Central
Economic-Administrative Tribunal
FIRST CHAMBER
DATE: 26 May 2021
PROCEEDINGS:
00-02545-2019; 00-06926-2019
CONCEPT: CORPORATE
INCOME TAX. I.SDES.
NATURE: SINGLE
GENERAL INSTANCE CLAIM
CLAIMANT: XZ SL -
NIF ....
REPRESENTATIVE:
... - NIF ...
ADDRESS: ... -
Spain
At Madrid, the
Court was convened as indicated above to rule in sole instance on the
above-mentioned claim, which is being processed under the general procedure.
The present claim
has been heard against the settlement agreement issued by the Regional
Inspection Unit of the Special Delegation of the AEAT of Madrid, for the
concept of CORPORATE TAX, years 2011 to 2013, derived from the non-conformity
report with reference number ..., with a tax debt to be paid in the amount of
1,570,291.77 euros, of which 1,339,429.59 correspond to tax and 230,862.18 to
late payment interest.
FACTUAL BACKGROUND
FIRST.- On
17/05/2019 the present claim, filed on 14/05/2019, was received by this Court.
SECOND.- The
commencement of the verification and investigation proceedings was carried out
by means of a communication dated 08/02/2017, notified on 08/02/2017.
Insofar as in the
course of the proceedings a procedure to check the market value of related
transactions took place, the valuation adjustment resulting from the same not
being the sole object of the regularisation carried
out, two assessments were issued for the same concept and periods -Corporate
Income Tax periods 2011-2012-2013, initiated on the same day: a first
Assessment -Act A02 no. . ...5-, which is the basis for the contested
agreement, limited to the settlement proposal arising from the valuation
adjustment made in the related transaction between the company XT SA (included
in the tax consolidation group a/07) and the Portuguese entity XZ SA, and a
second assessment A02-..., which documents the other elements resulting from
the tax adjustment made.
XZ SL's corporate
purpose is to participate in the businesses of other companies, to prepare and
carry out commercial, tax, financial, accounting and economic analyses and
studies, to provide IT services, to finance operations and businesses, to grant
loans and similar and to carry out leasing and similar operations. In addition
to the functions it performs directly, the company is the parent company of a
group of entities engaged mainly in the manufacture, distribution and sale of
pharmaceutical and cosmetic products and products intended for human
consumption, and the production of wine.
Of all the
entities making up the group, the following were audited:
- XZ SL, as the
parent company.
- XW SA (...) and
- XT SA (...).
The disputed
adjustment in the present complaint refers to the prices agreed in the sales of
generic medicines made by XT SA to a Portuguese subsidiary in the years under
review, on the grounds that the prices agreed do not correspond to those that
would have been agreed between independent parties under conditions of free
competition.
THIRD.- On
22/04/2019 the settlement agreement was notified, which resulted in a tax debt
to be paid in the amount of 1,570,291.77 euros, of which 1,339,429.59
corresponds to the tax liability and 230,862.18 to late payment interest.
As a result of the
verification carried out, disciplinary proceedings were initiated which ended
with the agreement notified on 23/10/2019, which resulted in a fine of
252,814.23 euros to be paid.
FOURTH - On
14/05/2019 and 14/11/2019, economic-administrative claims were lodged against
both resolutions, presenting the allegations that are analysed
below.
GROUNDS OF LAW
FIRST.- This Court
is competent to rule in accordance with the provisions of Law 58/2003, of 17
December, General Taxation Law (LGT), as well as the General Regulations for
the development of Law 58/2003, of 17 December, General Taxation Law, in
matters of administrative review (RGRVA), approved by Royal Decree 520/2005, of
13 May. None of the causes of inadmissibility foreseen in article 239.4 of the
LGT are present.
SECOND.- The
aforementioned claims are resolved cumulatively in accordance with the
provisions of article 230 of the LGT.
THIRD.- This Court
must rule on the following:
-Limitation due to
excess of the duration of the inspection period.
-Valuation of
transactions between related entities.
-Error in the
calculation of late payment interest.
