The current issue and full text archive of this journal is available atwww.emeraldinsight.com/0114-0582.htm

PAR22,2 IFRS in New Zealand: effects on

xnancial statements and ratiosWarwick Stent, Michael Bradbury and Jill Hooks

School of Accountancy, Massey University, Auckland, New Zealand

AbstractPurpose – The purpose of this paper is to examine the xnancial statement impacts of adopting NZIFRS during 2005 through 2008.Design/methodology/approach – The effects of NZ IFRS on the xnancial statements and ratios ofxrst-time adopters of NZ IFRS for a stratixed random sample of 56 listed companies is analysed. Intotal, 16 of these were early adopters and 40 of which waited until adoption of NZ IFRS becamemandatory. The analysis of the xnancial statement impact of NZ IFRS is conducted in the context ofthe accounting choice literature.Findings – The results show that 87 per cent of xrms are affected by NZ IFRS. The median andinter-quartile ranges indicate that for most xrms the impact of NZ IFRS is small. However, themaximum and minimum values indicate the impact can be large for some entities. The impact hasconsiderable effects on common xnancial ratios.Research limitations/implications – The usual limitations applicable to small samples apply.Practical implications – The xndings may be useful to regulators and policy makers reviewingxnancial reporting requirements.Originality/value – This study is the xrst to offer a comprehensive empirical analysis of the effectof adopting IFRS on xnancial statements in New Zealand, as well as on selected key ratios of interestto xnancial analysts. The data used are more recent than most IAS or IFRS studies around the worldand are stratixed to allow for comparison between voluntary/early adopters and mandatory/lateadopters.

Keywords Financial reporting, International standards, Management ratios, New ZealandPaper type Research paper

92

1. IntroductionThis paper examines the impact of international xnancial reporting standards (IFRS)on the xnancial statements of New Zealand listed companies. Specixcally, wedocument the impact of NZ IFRS on the following xnancial statement elements: assets,liabilities, equity, revenues and expenses. We then examine the xnancial statementimpact of NZ IFRS analysed by accounting standard (e.g. xnancial instruments,income taxes). Finally, we examine the impact of NZ IFRS on some common xnancialratios. Daske et al. (2008) note that the adoption of IFRS by over 100 countries is one of themost signixcant changes in world accounting history[1]. They also note that empiricalevidence on the consequences of mandatory IFRS is in its infancy and emphasise theneed for further evidence.

Pacixc Accounting ReviewVol. 22 No. 2, 2010pp. 92-107q Emerald Group Publishing Limited0114-0582DOI 10.1108/01140581011074494

The authors would like to extend their thanks to participants at a Massey University researchworkshop, especially Asheq Rahman, for helpful suggestions. Warwick Stent gratefullyacknowledges xnancial support from the New Zealand Institute of Chartered Accountants andthe Accounting and Finance Association of Australia and New Zealand.

This paper makes several contributions to the literature. It is the xrst to offer acomprehensive analysis of the xnancial statement effects of adopting IFRS in NewZealand. Similar to Hung and Subramanyam (2007), we report detailed xnancialstatement effects of adopting IFRS. This contribution may be useful to regulators andpolicy makers that are currently reviewing the application of IFRS for smaller entitiesin New Zealand. Second, studies examining the value relevance of IFRS (e.g. Daskeet al., 2008; Hung and Subramanyam, 2007; Barth et al., 2008) have found mixed results.A pre-requisite for IFRS to have value relevance is that IFRS must impact xnancialratios. Hence, a focus of this study is the impact of IFRS on common xnancial ratios.Furthermore, the potential impact of IFRS on ratios, will not only impact theassessment of value relevance but also analysts’ credit decisions (e.g. credit scoringmodels such as Altman, 1968) and contracting decisions by xrms that employ xnancialratios (e.g. debt covenants, compensation contracts). Third, this study is partitioned toallow for comparison, between voluntary early adopters and mandatory late adopters,of the xnancial statement effects of NZ IFRS. Understanding the impact of NZ IFRS isan important xrst step in seeking a deeper understanding of the motivations of earlyadopters of IFRS in New Zealand. Fourth, the investigation period covers xscal yearscommencing on, or after, 1 January 2005 through to 30 September 2008. This providesmore recent information on the impact of IFRS than is included in most recent studies.This is an important consideration in view of the signixcant, frequent and continuingamendments to IFRS since the 2005 “stable platform” was achieved. Fifth, most studiesexamine the switch to IFRS where previous GAAP was known to differ signixcantlyfrom the international standards, whereas old NZ GAAP is perceived to be relativelysimilar. Contrary to expectations, the analysis of effects in such a setting indicates thatIFRS information still conveys new information (Christensen et al., 2007). In summary,this paper makes a number of contributions with regard to the impact of IFRSadoption. The paper proceeds as follows. Section 2 provides background information on theNew Zealand adoption of IFRS. Section 3 briexy reviews the theory and recentliterature relevant to the impact of IFRS on xnancial statements. Section 4 contains adescription of the sample selection and data. Section 5 describes and discusses theresults and Section 6 provides a summary and conclusion.

