Business enterprises with
presence in multiple geographical locations either set up an independent legal
entity (recognised under the host state laws) or operate through extended
establishments such as branch offices, installation / project sites, personnel,
etc. (referred to as multi-location entity – MLE). Such MLEs give rise to
critical tax consequences in both the host state and the parent states. While
the general economic principle in framing fiscal laws for such presence is to
ensure avoidance of double taxation and non-taxation, due to a lack of
consensus on international transactions on the VAT / GST front this objective
is far from being achieved. The issue of taxation is relatively simpler when
the business enterprises set up an independent subsidiary and remunerate them
at arm’s length for all their activities. The complexity arises with presence
through extended establishments as the arrangements between the head office and
these establishments are not clearly discernible or documented for comparison
with external world transactions.
VAT laws globally have termed these extensions as
‘fixed establishment’ – this has been used as a tool to assist them in
identifying the end destination of the service or intangibles. The OECD VAT /
GST guidelines have suggested three approaches in taxation of MLEs for services
and intangibles and to reach the end consumption of the service: (i) Direct use
approach where the focus is on the establishment that uses the services; (ii) Direct delivery approach where the focus
is on the establishment to which the services are delivered (with a presumption that delivery is a good reference
point of use); and (iii) Recharge approach which factors the inefficiencies in
both approaches and requires the MLE to recharge the externally procured costs
to the establishment which ultimately uses the services in order to ensure that
the VAT chain continues to the state of consumption1.
1 It
is probable that the recharge approach has weighed heavily while drafting
Schedule 1 of the CGST Act
The Income Tax law (along with tax treaties) has
recognised the presence of foreign enterprises in India through business
connections / permanent establishments (PE) for the purpose of taxation. The
term ‘permanent establishment’ has been used with a dual purpose – (a) fixing
the taxing rights over the source of income pertaining to the extended presence
in India; and (b) application of Transfer Pricing Regulations for ascertainment
of arm’s length profits of permanent establishments through a process of profit
attribution. Being a nation-wide law, domestic branches really do not alter the
revenue situation from an income tax perspective and have no significance in
this respect.
Under the Indian GST law, not only do we have to
resolve transactions spanning across national borders, we also have to deal
with transactions over state borders. Article 286 of the Indian Constitution
empowered Parliament to play the role of a mediator for inter-state transactions.
Article 286 articulates the location of the supply and also defines the state
which would have exclusive jurisdiction to tax the supply transactions. To give
effect to this geographical jurisdiction, the IGST law, in addition to other
principles, included the concept of ‘fixed establishment’. The objective of
this term is to identify the taxable person, the source and destination of a
supply and the appropriate jurisdiction for taxing the transaction.
Before venturing into a comparative analysis of
fixed establishment (FE) with permanent establishment (PE), it is imperative
that the underlying objectives of both laws are understood clearly. The VAT /
GST laws have introduced the said concept in order to give effect to the
destination principle, i.e., taxation revenues ultimately accrue to the state
in which the services are consumed and the neutrality principle, i.e.,
non-resident and residents are treated in the same manner, while the Income Tax
laws introduced the concept of permanent establishments in line with the
‘source principle’ of taxation, i.e., the country from which the income has
been generated should have the right to tax the said income. This fundamental
divergence would act as a natural impediment to automatic application of the PE
definition to the FE definition.
FIXED ESTABLISHMENT UNDER GST
– CONCEPT
In general, the GST law defines the term FE in
contra-distinction with the term ‘place of business’. The phrase place of
business (PoB) refers to a place from where business is ordinarily carried out
and includes a warehouse, godown or any other place where the supplies are made
or received. FE has been defined to mean a place (other than the registered
place of business) which is characterised with a sufficient degree of permanence
in terms of human and technical resources in order to supply or receive and use
services for its own needs. The terms PoB and FE are relevant to decide the
‘from’ location and the ‘to’ location of the service activity and consequently
the right location of supplier and recipient of service.