FOURTH.- The
claimant firstly alleges that the inspection proceedings have exceeded the
legally authorised period and, therefore, the right
to settle and demand the tax debts corresponding to financial years 2011 and
2012 is time-barred. On 22 June 2015, the entity and the subsidiaries XW and XT
were notified of the commencement of partial verification and investigation
proceedings in respect of the IS corresponding to the 2010 financial year,
limited to the verification of the corrections to the accounting result. During
the course of the inspection proceedings, the Inspectorate agreed to extend the
scope to various items and years (2011 and 2012). On 20 July 2016, the
inspection proceedings were concluded by means of a Settlement Agreement. On 8
February 2017, the entity received from the same Regional Inspection Unit of
the Special Delegation of Madrid a communication from the same Regional
Inspection Unit of the Special Delegation of Madrid to initiate general
verification and investigation proceedings relating to the IS for the
2011-2012-2013 financial years. On 22 April 2019, it was notified of the
Settlement Agreement for the years 2011, 2012 and 2013. It argues that the
Inspectorate had more than 38 months to carry out the inspections for the years
2011 and 2012, which were carried out partially in the first stage and in a
general manner in the second stage. It takes the view that it should not be
acceptable that the same tax year can be the subject of successive partial
inspections which, taken together, exceed that maximum limit of 27 months. In
the applicant's view, the principle of a single procedure has been infringed in
that, instead of extending the inspection procedures of the first procedure,
the administration has combined two quasi-consecutive inspection procedures.
This way of acting constitutes a procedural fraud of the rules that determine
the maximum duration of the proceedings.
Article 150 of Law
58/2003 General Tax Law, as amended by Law 34/2015, states the following with
regard to the duration of inspection proceedings:
Term of inspection
proceedings.
1. The inspection proceedings
must be concluded within a period of:
a) 18 months, in
general.
b) 27 months, when
any of the following circumstances apply to any of the tax obligations or
periods subject to verification:
1.º That the
taxpayer's Annual Turnover is equal to or greater than that required to audit
its accounts.
2. The taxpayer is
part of a group subject to the tax consolidation system or the special system
for groups of entities that are being audited.
When audits are
carried out with various persons or related entities in accordance with the
provisions of Article 18 of Law 27/2014, of 27 November, on Corporate Income
Tax, the concurrence of the circumstances provided for in this letter in any of
them shall determine the application of this period to the inspection
procedures carried out with all of them.
The period of
duration of the procedure referred to in this paragraph may be extended in the
terms indicated in paragraphs 4 and 5.
According to the
contested settlement agreement, on 13 April 2016, an inspection report A02 ...3
was issued to the claimant for corporate income tax for the years 2010, 2011
and 2012, relating to a partial tax group audit No. a/07. With regard to
Corporate Income Tax for 2010, the audit was limited to the verification of:
- The negative
adjustments to the Taxable Base made in the determination of their respective
individual Taxable Bases by the following entities of Group a/07: XZ SL ..., XW
SA ... and XT SA ....
- The reinvestment
deduction credited by ...
- The Double
Internal Taxation Deduction and deductions for export activities, if any,
generated in 2010 by XZ SL.
- The deductible
amount for impairment of securities representing the shareholdings in the
capital of other entities by XZ SL.
- The income
attributable to the tax base of XT SA and the Tax Group corresponding to
transactions under the concept of co-developments that have involved the
transfer of dossiers relating to pharmaceutical specialities,
between the entity ZT SA ... and other non-resident entities, dependent on the
entity XZ SL ...
- Verification of
the deductions from the tax liability for export activity, pending application,
originating in the entities XZ SL (...) and XT SA (...).
With regard to
corporate income tax for 2011 and 2012, the sole scope of the verification was:
"The verification of the tax credits applied in 2011 and 2012,
corresponding to the deduction for export activities declared as having been
generated in previous periods".
The claimant
argues that the duration of this partial verification procedure should be added
to the one now under challenge, on the basis of the single procedure principle
that this Court has upheld.
It is true that
this Court has repeatedly ruled on the principle of the single procedure
alleged by the claimant, although these were cases in which, after the
commencement of an inspection procedure, the scope of the proceedings was
extended to other items or years; in these cases, we stated that the duration
of the proceedings is calculated from the first notification, that of the
commencement of the proceedings, but this is not the case here.