2. Background on the adoption of NZ IFRSIn 1974, the xrst of a new series of accounting standards issued by the New ZealandSociety of Accountants (i.e. SSAP 1 Disclosure of Accounting of Policies) carried theInternational Accounting Standards Committee crest (Bradbury, 1998). However, eventhat standard contained modixcations. Furthermore, in the following years, NewZealand developed its own standard setting agenda with regard to accounting issues. A decision was taken by the Financial Reporting Standards Board (FRSB)[2] in 1997to base new accounting standards on International or Australian accounting standards.These were modixed to ensure sector neutrality and consistency with other New Zealandpronouncements (Bradbury and van Zijl, 2006). On 21 October 2002, the AccountingStandards Review Board (ASRB) proposed that listed issuers in New Zealand shouldadopt IFRS and on 19 December 2002 they announced that adoption of IFRS was to bemandatory for reporting entities in New Zealand for periods beginning on or after1 January 2007. Unlike the European Union, Australia and many other countries which

IFRS in New Zealand

93

PAR22,2

94

opted for mandatory adoption in 2005, the ASRB allowed early adoption for periodsbeginning on or after 1 January 2005 (Bradbury and van Zijl, 2006). On 12 September 2007 the ASRB announced it had decided to delay mandatoryadoption of NZ IFRS for small companies that met specixed criteria, pending agovernment review of xnancial reporting requirements for small- and medium-sizedcompanies, which was to be commenced during mid-2008. A possible outcome of thisreview is that many entities may no longer be required by law to prepareGAAP-compliant xnancial statements (Sealy-Fisher, 2007). This review therefore hasmajor implications for a large number of New Zealand businesses. The xndings of thispaper may be informative to policy makers conducting this review. Since the ASRB’s announcement in 2002, there has been much in the way ofcomment by various authors, professional bodies, accounting xrms and commercialentities, about the impact of adopting IFRS (e.g. Dunstan, 2002; Ernst & Young, 2004).It is widely accepted that adopting IFRS may have signixcant implications.

3. Theory and literatureFields et al. (2001) provide a review of the accounting choice literature up until the1990s. We therefore begin with a summary of their review, followed by a review ofmore recent studies related to the impact of IFRS.

3.1 Summary of a review of accounting choice literature up until the 1990sFields et al. (2001) refer to three main categories of motivations for accounting choice:contracting, asset pricing and inxuencing external parties. They note that in general,researchers xnd that efxcient contracting incentives are effective in contractualarrangements, namely that managers select accounting methods to increasecompensation (the “bonus hypothesis”) and to avoid breaching debt covenants (the“debt hypothesis”). In their opinion the literature leaves some uncertainty as to whetherthese choices are made opportunistically or for value-maximising purposes. “Assetpricing”, concerns the economic consequences of accounting information on shareprices (market value of equity) and cost of capital. Fields et al. (2001) state that xndingsrelated to accounting choices being made for their effect on share prices are generallyunconvincing, mainly because of competing hypotheses, such as market efxciency andcontracting. They note that results are mixed as to whether increased levels ofdisclosure results in decreased cost of capital. The political cost hypothesis suggeststhat accounting choices are made to avoid transfers of wealth to external parties.Evidence from this research indicates that accounting choices are made to reduce ordefer taxes. However, the stock market effects of these actions is mixed.

3.2 Recent studies relating to impact of IFRSThis section briexy reviews more recent empirical studies examining the xnancialstatement effects of adopting IFRS (e.g. Daske et al., 2008; Lee et al., 2008; Barth et al.,2008; and Hung and Subramanyam, 2007). Daske et al. (2008) examine the economic consequences of mandatory IFRS adoptionfor a large sample of xrms across 26 countries covering xscal years ending on, or after,1 January 2001 through to 31 December 2005. Their xndings indicate that IFRSadoption appears to be associated with positive economic consequences for marketliquidity, cost of capital, and xrm value. These capital market effects are most

pronounced for voluntary adopters, both in the year when they switch and again later,when IFRS becomes mandatory. They caution that the initial impact is likely to be dueto self-selection and that the mandatory impact will be affected by omitted variablessuch as concurrent improvements to securities laws, regulatory enforcement, xrmgovernance, and reporting incentives. Support for the view that factors other than IFRScontribute signixcantly to capital market effects is evident in other studies (e.g. Leeet al., 2008; Barth et al., 2008). Lee et al. (2008) xnd, across a sample of 17 European countries, that the cost ofcapital only reduced in countries with high reporting incentives and enforcement.Barth et al. (2008) xnd that adoption of IFRS is associated with higher accountingquality. They study xrms from 21 countries applying IAS between 1994 and 2003 andxnd that there is less evidence of earnings management, more timely loss recognitionand more value relevance of accounting information for their sample xrms than for amatched sample of xrms applying non-US domestic GAAP. Hung and Subramanyam (2007) examine the xnancial statement effects of adoptingIAS during 1998 – 2002. They measure xnancial statement effects by direct comparisonof xnancial statements prepared under both IAS and German GAAP (referred to asHandelsgesetzbuch or “HGB”). They also contend that studies of the effects of IASbased on pre-1998 xnancial statements are unlikely to be representative. The core IASstandards were completed in 1998 and removed the choice of partial adoption (i.e.“cherry picking”) of IAS by requiring full implementation of IFRS. Hung andSubramanyam (2007) present two sets of analyses. First, the major accountingdifferences between HGB and IAS are analysed. They xnd that the adoption of IASresulted in “. . . widespread and signixcant changes to deferred taxes, pensions,property, plant and equipment, and loss provisions” (Hung and Subramanyam, 2007,p. 625). Overall, total assets and book value of equity were found to be signixcantlylarger under IAS than under HGB, while variations in book value and net income werefound to be signixcantly higher. Second, they examine the value relevance of bookvalues and net income, as well as the timeliness of income information. There is noevidence that IAS improves value relevance of book value or net income. There is weakevidence suggesting that IAS income has increased asymmetric timeliness (i.e. IASincorporates bad news into income in a more timely manner than HGB). These recent studies xnd mixed results for value relevance of IFRS. Barth et al.(2008), xnd more value relevance and Hung and Subramanyam (2007) xnd no evidencethat IFRS improves value relevance. Daske et al. (2008) note an increase in equityvaluations but only if they account for the possibility that the effects occur prior to theofxcial adoption date. This echoes comments by Fields et al. (2001) that xndingsrelated to the share price effect of accounting choices are generally unconvincing. Inassessing the value relevance of IFRS, a pre-requisite is that the adoption of IFRSshould have a signixcant impact on xnancial statements and common xnancialstatement ratios. Hence, the objective of this study is to provide descriptive evidence ofthe impact of IFRS. The study that is most relevant to ours is Hung and Subramanyam (2007). Theymake a case for a country-specixc approach, which considers the direct effects ofadopting IAS for the same set of xrm years. Such an approach would help to overcomeproblems associated with comparing across countries with different institutionalarrangements, as well as controlling for time-series differences. We improve on their