The PoB / FE concept also has some background in
the Service Tax enactment and the Place of Provision of Services Rules. With
similarly worded terms, the education guide on service tax explained the
objective behind the concept of fixed establishment as follows. It was stated
in paragraph 5.2.3 that the term ‘location’ was significant from the
perspective of service provider and recipient in order to ascertain the source
or rendition of a particular service. The paragraph also stressed that the
location also assists in identifying the jurisdiction of the field formation
(under Service Tax being a Central enactment) which would have domain to assess
the transaction. This can probably also support the point that the fixed
establishment concept would play a pivotal role in identifying the appropriate
state to which the taxing domain lies.
FIXED ESTABLISHMENT UNDER
EU-VAT – CONCEPT
The Indian GST definition of fixed establishment
has been influenced by the EU-VAT law. The EU Sixth VAT Directive (Article 43)
adopted the term ‘fixed establishment’ for ascertaining the place of supply of
services. The term was objectively expanded in 2011 by virtue of Articles 11(1)
and 11(2) of the implementing regulations (a procedural law). Prior to this
expansion, certain decisions analysed the definition of fixed establishment,
i.e., the Berkholz2 and ARO Lease3 cases. The Berkholz
case involved gaming machines installed by the taxpayer on on-going sea vessels
with intermittent presence of personnel for maintenance of the machines. The
ECJ on consideration of the cumulative requirement of permanence of human
and technical resources held that the FE was not established. Similarly, in
the ARO Lease case a company was engaged in leasing of cars to its customers in
Belgium (these cars were purchased by ARO, Netherlands from Belgian dealers and
delivered to the customers directly in Belgium); it was held by the ECJ that
neither the presence of cars of ARO in Belgium nor the self-employed intermediaries
(Belgium dealers) created a sufficient permanent human and technical presence
to constitute an FE.
With Articles 11(1) and (2) of the EU implementing
regulations, the EU-VAT law has categorised fixed establishment into ‘passive’
and ‘active’ fixed establishments4. The IGST law has adopted both in
one definition with the use of the disjunctive phrase ‘or’ in the last few
words of the definition (discussed in detail later). This may be considered as
essential because the concept of fixed establishment has been used to ascertain
both the location of cases where services are either received or provided from
the said fixed establishment.
After the introduction of the implementing
regulation, the Welmory case5 examined a situation where a Cyprian
company had appointed a Polish company to maintain and operate a website for
online auction. The entire process of auction was functioning through the
website maintained by the Polish company. The ECJ stated that the economic
activities of the Polish company and Polish customers does not by itself
constitute a fixed establishment and the services of the Polish company should
be viewed distinctly from that of the Cyprian company to the Polish customers.
PERMANENT ESTABLISHMENT UNDER
INCOME TAX – CONCEPT
In Income Tax, the domestic legislation used the
phrase ‘business connection’ which is of very wide amplitude, but taxpayers use
the benefit of a narrower term PE6 which is used in most tax
treaties (OECD, US or UN model). The term PE has been defined as a fixed place
of business through which the business of an enterprise is wholly or partly
carried out and includes the following: (a) place of management; (b) branch,
office, factory, workshop, warehouse, etc., (c) building, construction or
installation site; (d) provision of services through presence of personnel; and
(e) dependent agency. In case a foreign enterprise constitutes a PE in a host
state, the PE is to be granted a separate entity status and business profits
attributable to the PE at arms’ length are to be taxed in the state in which
the PE is constituted.
Having understood the broad concept of FE and PE in
GST / EU and Income Tax law, we now proceed to identify the points at which the
said terms would converge / diverge through the assistance of illustrations and
then tabulate the same for future reference.
2 ECJ
EC 4th July, 1985, 168/84, ECLI:EU:C:1985:299 (Günter Berkholz),
European Court Reports, 1985
3 ECJ
EC 17th July, 1997, C-190/95, ECLI:EU:C:1997:374 (ARO Lease),
European Court Reports, 1997
4 Passive
FEs being those which only receive / procure inputs / services (cost centres)
and Active FEs which also provide services (profit centres)
5 ECJ
EU 15th October, 2014, C-605/12, ECLI:EU:C:2014:2298 (Welmory),
Official Journal 2014, C 462
6 OECD
Model Convention on Tax Treaties as an example
CONSTITUTION OF A PE / FE –
COMPARATIVE ANALYSIS
(A) Fixed place PE (basic rule): Stability,
productivity and dependence
Article 5 of Income Tax treaties defines a fixed
place PE as ‘fixed place of business’ through which the business of an
enterprise is wholly or partly carried out. The fixed place PE rests on three
primary tests: (a) place of business; (b) location test; and (c) permanence
test. Article 5(2), in fact, enumerates instances such as branch, office,
factory, workshop, warehouse, etc., which by default satisfy the above tests,
especially the place of business test.