The claimant also
refers to our Resolution 5221/2010, dated 15/11/2012, but in this case the
issue raised was different, as it involved a case in which partial inspection
proceedings were initiated and, subsequently, without finalising
these proceedings, a new general proceeding was initiated which was extended,
in addition to the concept and exercise of the partial proceedings that were
not finalised, to other concepts and financial years.
We stated that, in the context of a single procedure initiated on a partial
basis, it was not possible to consider two procedures to have been opened, and
we should therefore have understood that the second notification had extended
the scope of the proceedings already initiated (from partial to general), as
well as modifying their scope, covering other concepts and financial years,
without there being, therefore, a new notification of initiation, which meant
that the calculation of the period of duration of the single procedure should
be carried out from the time of the first notification.
Our case is
different, since there is nothing to prevent the inspectorate, after completing
a partial verification procedure, from initiating another general verification
procedure, and the periods of duration should be calculated independently,
since these are two different procedures. This Court does not observe the
procedural fraud alleged in the actions of the inspectorate in these
proceedings that would suggest that this way of proceeding has been carried out
with the aim of unduly delaying the duration of the proceedings.
Thus, in the
proceedings that ended with the contested agreement, the maximum duration of
the inspection proceedings is 27 months, in accordance with section 1.b) 1 of
article 150 of Law 58/2003, of 17 December, General Tax Law, as the annual
turnover of the entity exceeds 5,700,000 euros, required to audit its accounts,
in accordance with the provisions of article 263 of Royal Legislative Decree
1/2010, on Capital Companies. Consequently, and given that the date of
commencement of the inspection activities was 08/02/2017, the maximum period of
27 months for the duration of the inspection activities, indicated in article
150 of the LGT, had not been exceeded when the settlement agreement was
notified on 22/04/2019. We must therefore dismiss the claimant's claims on this
point.
FIFTH.- Turning to
the substantive issues raised, in relation to the valuation of related-party
transactions, we must begin by pointing out the regulations applicable to the
case:
1.- CONVENTION
BETWEEN THE KINGDOM OF SPAIN AND THE PORTUGUESE REPUBLIC to avoid double
taxation and prevent tax evasion in matters of income tax and protocol, signed
in Madrid on 26 October 1993. Article 9 states
Article 9.1
ASSOCIATED
ENTERPRISES
1. Where
(a) an enterprise
of a Contracting State participates directly or indirectly in the management,
control or capital of an enterprise of the other Contracting State, or
(b) the same
persons participate directly or indirectly in the management, control or
capital of an enterprise of a Contracting State and of an enterprise of the
other Contracting State,
and in either case
the two enterprises are, in their commercial or financial relations, bound
together by conditions accepted or imposed which differ from those which would
be agreed between independent enterprises, profits which would have been realised by one of the enterprises in the absence of those
conditions, and which have not in fact arisen by reason of those conditions,
may be included in the profits of that enterprise and taxed accordingly.
2.- ROYAL
LEGISLATIVE DECREE, 4/2004, OF 5 MARCH, APPROVING THE REPEALED TEXT OF THE
CORPORATE TAX LAW (hereinafter, TRLIS), in the wording given by Law 36/2006, of
29 November, on measures for the prevention of tax fraud (applicable to tax
periods beginning on or after 1 December 2006).
"Article 16.
Related transactions
1. 1. Transactions
carried out between related persons or entities shall be valued at their normal
market value. Normal market value shall be understood to be that which would
have been agreed by independent persons or entities under conditions of free
competition.
2. The tax
authorities may check that transactions carried out between related persons or
entities have been valued at their normal market value and shall, where
appropriate, make the necessary valuation adjustments in respect of
transactions subject to this Tax, Personal Income Tax or Non-Resident Income
Tax which have not been valued at their normal market value, using the
documentation provided by the taxpayer and the data and information available
to it. The Tax Administration shall be bound by this value in relation to the
rest of the related persons or entities.
(...)
2. The related
persons or entities shall keep at the disposal of the Tax Administration the
documentation established by regulation.
3. The following
shall be considered to be related persons or entities:
(a) An entity and
its partners or participants.
(...)
d) Two entities
belonging to a group.
(...)