IFRS in New Zealand

95

PAR22,2

96

analysis as we use the mandatory IFRS/local GAAP reconciliations (now an IFRS 1requirement), while their study relies on voluntary IAS/local GAAP reconciliationdisclosures. Hence, their study is likely to have a self-selection bias. Hung andSubramanyam (2007) are only able to observe “book value” reconciliation information(i.e. equity adjustments) for 57 xrms; and “net income” reconciliation (i.e. proxt at endof prior period adjustments) for 31 xrms in their sample of 80 xrms. Also, the investigation period for this paper is more current than Hung andSubramanyam (2007). As noted earlier, this is important in view of the signixcant, andon-going amendments to IFRS since the “stable platform” was achieved in 2005. Hungand Subramanyam (2007) consider the adoption of IFRS where German GAAP wasknown to differ signixcantly from IAS, whereas old New Zealand GAAP is perceivedto be relatively similar. Christensen et al. (2007) xnd that in spite of IFRS beingrelatively similar to UK-GAAP, the IFRS reconciliations contained information thatanalysts considered relevant for xrm valuation and that xrms opportunistically tendedto delay unfavourable reconciliations.

4. Sample selection and dataFigure 1 reports the outcome of our sample selection procedures. We begin with all 161companies listed on the New Zealand Stock Exchange (NZX) on 1 March 2007. We thenstratify these into early and late adopters of NZ IFRS. Early adopters (EA) are the reporting entities that chose to adopt NZ IFRS for periodsbeginning on or after 1 January 2005, but before it became mandatory (periods beginningon or after 1 January 2007). Selection as an EA is only made once the “. . . explicit andunreserved statement of compliance . . . ” with NZ IFRS has been sighted in a xrstfull-year set of xnancial statements[3]. Our procedures identify an initial set of 48 EA. Theremaining total of 113 “non-EA” forms our initial population of Late Adopters (LA). Theseare reporting entities listed on the NZX, which chose to wait until periods beginning on orafter 1 January 2007, when it became mandatory to adopt NZ IFRS. We lose 12

Figure 1.Effect of sample selectioncriteria

observations where companies delisted from the NZX after 1 March 2007, resulting in areduced population of 101 LA. We conxrm that they are LA by ensuring that the “. . .explicit and unreserved statement of compliance . . . ” with NZ IFRS appears for the xrsttime in a xrst full-year set of xnancial statements beginning on or after 1 January 2007. For the 48 EA (Panel B), observations are discarded because the entity uses GAAPother than NZ IFRS (four observations); uses a functional currency other than NZdollars (2); has no prior year xnancial statements available as the xrst year of listing onthe NZX is also xrst year of application of NZ IFRS (1); or has no reconciliation becauseNZ IFRS was adopted from its xrst year of operation (1). As the data are hand collected from xnancial statements and footnotes, we make arandom selection, of approximately 40 per cent, from each of the EA and LApopulations. This results in a sample of 56 observations, comprising 16 EA and 40 LA.Our sample size is a trade off between the cost of collecting information and thebenexts of a larger sample. We gather two sets of xnancial statements for all observations: the xrst full-year NZIFRS xnancial statements and the year prior to adoption of NZ IFRS. Information on theadjustments made to the “pre-NZ IFRS” year xgures are extracted from the NZ IFRS/oldNZ GAAP reconciliations. NZ IFRS 1 requires comparatives to be restated and reconciledin the xrst year of adopting NZ IFRS. The reconciliations varied considerably in formatand level of detail supplied and it was important to determine and separate out whichxnancial statement elements were impacted by NZ IFRS, the amounts involved and theaccounting standards to which these impacts should be attributed. Table I reports descriptive statistics of the sample xrms. In this table we report oldNZ GAAP xgures (i.e. “OLD GAAP”) as this is the base from which we measure thedifferences due to the adoption of NZ IFRS. The mean total assets xgure is $772.1million, mean total equity is $477.6 million and mean net proxt is a loss of $149.2million. The sample data are not normally distributed and for the most part areleptokurtic and positively skewed. This results in the median being a better indicatorof central tendency than the mean. We therefore rely on non-parametric tests to analysestatistical differences in our sample data.

