Vienna Senate Branch Office (Vienna) 4

GZ. RV/2515-W/09,

co-discharged RV/2516-W/09

Appeal decision

The Independent Tax Tribunal decided on the appeal of W, x, represented by Freshfields Bruckhaus Deringer LLP 1010 Vienna, Seilergasse 16 of 3 October 2008 against the notices of the tax office Vienna 1/23 of 1 and 3 September 2008 regarding corporation tax for the years 2001 to 2005 as well as liability for capital gains tax pursuant to § 95 para. 2 of the Income Tax Act 1988

for the period 2001 to 2004 after the appeal hearing held on 24 July 2012 at Vordere Zollamtsstraße 7, 1030 Vienna:

The appeal against the assessments concerning corporate income tax for the years 2001, 2002 and 2003 is dismissed as unfounded.

The assessments are amended.

The appeal against the assessments of corporate income tax for the years 2004 and 2005 is upheld in part.

The assessments are amended.

The appeal against the assessments concerning liability for capital gains tax for the years 2001 and 2003 is dismissed as unfounded.

The notices are amended.

The appeal against the assessments concerning liability for capital gains tax for the years 2002 and 2004 is upheld.

The assessments are annulled.

The bases of assessment and the amount of the levies can be found at the end of the following reasons for the decision and form an integral part of this decision.

The due date of the additional amount of the levies determined by this decision can be seen from the accounting notice.

Reasons for decision

The W (distribution subsidiary of the Wx-group i.d.F. Bw.) is responsible for the group distribution of y in Austria. It is wholly owned by Z.

In the course of a comprehensive tax authority audit of the years 2001 to 2005, the intra-group transfer prices were re-determined for the years 2001 to 2004 due to a lack of arm's length. The adjustments to the results were treated as hidden distributions to the parent company pursuant to § 8 (2) KStG 1988.

The competent tax office, Vienna 1/23, followed the comments of the external audit (in the version of Ap.) and in the resumed proceedings issued new tax assessments regarding corporation tax for 2001 to 2005 and liability assessments regarding capital gains tax for the years 2001 to 2004.

In the report under point 4.3 (hidden distribution - transfer prices), the ap. stated that the group allocations had not been booked on an accrual basis and that the amount of the calculated group allocations had not been comprehensible.

The I/C sales had been invoiced without a profit mark-up at cost price plus 1.5% to 1.8% expenses. This procedure did not comply with the arm's length principle. With regard to the transfer prices, the complainant had submitted a study from 2000 according to which the operating margin median was 0.9%. The review had revealed a median of 1.53%. The ap. set the operating margin for 2001 at 1.5% and for the following years at 0.9% due to changed functions (outsourcing of accounts receivable, closure of half the IT department) for the years 2002 to 2004.

The applicant appealed against the above decisions in a letter dated 3 October 2008 and a supplement dated 2 December 2008.

It was responsible for the sale of group products of the Wx Group in Austria as a sales unit and was supplied by production companies at manufacturing costs (SIC standard inventory cost), which were determined annually in advance for all products and included the budgeted fixed and variable production costs.

On the basis of distribution agreements with the production companies, the applicant received a remuneration for distribution in the amount of 1% of net sales as well as interest on the capital employed.

The remuneration was defined as earnings before taxes (EBT or profit from ordinary activities). The target margins during the audit period would be between 1.20% and 1.58% of the annual net turnover.

If this margin was not achieved, the production units would be credited ('Purchase Price Adjustment' i.e. PPA). The amount of the credit is limited to 10% of the production costs. This corresponds to the total fixed costs of the production companies. The final transfer price could therefore not fall below 90% of the production costs.

Thus, the production companies would share the losses of the applicant on a pro rata basis.

If the profit was above the calculated target margin, this would remain entirely with the applicant.

Opportunities and risks from the business relationship would thus not be distributed to the disadvantage of the distribution company.

Ap. considered a fixed result (Earning before Interest and Taxes i.e. EBIT or operating result) of 1.5% (2001) or 0.9% (2002-2004) of net turnover to be appropriate and had adjusted the tax result accordingly.

The margins of 1.5% and 0.9% had been derived from a superficial database search, which the applicant had carried out to check the plausibility of a database analysis submitted by the applicant and which was formally incorrect, as it ignored the applicant's economic environment as well as the transfer pricing study submitted.

Contrary to what was stated in the report, the complainant had submitted detailed information on intra-group transactions.

In order to prove the arm's length nature of the transfer prices, a transfer pricing study prepared by Ernst&Young in November 2000 had been submitted with a description of the system and the methodology as well as a database analysis.

The transfer pricing system as well as the methodology for determining the transfer prices had been presented.

Transfer prices generally had to be determined in advance. In the case of the applicant, this had been the financial data for the years 1996-1998, which had been available on the occasion of the database analysis (version 10/2000).

Companies with a function and risk profile comparable to the applicant's had been sought and used to derive the range of profit margins customary at arm's length. The 7 companies included in the analysis had produced an EBIT margin with a median of 0.9%, a minimum value of -2.5% and a maximum value of 7.1%.

The EBT target margins set for the Bw. would range between 1.20% and 1.58% of net turnover. The EBIT derived from the EBT was therefore within the range of the comparable companies and the arm's length principle was thus respected.

The Wx Group had made high losses in Europe, especially at the beginning of the audit period, but also on average over the years audited. The group had struggled with very low profitability due to declining sales figures, which had prevented the complainant from achieving the set targets. The production units had therefore had to issue the applicant with additional credits, some of which were considerable.

Despite these credits, the applicant had achieved negative operating results in three out of five years, with the production units having to accept a loss from product sales of 10% of production costs at the same time.

The complainant subsequently presented the amount of the target EBT as well as the credits received by it per year and substantiated these numerically on the basis of related enclosures. The complainant also explained that the ap had also criticised the non-accrual accounting of PPAs in the audit period. An evaluation by the complainant had shown that the tax result in the audit period would have been reduced by a total of € 731,341.48 if the PPAs had been booked on an accrual basis.

The tax authorities recognise losses at distribution companies if the distributor accepts them for a reasonable period of time in order not to jeopardise long-standing business relationships and to maintain its own source of supply.

In the case of insolvency of its suppliers, a distribution company would risk the loss of its business basis. A termination of the business relationship with its supplier would lead to high costs, as fixed costs would continue to accrue over a certain period of time.

On the other hand, a production company would have no incentive to produce unless at least the variable costs were covered.

It would be in line with the arm's length principle to set transfer prices taking into account the production and distribution company.

The transfer pricing system of the Wx Group had already been subjected to an external audit in several European countries and had so far always been accepted.

The European market was dominated by the 4 largest market participants with a total share of 60%. In addition, new small companies and groups from the Far East were entering the market. The period under review was characterised by an economic downturn in the market, with margins also falling due to rising material costs.

The range of EBIT margins of the peer companies underpins the development. The range also included companies that had negative margins on average. The negative margins are representative and prove the generally low profitability of the industry. Although none of the comparable companies had losses in all years, on average negative margins were achieved.

The profitability of the Wx Group was behind that of its main competitors (B).

The profitability of the Wx Group in Europe was characterised by high losses (EBT according to US GAAP) in 2001 and 2002. The earnings situation had improved in the following years and had fallen again in 2005. On average, the Wx Group in Europe had generated an overall loss. During the audit period, not only the target margins of the applicant were within the range of the comparable companies, but also the EBIT margins actually achieved in four out of five years (between -0.69% and 2.17%).