The place of business test
postulates that some portion of the business activity of the MLE enterprise is
conducted through the physical office in India. The extent of business activity
conducted in India is ascertained through a FAR analysis (Functions performed,
Assets employed and Risks assumed) of the PE’s operations in India. But there
is one critical exclusion in terms of conduct of preparatory or auxiliary
activities – in effect, where a part of the business activity is conducted in a
territory and such part is preparatory or auxiliary in nature, the MLE is not
considered to have a permanent establishment in the said territory. In the
context of this Basic Rule PE, the Supreme Court in DIT vs. Morgan
Stanley [2007 (7) TMI 201] was examining whether back-office operations
of a subsidiary constitute a place of business of the parent company. The Court
held that the support function performed would not constitute an extension
of the business activity of the parent. Similarly, the mere fact that the
business of the parent company is outsourced or support functions are performed
by the Indian subsidiary, was considered as irrelevant for the purpose of concluding that the parent company is having
an establishment in India [ADIT vs. E Funds IT Solutions Inc. 2017 (10)
TMI 1011]. These decisions imply that the business activity should be
understood as an extension of the set-up already in place in the parent company
rather than an independent entity.
The location test requires that there has to be a
physical presence of the business in a specific place (though the place may be
mobile within the defined territory). This test warrants that the business
activity should be capable of being tagged to the physical territory either as
equipment, branch or any such physical infrastructure.
As for the permanence test,
the OECD commentary states that the term ‘fixed’ itself warrants a certain
degree of permanence at the location in which the physical infrastructure is
placed in the taxable territory. The phrase ‘fixed’, as an adjective, also
attaches a certain degree of permanence to the place of business which is a
prerequisite under this Article (also stated in the OECD commentary)7.
In the absence of a defined time period under the treaties, there is some
relativity in the way jurisprudence has developed to identify the degree of
permanence of an activity. While each case produced different results, a recent
decision of the Supreme Court in Formula One World Championship Ltd. vs.
CIT, Delhi [2017] 80 taxmann.com 347 (SC) examined a Formula One race
arrangement which lasted for approximately three weeks and held that such an
arrangement constituted a fixed place PE. The background to the case involved a
Formula One race conducted by FOWC which involved placement of the entire racing
infrastructure / equipment on Budd International Circuit and FOWC had complete
control over the schedule, equipment and personnel at the location. The Supreme
Court stressed on the disposal test (i.e., FOWC having exclusive and complete
control over the racing circuit) rather than the duration test for testing the
permanence of an activity. In other words, ‘fixed’ was interpreted by the
Supreme Court with reference to the exclusivity and control over the place
rather than the duration of usage (which is relative for each business
activity).
In comparison, the FE definition in IGST law also
relies on a ‘sufficient degree of permanence’. The legislature has used a very
subjective term probably keeping in view varied industry practices. For
example, the Formula One race arrangements which last only for three weeks in a
calendar year at a particular location, was considered as a sufficient degree
of permanence by the Court given the peculiarity of the sporting event.
Moreover, the Court stressed on the adequacy of control over the three-week
period rather than the duration of three weeks. Though speculative, it is
probable that the Court might not have reached the same result in case the
facts were for a long duration project (e.g., oil exploration activities,
etc.). Probably, this test would also be applicable for understanding the FE
definition. Whether the phrase ‘sufficient degree of permanence’ has to be
assessed keeping the disposal test or the duration test in mind, is a question
open for debate.