5. Results5.1 Descriptive statistics: impact of NZ IFRSDescriptive statistics of the impact of NZ IFRS for the xnancial statement elements (asper the NZ Framework) are presented in Table II.

Total assets

MeanStandard deviationMinimum25 percentileMedian75 percentileMaximum

Note: n ¼ 56

772,0841,702,230 164 20,254 92,099 438,9488,991,425

Total liabilities

294,827 730,193 12 4,616 25,419 180,2403,825,819

Total equity

477,6271,289,398 2 95 17,978 48,794 277,4108,738,489

Total revenue

303,610 671,058 6 9,871 52,234 320,9014,297,000

Net proxt

2 149,191 1,279,2332 9,542,816 76 3,140 35,722 214,000

IFRS in New Zealand

97

Table I. Descriptive statistics($000) of sample under OLD GAAP

PAR22,2

Panel A: magnitude of change aMeanStandard deviationMinimum25 percentileMedian75 percentileMaximum

Panel B: sign of changes (%)NegativePositiveNo change

Panel C: statistical tests bZ statisticp-value (two-tailed)

Totalassets

0.031 0.20620.640 0.000 0.002 0.020 1.230

255520

2.9400.003

Totalliabilities

0.2160.7140.0000.0000.0270.1515.130

47521

5.7130.000

Totalequity

2 0.070 0.5822 3.4402 0.0462 0.005 0.012 2.130

573013

2.1730.030

Totalrevenue

0.134 0.71320.99020.012 0.000 0.010 3.950

413623

0.1990.842

Netproxt

0.127 0.4682 1.0002 0.008 0.024 0.149 1.930

325513

2.5220.012

98

Table II.Impact of NZ IFRS onxnancial statementelements

Notes: n ¼ 56; aThe change is estimated as (NZ IFRS/OLD GAAP) – 1; bThe reported test statistic isa Wilcoxon test for equality of matched pairs

Panel A of Table II reports the magnitude of change in a particular xnancial statementelement due to the adoption of NZ IFRS. It is measured as: NZ IFRS/OLD GAAP - 1.For example, the mean (median) change in total assets due to the adoption of NZ IFRSis 3.1 per cent (0.2 per cent). Panel B reports the number of increases, decreases and nochanges. The adoption of NZ IFRS results in an increase in assets for 55 per cent (31/56)of the observations, a decrease for 25 per cent of observations and 20 per cent remainunchanged. Panel C reports Wilcoxon tests for difference in the distribution of avariable for matched pairs. That is, each xrm is compared with itself for differencesbetween OLD GAAP and NZ IFRS. The general increase in total assets due to theadoption of NZ IFRS is statistically signixcant at the 0.01 level. Table II reveals that the largest impact of NZ IFRS is for liabilities, where 75 percent of observations (42/56) report an increase in liabilities and only 4 per cent report adecrease. The impact of NZ IFRS is widespread as 87 per cent of xrms are affected.That is, only 13 per cent of xrms have no changes to equity or net proxt. Overall, theimpact of NZ IFRS signixcantly increases assets, liabilities and net proxt, butdecreases equity. The impact on revenue is not signixcant at conventional levels. Theinter-quartile range for most elements is small indicating that for most xrms, theimpact of NZ IFRS is small. However, the maximum and minimum values indicate thatthe effect of NZ IFRS can be quite substantial for some xrms. For example, theinter-quartile range for net proxt is 0.157 (0.149 – (2 0.008)) and the range is 2.930(1.930 – (2 1.000)).

5.2 Descriptive statistics: impact of NZ IFRS analysed by early (late) adopters and bysmall (large) entitiesTable III reports the impact of NZ IFRS on xnancial statement elements analysed byearly and late adopters. This Table is similar in presentation to Table II.

Totalassets

0.161

20.040 0.66223.44020.04420.0040.0002.130 582815

4.762

0.009

0.000

1.513 0.130

0.4010.0000.0000.0120.1251.590

68113

3.170 0.002

0.036

0.721

0.256

2.101

3.570

1.136

2.626

56386

441938

31636

255323

37325

0.3072 1.0002 0.1432 0.0260.0140.040

0.1692 0.0902 0.0030.0000.0000.600

0.32221.00020.0030.0720.1400.540

0.21820.0600.0000.0020.0231.230

0.8100.0000.0000.0350.1805.130

2 0.144

0.048

0.021

0.057

0.238

0.169 0.83720.99020.0250.0000.0163.950 404318

0.286 0.775

Early adopters (n ¼ 16) Total Total Totalliabilities equity revenue

Netproxt

Totalassets

Late adopters (n ¼ 40) Total Total Totalliabilities equity revenue

Netproxt

0.169 0.5132 0.9302 0.0090.0020.1821.930

335315

2.013 0.035

Panel A: magnitudeMeanStandarddeviationMinimum25 percentileMedian75 percentileMaximum

of change a2 0.035 0.1622 0.6402

0.0110.0070.0150.040Panel B: sign of changesNegative 25Positive 63No change 13

Panel C: statistical tests bZ statistic 0.847p-value (two-tailed) 0.397

Notes: aThe change is estimated as (NZ IFRS/OLD GAAP) – 1; bThe reported test statistic is a Wilcoxon test for equality of matched pairs