The authority had carried out a plausibility check although it had been provided with the possibly unavailable data of the Amadeus database from October 2006.

According to the report, Ap. had identified 53 comparable companies with a median EBIT margin of 1.53%.

The tax authorities' database search was unsuitable and had to be rejected from a formal point of view.

This was because the database search could not ensure comparability, as the Ap. had used an exclusively quantitative screening.

A qualitative search in which the results of the search steps were checked by researching websites or by telephone calls was indispensable, which had also been done in the case of the transfer pricing study by Ernst&Young. Furthermore, in order to ensure the independence of possible comparable companies, only those whose shareholders held less than 25% of the shares could be used. The Ap. assumed an independence criterion of 50%, which was not sufficient to exclude group companies.

Finally, the database study submitted (version 10/2000) used financial data for the years 1996 to 1998 for comparison. It had not been disclosed to the complainant on the basis of which years the external audit had determined the median of 1.53%. It should be noted, however, that in order to determine the price, it was necessary to refer to data that was already available prior to the priced transaction. The use of more recent comparative data could not be used to verify a database search based on older data. Neither the exact search strategy nor the selected comparable companies nor the calculation of the margin of 1.53% had been disclosed by the tax authorities.

Ap. commented on this in a letter dated 13 May 2009.

Not only the group-internal transfer prices were not at arm's length, but the entire group accounting including the group allocations and the internal deliveries of goods.

The group allocations were not comprehensible and on an accrual basis, and the intra-group deliveries of goods had been booked without a profit mark-up.

In line with the preliminary audit, Ap. had assumed a return on sales of 1.5% as being in conformity with external conduct and had based its calculation on the year 2001. For the years 2002 to 2005, the EBIT margin had been set at 0.9% due to reduced functions of the applicant. These percentages had been supported by an evaluation of the transfer pricing study.

An exact verification of the data submitted had not been possible, as old periods were no longer available in the database. Ap. had extended the comparison period by 1 year to 1999 (instead of 1996 to 1998) which had led to a range of 0.24% to 2.65% and a median of 0.98%.

A qualitative examination of the comparative enterprises used in the Ernst & Young study had shown that 3 comparative enterprises (C., V and D. respectively) had been deficient and could therefore not be used. Taking this result into account, the range was corrected to 0.84% to 3.28% with a median of 1.49%.

The ap. had carried out a comparison with purely quantitative screening in order to check this result.

On the basis of a sample size of 53 comparable enterprises for the period 1997 to 2005, a range of 0.79% to 3.10% and a median of 1.53% had resulted.

All deficiencies cited by the complainant were to be rejected as irrelevant.

The bandwidths cited by the complainant were not correct; rather, the values of -2.50% and 7.10% presented by the complainant were maximum and minimum values.

In the years 2001 to 2004, the company had consistently generated negative results from ordinary business activities. Even after correction of non-recognisable expenses from previous periods, there was still an overall loss.

A comprehensible calculation of the group apportionments had been requested several times, but had not been submitted by the time of the audit.

The applicant explained that as a distribution company it would accept losses in order not to jeopardise business relations and to maintain a source of supply. It would rather forego a profit in order to secure its own future business. From Ap.'s point of view, a positive functional benefit was to be attributed to Bw. It would not accept purchase prices if it could probably only sell the goods in question at a loss, as this would ultimately jeopardise its stock.

The requested adjustment of profit in 2005 in the amount of € - 134,121.54 in order to establish the period purity of the booked expenses was to be granted.

In a counterstatement of 2 July 2009, the complainant explained that the application of the Wx Austria transfer pricing guideline had been accepted by a large number of tax authorities.

The authority had failed to eliminate ambiguities by asking specific questions. It had also not in any way assessed a further transfer pricing study by Baker&McKenzie from 2005 submitted by the complainant.

The complainant did not regard the complainant as a distribution company with limited risk, but rather as a risk-free distribution company without the possibility of losses.

Considerable overall losses of Wx Europa in the audit period had been ignored, as had an increase in the applicant's equity capital.

While the ap. regarded intra-group deliveries of goods without a profit mark-up as not in conformity with arm's length conduct, it assumed on the other hand that group suppliers should not have any loss limits, which would lead to results that were not in conformity with arm's length conduct. In the audit period 2001 to 2005, Wx Europa had recorded a total loss of approximately € 82 million. It was not possible to grant a distribution company a permanent and unlimited profit guarantee.

In 2001, the result of a legally inadmissible manipulation of results from previous years had been corrected. There was a justified assumption that the disastrous result in that year was due to mismanagement. The applicant had received an equity injection of € 10 million in 2001, which had led to additional remuneration in the course of the PPA process. The ap. had accepted the remuneration of the risk capital, but not that the distribution company could not make losses, which was a contradictory argumentation.

The tax authorities had carried out audits in many European countries and, with the exception of Italy, the transfer pricing guidelines had been accepted in their entirety.

Possible mutual agreement procedures with several countries would lead to the repeated opening of years that had already been audited in detail and thus to lengthy and difficult processes.

Ap. had excluded 3 companies from the transfer pricing study prepared by Ernst&Young.

As far as C, one of the comparable companies, was concerned, the Ap.'s presentation could be followed. The concerns expressed about the other two companies that had been eliminated were not correct. The size of a company was not a relevant factor, the type of products sold was irrelevant for the TNNM (transaction-oriented net margins) method and the shareholders of V (another comparable company according to the study) were visible in the Amadeus database.

The complainant had not been able to reconstruct the figures determined by Ap. in its comparative study, as no data had been made available.

The statement that Ap. had not been given any documents concerning group apportionments was not correct. The latter had received a 'GSA-booklet 2005' on 30 July 2007 and documentation concerning the years 2001 to 2004 on 6 December 2007.

In a reminder dated 10 February 2012, the UFS explained to the complainant that two comparable companies (D, V) in the Ernst & Young transfer pricing study had not been taken into account due to unknown shareholdings.

The qualitative comparative study of the tax authorities was explained and the determined target EBT was subjected to a more detailed assessment. The economic situation in the audit period and the previous years showed that losses had been incurred in 3 out of 5 years and that the accumulated result from ordinary business activities had been negative in the period 1997 to 2009. With regard to the attribution of expenses from previous periods carried out by the ap., a difference to the presentation according to the 2001 annual financial statement was pointed out. In its submission of 16 April 2012, the complainant submitted documents on the two comparable companies that had been eliminated by Ap., which showed that, due to the proven shareholdings, it could not be assumed that Ap. had a controlling influence and was therefore an independent company.

With regard to the EB(I)T objective, it was explained that the margins determined by the applicant were confirmed by the transfer pricing study. The distribution agreement that had existed in the group since 1999 had been amended in 2008. Credible restructuring measures (outsourcing of accounts receivable accounting...) had a relevant influence on the EBIT margin from 2001 onwards, which was why it had been reduced.

The long-term losses of the applicant could be explained by the 90% limit on marginal costs. The Wx Group had high losses and lower profitability on average in the years audited.

Regarding the difference between the 'expenses from previous periods' of € 5,095,850.09 according to the 2001 annual accounts and the amount of € 4,552,695.09, i.e. € 543,155.00, attributed by the Bp. it was explained that this concerned the write-off of fictitious receivables in connection with a fictitious increase in turnover in 2001.

