6. Use of discretionary decisions and estimates

The consolidated financial statements require the Group Management Board to make discretionary decisions, estimates and assumptions that relate to the application of accounting policies and the amounts stated for the assets, liabilities, income and expenses. Actual results may differ from these estimates. Estimates and their underlying assumptions are reviewed on an ongoing basis. Revisions to estimates are recorded prospectively.

The most significant instances where discretion has been exercised and forecasts for the future have been used for these IFRS-consolidated financial statements are described below:

6.1 Impairment test

Ascertainment and allocation of goodwill

Goodwill arises from company mergers and acquisitions. It represents the difference between the acquisition costs and the proportional and current corresponding net fair value of identifiable assets, debts and specific contingent liabilities. In accordance with IAS 36, goodwill is not subject to amortisation, but reported at the acquisition cost less any accrued impairments.

For the purpose of the impairment test, UNIQA has allocated the goodwill to “cash-generating units” (CGUs). These CGUs are the smallest identifiable groups of assets that generate cash flows that are to the greatest possible extent independent from the cash generating units of other assets or other groups of assets. The impairment test implies a comparison between the amount that can be generated by selling or using each CGU, the present value of future cash flows, and the value to be covered, consisting of goodwill, the proportional net assets and any capital increases and internal loans. If the resulting value exceeds the realisable value of the unit based on the discounted cash flow method, an impairment loss is recognised. The impairment test was carried out in the fourth quarter of 2015. UNIQA has allocated goodwill to the CGUs listed below, which coincide with the countries in which UNIQA operates. As an exception to this, the Austrian companies were considered individually, and Salzburger-Landesversicherung AG and UNIQA Österreich Versicherungen AG were considered as a group, as was the Sigal Group, in which the three countries of Albania, Kosovo and Macedonia were combined as one CGU, due to their similar development and organisational connection:

• Albania/Kosovo/Macedonia as “Sigal Group” sub-group (SEE)
• Bosnia and Herzegovina (SEE)
• Bulgaria (SEE)
• Italy (WE)
• Croatia (SEE)
• Liechtenstein (WE)
• UNIQA Austria (AT)
• Raiffeisen Insurance Austria (AT)
• FinanceLife (AT)
• Poland (CE)
• Romania (EE)
• Russia (RU)
• Switzerland (WE)
• Serbia (SEE)
• Montenegro (SEE)
• Slovakia (CE)
• Czech Republic (CE)
• Ukraine (EE)
• Hungary (CE)
• UNIQA Re

Determination of the capitalisation interest rate

The assumptions with regard to risk-free interest rate, market risk premium and segment betas made for determination of the capitalisation rate are consistent with the parameters used in the UNIQA planning and controlling process and are based on the capital asset pricing model.

In order to depict the economic situation of income values as accurately as possible, considering the volatility on the markets, the capitalisation rate was calculated as follows: a uniform, risk-free interest rate according to the Svensson method (German treasury bonds with 30 year maturities) was used as a base interest rate.

The beta factor was determined on the basis of the monthly betas over the last five years for a defined peer group. The betas for the non-life, life and health segments were determined using the revenues in the relevant segments of the individual peer group companies. The health insurance segment, which is strongly focused on the Austrian market, is operated in a manner similar to life insurance. A uniform beta factor for personal injury insurance is therefore used in relation to the life and health insurance lines.

The market risk premium was determined on the basis of the current standards issued by the Austrian Chamber of Public Accountants and Tax Advisors (Kammer der Wirtschaftstreuhänder). The calculations published by Damodaran were used to determine the country risk premium. The country risk premium in accordance with the Damodaran method is calculated as follows: Starting from the rating of the country concerned (from Moody’s), UNIQA obtains the yield spread from credit default swap spreads with the same rating as risk-free US government bonds, and adjusts this spread for the volatility difference between equity and bond markets. UNIQA also assumes that country risk will decline over the next few years on the basis of subsequent trends.

The calculation also factored in the inflation differential for countries outside the eurozone. In general, the inflation differential represents inflation trends in different countries and is used as a key indicator in assessing competitiveness. In order to calculate the inflation differential, the deviation of the inflation forecast for the country of the CGU in question in relation to the inflation forecast for a risk-free environment (Germany, in this case) was used. This is adjusted annually in the detailed planning by the expected inflation, and is subsequently applied for perpetuity with the value of the last year of the detailed planning phase.

Impairment test for goodwill – ascertainment of the recoverable amount

UNIQA calculates the recoverable amount of the CGUs with goodwill allocated on the basis of value in use by applying generally accepted valuation principles by means of the discounted cash flow method (DCF). The budget projections (detailed planning phase) of the CGUs, the estimate of the long-term net profits achievable by the CGUs and long-term growth rates (perpetuity) are used as the starting point for determination of the capitalised value.