99

Table III.Impact of NZ IFRS on xnancial statement elements

IFRS in New Zealand

PAR22,2

100

The impact of NZ IFRS on EAs is signixcant for total liabilities (at the 0.01 level) andequity (at the 0.05 level). Total assets, revenue and net proxt are not statisticallydifferent under NZ IFRS compared to OLD GAAP. On the other hand, total assets andtotal liabilities of LAs are signixcantly higher (at the 0.01 level), as is net proxt (at the0.05 level). Total equity and total revenue are not statistically different under NZ IFRS.The minimums are generally much lower and maximums much higher for the LAs ascompared to the EAs. These results are consistent with early adopting xrms beingthose xrms on which NZ IFRS have a lower impact. We performed a similar analysis (untabulated) by small and large xrm, using themedian xgure for total assets under OLD GAAP as our cut-off point (i.e. “small”, $92,099,000 , “large”). We xnd that the impact of NZ IFRS on the small xrms is notsignixcant, at conventional levels, for any xnancial statement elements except totalliabilities (at the 0.01 level). This contrasts strongly with the xndings for the largexrms, where the impact of NZ IFRS is found to be signixcant for all xnancial statementelements except total revenue. Total assets and total liabilities are signixcant at the0.01 level, and total equity and net proxt at the 0.05 level for the large xrms. Minimumand median xgures, as well as the proportion of positive sign changes are generallymuch higher for the large xrms than for the small xrms. Thus, small listed xrms areless affected by NZ IFRS than large listed xrms.

5.3 Descriptive statistics: impact of NZ IFRS by accounting standardsTable IV reports the impact of the adoption of NZ IFRS analysed by the accountingstandard that caused the accounting change and is similar to Tables II and III, exceptthat it is presented in landscape to provide sufxcient detail. In Table IV we report thebalance sheet impact as well as the income statement impact, separately showingincreases and decreases in revenues and expenses. The data for this analysis are extracted from the NZ IFRS/OLD GAAPreconciliations required by NZ IFRS 1. The most informative reconciliations providefull balance sheets (at date of transition and at end of prior period) and an incomestatement (at end of prior period), each of which disclose OLD GAAP, NZ IFRS andreconciliation adjustments for every line item in these xnancial statements, togetherwith detailed notes explaining the adjustments, with specixc reference to the relevantNZ IFRS standards. Less informative reconciliations simply reconcile Equity (at date oftransition and at end of prior period) and Proxt (at end of prior period) under OLDGAAP to NZ IFRS, showing material adjustments with relatively little in the way ofdetailed explanation and more general references to “NZ IFRS” rather than specixcstandards. As NZ IFRS comprises in excess of 40 standards, we restrict the tables presented inthis paper to separate consideration of only those standards that have non-zero effectson 10 per cent or more of our total sample. The most striking feature of Table IV is that, with the exception of the income taximpact on equity, the median observation is zero across all xnancial statementselements and accounting standards. All but three of the 25th percentiles are also zero.The last column in Table IV reports the percentages of no change. The range of “noimpact” is from 57 per cent to 100 per cent. This indicates that for most xrms a specixcstandard of NZ IFRS has no impact. However, the minimums and maximums indicatethat a specixc NZ IFRS can be very material for a small number of xrms.

Minimum $000

(NZ IAS 32/39)(NZ IAS 19)(NZ IFRS 3)(NZ IAS 12)(NZ IAS 32/39)(NZ IAS 19)(NZ IFRS 3)(NZ IAS 12)(NZ IAS 32/39)(NZ IAS 19)(NZ IFRS 3)(NZ IAS 12)(NZ IAS 32/39)(NZ IAS 19)(NZ IFRS 3)(NZ IAS 12)(NZ IFRS 2)(NZ IAS 32/39)(NZ IAS 19)(NZ IFRS 3)(NZ IAS 12)(NZ IFRS 2)

268,00021,04926,89821,599217,44521,00002243,0002543,736218,00226,8982663,457292,04002174002118,0002158294,382241,70621,134

000110

0000000002107022,63300000000260

00000000000250000000000

3180035410703,1541870000000

101,3003,00094,38413,700545,59318,0021,287664,84189,7004,00094,384243,000133,0004,0004,1300035,3941,0365,81028,2601,000

26113161421124242131181100996158

713752021221328301008613152

25thpercentile $000

Median $000

Maximum % $000 positive

%negative

75thpercentile $000

% nochange 5798847781779974647784618999981001008385817090

AssetsAssetsAssetsAssetsLiabilitiesLiabilitiesLiabilitiesLiabilitiesEquityEquityEquityEquityRevenues/incomeRevenues/incomeRevenues/incomeRevenues/incomeRevenues/incomeExpensesExpensesExpensesExpensesExpenses

Financial instrumentsEmployee benextsBusiness combinationsIncome taxesFinancial instrumentsEmployee benextsBusiness combinationsIncome taxesFinancial instrumentsEmployee benextsBusiness combinationsIncome taxesFinancial instrumentsEmployee benextsBusiness combinationsIncome taxesShare base paymentsFinancial instrumentsEmployee benextsBusiness combinationsIncome taxesShare based payments