3.- OECD
GUIDELINES APPLICABLE TO MULTINATIONAL COMPANIES AND TAX ADMINISTRATIONS IN THE
AREA OF TRANSFER PRICING in its 2010 version. With regard to these Guidelines,
it should be noted that the explanatory memorandum of Law 36/2006 of 29 November,
on measures for the prevention of tax fraud (which amends Spanish tax
regulations on related-party transactions or transfer pricing) states that the
Guidelines will have interpretative value for these regulations. The second
objective is to adapt Spanish transfer pricing legislation to the international
context, in particular to the OECD guidelines on the subject and to the
European Forum on Transfer Pricing, in the light of which the amended
legislation must be interpreted'.
According to the
settlement agreement, in group X the manufacturing activity of generic
medicines is carried out solely by the entity XT SA, which belongs to the
Spanish tax consolidation group. This entity manufactures products both for its
own marketing and for sale to foreign subsidiaries and to independent Spanish
or foreign third parties.
The inspection
notes that it has not found in the study of the margins obtained by XT SA on
its sales of generic medicines to the Italian and French subsidiaries the
discrepancies and inconsistencies that will be indicated below, and which
result from the study of the margins applied in the sales of generic medicines
to the Portuguese subsidiary of the group.
In the contracts
signed between XT SA and the third-party marketing entities (not related), a
Supply Price is established to guarantee a margin to the third-party marketing
entity, which is usually, as explained in the documentation on transactions
with related parties, around 60%-80% of the Laboratory Sales Price (LSP, which
is set by the Administration) and a minimum price, as stated in the said
documentation, below which the sale is not made and which is independent of the
LSP. These conditions, in sales to third parties, differ substantially from
those agreed with group companies through so-called 'intercompany supply'
contracts.
In the case of the
contract between XT SA and X PORTUGAL, it is stated that the supplier will
invoice all the products delivered at "Ex - Works" prices, based on
the total cost of the packaged product, increased by 8%, and that,
additionally, the supplier will charge the cost of the wooden pallets used in
the packaging, which will be invoiced at a price of 6.86 euros per pallet. The regularisation refers to sales of generic medicines to this
subsidiary in Portugal.
On the other hand,
the inspection notes that, in the case of the subsidiaries in France and Italy,
the number of product references is very limited, reaching a total number of 27
product references in the case of France and 22 product references in the case
of Italy; with regard to Portugal, the number of product references whose
marketing is covered by the contract for the supply of medicines is much more
numerous, reaching 158 references and formats of different generic medicines.
In other words, the number of generic drug product references whose marketing
and distribution is covered by the contract with Portugal is almost six times
higher than the contracts with the subsidiaries in France and Italy,
respectively. The differences in the number of products/packages also result in
significant differences between the sales made by XT to the Portuguese
subsidiary of group X, compared to those made to the group's subsidiaries in
Italy and France, which are summarised in the tables
on page 80 of the settlement agreement, from which the Commission concludes
that they can hardly be considered equivalent or, as the case may be,
comparable markets.
According to
Article 16.1.1.1 of the TRLIS, "normal market value shall be understood to
be that which would have been agreed by independent persons or entities under
conditions of free competition". As regards the methods to be applied to
determine the market value, the method considered by the entity to be the most
appropriate for the market valuation in this related-party transaction is the
"net margin of the set of transactions" method, as provided for in
Article 16.4.2.2 b) of the TRLIS:
"(b) Net
margin method of the set of transactions, whereby transactions carried out with
a related person or entity are attributed the net result, calculated on costs,
sales or the most appropriate magnitude depending on the characteristics of the
transactions, which the taxpayer or, where appropriate, third parties would
have obtained in identical or similar transactions carried out between
independent parties, making, where necessary, the necessary corrections to
obtain equivalence and consider the particularities of the transactions".
In relation to the
selection of this method, the documentation on related-party transactions
provided by the taxpayer states the following:
"Firstly, a
search was carried out for TUFs that would allow the application of the CUP
method to assess the market value of the transactions under analysis. Although
XT SA carries out transactions with independent third parties, the products it
sells to them and their affiliates are not always the same. Similarly, the
geographic markets in which these products are distributed are different, which
is important in this industry, given that the level of market penetration and
the response of the health authorities in each country must be assessed. In
addition, the volume of products sold to related third parties is very
different, which implies that the terms and conditions of distribution are not
comparable.