7 CIT
vs. Visakhapatnam Port Trust 1983 (6) TMI 31
Further, the disposal test
should also be addressed factoring in the nature of supply which is under
consideration, for example, presence of a few weeks would be sufficient for
rendering legal advisory services on a particular issue but a presence of even
a few months would not be sufficient while rendering the very same advisory
services for construction of a building. The service tax education guide states
that the relevant factor is the ‘adequacy’ of the human and technical resources
to render the ‘service’ for deciding whether there is sufficient permanence in
the activity. In the absence of detailed jurisprudence on this subject, one may
imbibe the settled principle under Income Tax while interpreting the FE
definition.
(B) Service PE (Extended PE): Physical presence of
employees
The OECD Model Convention
provides for a service PE if the aggregate presence of the personnel in the
taxable territory is beyond 183 days in a 12-month period8. The
nature of services that are rendered by the personnel could be wide in range.
Physical presence of personnel in the taxable territory rendering a service,
irrespective of the type of service, would be prominent in deciding the
constitution of a service PE. But the Supreme Court in Morgan Stanley
(Supra) differentiated the services rendered through own employees of
MS USA and deputed employees of MS USA. The Court held that in the case of own
employees performing supervisory / stewardship activities, there is no service being rendered by the presence
of employees in India to the Indian subsidiary, rather the presence of
personnel is for the sole benefit of the parent company in order to ensure
quality control, confidentiality, etc. On the other hand, the employees of MS
India deputed for assisting the operations of MS India were considered as an identifiable service activity to the
Indian subsidiary, hence a service PE was held to be constituted.
8 UN
Model narrows the test to a 6-month period
Under GST, the reference to
personnel is made in the definition of fixed establishment by use of the phrase
‘human resources’. But this aspect is not a stand-alone requirement for
constitution of a fixed establishment. The human capital is required to be
equipped with a degree of permanence and technical resources (depending on the
nature of work) in order to constitute a fixed establishment. The permanence
test applicable to Fixed Place PE may be adopted while examining the FE arising
out of service personnel. It is also important that the human capital should be
‘capable of availing and rendering the service activity’ to the third party
while being present in the taxable territory.
We may recall that the EU-VAT in Berkholz’s case
held that the presence of personnel to maintain the gaming equipment on an
intermittent basis did not constitute a fixed establishment in the foreign
territory. Probably, the IGST law may also have to follow suit and despite a
service PE being constituted under Income Tax, if the human capital is using
technical resources of the end customer there may be a ground to contend that a
fixed establishment is not constituted. From a practical standpoint, many MNCs
appoint personnel for maintenance and on-site repair of machinery. Where repair
and maintenance is partially conducted from the Indian territory with a
significant portion being conducted remotely, one may view the same as not
constituting an FE in India, but where the said personnel (with their
equipment) are capable of rendering the service exhaustively, the said
personnel can be said to have constituted a PE in India.
(C) Agency PE (non-obstante rule):
Dependence (DAPE)
The concept of agency PE was introduced to address quasi-presence
of foreign enterprises through dependent representatives in the taxable
territory. Notwithstanding the requirements of a basic rule PE, Article 5
provides for formation of a dependent agency PE where, (a) the person
habitually concludes contracts or even possesses the authority to conclude
contracts; (b) maintains inventory of goods on behalf of its principal; or (c)
habitually secures orders wholly or primarily for its principal in India. The
primary requirement is that a principal-agency relationship is visible and such
agent, if at all, should be one who is NOT operating in his independent
capacity in his ordinary course of business. Evidently, a principal-agent
relationship would be established if the agent acts in a representative
capacity with an ability to bind its principal to its actions.
The Bombay High Court in CIT vs. Taj TV
Limited [2020 (3) TMI 500] examined whether revenues under the
exclusive distribution agreement by Taj TV Limited involving fees from cable
operators and granting them viewing and relay rights, were under an agency
relationship. Under the distribution agreement, Taj TV was given independent
rights to market and promote the TV channels and negotiate and conclude
contracts with sub-distributors. In this decision, the Court observed that Taj
TV under sub-distributor agreement and the cable-operator agreement, acted in
its own capacity and the foreign enterprise did not have any privity in
deciding the pricing of the rights. Hence it was concluded that there was a
principal-to-principal relationship. In contrast, the very same Bombay High
Court in DIT Mumbai vs. B4U International Holdings Limited 2015 (5) TMI
277 on those specific facts held that the Indian entity did not have
any authority to conclude contracts on its own and was under the direction and
control of its principal and hence constituted a DAPE in India.