Note: n ¼ 56

101

Table IV. Impact of NZ IFRS by xnancial statementelement and accounting standard

IFRS in New Zealand

PAR22,2

102

Further analysis of Table IV shows that the adjustments under NZ IAS 32 and 39(xnancial instruments) are the most frequent, with 33 per cent non-zero observationsaffecting assets, 19 per cent affecting liabilities, 36 per cent affecting equity, and 28 percent affecting net proxt (11 per cent revenues and 17 per cent expenses). Furthermore,the impact is both positive and negative, but the positive impact is typically more thantwice as frequent as the negative impact. These xndings are consistent withexpectations noted in Ernst & Young (2004, p. 1) who predicted that xnancialinstruments would be the “ . . . area most heavily impacted by the move to IFRS”. Hungand Subramanyam (2007) report a relatively infrequent 23 per cent adjustmentobservations for xnancial instruments in their sample, 16 per cent of which werepositive. This is consistent with the results of Berkman et al. (1997), who xnd that NewZealand xrms are large users of xnancial instruments relative to US xrms. The signixcant increase in liabilities under NZ IFRS noted in Table II can beattributed mainly to NZ IAS 12 Income Taxes. This standard increases liabilities in 24per cent of the observations, decreases equity in 28 per cent of the observations andincreases assets in 16 per cent of the observations. The impact on current year proxtsis, however, mixed with Table IV showing both increases and decreases in expenses of15 per cent. The income tax adjustments for our sample are due mainly to deferred taxdifferences, which arise because NZ IAS 12 adopts a “balance sheet approach”, whichis signixcantly different to the “income statement approach” formerly used under OLDGAAP. While the inter-quartile range for the impact of NZ IAS 12 on equity is $2.6million the range is $906 million. These results provide an insight as to the variation inimpact of the deferred tax adjustments across the sample xrms, as well as a reminderthat these adjustments may be both book value – increasing and decreasing. These xndings are consistent with the expectation that the amount of deferred taxassets and liabilities reported in the balance sheet would increase (Ernst & Young,2004). Our results reveal that the increases are considerably larger and more frequentfor liabilities than for assets. These xndings are also consistent with those of Hung andSubramanyam (2007). However, they report frequencies of 95 per cent of observationsaffecting liabilities and 81 per cent affecting net proxt. This supports the view that oldNZ GAAP is relatively closer to NZ IFRS than German GAAP was to IAS. Employee Benexts (under NZ IAS 19) also contribute to the signixcant increase inliabilities under NZ IFRS. Liabilities increase and equity decreases for 21 per cent of theobservations as a result of a number of new requirements under NZ IAS 19. Forexample, employers are required to recognise an asset or liability in respect of anyemployee dexned benext plans; they are also required to accrue for both vested andnon-vested employee benexts such as long service leave and sick leave. Hung andSubramanyam (2007) report that employee benexts are the second most frequentadjustment in their sample at 72 per cent; of which 67 per cent result in negativeadjustments for equity. Business Combinations increase equity and assets by 13 per cent and decreaseexpenses also by 13 per cent. These adjustments arise mainly as a result of therequirements in NZ IFRS 3 that goodwill should not be amortised, but should besubject to impairment testing. The predominance of increases in assets (and hence alsoin equity) and decreases in expenses are consistent with expectations expressed in theErnst & Young (2004) report. Hung and Subramanyam (2007) report similar results fortheir sample.

Share-based payments are the last category of adjustments, which have non-zeroeffects on 10 per cent or more of our total sample. These adjustments affect onlyexpenses, increasing them for 8 per cent of observations and decreasing them for 2 percent of observations. The minimum and maximum values indicate that theseadjustments are relatively small in comparison to the other adjustments.

5.4 The effect of NZ IFRS on key ratiosTable V presents results of our analyses into the effect of NZ IFRS on key ratios, whichserve as proxies for variables that may inxuence or be of interest to analysts. Wechoose xve key ratios[4]: (1) return on equity (net proxt to equity); (2) return on assets (net proxt to total assets); (3) leverage (total liabilities to equity) (4) asset turnover (revenue to total assets); and (5) return on sales (net proxt to revenue).

These ratios rexect the main ratios in the Du Pont analysis[5]. Under NZ IFRS the median return on equity increases from 9.2 per cent to 11.9 percent (see Panel A). Return on equity increases for 64 per cent of observations anddecreases for 25 per cent of observations (see Panel B). The change in this ratio issignixcant at the 0.01 level. The median return on assets increases from 4.7 per cent to5.0 per cent (56 per cent increase and 33 per cent decrease). The large impact of NZIFRS on liabilities has an effect on leverage. Median leverage increases from 60.2 percent to 69.7 per cent (64 per cent increase and 24 per cent decrease). The differences forleverage are statistically signixcant at the 0.01 level. The median for asset turnoverdecreases from 78.1 per cent to 69.5 per cent rexecting the general increase in totalassets under NZ IFRS (30 per cent increase and 57 per cent decrease). The median forreturn on sales increases from 5.8 per cent to 6.2 per cent (66 per cent increase and 25per cent decrease). The differences for return on sales are statistically signixcant at the0.01 level. Table V indicates, xrst, that the “no-change” effect of NZ IFRS on xnancial ratios issmall (9 to 13 per cent). Second, the impact of NZ IFRS does not simply result in auniform “jump” in xnancial statement ratios but has a xrm-specixc effect. For somexrms a particular ratio may increase, while for other xrms that ratio may decrease. In Panels C and D of Table V we analyse the changes in xnancial ratios resultingfrom the move to NZ IFRS for early adopters and late adopters. The results are similarto Panel A in the sense that for most ratios the percentage of ratio increases is almostalways twice that of decreases. However, changes in return on assets, leverage andasset turnover are not statistically signixcant (at conventional levels) for earlyadopters. For late adopters the increase in leverage is signixcant at the 0.01 level.Return on equity and return on sales are signixcant for both early and late adopters atthe 0.05 level. In general, results are more strongly signixcant for late adopters than forearly adopters. We perform a similar analysis (untabulated) on small and large xrms, using the sizecriterion in section 5.2. The trends for ratios discussed for Panel A hold for both thesmall and the large xrms, but are more pronounced for the large xrms. The proportion

