The Bombay High Court has
recently given an important judgment in Harkirat Singh v Rabobank.
It has revisited the law on the implied exclusion of Part I of the Indian
Arbitration and Conciliation Act, 1996, which we have considered this on a
number of occasions on this Blog. The
point remains of great practical importance even though Bhatia
International has been
overruled because it is a live issue for any dispute arising under an
arbitration agreement entered into before 6 September 2012.
recently given an important judgment in Harkirat Singh v Rabobank.
It has revisited the law on the implied exclusion of Part I of the Indian
Arbitration and Conciliation Act, 1996, which we have considered this on a
number of occasions on this Blog. The
point remains of great practical importance even though Bhatia
International has been
overruled because it is a live issue for any dispute arising under an
arbitration agreement entered into before 6 September 2012.
As readers will know, the context
in which this problem arises is the Supreme Court’s conclusion in Bhatia
International that Part I of
the 1996 Act applies even if the seat of arbitration is outside India. In
paragraph 32 of its judgment, the Court said that the parties could, by
agreement, expressly or impliedly exclude Part I. At first sight, this seems no
more than a question of construction which should be approached in the light of
the modern approach to interpretation, which is associated in particular with Lord
Hoffmann. However, it has given rise to a long and (one might respectfully
observe) unsatisfactory line of authorities which are not entirely consistent. Indeed,
one can find support in this line of cases for (at least) all of the following
propositions: (i) the choice of a
foreign law to govern the main contract does not impliedly exclude Part I (Citation, Indtel);
(ii) the choice of a foreign seat does
not impliedly exclude Part I (Yograj, Dozco—see
further below); (iii) the choice of
a foreign seat coupled with a foreign
law governing either the arbitration agreement (Reliance, Videocon) or the conduct of the
arbitration (Yograj) impliedly excludes Part I. It
is not clear that these propositions are correct. For one thing, if, as the
Supreme Court appears to accept, the choice of a foreign lex arbitri impliedly excludes Part I, it is difficult to see why
this is the case only when the foreign lex
arbitri is chosen, as in Yograj (through Rule 32 of the SIAC
Rules 2007*). As Hamblen J has
recently pointed out, the choice of a foreign seat is itself usually an implied choice of that country’s law as the lex arbitri: so the Supreme Court’s own
reasoning dictates that the mere choice of
a foreign seat impliedly excludes Part I, though this is a conclusion it has thus
far rejected.
in which this problem arises is the Supreme Court’s conclusion in Bhatia
International that Part I of
the 1996 Act applies even if the seat of arbitration is outside India. In
paragraph 32 of its judgment, the Court said that the parties could, by
agreement, expressly or impliedly exclude Part I. At first sight, this seems no
more than a question of construction which should be approached in the light of
the modern approach to interpretation, which is associated in particular with Lord
Hoffmann. However, it has given rise to a long and (one might respectfully
observe) unsatisfactory line of authorities which are not entirely consistent. Indeed,
one can find support in this line of cases for (at least) all of the following
propositions: (i) the choice of a
foreign law to govern the main contract does not impliedly exclude Part I (Citation, Indtel);
(ii) the choice of a foreign seat does
not impliedly exclude Part I (Yograj, Dozco—see
further below); (iii) the choice of
a foreign seat coupled with a foreign
law governing either the arbitration agreement (Reliance, Videocon) or the conduct of the
arbitration (Yograj) impliedly excludes Part I. It
is not clear that these propositions are correct. For one thing, if, as the
Supreme Court appears to accept, the choice of a foreign lex arbitri impliedly excludes Part I, it is difficult to see why
this is the case only when the foreign lex
arbitri is chosen, as in Yograj (through Rule 32 of the SIAC
Rules 2007*). As Hamblen J has
recently pointed out, the choice of a foreign seat is itself usually an implied choice of that country’s law as the lex arbitri: so the Supreme Court’s own
reasoning dictates that the mere choice of
a foreign seat impliedly excludes Part I, though this is a conclusion it has thus
far rejected.