The secondary requirement of constitution of a DAPE
is that the agent should not be of independent status. ‘Legal and economic
independence’ are the two tests which are usually applied for this clause.
International commentaries suggest that if the agent is responsible for its own
actions, takes risks and has its own special skill and knowledge to render the
agency service, such agent would be termed as an independent agent. In certain
AAR rulings9 under Income Tax, the dependency agency test was
applied and held not to be fulfilled on the ground that similar services were
being provided to multiple FIIs and not just one principal. Moreover, the
treaties itself provides for an exclusion where the agent renders its services
to multiple principals, evidencing that it has risk abilities, own skill and
knowledge, etc. to act independently.
As regards the authority to conclude contracts,
paragraph 33 of the OECD commentary states that a person who is authorised to
negotiate all elements of a contract in a binding way on the enterprise can be
said to have the authority to conclude contracts and the mere fact that the
person has attended and participated in negotiations is not sufficient
authority under this PE rule10.
9 XYZ/ABC Equity Fund vs. CIT (2001) 116 Taxman
719 (AAR); Fidelity Services Series VIII in (2004) 271 ITR 1 (AAR)
10 India
has expressed its reservations to the OECD commentary on this aspect and
submitted that mere participation in negotiations is also evidence of authority
to conclude contracts
In contrast, the IGST law does not have a specific
case of agency for the purpose of fixed establishment for a supplier or its
recipient. It would be interesting to note that section 2(105) of the CGST Act
has extended the definition of a supplier to include its agent as well who is
engaged as supplier of services. By virtue of this inclusion, the location of
the supplier would have to be adjudged after taking cognisance of the presence
of an agent in India. As an illustration, if Taj TV was appointed as an agent
of Taj Mauritius for distribution of channels, the location of the supplier for
the services rendered by Taj Mauritius (i.e., channel access fee) would be
liable to be attributed to the Taj TV (as an agent) of Taj Mauritius resulting
in Taj Mauritius being considered as present in India through the FE-agency
relationship. Unlike Income Tax, the agency relationship need not always be one
who is dependent on the principal. The place of business of the agent would be
considered as the place of business of the principal (to the extent of the
agency relationship) and fixed as the location of supplier of such services. Of
course, the basic FE rule under GST such as permanence, human and technical
resources would still have to be satisfied by the agent in order to qualify as
an FE in India.
The location of supplier and
recipient definitions has been limited only for the service activity and not
transactions of supply of goods. One of the reasons may be that the agency FE may
not be applicable to selling / purchasing agents of goods. This is because
agency transactions in respect of supply of goods are covered under Schedule I
of the CGST Act which deems the principal and agent as separate beings. Once
the agent is equated to a purchaser of goods, by way of a deeming fiction, the
agent would be considered as a quasi-supplier for other purposes of the
Act and not as a representative of its principal. Hence, an agency PE under
Income Tax for supply of goods would not translate into an FE for the very same
principal under GST.
(D) Construction / Installation PE
Tax treaties recognises any construction,
installation, project site, etc., including supervisory activities11
of a foreign enterprise as being construction / installation PEs. The treaties
(OECD – 12 months; UN – 6 months) specify the period beyond which the PE is
said to be constituted. All planning and designing activity prior to physical
visit to the site would be included in the PE (UN Model). The Tribunal in SAIL
vs. ACIT (2007) 105 ITD 679 held that supervisory services provided by
a company even though it was itself not engaged in installation activity would
be sufficient to constitute an installation PE.
The IGST law has not carved out a specific instance
of a construction / installation FE. The fixed establishment definition itself
encompasses any presence of human and technical resources as sufficient grounds
to constitute an FE. Moreover, construction activities generally entail direct
procurement and supply of services and hence all the ingredients of
constituting an FE stand established. Unlike the situation in SAIL’s case above
which involved mere provision of supervisory services, such activities may not
constitute an FE in GST law.