In a statement of 18 May 2005, the representative of the tax office explained that the concerns expressed against the inclusion of V and D could be dropped. The reason was that at the time of the research in March 2008, no 'shareholder' information had been available. On the basis of the documents submitted, both companies could be left in the sample.

In response to a letter from the UFS on the protection of the right to be heard, the complainant submitted on 22 June 2012 that the transfer pricing study by Ernst & Young had been prepared on the basis of data for the years 1996 to 1998, as results for 1999 were not yet available. The inclusion of the year 1999 in the calculation (by the tax office) led to incorrect results, as the complainant had not been able to use them at that time.

Furthermore, in the opinion of the complainant, no correction had to be made to the median, but rather to the lowest value of the determined range. Both Article 9 of the DTA and the OECD Transfer Pricing Principles 2010 (as amended by the OECD TPP) were relevant in the interpretation of the arm's length principle under section 6(6) of the Income Tax Act 1988. A range of arm's length prices resulted from the OECD TPPs, which could be explained not only by functions and risks, but also by different corporate strategies or objectives. If a range of comparative prices is set, then in principle any price within the range must be recognised. If the initially fixed value was outside the range, a mandatory adjustment to the median was not justified, as there was no objective justification for differentiating according to whether the original price was within or outside the range.

The application of the median was only permissible if the concrete circumstances showed that this mean value could be derived from the arm's length comparison. The tax office had at no time presented concrete circumstances that would show that only the median was customary in an arm's length comparison.

The tax office had narrowed the range by using interquartiles, which Ap. regarded as a possibility to eliminate possible deficiencies, implicitly confirming the arm's length nature of all interquartile values.

In the appeal hearing held on 29 May 2012, it was additionally stated that the non-recognition of the group levy was unfounded and unlawful from the applicant's point of view. The applicant was not a risk-free distribution company, which implied that it was also allowed to make losses. The tax office had failed to reach agreement with the relevant foreign authorities, which was why, in the event of non-recognition of the losses in Austria, costly and possibly lengthy proceedings would have been necessary.

and possibly lengthy mutual agreement procedures. Agreement had largely been reached on the application of the target margin discussed in the proceedings (note: regarding the Ernst & Young study). Only the inclusion of the year 1999 was questionable. An adjustment had to be made in eventu to the lowest value of the range.

The appeal was considered:

§ 6 Z 6 EStG 1988 in the version until 30.12.2004 reads:

§ 6. The following shall apply to the valuation of individual business assets:

... 6.

If assets of a business (permanent establishment) situated in Austria are transferred abroad to another business (permanent establishment), the assets transferred abroad shall be valued at the values that would have been valued in the case of a supply or other service to a business that is completely independent of the taxpayer, if

- the foreign business is owned by the same taxable person,

- the taxpayer is a co-entrepreneur of the foreign business,

- the taxpayer has a substantial shareholding, i.e. more than 25%, in the foreign corporation, or

- the same persons exercise the management or control of both establishments or have an influence on them.

This also applies if a business (permanent establishment) located in Austria is transferred abroad, and if assets or businesses (permanent establishments) are transferred or relocated from abroad to Austria.

§ 6 Z 6 EStG 1988 as amended from 31 December 2004 reads:

... 6.

a) If assets of a business (permanent establishment) located in Austria are transferred abroad to another business (permanent establishment) or if businesses (permanent establishments) located in Austria are relocated abroad, the assets transferred abroad shall be assessed at the values that would have been assessed in the case of a supply to a business completely independent of the taxpayer if

- the foreign business is owned by the same taxable person,

- the taxpayer is a co-entrepreneur of the foreign and/or the domestic business,

- the taxable person has a substantial shareholding, i.e. more than 25%, in the foreign corporation or the foreign corporation has a substantial shareholding in the taxable person, or

- the same persons exercise management or control or have influence over both businesses.

This shall apply mutatis mutandis to other services.

b) Upon application, the tax liability incurred shall not be assessed until the actual sale or other withdrawal of the assets from the business assets in the following cases under a):

1. in the case of transfer of assets within a business of the same taxable person; or

2. in the case of transfer of businesses or permanent establishments, provided that in both cases the transfer or relocation - takes place to a State of the European Union or - to a State of the European Economic Area with which there is comprehensive administrative and enforcement assistance with the Republic of Austria.

The unassessed tax debt shall be decided in the tax assessment notice. A later transfer or relocation

- to a state which is not a member of the European Union, or

- to a state of the European Economic Area with which there is no comprehensive administrative and enforcement assistance with the Republic of Austria shall be deemed to be a sale. The sale or other withdrawal of the assets from the business assets shall be deemed to be a retroactive event within the meaning of section 295a of the Federal Tax Code. Decreases in value occurring between the transfer or relocation and the sale or other withdrawal shall be taken into account at most to the extent of the basis of assessment at the time of the transfer or relocation. § Section 205 of the Federal Tax Code shall not apply.

c) If, within the meaning of lit. a, assets or businesses (permanent establishments) are transferred or relocated from abroad to Austria, the values shall be assessed which would have been assessed in the case of a supply to a business completely independent of the taxpayer. This shall apply mutatis mutandis to other supplies. If, in cases of non-assessed tax liability within the meaning of lit. b or due to a reorganisation within the meaning of the Reorganisation Tax Act, assets are repatriated from abroad to Austria or businesses (permanent establishments) are relocated, the book values before repatriation or relocation shall be decisive.

transfer are decisive. The subsequent sale or other withdrawal shall not be deemed a retroactive event within the meaning of section 295a of the Federal Tax Code. If the taxpayer proves that increases in value have occurred in the EU/EEA area, these shall be deducted from the proceeds of the sale.

§ Section 8 (1)+(2) KStG 1988 reads:

(1) In determining income, contributions and contributions of any kind shall be excluded to the extent that they are made by persons in their capacity as partners, members or in a similar capacity.

(2) For the purpose of determining income, it shall be irrelevant whether the income is earned

- distributed or withdrawn by way of open or hidden distributions, or

- withdrawn or

- used in any other way.

§ Section 93(1) and (2)(1)(a) of the Income Tax Act 1988 reads:

(1) In the case of domestic investment income (para. 2) as well as investment income received in Austria from debt securities (para. 3), income tax shall be levied by deduction from the investment income (investment income tax).

(2) Domestic investment income shall be deemed to exist if the debtor of the investment income has a domicile, management or registered office in Germany or is a branch in Germany of a credit institution and the following investment income is concerned:

1. a) Profit shares (dividends), interest and other payments from shares, shares in limited liability companies.

§ 94 Z 2 EStG 1988 reads:

The person obliged to deduct (section 95(3)) shall not deduct any capital gains tax:

....

2. under the following conditions in the case of the investment income of corporations within the meaning of

3. section 1 subsection 2 of the Corporation Tax Act 1988:

- It concerns shares in profits (dividends), interest and other emoluments from shares, shares in limited liability companies or in commercial and industrial cooperatives and

- the corporation holds at least one quarter of the share capital or share capital directly.

§ Section 95 (1+2) EStG 1988 reads:

(1) The capital gains tax shall be 25%.

(2) The debtor of the capital gains tax is the recipient of the capital gains. The capital gains tax shall be withheld by deduction. The person obliged to deduct (para. 3) shall be liable to the Federal Government for the withholding and payment of the withholding tax.