IFRS in New Zealand

103

104

PAR22,2

Panel A: ratio comparisons (all observations) aMeanStandard deviationMinimum25 percentileMedian75 percentileMaximumPanel B: number of changes and statistical tests(all observations) bDecreases (%)Increases (%)No change (%)Z statisticp-value (two-tailed)Panel C: number of changes and statistical tests(early adopters) bNegative (%)Positive (%)No change (%)Z statisticp-value (two-tailed)Panel D: number of changes and statistical tests(late adopters) bNegative (%)Positive (%)No change (%)Z statisticp-value (two-tailed)

0.0851.56125.5800.0210.0920.2319.730

20.0930.56723.34020.0130.0470.1160.320

20.0900.61223.85020.0230.0500.1030.320

0.2501.40122.0500.0300.1190.2179.730

0.6371.11426.2700.3100.6021.1282.760

0.6771.19826.2700.3770.6971.1792.670

1.0461.1790.0000.2310.7811.5107.030

256411 2.9950.003

335611 1.7360.083

246413 3.2780.001

573013 2.0640.039

207371.9770.048

276771.4750.140

276771.6010.109

632513 0.6590.510

286013 2.1970.028

355313 1.0970.272

236315 2.7680.006

553313 1.1930.233

Notes: Key ratios are dexned as follows: ROE is return on equity which equals net proxt divided by book value of equity; ROA is return on assets, which equals netproxt divided by total assets; LEV is leverage, which equals total liabilities divided by book value of equity; ATO is asset turnover, which is revenue to total assets;ROS is return on sales, which is net proxt to revenue. a ratios under OLD GAAP and NZ IFRS are reported, rather than the change to these ratios, for ease ofreference to base xgures. b The reported test statistic is a Wilcoxon test for equality of matched pairs

Table V.Descriptive statistics onkey ratios

ROEOLDGAAP

ROE NZIFRS

ROAOLDGAAP

ROA NZIFRS

LEVOLDGAAP

LEV NZIFRS

ATOOLDGAAP

ATO NZIFRS

ROSOLDGAAP

ROS NZIFRS

0.901 2198.451 2194.2180.847 1,440.356 1,440.4930.000 210,781.270 210,781.2700.196 20.020 20.0140.695 0.058 0.0621.472 0.135 0.1382.720 1.680 3.030

256692.7180.007 256962.1130.035 256510 2.4350.015

of positive sign changes are generally much higher for the large xrms than for thesmall xrms. We xnd that the impact of NZ IFRS on the small xrms is relativelyinsignixcant – only asset turnover (at 0.05) and return on sales at (0.10) showsignixcant change at conventional levels. This again contrasts strongly with thexndings for the large xrms, where the impact of NZ IFRS is found to be signixcant forall ratios except asset turnover. Return on equity and leverage are signixcant at the0.01 level, while return on assets and return on sales are signixcant at the 0.05 level forthe large xrms. These results are consistent with earlier xndings relating to xnancialstatement elements, indicating that small listed xrms are less signixcantly affected byNZ IFRS than large listed xrms. The results suggest that the impact of NZ IFRS on ratios is both extensive andcomplex. Analysts will not be able to apply a simple transformation of OLD GAAPratios to NZ IFRS. This has implications for the cost of xnancial analysis, the cost ofvaluations and the use of ratios in contracting. Furthermore, there are differencesbetween early and late adopters. This is consistent with early adopters self-selectingbased on xrm specixc characteristics and the xnancial consequences of adoption.

6. Summary and conclusionIn this paper we examine the impact of NZ IFRS on xnancial statement elements(assets, liabilities, equity, revenues/income and expenses/losses) and on key xnancialratios. The results show that NZ IFRS affects 87 per cent of entities in our sample andfor most xnancial statement elements the changes are statistically signixcant. Themedian and the inter-quartile range indicate that the impact of the move to NZ IFRS is,for most entities, very small. However, the minimum and maximum values indicatethat the impact can be material for some companies. The xnancial statement element most affected by the move to NZ IFRS is liabilities(increases for 75 per cent of companies), followed by equity (decreases for 57 per cent ofcompanies). Income taxes and employee benexts are the main reasons for the increasesin liabilities. Financial instruments are the most common reason for increases in assets(26 per cent of observations). They impact both positively and negatively on assets andliabilities. The net effects of these xnancial instruments impacts are that observationsfor equity increase twice as frequently as they decrease. The results are generallyconsistent with the expected impacts of IFRS (Ernst & Young, 2004). However, in someinstances they differ from the impact of IFRS in Germany (Hung and Subramanyam,2007). In particular, adjustments for xnancial instruments were more frequent in NZ. The move to NZ IFRS also has a considerable impact on common xnancialstatement ratios. The median for each of four ratios increases under NZ IFRS (return onequity, return on assets, leverage and return on sales) and decreases for the remainingratio investigated (asset turnover). This has implications for xnancial analysis,valuation and credit decisions and contracting agreements that employ accountingratios. Our results are important for accounting policy makers who have deferred theapplication of NZ IFRS for smaller xrms (Sealy-Fisher, 2007). They indicate that somexrms will be signixcantly affected by the adoption of NZ IFRS – the currentdifferential reporting exemption for deferred tax, for example, would therefore be amajor concession if NZ IFRS were to be adopted by smaller xrms. Our results alsoindicate that small listed xrms are less signixcantly affected by NZ IFRS than large

