This article was downloaded by: [Colorado College]On: 08 October 2014, At: 16:58Publisher: RoutledgeInforma Ltd Registered in England and Wales Registered Number:1072954 Registered office: Mortimer House, 37-41 MortimerStreet, London W1T 3JH, UK

European AccountingReviewPublication details, including instructionsfor authors and subscription information:http://www.tandfonline.com/loi/rear20

Earnings manipulation:cost of capital versustax. A commentaryNiclas HellmanPublished online: 22 Oct 2010.

To cite this article: Niclas Hellman (1999) Earnings manipulation: costof capital versus tax. A commentary, European Accounting Review, 8:3,493-497, DOI: 10.1080/096381899335907

To link to this article: http://dx.doi.org/10.1080/096381899335907

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy ofall the information (the “Content”) contained in the publicationson our platform. However, Taylor & Francis, our agents, and ourlicensors make no representations or warranties whatsoever asto the accuracy, completeness, or suitability for any purpose ofthe Content. Any opinions and views expressed in this publicationare the opinions and views of the authors, and are not the viewsof or endorsed by Taylor & Francis. The accuracy of the Contentshould not be relied upon and should be independently verified

with primary sources of information. Taylor and Francis shall notbe liable for any losses, actions, claims, proceedings, demands,costs, expenses, damages, and other liabilities whatsoever orhowsoever caused arising directly or indirectly in connection with,in relation to or arising out of the use of the Content.

This article may be used for research, teaching, and privatestudy purposes. Any substantial or systematic reproduction,redistribution, reselling, loan, sub-licensing, systematic supply,or distribution in any form to anyone is expressly forbidden.Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

Dow

nloa

ded

by [

Col

orad

o C

olle

ge]

at 1

6:58

08

Oct

ober

201

4

Earnings manipulation: cost of capitalversus tax. A commentary

Niclas HellmanStockholm School of Economics

ABSTRACT

The paper by Eilifsen, Knivs� å and Sættem, in this issue of the journal, providessome interesting research results in the � eld of accounting and taxation. Contraryto most European research in this area, Eilifsen, Knivs� å and Sættem apply atheoretical, deductive, model-based approach in the positive accounting tradition,and reach the somewhat provocative conclusion that linking taxable income toaccounting income reduces managers’ incentives to manipulate (overstate) earn-ings. Their theoretical results, supporting a view of the link between taxableincome and accounting income as an automatic safeguard against overstatementsof earnings, is a meritorious contribution to the current debate. For example, theirresults should be of interest for legislators and other authorities who consider achange to less dependence between accounting and taxation. However, from myperspective, the paper by Eilifsen, Knivs� å and Sættem also gives rise to somecriticism related to (i) the use of a model-based approach, and (ii) the applicabilityof the earnings manipulation approach.

CRITICISM RELATED TO THE USE OF A MODEL-BASEDAPPROACH

As I read the paper by Eilifsen, Knivs� å and Sættem, I reacted against theiruse of the earnings concept and the model situation. I believe that these aretoo simpli� ed and too speci� c, respectively, in order to justify the conclu-sions made regarding the real-world relationships between taxable income,accounting income and managers’ earnings manipulation incentives. At thesame time, I realize that this criticism is related to the use of model-basedresearch approaches in general, and that the approach also has advantages.Assumptions need to be made in order to simplify the real-world phenomenonenough to make it possible to model. On the basis of this, Eilifsen, Knivs� åand Sættem generate certain theoretical results. The critical issue is to whatextent the simpli� cations reduce the possibilities of drawing conclusionsabout the more complex real-world phenomenon.

Address for correspondenceStockholm School of Economics, Box 6501, S-113 83 Stockholm, Sweden.

The European Accounting Review 1999, 8:3, 493–497

© 1999 European Accounting Association 0963–8180

Dow

nloa

ded

by [

Col

orad

o C

olle

ge]

at 1

6:58

08

Oct

ober

201

4

The use of the earnings concept

The paper concerns the overstatement and understatement incentives regard-ing earnings, that follow from the commercial accounting and taxationperspectives, respectively. However, the Eilifsen, Knivs� å and Sættem modelis a one-period model, which means that they pass over the dynamic featureof earnings with respect to overstatements and understatements. For example,an investment with an economic life of � ve years that is directly charged toexpenses will cause an understatement of earnings in year 1, but as aconsequence earnings will be overstated in years 2–5. Eilifsen, Knivs� å andSættem admit that the model ‘. . . does not highlight the intertemporalconsiderations of earnings management’, as they put it. Applying a widerperspective, I believe the issue is more critical than that. Allocating revenuesand expenses to the right periods is the key element of earnings measure-ment, and in a one-period model this is not captured.