The earning power is determined by discounting the future profits with a suitable capitalisation interest rate after assumed retention to strengthen the capital base. In the process, the earning power values are separated by balance sheet segments, which are then totalled to yield the value for the entire company.

Taxes on operating income were set at the average effective tax rate of the past three years.

Cash flow forecast (multi-phase model)

Phase 1: Five-year company planning

The detailed company planning generally encompasses a period of five years. The company plans used for the calculation are the result of a structured and standardised management dialogue at UNIQA with the participation of UNIQA International, in combination with the reporting and documentation processes that are integrated into this dialogue. The plans are formally approved by the Group Management Board and also include material assumptions regarding the combined ratio, investment income, market shares and the like.

Phase 2: Perpetuity growth rate

The last year of the detailed planning phase is used as the basis for determining the cash flows in phase 2. The growth in the start-up phase leading up to phase two was determined using a projection of the growth in insurance markets. It was assumed that the insurance markets would come into line with the Austrian level in terms of density and penetration in 40 to 60 years.

The capitalisation rate for all CGUs is listed below:

The following discount rates were applied in 2014:

Uncertainty and sensitivity

Various studies and statistical analyses were used as sources to provide a basis for determining the growth rates in order to consistently and realistically reflect the market situation and macroeconomic development.

The reference sources included the following studies and materials:

• Internal research
• Damodaran – country risks, growth rate estimations, multiples

Sensitivity analyses of financial instruments

In order to substantiate the results of the calculation and estimation of the value in use, random sensitivity analyses with regard to the capitalisation rate and the main value drivers are performed.

These analyses show that sustained surpluses on the part of the individual CGUs are highly dependent on the actual development of these assumptions within the individual national or regional economies (GDP, insurance density, purchasing power parities), particularly in the CEE markets, as well as the associated implementation of the individual profit goals. These forecasts and the related assessment of how the situation in the markets will develop in the future, under the influence of the continuing financial crisis in individual markets, are the largest uncertainties in connection with measurement results.

In the event that the recovery from the economic crisis turns out to be much weaker and slower than assumed in the business plans and fundamental forecasts, and the insurance markets therefore develop completely differently from the assumptions made in those business plans and forecasts, the individual CGUs may be subject to unscheduled impairment losses. Despite slower economic growth, income expectations have not changed significantly compared to previous years.

A sensitivity analysis shows that if there is a rise in interest rates of 50 basis points in the countries of Romania and Hungary, there could be a convergence between the value in use and the carrying amount or a value in use that is lower than the carrying amount. If there were a stronger rise in interest rates of 100 basis points or more, Bosnia-Herzegovina, Italy, Serbia and Slovakia would also be affected. If the underlying cash flows change by –5.0 per cent, there will also be a risk of a convergence or a value in use that is lower than the carrying amount in Hungary. This list expands to include Bosnia-Herzegovina, Italy, Romania and Slovakia when there is a deviation of more than –10.0 per cent in the cash flows.

In 2015, due to exchange rate effects and necessary adjustments of the discount rate as a result of the changes in the economic environment, impairment losses were recognised in the amount of € 13.1 million for Ukraine.

The following table shows the recoverable amounts at the time of the impairment test for all CGUs with the necessary goodwill.

Backtesting

Backtesting is regularly carried out on the planning for the individual countries. The objective is to obtain information for internal purposes on the extent to which the operating units plan their results accurately and on the extent to which details useful with regard to subsequent development are highlighted. Backtesting is intended to help draw conclusions that can be applied to the latest round of planning, in order to enhance the planning accuracy of forthcoming financial plans.

6.2 Investment property

The fair value of investment property within the scope of the impairment test in accordance with IAS 36, as well as for the disclosures according to IFRS 13, is determined by using a report prepared by independent experts. These expert reports are prepared based on earned value and asset value methods. This is typically done using the DCF method by discounting expected future payments from the relevant land and buildings. It requires making assumptions, principally concerning the discount rate, the exit yield, the expected utilisation (vacancy rate), the development of future rental charges, and the condition of the land and buildings. For this reason, all measurements of the fair value for the land and buildings come under level 3 of the hierarchy according to IFRS 13. The nature of the measurement procedures stated above is that they respond sensitively to the underlying assumptions and parameters. For instance, any reduction in the discount rate applied would result in an increase in the values ascertained for the land and buildings if the other assumptions and parameters remained unchanged. Conversely, any reduction in the expected utilisation or the expected rental charges would, for instance, result in a decrease in the values ascertained for the land and buildings if the other assumptions and parameters remained unchanged. The measurement-related assumptions and parameters are ascertained carefully at each key date based on the best estimate by management and/or the experts in view of the current prevailing market conditions. These estimates are updated at each key measurement date.