According to the case law of the VwGH, contracts between close relatives are only to be recognised for tax purposes if they

- are sufficiently expressed to the outside world

- have an unambiguous content and

- would have been concluded between strangers under the same conditions. This principle also applies to contracts concluded between corporations and their shareholders (cf. VwGH Zl. 2005/15/0073 of 17.4.2008 with further references).

1. group allocations

When making the transfer price adjustments (according to the Ap. report, para. 4.3.), the Ap. findings under the title 'Hidden distribution - transfer prices' argued for additions, among other things, with reference to group allocations that were not booked on an accrual basis and whose amount was not comprehensible, as well as to I/C sales (sales within the group) that did not stand up to an arm's length comparison.

However, it subsequently based its calculation exclusively on what it considered to be non-arm's length returns on sales when determining the transfer prices. Since the other findings (group allocations, I/C turnover) are not specified in more detail in terms of figures and are not reflected in the calculation of the transfer prices or in other allocations, they must be ignored in the following. It should be noted that the ap. referred to incomprehensible documents for the calculation of the group allocations, whereas the complainant referred to documents submitted during the proceedings. Documents on the group apportionments can be found in the work sheet (specifically in the filing of the EM body used in the procedure to evaluate the transfer pricing studies).

2. correction of the PPA (Purchase Price Adjustments) - period shifts

The complainant explains in its appeal that the ap. had objected to the fact that the credit notes had not been booked on an accrual basis and subsequently determines the effects of booking on an accrual basis.

It submits comprehensible documents on the necessary corrections of the PPA payments to establish accrual-based taxation for the years 2001 to 2005.

In the course of the statement on the appeal, Ap. explains that the requested adjustment of profits for 2005 in the amount of € -134,121.54 is justified and that the appeal should be upheld on this point.

Although the numerical corrections requested in the notice of appeal do not include the changes in results described below (which are to be carried out as a result of booking unrelated to the accounting period), the applicant must make a corresponding adjustment to establish the purity of the accounting period, since it calculates the profit in accordance with § 5 EStG 1988 - which is immanent to the principle of taxation on an accrual basis.

Correction of PPA adjustments

Korrektur PPA-Adjustments	2001	2002	2003	2004	2005
	€	€	€	€	€
Gutschriften	11.146.746,88	2.237.212,15	1.685.530,96	1.313.665,05	1.128.737,21
Anpassung betreffend Vorjahr	-671.234,07	-82.000,00	-36.000,00	81.786,41	-74.014,13
Anpassung betreffend Folgejahr	82.000,00	36.000,00	-81.786,41	74.014,13	-60.107,41

korrekte Gutschriften 10.557.512,81 2.191.212,15 1.567.744,55 1.469.465,59 994.615,67

Ergebniskorrektur -589.234,07 -46.000,00 -117.786,41 155.800,54 -134.121,54

3. transfer pricing in the group

Bw. is a distribution subsidiary of the Wx Group and is wholly owned by Z. Deliveries to it are made by production companies of the Group located in Germany, Italy, France, Slovakia, Poland and Sweden with which it has concluded distribution agreements to determine transfer prices.

It is then supplied at standard inventory cost (SIC), which is determined annually in advance for all products and includes the budgeted fixed and variable production costs.

To compensate for its activities, Bw. receives a remuneration for distribution in the amount of 1% of net sales with third parties and a return on investment in the amount of a 10-year Eurobond (risk-free interest rate).

This results in a target margin (return) from the profit on ordinary activities (EBT), taking into account the turnover achieved.

If the Bw. does not reach this target margin, the production companies have to pay credits (PPA) to the Bw., which are limited to 10% of the production costs (90% clause) and, according to the Bw., correspond to the total fixed costs of the production companies.

Due to the limited amount of compensation payments, it can happen that the company does not reach the target EBT margin.

During the audit period this was the case in 2001, 2002 and 2004.

In order to prove the arm's length nature of the transfer prices in the group, the target EBT margins set were compared by the applicant with the margins of companies in a transfer price study by Ernst & Young from the year 2000, which used figures from the years 1996 to 1998(9) .

3.1 Arm's length of distribution agreements

First of all, it must be examined whether the distribution agreement can lead to underlying results (possibly losses) for the applicant that deviate from the defined target EBT margins, which is contrary to the arm's length principle.

The applicant assumes an arm's length agreement and bases its view on the following arguments:

3.1.1 Profitability of Wx Europe

It explains that the entire group of companies is less profitable than its competitors.

Among other things, it explains that the result of the Wx Europe group before tax (EBT) amounts to (accumulated 2001-2005) € - 81,727,000.

The figures broken down by year do not refer to the operating profit (EBIT) on which the transfer pricing studies are based, but rather also include the financial result and are therefore of limited comparability, since a 'more positive' operating profit (EBIT) must be assumed in the case of high interest expenses of the group (analogous to those of the applicant).

Under the distribution agreements in force since 1999, the applicant was in a situation which, if one looks at several years (also beyond the audit period), led to accumulated losses (cf. below profitability of Wx Austria).

In 1998 it received a shareholder subsidy in the amount of ATS 50 million and in 2001 a further subsidy in the amount of € 10 million, which made it possible to maintain the loss situation over a longer period of time.

The applicant explains that it had consciously accepted losses in order to be able to maintain long-term business relations with suppliers.

Obviously, the group's internal goal of a comprehensive market presence in Europe is at the centre of the considerations of the group's management, for which they are also prepared to bear losses.

Strategies to defend high market shares, especially in times of recession or strong competitive pressure, are to be assigned to the affiliated group supplier from a business management point of view and in accordance with the arm's length principle (cf. Macho/Steiner/Spensberger Verrechnungspreise kompakt2 p 253).

A comparable (independent third-party) business would not have survived such a long period without subsidies.

It can be assumed that a comparable company would have renegotiated purchase prices or terminated the supply relationship, otherwise there would have been a market shakeout. The arm's length principle is obviously overlaid by the group strategy.

If the complainant refers to the lower profitability of the Wx Group in comparison to its competitors, this is a consequence of the general entrepreneurial risk, which refers to risks that have their cause in the present competitive situation, productivity, etc., and which are to be attributed to those group companies. The Wx Group's profitability is a consequence of the general entrepreneurial risk, which refers to risks that have their origin in the current competitive situation, productivity and similar factors and are to be attributed to those group companies that have a corresponding disposition capability.

3.1.2 Profitability of Wx Austria

The complainant explains that the margins achieved during the audit period must be assessed against the background of the poor economic situation in the household appliance sector.

This is to be countered by the fact that the unchanged application of the distribution agreements from 1999 to 2007 led to a declared cumulative negative operating profit (EBIT) of € - 5,078,724.71.

Taking into account all available data from previous and subsequent years, the cumulative negative operating result for the period 1997 to 2009 amounts to € - 6,228,664.75 and thus a loss which, when the financial result is included, increases considerably (to approx. € - 13.7 million EBT).

Although it is conceivable that losses are accepted in exceptional cases, the development over time shows that losses due to the agreements concluded could not be made up and compensated for even in comparatively 'good years'.