The model situation

The model situation is an original stockholder (identical to the manager)who, in order to � nance new investments, decides to sell out a proportion ofhis or her shares to new investors. The proportion sold out depends on themarket value of the � rm, which in turn might depend on what incomeestimate (Z) the � rm discloses regarding the period. In turn, this incomeestimate (Z) might differ from the manager’s own income estimate (Y). Interms of real-world situations, this looks like an IPO situation, whereprofessional owners seek to get as high an IPO price as possible, with lessemphasis on the long-term share price development (Bergstrom et al., 1995;Hogfeldt, 1997). This is a type of situation where the current owner/managercould have an incentive to overstate the expected income � gure for the nextperiod. The common empirical � nding, that IPOs tend to earn signi� cantlylower returns on investment than benchmark indices (Ritter, 1991; Levis,1995), supports this idea, although there are other possible explanations.However, such an IPO situation seems like a very special case of owner–manager relations, whereas the purpose formulated in the paper by Eilifsen,Knivs� å and Sættem refers to earnings manipulation incentives in general.First, there are both understatement and overstatement incentives. Forexample, it does not seem uncommon that newly appointed CEOs chargelarge non-recurring costs to expenses, knowing that there will then be fewercosts left during periods to come (understatement incentive). Second, whatdetermines the extent to which managers have incentives for earningsunderstatements, or earnings overstatements, seems to be a rather complexissue. Johansson and Ostman (1995: 26–7) describe this as follows:

. . . there is a continuous pressure on the managers responsible to always presenta satisfactory picture of their activity, a period measure. This may stimulate twoforms of behaviour. The � rst is to create margins of safety for the future. In

494 The European Accounting ReviewD

ownl

oade

d by

[C

olor

ado

Col

lege

] at

16:

58 0

8 O

ctob

er 2

014

favourable situations, the immediate impression does not involve a strain onaccounting numbers, whether reported or budgeted. The position in the face of anuncertain future tends to be strengthened. Some positive components of incomeare ‘saved’ for coming periods. In the reverse situation, there is a tendency to over-use positive arguments. Somebody thinks his unit is close to or below aquestionable level and therefore he must take care of the immediate impression,and the future impression will be of secondary importance. Margins of safety, ifany, diminish. Excessively positive reports are delivered; they may in turn createnew problems in terms of the expectations that are then created. . . . It is a commonobservation both for internal and external information, that good units are oftenbetter and bad units are often worse than they seem to be. In good times there isa tendency to employ conservative reporting; in bad times the contrary may betrue.

Neither understatements, nor the dynamics described in the above quotation,seem to be taken into account in the model situation. I believe that thisreduces the possibilities of drawing conclusions linked to earnings manipula-tion incentives in general.

Another problem with the model situation used by Eilifsen, Knivs� å andSættem is that it concerns income estimates, not reported income. First,accounting and tax rules really only apply to reported income. Second, itcannot be excluded that managers act differently when they provide the stockmarket with an income estimate, compared to when they close the company’sbooks.

THE APPLICABILITY OF THE EARNINGS MANAGEMENTAPPROACH

In the accounting and taxation literature, different countries have beenclassi� ed on a scale from strong links to weak links between accounting andtaxation. (See Hoogendoorn, 1996; Lamb et al., 1995. For an overview ofolder literature, see Nobes and Parker, 1995.) Eilifsen, Knivs� å and Sættemsuggest that the risk of overstatements of income will be lower in countrieswhere the commercial accounting is more strongly linked to taxation. Fromone point of view this seems very plausible. In order to minimize tax costs,companies in such countries could be expected to choose the most con-servative accounting treatment in cases where they have a choice. In Sweden,a country which has a medium-to-strong link between accounting andtaxation (Artsberg, 1996; Hoogendoorn, 1996), only what occurs as costs inthe commercial accounting will be deductible in the tax return form. Forexample, the Swedish telecommunication company Ericsson charges allits R&D costs to expenses, although it is allowed, according to Swedishcommercial accounting standards, to capitalize certain R&D costs. However,in Ericsson’s reconciliation of � nancial statements to US GAAP, softwaredevelopment must be capitalized. As a consequence, Ericsson has shown asubstantially lower earnings � gure according to Swedish GAAP, comparedwith US GAAP, over the last few years. In terms of the reasoning underlying