In this respect,
the OECD guidelines state in paragraph 1.33 that "The application of the
arm's length principle is generally based on a comparison of the terms and
conditions of a tied transaction with the terms and conditions of transactions
between independent enterprises. Comparability means that none of the
differences (if any) between the situations being compared can significantly
affect the terms and conditions analysed in the
methodology.
Finally, the
method selected in the documentation for fiscal year 2009 was applied. The
method selected in the documentation was the TNMM, which takes into account the
operating profit earned by functionally comparable independent companies.
For the
application of this method, the Orbis database was used and a search was
carried out based on the following criteria:
(...)"
These search
criteria resulted in a final sample of 26 comparable companies.
The interquartile
ranges determined in the taxpayer's own related-party transactions
documentation for the case of the generic medicines manufacturing function for
further distribution are as follows:
CFCMU 2011 2012
2013
Maximum 49.16%
46.31% 39.75% Upper quartile
Top Quartile
15.06% 13.06% 14.47% 14.47% Median 12.74% 10.06% 10.47% 10.47% 10.47% 10.47
Median 12.74%
10.43% 9.09% Middle 12.74% 10.43% 9.09% Lower quartile
Lower Quartile
7.58% 5.64% 4.00% 4.00% Minimum
Lowest -3.12%
-7.02% -7.58% -7.58% -3.12% -7.02% -7.58% -7.58% -3.12
Accepting the method
and the way in which the samples are selected by the taxpayer, but comparing
these data with the results of the information supplied by the entity, the
inspection analysis focused on the study of the costs incurred by XT SA in the
manufacture of generic medicines and the margins obtained in the sale to
subsidiaries.
On the basis of
the information supplied by the taxpayer on the costs and prices of XT SA's
supplies to the group's Portuguese subsidiary, the inspectorate obtained the
gross margins, broken down by product, set out on pages 25 et seq. of the
settlement agreement.
A statistical
study of these values shows a wide disparity; for example, in 2011, from a
margin of -47.99% as a minimum value to 284.37% as a maximum value. Thus,
although the median of the values calculated in this financial year was 7.71%,
close to the 8% that the taxpayer states in the transfer pricing documentation
it applies, the inspection considers it necessary to calculate a more detailed
and realistic analysis in which the study of margins takes into account the
correct weighting according to the sales volume of each product.
As a result, for
the 2011 financial year, an aggregate margin of 7.94% is obtained, considering
the weighted weighting of the sales of each product, which is close to the
median value obtained in the unweighted margin study of 7.71% and is also close
to the value of 8%, which the taxpayer states is applied in the transfer
pricing documentation.
A similar analysis
was performed for 2012 and 2013, concluding that, for 2012, the data obtained
implies a negative aggregate margin of -20.52%, which is far from both the
median value obtained in the unweighted margin study, 8.03%, and the value of
8% that the taxpayer states is applied in the transfer pricing documentation.
For the 2013 financial year, the data obtained imply obtaining a positive
aggregate margin of 3.63%, considering the weighted weight of the sales of each
product, which is very far from both the median value obtained in the study of
unweighted margins, 8.25%, and the value of 8% stated by the taxpayer in the
transfer pricing documentation.
Furthermore, it
can be seen that, for 2011, the margin obtained is within the interquartile
range determined in the taxpayer's own documentation on related-party
transactions, as the average gross margin, weighted according to sales, is
slightly higher than the 7.58% of the lower quartile obtained in the previous
table; However, for the years 2012 and 2013 this circumstance does not occur,
so that, in the opinion of the inspection, given that the margin effectively
applied is outside the interquartile range, in order to adapt the margin to be
applied to a central tendency value within the range, the central tendency
magnitude considered appropriate is the median.