According to BFH v. 17.1.2001 I R 103/00, a body of experience can be taken into account according to which an independent distribution company will not distribute products with which it only makes losses in the long run. It would stop the distribution of loss-making products in time or look for other products whose distribution promises profits'. According to the Baker&McKenzie study, the production companies (International Product Centre, 'IPC') within the group are 'full-fledged, risk-bearing manufactures', i.e. independent risk-bearing production companies, whereas the sales companies (National Sales Organisation, 'NSO') are 'limited risk distributors', whose tasks lie in the area of sales activities, marketing and distribution.

This includes procurement planning, warehousing, pricing policy, promotional activities and all functions related to sales and distribution.

The sales organisations assume, among other things, customer service within and outside the warranty period and bear the risk of bad debts on the buyer side.

The exchange rate risk for deliveries of goods in the Group, on the other hand, is borne by the production companies, as is the warranty risk, whereby payments by the sales companies are reimbursed.

The Baker&McKenzie study does not go into detail on the functions that led to a reduction of functions at the Bw. in 2002 (accounts receivable, telephone switchboard, IT department) and does not differentiate the (comparison) distribution companies according to these criteria.

Instead, the study derives a differentiation of the comparable companies, which in the present case should lead to an individually adjusted (comparative) return on capital of the applicant, exclusively from the factors inventory and the accounts receivable and payable structure.

In response to the UFS's supplementary request, the applicant explained that the restructuring measures taken (outsourcing of accounts receivable, closure of half of the IT department...) had a relevant influence on the EBIT margin and the ap also referred to the reduced functions to justify the (reduced) EBIT margin applied by it in the calculation according to the report as of 2002.

On the other hand, the applicant did not change the distribution agreements due to these measures (there was no reduction in the transfer price calculation) and also the Baker&McKenzie study presented (as shown above) does not use such functions to differentiate the EBIT margin, so that it can be assumed that, contrary to the presentation, the measures only had a minor influence.

This is also because any savings resulting from the reductions in functions (e.g. accounts receivable) had to be taken over by third parties or Group companies, and these had to be compensated in part by higher Group allocations.

The functions and risks described above do not justify distribution agreements that do not ensure that the applicant, as a limited risk distributor, will not be able to achieve an overall (cumulative) positive operating result over a reasonable (foreseeable) period of time. This is also the case if this would be associated with higher losses for the independent production companies.

The applicant states that it retains the entire profit if it is above the target margin, whereas losses must be (partly) borne by the production companies in the manner described above, which leads to a balanced relationship between opportunities and risks.

It overlooks the fact that in the case of correspondingly high profits it receives remuneration for the sales activities in the form described above and that the EBIT margin could also be above the range of the comparable companies within the framework of the arm's length method chosen by it, which would also have to lead to corrections (with reversed signs).

Furthermore, it is argued that the market was characterised by declining sales figures from 2001 onwards and only recovered in 2005.

The submitted document shows a decline in turnover between 2001 and 2003 in the range of 2% to 4% for the z' sector.

If one compares the results of comparable enterprises of this period according to the market study by Baker&McKenzie with those of the applicant, it emerges that these enterprises even achieved a higher return on capital (EBIT margin median 2.6%) in the years 2002 to 2004 than in previous years (cf. the comparative study by Ernst&Young, which is based on data material of the years 1996 to 1998/1999 (median 0.9% or corrected (without C cf. below) 1.49%).

The statement that the (comparatively small) decline in turnover would have led to a higher drop in earnings cannot be accepted.

The complainant's counterstatement also states that legally inadmissible manipulations of results from previous years were corrected in 2001.

According to the complainant, there was also the assumption that the result in this year was due to mismanagement.

This is to be countered by the fact that the expenses booked in 2001 under 'Expenses from previous periods' were taken into account (added) by Ap. when calculating the EBIT margin for this year (cf. below).

No further indications of mismanagement in 2001 were given. 3.1.3.

3.1.3 Internationally recognised transfer prices

The complainant explains that the foreign customary nature of the Wx Transfer Pricing Guidelines has been examined and assessed in detail by a large number of European and non-European tax authorities.

The view of the tax authorities is in contrast to that of numerous foreign tax authorities.

The applicant is supplied by production companies from Germany, Italy, France, Slovakia, Poland and Sweden.

The most important suppliers are the German and Italian production companies.

Although the German tax authorities would in principle take the view that a distribution company could not in principle make losses, the tax audit authority or the competent tax office in Germany had explicitly determined that this principle could not be applied to the Wx-Europe Transfer Pricing Directive.

The Italian tax authority was the only one that did not accept the transfer pricing guidelines in their entirety and expressed reservations about the fact that the transfer prices after PPA payments would be below the total costs (variable and fixed).

The economic situation of the Wx-Group Europe, which had made a total loss during the audit period, had been pointed out.

A mutual agreement procedure would involve enormous difficulties, especially if the tax authorities had insight into the results of the factories and the distribution company.

It is to be countered that the Bw. is a wholly-owned subsidiary of Z, which is based in Germany, and the authorities there would have been able to obtain all information concerning the applicant via the parent company.

The assessment of whether transfer prices of a company domiciled in Austria comply with the arm's length principle must (initially) be made by the domestic tax authorities.

If several countries are affected by supply transactions, § 6 (6) of the Income Tax Act 1988 must be used as the relevant domestic standard.

In order to eliminate any multiple charges resulting from domestic corrections, these would have to be combined with corresponding adjustments of results abroad. Art. 9 (2) DTA-MA provides for corresponding corrections.

If, in the opinion of the person concerned, the measure of a Contracting State does not lead or cannot lead to taxation in accordance with the agreement, Art. 25 DTA-MA provides for this opinion to be submitted to the competent authority, which must endeavour to eliminate difficulties and doubts arising from the interpretation of the agreement with the competent authority of the Contracting State concerned.

As far as the complainant complained during the oral hearing that the authority had failed to reach an agreement with the foreign authorities by way of mutual agreement 'so that in the event of non-recognition of the losses in Austria, costly and possibly lengthy mutual agreement procedures with the respective states' would be threatened, it must be pointed out that according to Art. 25 DTA-MA, the complainant should have applied for such a procedure 'within three years after the first notification of the measure...which leads to taxation not in accordance with the agreement'.

The reference to the tax authorities having 'insight into the results of the factories and the distribution company' is incomprehensible, as it should be in the applicant's interest to set transfer prices that are generally accepted (i.e. also by the Austrian tax authorities). If this is not the case due to differences of opinion with the the authority, the procedure described above offers the possibility of finding a solution that is satisfactory for both sides.

The reference in the counterstatement that periods from 2001 onwards are affected and that these years are closed for tax purposes in the European countries concerned and that their resumption is protracted is not helpful either, as audit procedures by the tax authorities usually concern longer periods in the past.

A corresponding time limit cannot be derived from Article 25 of the OECD Model Tax Convention (as amended by the OECD-MA).

If the complainant shows no interest in initiating a possible mutual agreement procedure, it cannot be the task of the tax authority to coordinate transfer prices with all countries with which the complainant has group supply relationships (without a request), for example by way of administrative assistance agreements.

It follows from this that although the target EBT margins derived from the distribution agreements may correspond to the results of the arm's length comparison in individual years, due to the restriction (90% clause) on the remuneration granted, the margin actually achieved may fall outside the range of comparable companies in individual years, but may also result in accumulated losses over longer periods (several years), which is why the distribution agreements in their present form (specifically the compensation payments stipulated therein) and the transfer prices derived from them must not be used for tax purposes. the transfer prices derived from them do not stand up to arm's length comparison.