Earnings manipulation: cost of capital versus tax 495D

ownl

oade

d by

[C

olor

ado

Col

lege

] at

16:

58 0

8 O

ctob

er 2

014

the Eilifsen, Knivs� å and Sættem model, Ericsson could be seen as having anincentive to understate income in Sweden, where it pays a large proportion ofits taxes, and an incentive to overstate income internationally where it obtainsa large proportion of its capital.1 However, although this interpretation isconsistent with the model, it does not seem plausible. I believe Ericssoncharges all its R&D costs to expenses in its Swedish accounts primarily inorder to minimize its tax costs, and it capitalizes software development costsaccording to US GAAP because it is compulsory to do so. This seems likecommon sense, and I see little need for an explanation involving earningsmanipulation incentives.

Applying a wider perspective, I believe the basic view of managers aspeople who primarily seek to manipulate � gures is to be deplored. Managers’manipulation incentives is only one aspect that affects the preparation anduse of accounting information. After all, accounting is a language used todescribe the � rm’s business activities in a reliable way and, as such, it is avery important instrument for the management, both internally and ex-ternally. If users perceive that the accounting has been manipulated, theymight start relying on some other information system. In turn, this mightcause problems for the managers in terms of decreased trust and lessoverview and control. Therefore, I believe managers also have an incentive toavoid manipulation of the accounting. Earnings manipulation papers in thepositive accounting tradition have become a vast literature over the last fewyears, and critique is rare (Sterling, 1990; Chambers, 1993). However, for thefuture of accounting research, I believe it is important that the basicassumptions we make are not biased towards some particular aspect of thepreparation and use of accounting information, such as managers’ incentivesto manipulate earnings. Otherwise, why should people outside the scienti� ccommunity be interested in the research results from a discipline which holdsthe basic view that managers are people who manage earnings rather than� rms?

NOTE

1 The authors provided me with this interpretation during the review process.

REFERENCES

Artsberg, K. (1996) ‘The link between commercial accounting and tax accounting inSweden’, European Accounting Review, 5 (Supplement): 795–814.

Bergstrom, C., Hogfeldt, P. and Westin, A. (1995) ‘The role of venture capitalists ininitial public offerings: an explorative study of US and Swedish evidence’,Research in International Business Finance, 12: 153–83.

Chambers, R. J. (1993) ‘Positive accounting theory and the PA cult’, Abacus, 29(1):1–26.

Eilifsen, A., Knivs� å, K. H. and Sættem, F. (1999) ‘Earnings manipulation: cost ofcapital versus tax’, European Accounting Review, this issue.

496 The European Accounting ReviewD

ownl

oade

d by

[C

olor

ado

Col

lege

] at

16:

58 0

8 O

ctob

er 2

014

Hogfeldt, P. (1997) ‘Vad vet vi om svenska borsintroduktioner? ’ [What do we knowabout Swedish IPOs?], in Bergstrom, C. and Bjork, T. (eds) Från option-sprissattning till konkurslagstiftning – Essaer om � nansiella teorier, institutioneroch marknader. Stockholm: EFI at Stockholm School of Economics, pp. 31–61 (inSwedish).

Hoogendoorn, M. N. (1996) ‘Accounting and taxation in Europe – a comparativeoverview’, European Accounting Review, 5 (Supplement): 783–94.

Johansson, S.-E. and Ostman, L. (1995) Accounting Theory – Integrating Behaviourand Measurement. London: Pitman.

Lamb, M., Nobes, C. and Roberts, A. (1995) ‘The in� uence of taxation onaccounting: international variations’, Paper presented at the EIASM Workshop onAccounting in Europe, No. 4, Geneva.

Levis, M. (1995) ‘Seasoned equity offerings and short and long run underperfor-mance of initial public offerings’, European Financial Management, 1(2):125–46.

Nobes, C. W. and Parker, R. (1995) Comparative International Accounting, 4th edn.New York: Prentice-Hall.

Ritter, J. R. (1991) ‘The long run underperformance of initial public offerings’,Journal of Finance, 46(1): 3–27.

Sterling, R. R. (1990) ‘Positive accounting: an assessment’, Abacus, 26(2):97–135.

Earnings manipulation: cost of capital versus tax 497D

ownl

oade

d by

[C

olor

ado

Col

lege

] at

16:

58 0

8 O

ctob

er 2

014