11 Only in UN model treaties
(E) Significant Economic Presence (SEP): Digital
footprint
With increasing digital presence globally, BEPS
Action Plan 1 identified the problems of taxation due to advancement of digital
economy. The concept of significant economic presence (SEP), which could be
represented by the digital footprint of an enterprise, was given legal sanctity
in the Income Tax law. The law has prescribed minimum thresholds on the basis
of digital footprint for constituting significant economic presence in the
country – the thresholds could take the form of number of digital users,
revenue per user, etc. The BEPS Action Plan as well as the memorandum
suggesting this amendment stated that the traditional concept of permanent
establishment requiring physical presence in the taxable territory cannot be
possibly applied to digital transactions and hence alternative criteria such as
the above are necessary for viewing significant economic presence.
The Indian GST law has not considered
this aspect for the purpose of defining fixed establishment. The law has
defined a term ‘Online Database and Information Access or retrieval service
activity’ to tax all digital economy transactions. In fact, this law has
alternatively chosen to place a tax liability by adopting a different approach
and treating this transaction as an import of service into India and taxing the
same in the hands of the business user. In case of B2C transactions, the GST
law has placed the obligation on the foreign company to identify a
representative in India for discharging the tax burden on such transaction.
In summary, the GST FE concept and the Income Tax PE concept converge and
diverge at multiple points which can be tabulated as follows (see Table 1):
ATTRIBUTION / RECHARGE
APPROACH
Income Tax attributes income /
profit to the PE based on the FAR approach. This would involve estimations and
approximations; profitability is not traceable to the transaction level.
Table 1
|
Aspect
|
Convergence
|
Limited Divergence
|
Complete Divergence
|
|
Fixed
PE
|
Degree
of permanence
|
Disposal
test which does not seem to be a clear prerequisite in FE; no exclusion for
preparatory / auxiliary activity in FE
|
Capability
of receipt and / or supply of services is not required in PE
|
|
Service
PE
|
Rendition
of services
|
Duration
test for PE and FE may be differently viewed
|
Presence
of technical resources are essential for FE
|
|
Agency
PE
|
Agent
constituting PE / FE
|
Agency
in respect of goods not part of FE test due to deeming fiction in Schedule I
|
Dependency
not an FE test ~ multi service agents constituting FE
|
|
Construction
PE
|
Construction
site being place of business
|
Ancillary
activities may form a Construction PE but not an FE
|
No
time limit specified for FE and relative to nature of work
|
|
SEP
|
NIL
|
NIL
|
Treated
as OIDAR and liable for RCM / FCM as the case may be
|
In GST, India has adopted the
recharge approach in cases of service transactions. For example, with respect
to services directly connected to an FE, the requirement under GST law is to
‘bill from’ the FE / ‘bill to’ the FE which is most directly concerned with the
supply. The FE would then have to recharge the relevant establishment in the
organisation which has consumed the service even partially. In such a manner,
the tax revenues would follow the contractual obligations (with the presumption
that economics would assist in reaching the point of consumption) and hence the
importance of identification of the relevant FE would assume high significance.
However, in cases of agency of supply of goods, the
requirement of the law is to deem them as distinct entities at the outset
itself and equate them to a seller or purchaser of goods. By this approach, the
VAT chain passes through the agent and attempts to reach the end consumption of
goods. This activity takes place at the transaction level rather than
the entity level which is the case in Income Tax. The global apportionment
approach / FAR analysis adopted for attribution of profits to the PE would not
serve any purpose for GST.
In summary, it is observed that though both the concepts are
interconnected at multiple points, there are bound to be divergent answers in
view of the basic structure of both the Income Tax and the GST laws being
different. This concept would be relevant for outbound as well as inbound
presence and hence a balanced interpretation of this concept is essential from
the perspective of all stakeholders. In any case, the attempt should be to
identify the last point in the value chain where the consumption actually takes
place. This destination principle, though unwritten in the law, is the core of
the GST system in place and violation of this principle at the cost of legal
interpretation may cause chaos in the economic distribution of the wealth of
the nation.