3.2 Comparative study, arm's length margin

Based on the distribution agreements found to be at arm's length, the next step is to clarify the range within which the margin achieved must move in order to meet the arm's length requirement.

Then it must be examined whether the margins actually achieved by the complainant correspond to the arm's length principle, since the lack of arm's length in the distribution agreements does not necessarily mean that the results achieved by the complainant are not arm's length. Furthermore, the value to which the margin is to be adjusted in the case of an identified need for correction (i.e. if the result is outside the range) must be presented.

3.2.1 Bandwidth, comparative study

Transfer pricing is usually tested using the methods set out in the OECD CPCs, which serve as an interpretative aid to compliance with the arm's length principle.

The OECD CPCs concretise these according to the principle of 'dealing at arm's length'.

Accordingly, prices are to be determined as if they had been agreed between independent enterprises.

The Austrian tax authorities regard the OECD transfer pricing principles as an internationally recognised interpretation of Art. 9 of the OECD-MA (cf. 'Die neuen Verrechnungspreisrichtlinien der OECD Bd. 4 S13 with reference to Loukota, Internationale Steuerfälle (1989) Rz. 738 u.a.) as does the applicant.

Although the 1995 OECD TPPG were valid at the time the comparative study was prepared, the provisions set out in the 2010 OECD TPPG can, according to the above opinion, be included in the considerations and used for interpretation.

This is at least insofar as the OECD CPC 2010, in relation to the predecessor provision, does not have a deviating but rather a clarifying and precise meaning.

In the literature (Linde, Die Neuen Verrechnungspreisrichtlinien der OECD Bd. 4 S 27), the opinion is expressed that the application of the 'later' transfer pricing principles is much more in line with their purpose, 'namely a concretisation of the arm's length principle pursuant to Art 9 OECD-MA'.

The UFS therefore subsequently consulted the OECD-VPG 2010 for the interpretation of § 6 Z 6 EStG 1988 and Art. 9 OECD-MA.

In order to examine the arm's length nature of the transfer prices set by the applicant, it used a study prepared by Ernst&Young, which was based on a database analysis of the Amadeus database (version October 2000) for the years 1996 to 1998 and thus on an arm's length comparison.

According to the considerations explained in more detail therein, the study used 7 companies comparable to the applicant. An evaluation of the operating margin (EBIT) results in a median of 0.9% with an upper and lower (quartile) range of 0.3% to 2.5%.

The ap. reviewed the comparable companies used by the applicant and initially eliminated 3 companies for reasons explained in more detail.

The inclusion of C in the comparative study was not accepted because the company was only founded in 1997 and therefore a start-up loss can be assumed.

The complainant followed the concerns of the ap on this point.

The remaining concerns expressed by the applicant against two other comparative farms were withdrawn after the applicant had submitted further documents.

3.2.1.1 Comparative study (plausibility check) by Ap.

In order to verify the results of the submitted comparative study, Ap. carried out its own quantitative database research as a plausibility check, assuming the required independence of the comparable companies from a participation share of affiliated companies of 25% or 50%. As the complainant points out with reference to Steiner/Macho Verrechnungspreise - Dokumentation durch Datenbankstudien, ÖStZ 7/2008, according to the 'conservative approach' an exclusion of comparable companies is already to be carried out at a 25% shareholding. From this percentage on, one can no longer speak of independence (25% criterion).

Using its own database analysis, Ap. arrived at an EBIT median of 1.2% (with a quartile range between 0.44% and 4.36%) with an independence criterion of 75% and an EBIT median of 1.53% (with a quartile range between 0.79% and 3.10%) with an independence criterion of 50%.

The Ap.'s 'screening' is based on the results of comparable companies from 1997 to 2005.

The ap. corrected the operating result of the applicant by assuming an arm's length return on sales of 1.53% in 2001 (median according to the plausibility check according to the "50% criterion") and a return on sales of 0.9% for the years 2002 to 2004 (median according to the study submitted by Ernst & Young) and adding the difference to the operating profit achieved to the profit.

The complainant contests the application of the comparative study prepared by Ap. on which the calculation of the operating result (in 2001) was based with several arguments.

On the one hand, there was a purely quantitative screening, which did not allow any conclusions to be drawn about comparable function and risk profiles.

In addition, in the opinion of the complainant, the independence criterion of at least 50% on which the comparative calculation is based (and which was used to calculate the operating result for 2001) does not meet the requirement of ensuring the independence of comparable companies.

In addition, since transfer prices are set in advance, it is necessary to use data that is available before the priced transaction is carried out. It should be noted that the figures on which the comparative study is based cover the years 1997 to 2005.

In the opinion of the UFS, the arguments of the complainant are at least partially justified, which is why the study prepared by Ap. is unsuitable for determining the arm's length transfer price and could at best be used by the complainant to check the plausibility of the study submitted by Ernst & Young.

In this context, however, it must be noted that the objections of the complainant, insofar as they relate to the years 2002 to 2004, are unfounded, as the complainant used the median of the Ernst & Young study submitted by it and not the comparative study (prepared by the complainant) to calculate the operating result for these years.

With regard to the criticism by the complainant that the study submitted by it, whose EBIT margins were based on values from the years 1996 to 1998, was extended by Ap. to values from 1999, the following should be noted:

According to OECD-VPG 2010 point 3.69, transfer pricing documentation is to be prepared 'on an ex-ante basis using information reasonably available to it at that time'.

It should be noted that the transfer pricing study by Ernst & Young (October 2000) submitted by the applicant contains statistical evaluations based on the years 1996 to 1998, but the study also contains values for comparable companies for the year 1999.

In the view of the UFS, these values, which are accessible and known to the complainant, can be included in any evaluations/analyses without hesitation.

When the complainant states that the target EBT resulting from the distribution agreements is between 1.2% and 1.58% and thus within the range of the comparable companies in its study by Ernst & Young, which was between -2.5% and 7.1%, this cannot be accepted.

Firstly, these are target margins, which represent results from ordinary business activities (EBT) in relation to turnover and not operating results (EBIT) as they underlie transfer pricing studies, which is why the results are not comparable.

Secondly, the calculated target value margins represent target values that are not achieved in part (in the audit period in 3 out of 5 years) due to the compensation payments of the suppliers to the applicant, which are limited to a maximum of 10% of the manufacturing costs.

Thirdly, it must be taken into account that one of the companies included in the Ernst&Young study (C) must be excluded from the comparative study and thus a new range must be used.

Fourthly, the range according to Bw. (-2.5% to 7.1%) are average values of the enterprises included in the comparative study that are not restricted by quartile limits (see below), whereby it is to be examined whether a corresponding restriction is to be carried out or not.

3.2.2 Narrowing the range by quartiles?

The Ernst & Young study (point 6.2.) states that according to the OECD guidelines, comparator companies must be adjusted in case of reasonable and significant differences. If this is not possible, statistical methods can also be used to determine the range. According to the transfer pricing study, since only limited information was available on the comparable enterprises, it was not possible to make the appropriate adjustments, which is why an interquartile range was used, which resulted in profitability values between 0.3% and 2.5% with a median of 0.9%. For the calculation of the restriction of the range it is to be stated that for this purpose 'the upper as well as the lower 25% of the results are eliminated and thus the range is restricted to the area between the 25th and the 75th percentile' (cf. Macho/Steiner Verrechnungspreise - Dokumentation durch Datenbankstudien ÖStZ 7/2008).