SIXTH.- The
claimant argues that we are not faced with an improper determination of the
transfer pricing policy. Firstly, it points out that there has been a one-off
error that has been repeatedly explained to the inspectorate. The numerical
disparity recorded for a given product is an error in the computer system in
reflecting the production costs of a given product (and more specifically in
the number of packing cases indicated in the recipe) which is sold to the
subsidiary located in Portugal. It was that error in the number of packing
cases which led to the appearance of a unit cost which did not correspond to
reality. This error was subsequently corrected by the company. Annually (at the
beginning of the year) the prices for each component are reviewed and updated
and with the update the costs for the new year are calculated. This calculation
is only modified during the year (the system only allows changes to be
registered each month) to correct possible errors detected or to update the
cost calculation as a consequence of important changes in the components (in
the case of variations equal to or greater than 10%). Accordingly, the final
production costs are usually more representative in the middle of the year,
which is why the costs provided for inspection are as of June. As was stated at
the hearing and again at the Preliminary Hearing, when a change in the bulk of
the tablet was included in June 2012 (from "..." to "..."),
the number of packing cases was also changed by mistake. In July 2012, when it
was discovered that the prescription in question was inconsistent, the process
department of the industrial area analysed it and
found that 152,680 boxes had been entered for packing 10,000 units (units of
boxes of medicine, the sales presentation of which is 60 tablets per box) in
the prescription for material 10522. In accordance with the above, the
complainant considers that it is sufficiently proven that, according to the
volume of sales, it is illogical that 152,680.051 "3073" packaging
units (boxes) are needed for 10,000 sales copies, which shows that it was a
simple error in the entry of the number of boxes, the correct quantity being
152 packaging units for the 10,000 sales copies (65 sales units per 1 packaging
box).
The complainant
further alleges that the margin applied is in accordance with the transfer
pricing policy, and that no other margin has been justified by the
inspectorate. It considers that the margin actually applied was 8.03% in 2012
and 8.25% in 2013. The inspection carries out an analysis in which the margin
study takes into account the weighting according to the sales volume of each
product. However, even when calculating the margin obtained by weighting
according to sales volume, the aggregate margins obtained are within the interquartile
ranges applicable to the generic drug manufacturing activity defined in the
transfer pricing study: In 2012, if we consider the weighted weighting of the
sales of each product and calculate the result obtained by isolating the effect
associated with the product data that has been provided to the Inspectorate in
error, the overall gross margin obtained would have been 12.05%. In 2013, if
the aggregate margin is below the range, an adjustment to the lower part of the
interquartile range would be appropriate. In this case, taking into account
that the positive aggregate profit margin is 3.63% and, as can be seen from the
price report, the lower interquartile range in the 2013 financial year is
4.00%, the adjustment should have been made in respect of that 4% and not in
respect of the median, since this adjustment to the median is only permitted in
the event that defects in comparability have been duly accredited, which has
not occurred in this case.
Thus, with regard
to the alleged error, this is a question of proof. The interested party argues
that the error explained has undermined the margin calculated by the inspection
in 2012, but this Court, despite the detailed explanations given by the
claimant in its written pleadings, cannot consider it proven that the data used
by the inspection, previously provided by the taxpayer, contains the alleged
error and that this has led to an error in the calculation of the margin of
that product.
Examination of the
file and the settlement agreement shows that the data on costs were requested
by the inspectorate with reference to annual data, which is not consistent with
the claimant's statements concerning a one-off error in a specific product in a
specific month.
On the other hand,
if in July 2012 the alleged error was detected and corrected, it is not logical
that the information requested is provided as of June 2012, the month in which
the error occurred, and the inspection was not warned on any of the three times
that this data was requested.
With regard to the
way in which the inspection carried out the calculation of the margin,
weighting the costs of each product by the volume of sales, we consider that
this way of carrying out the study is more in line with reality. As we have
seen, the method applied is that of the net margin of the set of operations, so
that the volume of sales, different for each product, determines to what extent
the margin of each product contributes to the margin of the set of operations,
and the product that has been sold more than the one that has been sold in
smaller quantities must be taken into account to a greater extent, which in the
study carried out by the inspection translates into the weighting of the
margins by the volume of sales, a calculation that we understand to be more in
line with reality than the one carried out by the taxpayer. This leads us not
to accept, as indicated by the inspectorate, the 8% margin that the entity
claims to have applied.
At this point, the
claimant argues that the inspectorate has not justified the application of the
median. We must turn to the OECD Transfer Pricing Guidelines in the analysis of
this issue:
3.60 If the
relevant terms of the controlled transaction (e.g. price or margin) are within
the arm's length range, no adjustment is necessary.