6.3 Deferred tax assets

As at 31 December 2015 UNIQA had deferred tax assets amounting to € 9.4 million (netted out), of which € 11.7 million were attributable to tax loss carryforwards. The deferred tax assets result from tax loss carryforwards, impairment in accordance with section 12 of the Corporation Tax Act (KStG), and from deductible temporary differences between the carrying amounts of the assets and liabilities in the consolidated statement of financial position and their tax values. Deferred tax assets are accounted for, provided that it is likely that there will be adequate taxable profit for them to be realised. An assessment of the ability to realise deferred tax assets requires an estimate of the amount of future taxable profits. The amount and type of these taxable earnings, the periods in which they are incurred, and the available tax planning measures are all taken into account. The corresponding analyses and forecasts, and ultimately the determination of the deferred tax assets, are carried out by the local tax and finance experts in the relevant countries. Because the effects of the underlying estimates may be significant, there are internal group procedures that guarantee the consistency and reliability of the evaluation process. The resulting forecasts are based on business plans that are reviewed and approved through proper procedures. Especially convincing evidence regarding accounting for deferred tax assets is required under internal group policies if the relevant group company has suffered a loss in the current or a prior period.

6.4 Provisions for defined benefit obligations

For this purpose, reference is made to the statements and sensitivity analyses in the Notes to this balance sheet item under Note 28.

6.5. Technical provisions

Reference is made here to the statements and sensitivity analyses under Note 7.5.

6.6 Investments

Reference is made here to the statements and sensitivity analyses under Notes 7.5 and 13.

6.7 Other discretionary judgements and assumptions regarding the future

The investment in STRABAG SE, the portfolio of asset-backed securities (ABS), and the assets related to Hypo Group Alpe Adria (HGAA) and HETA Asset Resolution AG (HETA) continue to be carefully monitored.

As at 31 December 2015, UNIQA held a 13.8 per cent stake in STRABAG SE (31 December 2014: 13.8 per cent). UNIQA is continuing to treat STRABAG SE as an associate due to contractual arrangements. The carrying amount of the investment in STRABAG SE at 31 December 2015 amounted to € 463.0 million (31 December 2014: € 456.5 million).

UNIQA held 0.7 per cent (2014: 1.8 per cent) of its investments in ABS. There were several developments in 2015 that resulted from the continued improvement in the liquidity situation of ABS, which influenced UNIQA’s decision to depart from its forecast of expected losses based on internal valuations developed during crisis periods. The switch to a market data-based measurement of the ABS portfolio resulted in a reclassification of the ABS portfolio from category two to category three. This resulted in an increase in market value of € 5.3 million.

The Act to restructure the state-owned Hypo Alpe Adria International AG (HaaSanG), which took effect on 1 August 2014, and which resulted in the cancellation of the subordinate liabilities of Hypo Alpe Adria International AG (HAA) without consideration of existing guarantees of the federal state of Carinthia, resulted in a complete impairment for the bonds held by UNIQA in 2014, with a nominal value of € 36 million and a loss of € 34.1 million. The legal steps taken by UNIQA and other creditors against these statutory measures culminated in a decision by the Austrian Constitutional Court on 3 July 2015, G239/2014 inter alia, whereby HaaSanG, along with the order by the Austrian Financial Market Authority (FMA) regarding the implementation of restructuring measures in accordance with section 7(2) HaaSanG in conjunction with sections 3 and 4(1) HaaSanG (HaaSanV), were lifted in their entirety. As a consequence, the subordinate bonds with a nominal value of € 36 million that had been derecognised in 2014 were included on the balance sheet once again as at 30 September 2015. The value of these bonds on the balance sheet as at the reporting date of 31 December 2015 was € 6.8 million.

On 1 March 2015 the FMA passed a decision on a debt moratorium for Heta Asset Resolution AG (HETA) with immediate effect until 31 May 2016, whereby no interest payments or repayments may be made inter alia on bond liabilities, subordinated capital or bonded loans. Aside from the subordinated liabilities, senior bonds guaranteed by the federal state of Carinthia and held by UNIQA with a nominal value of € 25.0 million are affected by this measure. The value of these bonds as at the reporting date amounts to € 17.1 million.

A subordinated liability of HETA collateralised by the Republic of Austria with a balance sheet value of € 10.6 million (nominal value: € 10.0 million) is the only one not affected by the statutory and supervisory measures.