The presentation of the authors of the comparative study (Ernst&Young - who calculated a restricted range (0.3%-2.5%)) was not followed by the applicant in the appeal with the argumentation that the target values would lie within the (non-restricted (-2.5% to 7.1%)) range. range, was not followed.

The OECD VPG 2010 state under point 3.57 on the question of a range to be restricted according to statistical methods:

"It may also be the case that, although every effort has been made to exclude items that exhibit a lower degree of comparability, a range of values is obtained for which it is assumed that, given the process used to select the comparables and the limitations in the available information on the comparables, there remain comparability deficiencies that cannot be identified and/or quantified and are therefore not adjusted. In such cases, when the range contains a significant number of such observations, statistical tools that take into account the central tendency to narrow the range (e.g. interquartile range or other percentiles) can help improve the reliability of the analysis'.

The Austrian Transfer Pricing Guidelines VPR 2010 para 67 state:

'It is international practice to narrow the bandwidth by forming quartiles in such a way that the smallest and largest values are each eliminated to the extent of 25% of the total amount of comparative values'.

In the present case there is a study with 7 comparative farms (=observations), which had to be reduced to 6 comparative farms (due to losses in the foundation of the C).

In the opinion of the Senate, such a small sample does not (yet) meet the requirement of a 'considerable number of observations' as set out in point 3.57 OECD VPG 2010, which is why the application of quartiles for narrowing the bandwidth is not to be applied in the given case.

3.2.2.1 Calculation of the range and the median of the adapted Ernst&Young study If the values of the submitted study are used as a basis for these considerations, disregarding company C, the range and the median are calculated as follows using the Excel calculation programme (Microsoft):

Vergleichsbetrieb	EBIT Margen	EBIT Margen	EBIT Margen	EBIT Margen	EBIT Margen
	1996	1997	1998	1999	Durchschnitt
D	1,60	4,40	3,00	4,20	3,30
Ex	0,20	0,70	n.a.	n.a.	0,45
Lx	10,70	3,50	n.a.	n.a.	7,10
Rx	0,50	1,60	0,50	1,30	0,98
Vx	1,70	2,30	n.a.	n.a.	2,00
V	0,00	0,00	0,10	n.a.	0,03

Minimumwert	0,00	0,00	0,10	1,30	0,03
unteres Quartil	0,28	0,93	0,30	2,03	0,58
Median	1,05	1,95	0,50	2,75	1,49
oberes Quartil	1,68	3,20	1,75	3,48	2,98
Maximalwert	10,70	4,40	3,00	4,20	7,10
unlimitierte Bandbreite
Untergrenze	0,00	0,00	0,10	1,30	0,03
Obergrenze	10,70	4,40	3,00	4,20	7,10

The unlimited range of the EBIT margins of the comparable companies thus lies between 0.03% and 7.10% with a median of 1.49%.

3.3 Comparison of the EBIT margins achieved with the customary range according to the (adapted) study

As shown above, the calculated range should not be compared with the planned target margins according to the distribution agreements (which lie between 1.2% and 1.58%) but with the actually achieved EBIT margins.

Before applying the EBIT margin, Ap. adjusted the operating result when calculating the net return, taking into account that in 2001 expenses from previous periods were neutralised (cf. calculation below) and in 2002 restructuring costs (among others from the outsourcing of accounts receivable accounting, the closure of the Austrian telephone exchange and half of the IT department) in the amount of € 304,000 were incurred.

The corrections were basically correct and represent an adjustment to the individual circumstances of the business because it can be assumed that such expenses were not incurred in comparable businesses.

In addition, the EBIT had to be corrected by those values that resulted in profit changes according to the appraisal report (findings of the appraisal report) or were pointed out by the complainant (specifically, period shifts from PPA adjustments, see above) and are necessary to take into account a comparable operating result.

This results in the following picture:

	2001	2002	2003	2004	2005
	€	€	€	€	€
Nettoverkäufe lt. GuV	40.426.230,25	32.340.926,03	30.543.443,84	29.763.006,48	29.117.569,95
EBIT lt. GuV Korrektur Aufwand aus Vorperioden	-7.457.634,60 4.552.695,09	-224.243,41	107.351,26	-45.666,48	630.606,67
Restrukturierungskosten sonst. Korrekturen laut Ap.-Bericht	-11.844,39	304.000,00 4.689,81		-9.220,38	15.670,00
Ergebniskorrekturen aus PPA-Adjustments lt. Bw.	-589.234,07	-46.000,00	-117.786,41	155.800,54	-134.121,54
EBIT korrigiert lt. UFS	-3.506.017,97	38.446,40	-10.435,15	100.913,68	512.155,13
EBIT-Marge lt. GuV	-18,45%	-0,69%	0,35%	-0,15%	2,17%
EBIT-Marge korrigiert	-8,67%	0,12%	-0,03%	0,34%	1,76%

adaptierte Bandbreite für
EBIT-Marge lt. Studie	0,03%-7,1%	0,03%-7,1%	0,03%-7,1%	0,03%-7,1%	0,03%-7,1%
EBIT-Marge
innerhalb/außerhalb d.
Bandbreite	außerhalb	innerhalb	außerhalb	innerhalb	innerhalb

The evaluation of the achieved net returns (EBIT margins) leads to the result that the margins achieved in 2001 and 2003 are outside the range and an adjustment of the EBIT margin of these years has to take place. In the other years, the EBIT margin achieved is within the range, which is why no adjustment is required.

3.4 Adjustment to the median

According to OECD CPC 2010 paragraph 3.55, transfer pricing is not an exact science, so there are many situations where the application of the most appropriate method or methods will produce a range of values, all of which are relatively equally reliable.

On the question of selecting the most appropriate point within the range, the OECD CPD 2010 states in paragraph 3.61: 'In general, and to the extent that a choice between different points in the range is possible, such adjustments should be aimed at the point within the range that best reflects the facts and circumstances of the intercompany transaction concerned'.

The OECD CPD 2010 further states in para 3.62:

'In determining this point, where the range has a relatively equivalent and high reliability of results, it could be argued that any point within the range satisfies the arm's length principle. Where comparability deficiencies remain, as discussed in paragraph 3.57 (see above), it may be appropriate to use central tendency statistical tools to determine this point (e.g. the median, means or weighted averages, etc., depending on the precise characteristics of the data series) in order to minimise the risk of error from unknown or unquantifiable remaining comparability deficiencies'.

The Ernst & Young transfer pricing study to be used states in point 6.2 that it was not possible for the compilers to make precise adjustments:

'Due to the limited amount of balance sheet information available for the comparable companies and variations in accounting principles, it was not possible to accurately make adjustments to the comparable company data'.

For this reason, an interquartile range narrowing was made in the transfer pricing study, which, as explained above, is not in line with the OECD CPC 2010 paragraph 3.57 due to the small sample.

In paragraph 3.62, the OECD CPG 2010 provides for the use of statistical instruments with a central tendency without narrowing the range as a further possibility to improve results that are affected by comparability deficiencies.

OECD CPD 2010 para. 3.55 assumes that all values within the bandwidth are relatively equally reliable, but with the restriction that if there are comparability deficiencies (para. 3.57), the reliability of the information can be increased by applying statistical methods.