IFRS in New Zealand

105

PAR22,2

106

listed xrms. If non-listed xrms are similarly affected, it suggests limited benexts(relatively little change in xnancial information) as a result of smaller xrms moving toNZ IFRS. This perspective is relevant to the discussion documents recently released bythe Ministry of Economic Development (2009) and the ASRB (2009), regarding aproposed new statutory framework for xnancial reporting in New Zealand and inparticular to the Ministry of Economic Development (2009, p. 12) conclusion that“. . .the requirements to prepare xnancial statements should be removed for all but the1-2 per cent of companies that are issuers, large and/or do not have separation [ofownership and management]”. Furthermore, the impact of NZ IFRS on xnancial ratiosindicates that there is no simple transformation that will make OLD GAAP ratioscomparable with NZ IFRS ratios. This has implications for accountants, advisors,bankers and managers of xrms that are required to adopt NZ IFRS. Finally, our results show that the impact of NZ IFRS on early and late adopters isquite different. This suggests that early adopters have self-selected and that futureresearch might examine the causes and consequences of early adoption. It alsosuggests possibilities for research that could reveal insights into accounting choicesconcerning IFRS adoption and their association with opportunism (including thepossibility of earnings management) and value maximising behaviour.

Notes 1. International Accounting Standards (IAS) were signixcantly developed and renamed after 2001 to become IFRS. For convenience, we simply refer to IFRS to include both IAS and IFRS. When the context is more specixc we use IAS.

2. The FRSB develops accounting standards which are then submitted to the ASRB, a statutory body that has legal authority to review and approve the standards.

3. This statement is required in terms of NZ IFRS 1 First Time Adoption of New Zealand Equivalents to International Financial Reporting Standards.

4. It is normal to use EBIT (earnings before interest and taxation) in estimating return on assets. However, the OLD GAAP/NZ IFRS reconciliations only provide details for net proxt. Hence we are unable to estimate EBIT. Similarly, we are unable to estimate an Altman-type Z score model.

5. In the Du Pont analysis return on sales and asset turnover are components of return on assets. Return on assets and leverage are components of return on equity.

ReferencesAccounting Standards Review Board (2009), “Proposed application of accounting and assurance standards under the proposed new statutory framework for xnancial reporting: discussion document”, Accounting Standards Review Board, Wellington, September.Altman, E. (1968), “Financial ratios, discriminant analysis and the prediction of corporate bankruptcy”, Journal of Finance, Vol. 23, pp. 589-609.Barth, M.E., Landsman, W.R. and Lang, M.H. (2008), “International accounting standards and accounting quality”, Journal of Accounting Research, Vol. 46 No. 3, pp. 467-98.Berkman, H., Bradbury, M.E. and Magan, S. (1997), “An international comparison of derivative usage”, Financial Management, Winter, pp. 69-73.Bradbury, M. (1998), “Harmonising with overseas accounting standards: a New Zealand perspective”, Australian Accounting Review, November, pp. 18-23.

Bradbury, M. and van Zijl, T. (2006), “Due process and the adoption of IFRS in New Zealand”, Australian Accounting Review, Vol. 16 No. 39, pp. 86-94.Christensen, H.B., Lee, E. and Walker, M. (2007), “Do IFRS/UK-GAAP reconciliations convey new information?”, paper presented at the SSRN eLibrary, available at: http://ssrn.com/ paper¼997800Daske, H., Hail, L., Leuz, C. and Verdi, R. (2008), “Mandatory IFRS reporting around the world: early evidence on the economic consequences”, Journal of Accounting Research, Vol. 46 No. 5, pp. 1085-142.Dunstan, K.L. (2002), “The case for the use of international xnancial reporting standards in New Zealand”, a briexng paper prepared on behalf of the New Zealand Securities Commission, Centre for Accounting Governance and Taxation Research School of Accounting and Commercial Law Victoria University of Wellington, Wellington.Ernst & Young (2004), IFRS Impact Report, Ernst &Young, Southampton.Fields, T.D., Lys, T.Z. and Vincent, L. (2001), “Empirical research on accounting choice”, Journal of Accounting and Economics, Vol. 31 Nos 1-3, pp. 255-307.Hung, M. and Subramanyam, K.R. (2007), “Financial statement effects of adopting international accounting standards: the case of Germany”, Review of Accounting Studies, Vol. 12 No. 4, pp. 623-57.Lee, E., Walker, M. and Christensen, H.B. (2008), “Mandating IFRS: its impact on the cost of equity capital in Europe”, No. RR105, ACCA, Manchester.Ministry of Economic Development (2009), “The statutory framework for xnancial reporting” discussion document, Wellington, September.Sealy-Fisher, V. (2007), “FRSB developments”, Chartered Accountants Journal of New Zealand, Vol. 86 No. 9, pp. 45-6.

Corresponding authorWarwick Stent can be contacted at: w.j.stent@massey.ac.nz

IFRS in New Zealand

107

To purchase reprints of this article please e-mail: reprints@emeraldinsight.comOr visit our web site for further details: www.emeraldinsight.com/reprints