En komparativ analyse af IAS 39 & IFRS 9: Med fokus på indregning og måling - tab på udlån

Emil Linde & Jesper Møller Pedersen

Student thesis: Diploma thesis


Since the original version of International Accounting Standard 39 (IAS 39) was introduced in 1999, it has been subject to criticism from politicians as well as accounting professionals in both Denmark and internationally. The criticism as regards the accounting for impairment of loans and other financial assets has accelerated since the commencement of the credit crisis. The criticism has centered around the current impairment model for financial assets. While taking into account incurred losses, the model does not reflect expected future credit losses. Critics have argued that the impairment is recognised too late with the incurred loss model in comparison with the expected loss model. This has led The International Accounting Board (IASB) to initiate the implementation of a new standard, International Financial Reporting Standard 9 (IFRS 9), setting out to formulate a simpler standard that is easier applied by reporting companies, besides changing the impairment methodology. In our study, we have carried out a comparative analysis of IAS 39 and IFRS 9, placing key emphasis on the material changes in the impairment methodology. The most fundamental changes are “the two measurement approach” and “the expected loss model”. The two measurement approach allows only two different measurements, fair value and amortised cost – which is a significant shift in comparison to IAS 39 where four different categories of initial recognition and measurement are applied. The aim of this simplification is to lessen recognition doubts as fair value is to be used unless an asset has loan characteristics. The expected loss model also represents a significant change of practice and will lead to earlier recognition of impairment losses. The credit premium for riskier loans will no longer be accounted for as income before it is evident that no losses will occur (the repayment of the loan). These changes have attracted massive feedback from all over the world. The views on the exposure drafts differ widely among the respondents. However, some of the proposals have received predominantly negative feedback, leading to certain ideas being withdrawn, including the proposal that would prohibit reclassification between fair value and amortised cost after initial recognition. In analysing IAS 39 and IFRS 9, this thesis focuses on recognition and measurement of impairment losses on loans and the impact on the financial reporting of the banking sector. To be able to perform precise analysis, we have chosen to focus on group 1 banks as defined by the Danish Financial Supervisory Authority. This group of banks accounts for a major share of total loans in Denmark and thus offers an interesting analytical case. We have carried out detailed analysis of FIH Bank’s financial statements for 2009. The profit and loss account as well as the balance sheet have been analysed in detail and the effects of conversion from IAS 39 to IFRS 9 have been estimated. These estimates are based on data from the 2009 risk report as well as interviews with key persons. We conclude that conversion will not lead to significant changes in classifications and initial recognition of assets as loans are already largely recognised at amortised cost. However, the implementing phase will involve a sizeable increase in impairment losses as both actually incurred losses and expected losses are being provided for. This effect will gradually diminish until an appropriate reserve has been built up. Once IFRS 9 is fully implemented, the level of impairment losses will be the same as with IAS 39 seen over an economic cycle. However, impairment losses will be less volatile from year to year using the IFRS 9 rules. Furthermore, it is our view that IFRS 9 and the rules described in the exposure draft of November 2009 are easier to apply than their predecessor.

EducationsGraduate Diploma in Financial and Management Accounting, (Diploma Programme) Final Thesis
Publication date2010
Number of pages109