In the view of the UFS, the use of the median in the event that the EBIT margin achieved is outside the range is to be applied in the present case because, according to the study, there is no 'highly reliable' range (cf. Loukota/Jirousek comments on the criticism of the Transfer Pricing Guidelines 2010 ÖStZ 2011) due to comparability deficiencies.

Insofar as the applicant assumes that the correction of the EBIT margin to the median value constitutes an impermissible punitive taxation and possibly seeks an adjustment to the lower bandwidth value, whereby it recognisably refers to a decision of the BFH of 17 October 2001 I R 103/00, according to which an estimate is based on the upper or lower value of the bandwidth of arm's length transfer prices, which is more favourable for the taxpayer. In addition to the existing comparability deficiencies, which in themselves justify an adjustment to the median, reference should also be made to the transfer pricing study by Baker&McKenzie from 2005, which was also submitted by the applicant.

It may be true that transfer prices have to be fixed in advance, but in the case at hand no transfer prices were fixed per transaction carried out; instead, distribution agreements had been concluded in unchanged form since 1999 and the arm's length nature of these agreements was justified by the results of comparative company studies.

From the above point of view, it is permissible to use a study (Baker&McKenzie) for the further assessment of the arm's length nature of the EBIT margin, which was prepared at a time (here 31 December 2005) that follows the period in which the net returns to be assessed were generated (2001 to 2005), but which refers to data material that originates from this period (2002 to 2004).

This is because a comparison of the net returns achieved in the period under review (2001 to 2005) with comparable enterprises based on data from the years 1996 to 1999 can at best be used for planning purposes, but subsequent significant developments in the period under review (e.g. economic downturns...) are not (or cannot be) taken into account.

According to Baker&McKenzie, the data material used in this process led to the result of comparable net yields with a median of 2.3% and a quartile range between 1.3% and 3.9%.

An appendix to this study, which was prepared especially for the company and deals with the special features of inventory adjustment, accounts receivable and accounts payable, shows a comparable median EBIT return for the company of 2.6% with a quartile range of 1.5% to 4.1%.

The values shown were achieved by comparable companies in the audit period and are consistently above the adapted median according to the transfer pricing study by Ernst & Young, which is why the adjustment to the lower range requested by the applicant is also unjustified for this reason.

If the UFS bases its assessment of the arm's length transfer price on the Ernst & Young study and uses the median achieved there, this is because it follows the applicant's argumentation regarding the price determination required in advance and for this reason bases its considerations regarding comparable net returns on the modified Ernst & Young transfer price study.

There are no other particular influencing factors that would make an adjustment of this study necessary.

In view of the above considerations, the UFS assumes that the median net return of 1.49% determined in the modified comparative study by Ernst & Young submitted by the applicant is appropriate and should be applied for the audit period.

3.5 Calculation of EBIT on the basis of the arm's length EBIT margin

The EBIT margin must be calculated after taking into account all other findings. This includes the period shifts described by the applicant as well as the other findings of Ap.

	2001	2003
	€	€
Nettoverkäufe lt. GuV	40.426.230,25	30.543.443,84
Median EBIT-Marge lt. Studie = Marge lt. UFS	1,49%	1,49%
EBIT lt. UFS = Nettoverkäufe mal 1,49%	602.350,83	455.097,31
EBIT korrigiert lt. UFS (s. oben)	-3.506.017,97	-10.435,15
Differenz=Zurechnung	4.108.368,80	465.532,46

4. hidden distribution

Hidden distributions are contributions to persons directly or indirectly involved in a corporation that lead to a reduction in the profits of the corporation and that would not have been granted to third parties unrelated to the corporation (cf. VwGH v. 20.9.1983, 82/14/0273).

The term "hidden distribution" encompasses all benefits (granting of advantages) to those directly or indirectly involved in a corporation which are not readily recognisable as distributions and which lead to a reduction in the income of the corporation and which are not granted to third parties unrelated to the corporation.

As the Austrian distribution subsidiary of the Wx Group, the Bw. is wholly owned by Z.

Since it was determined that the agreed prices in two years were not within the range of reasonable arm's length prices, the adjustments to the transfer prices made by means of profit adjustments constitute hidden distributions to the parent company pursuant to section 8 (2) KStG 1988, with the capital gains tax being borne by the applicant.

The hidden distribution is calculated as follows:

	2001	2003
verdeckte Ausschüttung vA netto = Zurechnung Verrechnungspreis	€ 4.108.368,80	€ 465.532,46
33,33% KESt	1.369.456,27	155.177,49
vA brutto	5.477.825,07	620.709,95
The bases of taxation are as follows:
	2001	2002	2003	2004	2005
	€	€	€	€	€
Summe Korrekturen laut
Ap.-Bericht	7.521.156,53	216.001,55	167.539,73	304.313,16	15.670,00
abzüglich Korrektur			-
Verrechnungspreise lt. Ap.	-2.980.305,83	-211.311,74	167.539,73	-313.533,54
zuzügl. Korrektur
Verrechnungspreise lt. UFS	4.108.368,80	0,00	465.532,46	0,00
Korrektur
Periodenverschiebungen			-
PPA Adjustment lt. Bw.	-589.234,07 -46.000,00 117.786,41 155.800,54 -134.121,54
Summe Korrekturen laut
UFS	8.059.985,43 -41.310,19 347.746,05 146.580,16 -118.451,54
	öS

110.907.817,48.

Einkünfte aus Gewerbebetrieb	2001	2002	2003	2004	2005
	öS	€	€	€	€
Laut Erklärung/Veranlagung	-111.133.504,00	300.004,60	-55.578,84	-30.900,34	566.990,09
Korrekturen lt. UFS s.o.	110.907.817,48 -41.310,19 347.746,05 146.580,16 -118.451,54
Einkünfte aus Gewerbebetrieb laut UFS	-225.686,52 258.694,41 292.167,21 115.679,82 448.538,55

	2001 öS	€
Einkünfte aus Gewerbebetrieb	-225.686,52	-16.401,28
Gesamtbetrag d. Einkünfte	-225.686,52
Einkommen	0,00
Köst gem. § 22 KStG 1988	0,00
Diff. Zur Mindestkörperschaftsteuer	24.080,00
Körperschaftsteuer	24.080,00	1.749,96

Einkünfte aus Gewerbebetrieb	258.694,41
Gesamtbetrag d. Einkünfte	258.694,41
Verlustabzug	-16.401,28
Einkommen	242.293,13
Köst 34% v. 242.293,13	82.379,66
anrechenbare Mindestkörperschaftsteuer	-1.750,00
Körperschaftsteuer	80.629,66

Einkünfte aus Gewerbebetrieb	292.167,21
Gesamtbetrag d. Einkünfte	292.167,21
Einkommen	292.167,21
Köst 34% v. 292.167,21	88.570,51
Körperschaftsteuer	88.570,51

Einkünfte aus Gewerbebetrieb	115.679,82
Gesamtbetrag d. Einkünfte	115.679,82
Einkommen	115.679,82
Köst 34% v. 115.679,82	39.331,14

Einkünfte aus Gewerbebetrieb	448.538,55
Gesamtbetrag d. Einkünfte	448.538,55
Einkommen	448.538,55
Köst 25% v. 448.538,55	112.134,64
Körperschaftsteuer	112.134,64
Kapitalertragsteuer 2001	2003
€	€
vA netto = Zurechnung Verrechnungspreis 4.108.368,80	465.532,46
33,33% KESt 1.369.456,27	155.177,49