This is what
happened in 2011, so the inspection does not regularise
this year. For the years 2012 and 2013, the margin calculated by the
inspectorate is outside the range, so we should refer to the following
paragraphs:
3.61 If the
relevant terms of the controlled transaction (e.g. price or margin) are outside
the arm's length range determined by the tax administration, the taxpayer
should be given the opportunity to argue how the terms of the controlled transaction
satisfy the arm's length principle, and whether the result falls within the
arm's length range (i.e. the arm's length range is different from that
determined by the tax administration). If the taxpayer is unable to demonstrate
these facts, the tax administration must determine the point within the arm's
length range to which to adjust the condition of the controlled transaction.
3.62 In
determining this point, where the range comprises highly reliable and
relatively equal results, it can be argued that either satisfies the arm's
length principle. Where some defects in comparability persist, as discussed in
paragraph 3.57, it may be appropriate to use measures of central tendency that
allow this point to be determined (eg median, mean or
weighted mean, depending on the specific characteristics of the data) in order
to minimise the risk of error caused by defects in
comparability that persist but are not known or cannot be quantified.
It is on the basis
of this last paragraph that the Inspectorate justifies the application of the
median for the years 2012 and 2013. However, the Guidelines provide for the
application of the median "where defects in comparability persist, as
discussed in paragraph 3.57":
3.57 It may also
happen that, despite having made every effort to exclude items with a lower
degree of comparability, a range of figures is reached for which it is
considered that, taking into account the process used to select comparables and
the limitations of the information available about them, they still contain
some defects in comparability that cannot be identified or quantified, and are
therefore not susceptible to adjustment. In these cases, if the range has been
obtained across a large number of observations, statistical tools that allow
the range to be narrowed by reference to central tendency (e.g. interquartile
range or other percentiles) can help to improve the reliability of the
analysis.
In the present
case, the inspectorate has accepted the comparability study of the company
without noting any shortcomings in the study. It only notes, perhaps as a
justification for the unreliability of the company's information, that:
It should be
clear, therefore, that, according to the background information in the file, at
no time has group X commissioned or agreed to have its costs and other elements
determining the group's internal data, including its own costs, verified by an
independent third party, prior to their provision to the entity responsible
(...) for preparing the documentation on related-party transactions, provided
in the course of this verification.
This observation,
in any event, referring to the entity's costs, does not constitute a defect in
the comparability of the study carried out by the taxpayer, since, as we have
seen, the method and the search for comparable companies have been admitted by
the inspection, as have the interquartile ranges resulting from this selected
sample.
In conclusion,
once it has been established that the appellant's margins in the years under
discussion are outside the lowest interquartile range, the corresponding
adjustment should indeed be made. However, the fact that this is the case does
not, without more, allow the median to be applied in the terms provided for in
Rule 3.62, since the application of that rule is not justified by the fact of
being outside the arm's length range, but by the existence of 'shortcomings in
comparability', which have not been explained by the inspectorate, so that the
application of the median is not justified.
This is in line
with what is stated in the Judgment of the Audiencia Nacional of 06-03-2019
(appeal number 353/2015), which, after some illustrative reasoning in this
regard, concludes (only this conclusion is extracted, without prejudice to the
interest of the full Judgment, to which we refer):
"Now, in our
opinion, it is clear that, if the ROS is outside the limits of the
inter-quantile range, the corresponding adjustment must be made, since only
from 2.1% is the company within the comparable market margins. In order for the
median to be applied, however, there must also be 'comparability defects'.
In the present
case, no defects of comparability have been revealed in the study carried out
by the taxpayer, which, as we have said, has been accepted by the inspectorate,
so that the application of the median is not in accordance with the law and,
consequently, we must uphold the claimant's claims and annul the settlement.
SEVENTH- The
interested party also alleges errors in the calculation of interest for late
payment.
Once the
settlement has been annulled, it is not appropriate to analyse
the calculation of the interest for late payment.
EIGHTH - Once the
assessment has been annulled, the penalty arising therefrom must also be
annulled.
In view of the
foregoing
This
Economic-Administrative Tribunal agrees to uphold the present complaint,
annulling the contested acts.