFRS 15 has enhanced users’ understanding of companies’ activities and overallbusiness model. The present study examines only the first year of implementing a new accountingstandard. Longer-term studies, both interview-based and quantitative, are desirable. We haveinvestigated the real effects of a single new standard. The more general framework that wehave developed in this paper could be used to examine other new or amended standards,which may reveal a wider range of real effects.

AcknowledgementThe authors thank Robert Hodgkinson, Alison Dundjerovic and Gillian Knight (ICAEW) for their supportduring the writing of this paper, Richard Veysey for his practitioner’s view, and the interviewees who con-tributed to the research. The authors are grateful for comments from Andrew Lennard, Chris Nobes, and par-ticipants at the ICAEW Information for Better Markets Conference 2019, and from an anonymous reviewer.

Disclosure statementNo potential conflict of interest was reported by the author(s).

ORCID

Christopher J. Napier http://orcid.org/0000-0002-2037-0477Christian Stadler http://orcid.org/0000-0002-3641-0824

ReferencesAccounting Standards Board, 1994. Financial reporting standard 5 (FRS 5). Reporting the substance of

transactions.Accounting Standards Board, 1999. Statement of principles for financial reporting.Accounting Standards Board, 2003. Amendment to FRS 5 ‘reporting the substance of transactions’. Revenue

recognition (Application Note G).Ball, R. 1972. Changes in accounting techniques and stock prices. Journal of Accounting Research, 10

(Empirical research in accounting: selected studies), 1–38.Barker, R., 2010. On the definitions of income, expenses and profit in IFRS. Accounting in Europe, 7 (2),

147–158.

Accounting and Business Research 501

Barth, M.E., Cahan, S.F., Chen, L., and Venter, E.R., 2017. The economic consequences associated withintegrated report quality: capital market and real effects. Accounting, Organizations and Society, 62,43–64.

Benson, H., 1989. Accounting for Life. London: Kogan Page.Bereskin, F.L., Hsu, P.-H., and Rotenberg, W., 2018. The real effects of real earnings management: evidence

from innovation. Contemporary Accounting Research, 35 (1), 525–557.Briloff, A.J., 1972. Unaccountable Accounting. New York: Harper and Row.Briloff, A.J., 1976. More Debits Than Credits: The Burnt Investor’s Guide to Financial Statements.

New York: Harper and Row.Camfferman, K. and Zeff, S.A., 2015. Aiming for Global Accounting Standards: The International

Accounting Standards Board, 2011–2011. Oxford: OUP.Christensen, H.B., Floyd, E., Liu, L.Y., and Maffett, M., 2017. The real effects of mandated information on

social responsibility in financial reports: evidence from mine-safety records. Journal of Accounting andEconomics, 64 (2-3), 284–304.

Christensen, H.B. and Nikolaev, V.V., 2017. Contracting on GAAP changes: large sample evidence. Journalof Accounting Research, 55 (5), 1021–1050.

Dechow, P.M., Ge, W., Larson, C.R., and Sloan, R.G., 2011. Predicting material accounting misstatements.Contemporary Accounting Research, 28 (1), 17–82.

De Lacy, J., 1995. Romalpa theory and practice under retention of title in the sale of goods. Anglo-AmericanLaw Review, 24 (3), 327–368.

Dicksee, L.R., 1903. Advanced Accounting. London: Gee & Co.Dou, Y., Ryan, S.G., and Xie, B., 2018. The real effects of FAS 166/167 on banks’ mortgage approval and

sale decisions. Journal of Accounting Research, 56 (3), 843–882.Dou, Y. and Zou, Y., 2019. The real effects of geographic lending disclosure on banks.

Available from: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3435099 [Accessed 29November 2019].

Ernstberger, J., Link, B., Stich, M., and Vogler, O., 2017. The real effects of mandatory quarterly reporting.The Accounting Review, 92 (5), 33–60.

European Community, 1978. Fourth council directive of 25 July 1978 based on article 54 (3) (g) of the treatyon the annual accounts of certain types of companies (78/660/EEC) (Fourth Directive on CompanyLaw).

Fields, T.D., Lys, T.Z., and Vincent, L., 2001. Empirical research on accounting choice. Journal ofAccounting and Economics, 31 (1-3), 255–307.

Financial Accounting Standards Board, 1980. Statement of financial accounting concepts no. 3 (SFAC 3).Elements of financial statements of business enterprises.

Financial Accounting Standards Board, 1984. Statement of financial accounting concepts no. 5 (SFAC 5).Recognition and measurement in financial statements of business enterprises.

Financial Accounting Standards Board, 2008. Preliminary views on revenue recognition in contracts withcustomers.

Financial Accounting Standards Board, 2014. Accounting standards classification topic 606 (ASC 606).Revenue from contracts with customers.

Financial Reporting Council, 2018. Financial reporting standard 102 (FRS 102). The Financial ReportingStandard applicable in the UK and Republic of Ireland.

Francis, J., 2001. Discussion of empirical research on accounting choice. Journal of Accounting andEconomics, 31 (1–3), 309–319.

Hart, O., 2009. Regulation and Sarbanes-Oxley. Journal of Accounting Research, 47 (2), 437–445.Healy, P.M. and Wahlen, J.M., 1999. A review of the earnings management literature and its implications for

standard setting. Accounting Horizons, 13 (4), 365–383.Hines, R.D. 1988. Financial accounting: in communicating reality, we construct reality. Accounting,

Organizations and Society, 13 (3), 251–261.Institute of Chartered Accountants in England and Wales, 1964. Recommendation on accounting principles

23 (N23). Hire purchase, credit sale and rental transactions. In: Zeff, S.A., ed., 2009. Principles BeforeStandards: The ICAEW’s ‘N Series’ of Recommendations on Accounting Principles 1942–1969.London: ICAEW, 142–150.

Institute of Chartered Accountants in England and Wales, 1971. Survey of Published Accounts 1969–70.London: ICAEW.

Institute of Chartered Accountants in England and Wales, 1976. Accounting recommendation V24.Accounting for goods sold subject to reservation of title.

502 C.J. Napier and C. Stadler

International Accounting Standards Board, 2010. Exposure draft ED/2010/6. Revenue from contracts withcustomers.

International Accounting Standards Board, 2011. Exposure draft ED/2011/6. Revenue from contracts withcustomers.

International Accounting Standards Board, 2014. International financial reporting standard 15 (IFRS 15).Revenue from contracts with customers.

International Accounting Standards Board, 2018. Conceptual framework for financial reporting.International Accounting Standards Committee, 1979. International accounting standard 11 (IAS 11).

Accounting for construction contracts.International Accounting Standards Committee, 1982. International accounting standard 18 (IAS 18).

Revenue recognition.International Accounting Standards Committee, 1989. Framework for the preparation and presentation of

financial statements.International Accounting Standards Committee, 1993a. International accounting standard 11 (IAS 11).

Construction contracts.International Accounting Standards Committee, 1993b. International accounting standards 18 (IAS 18).

Revenue.Kanodia, C., 1980. Effects of shareholder information on corporate decisions and capital market equilibrium.

Econometrica, 48 (4), 923–953.Kanodia, C., 2006. Accounting disclosure and real effects. Foundations and Trends® in Accounting, 1 (3),

167–258.Kanodia, C. and Mukherji, A., 1996. Real effects of separating investment and operating cash flows. Review

of Accounting Studies, 1 (1), 51–71.Kanodia, C. and Sapra, H., 2016. A real effects perspective to accounting measurement and

disclosure: implications and insights for future research. Journal of Accounting Research, 54 (2),623–676.

Leftwich, R., 1981. Evidence of the impact of mandatory changes in accounting principles on corporate loanagreements. Journal of Accounting and Economics, 3 (1), 3–36.

Levitt, A., 1998. The ‘numbers game’. Speech at New York University Center for Law and Business, 28September. Available from: https://www.sec.gov/news/speech/speecharchive/1998/spch220.txt[Accessed 30 November 2019].

Leuz, C. and Wysocki, P.D., 2016. The economics of disclosure and financial reporting regulation: evidenceand suggestions for future research. Journal of Accounting Research, 54 (2), 525–622.

Nobes, C.W., 2012. On the definitions of income and revenue in IFRS. Accounting in Europe, 9 (1), 85–94.Oxford English Dictionary, 2011. Revenue. Oxford English Dictionary. Oxford: OUP. On-line edition.

Available from: https://www.oed.com/view/Entry/164734 [Accessed 30 November 2019].Paton, W.A., 1922. Accounting Theory: With Special Reference to the Corporate Enterprise. New York:

Ronald Press.Paton, W.A. and Littleton, A.C., 1940. An Introduction to Corporate Accounting Standards. New York:

American Accounting Association.Previts, G.J. and Merino, B.D., 1998. A History of Accountancy in the United States: The Cultural

Significance of Accounting. Columbus: Ohio State University Press.Roberts, C.B., Salter, S.A., and Kantor, J., 1996. The IASC comparability project and current financial

reporting reality: an empirical study of reporting in Europe. The British Accounting Review, 28 (1),1–22.

Roychowdhury, S., 2006. Earnings management through real activities manipulation. Journal of Accountingand Economics, 42 (3), 335–370.

Trainer, D., 2019. How companies implemented the new revenue recognition standard. Forbes, 10 October.Available from: https://www.forbes.com/sites/greatspeculations/2019/10/10/how-companies-implemented-the-new-revenue-recognition-standard/#4375caba920e [Accessed 30 November 2019].

Turner, L.E., 1998. Towards higher values in financial reporting. Speech at the Financial Executives Institute1998 Annual Conference, 22 October. Available from: https://www.sec.gov/news/speech/speecharchive/1998/spch231.htm [Accessed 30 November 2019].

Wüstemann, J. and Kierzek, S., 2005. Revenue recognition under IFRS revisited: conceptual models, currentproposals and practical consequences. Accounting in Europe, 2 (1), 69–106.

Zeff, S.A., 1978. The rise of ‘economic consequences’. Journal of Accountancy, 146 (6), 56–63.Zeff, S.A., 2018. An Introduction to Corporate Accounting Standards: detecting Paton’s and Littleton’s

influences. Accounting Historians Journal, 45 (1), 45–67.

Accounting and Business Research 503

<< /ASCII85EncodePages false /AllowTransparency false /AutoPositionEPSFiles false /AutoRotatePages /PageByPage /Binding /Left /CalGrayProfile () /CalRGBProfile (Adobe RGB 50199851) /CalCMYKProfile (U.S. Web Coated 50SWOP51 v2) /sRGBProfile (sRGB IEC61966-2.1) /CannotEmbedFontPolicy /Error /CompatibilityLevel 1.3 /CompressObjects /Off /CompressPages true /ConvertImagesToIndexed true /PassThroughJPEGImages false /CreateJobTicket false /DefaultRenderingIntent /Default /DetectBlends true /DetectCurves 0.1000 /ColorConversionStrategy /sRGB /DoThumbnails true /EmbedAllFonts true /EmbedOpenType false /ParseICCProfilesInComments true /EmbedJobOptions true /DSCReportingLevel 0 /EmitDSCWarnings false /EndPage -1 /ImageMemory 524288 /LockDistillerParams true /MaxSubsetPct 100 /Optimize true /OPM 1 /ParseDSCComments false /ParseDSCCommentsForDocInfo true /PreserveCopyPage true /PreserveDICMYKValues true /PreserveEPSInfo false /PreserveFlatness true /PreserveHalftoneInfo false /PreserveOPIComments false /PreserveOverprintSettings false /StartPage 1 /SubsetFonts true /TransferFunctionInfo /Remove /UCRandBGInfo /Remove /UsePrologue false /ColorSettingsFile () /AlwaysEmbed [ true ] /NeverEmbed [ true ] /AntiAliasColorImages false /CropColorImages true /ColorImageMinResolution 150 /ColorImageMinResolutionPolicy /OK /DownsampleColorImages true /ColorImageDownsampleType /Bicubic /ColorImageResolution 300 /ColorImageDepth -1 /ColorImageMinDownsampleDepth 1 /ColorImageDownsampleThreshold 1.50000 /EncodeColorImages true /ColorImageFilter /DCTEncode /AutoFilterColorImages false /ColorImageAutoFilterStrategy /JPEG /ColorACSImageDict << /QFactor 0.90 /HSamples [2 1 1 2] /VSamples [2 1 1 2] >> /ColorImageDict << /QFactor 0.40 /HSamples [1 1 1 1] /VSamples [1 1 1 1] >> /JPEG2000ColorACSImageDict << /TileWidth 256 /TileHeight 256 /Quality 15 >> /JPEG2000ColorImageDict << /TileWidth 256 /TileHeight 256 /Quality 15 >> /AntiAliasGrayImages false /CropGrayImages true /GrayImageMinResolution 150 /GrayImageMinResolutionPolicy /OK /DownsampleGrayImages true /GrayImageDownsampleType /Bicubic /GrayImageResolution 300 /GrayImageDepth -1 /GrayImageMinDownsampleDepth 2 /GrayImageDownsampleThreshold 1.50000 /EncodeGrayImages true /GrayImageFilter /DCTEncode /AutoFilterGrayImages false /GrayImageAutoFilterStrategy /JPEG /GrayACSImageDict << /QFactor 0.90 /HSamples [2 1 1 2] /VSamples [2 1 1 2] >> /GrayImageDict << /QFactor 0.40 /HSamples [1 1 1 1] /VSamples [1 1 1 1] >> /JPEG2000GrayACSImageDict << /TileWidth 256 /TileHeight 256 /Quality 15 >> /JPEG2000GrayImageDict << /TileWidth 256 /TileHeight 256 /Quality 15 >> /AntiAliasMonoImages false /CropMonoImages true /MonoImageMinResolution 1200 /MonoImageMinResolutionPolicy /OK /DownsampleMonoImages true /MonoImageDownsampleType /Average /MonoImageResolution 300 /MonoImageDepth -1 /MonoImageDownsampleThreshold 1.50000 /EncodeMonoImages true /MonoImageFilter /CCITTFaxEncode /MonoImageDict << /K -1 >> /AllowPSXObjects true /CheckCompliance [ /None ] /PDFX1aCheck false /PDFX3Check false /PDFXCompliantPDFOnly false /PDFXNoTrimBoxError true /PDFXTrimBoxToMediaBoxOffset [ 0.00000 0.00000 0.00000 0.00000 ] /PDFXSetBleedBoxToMediaBox true /PDFXBleedBoxToTrimBoxOffset [ 0.00000 0.00000 0.00000 0.00000 ] /PDFXOutputIntentProfile (None) /PDFXOutputConditionIdentifier () /PDFXOutputCondition () /PDFXRegistryName () /PDFXTrapped /False /Description << /ENU () >>>> setdistillerparams<< /HWResolution [600 600] /PageSize [595.245 841.846]>> setpagedevice