In Harkirat Singh, a Share Subscription
Agreement (‘SSA’) designated London
as the seat of arbitration and Indian law as the law governing the main
contract. It also chose the UNCITRAL Arbitration Rules. In due course, the
Tribunal passed an award which neither party challenged in the English courts
(the supervisory court). Mr Singh then filed a petition under section 34 of the
1996 Act in the Bombay High Court seeking to have the award set aside as
contrary to Indian public policy. Notably, the SSA did not specifically choose
the English Arbitration Act 1996 but it was plain that that Act was the lex arbitri in view of the choice of
London as the seat and this is likely to be the case even if the 1996 Act had
not contained any mandatory provisions. The Bombay High Court held that the
parties had impliedly excluded Part I of the Act. It thought that this
conclusion was consistent with Yograj even though (as it recognised) the parties had specifically chosen
the lex arbitri:
Agreement (‘SSA’) designated London
as the seat of arbitration and Indian law as the law governing the main
contract. It also chose the UNCITRAL Arbitration Rules. In due course, the
Tribunal passed an award which neither party challenged in the English courts
(the supervisory court). Mr Singh then filed a petition under section 34 of the
1996 Act in the Bombay High Court seeking to have the award set aside as
contrary to Indian public policy. Notably, the SSA did not specifically choose
the English Arbitration Act 1996 but it was plain that that Act was the lex arbitri in view of the choice of
London as the seat and this is likely to be the case even if the 1996 Act had
not contained any mandatory provisions. The Bombay High Court held that the
parties had impliedly excluded Part I of the Act. It thought that this
conclusion was consistent with Yograj even though (as it recognised) the parties had specifically chosen
the lex arbitri:
We are of the clear
view that in the present case, the parties, having chosen the juridical seat of
arbitration at London, had subjected themselves to [sic] the English law by
virtue of section 2, read with section 4 and sections 67 and 68 of the English
Arbitration Act 1996. We are, therefore, of the firm view that the award in
question could be challenged by the appellant only in the English courts…and
not under section 34 of the Arbitration and Conciliation Act 1996…It is true,
as submitted by Mr Advani, that in Videocon
and Dozco the agreement was specific
with reference to the law governing the arbitration. However, that would make
no difference. All these judgments clearly lay down, including the judgment in Yograj, that once the parties have
chosen the law governing the arbitration
to be foreign law, then Part I is necessarily excluded.
view that in the present case, the parties, having chosen the juridical seat of
arbitration at London, had subjected themselves to [sic] the English law by
virtue of section 2, read with section 4 and sections 67 and 68 of the English
Arbitration Act 1996. We are, therefore, of the firm view that the award in
question could be challenged by the appellant only in the English courts…and
not under section 34 of the Arbitration and Conciliation Act 1996…It is true,
as submitted by Mr Advani, that in Videocon
and Dozco the agreement was specific
with reference to the law governing the arbitration. However, that would make
no difference. All these judgments clearly lay down, including the judgment in Yograj, that once the parties have
chosen the law governing the arbitration
to be foreign law, then Part I is necessarily excluded.
The Bombay High Court’s analysis
is largely correct in principle. However, it is respectfully submitted that three
conclusions are questionable. First, the Court relies on the fact that the English
Act is applicable in terms in order to find that Part I is impliedly excluded:
indeed it refers to particular provisions of the English Act to establish that
that Act applies to an arbitration with an English seat. The better analysis is
that the choice of London as a seat has a negative
significance: regardless of the applicability of the English Act, it
demonstrates that the parties did not wish
the Indian courts to supervise the arbitration. Secondly, the Bombay High
Court’s conclusion that the choice of a London seat is on its own enough to
impliedly exclude Part I (though correct in principle) is inconsistent with
four Supreme Court cases: Videocon, Yograj, Dozco and Reliance. It is inconsistent with Reliance, Videocon and Dozco because
there the Supreme Court held that the parties had impliedly excluded Part I by
choosing, in addition to a foreign
seat, a foreign law (English and Korean respectively) to govern the arbitration
agreement. It is inconsistent with Yograj
because the Supreme Court in that case held that the parties had impliedly
excluded Part I by choosing, in addition to
a foreign seat (Singapore), the SIAC Rules 2007 to govern the arbitration, which
provided that the lex arbitri would
be the Singapore International Arbitration Act 2002. Thirdly, the Bombay High
Court does not distinguish between the law governing the arbitration and the law governing the arbitration agreement: for jurisdictional questions
(such as implied exclusion), it is the former that is likely to be decisive.
is largely correct in principle. However, it is respectfully submitted that three
conclusions are questionable. First, the Court relies on the fact that the English
Act is applicable in terms in order to find that Part I is impliedly excluded:
indeed it refers to particular provisions of the English Act to establish that
that Act applies to an arbitration with an English seat. The better analysis is
that the choice of London as a seat has a negative
significance: regardless of the applicability of the English Act, it
demonstrates that the parties did not wish
the Indian courts to supervise the arbitration. Secondly, the Bombay High
Court’s conclusion that the choice of a London seat is on its own enough to
impliedly exclude Part I (though correct in principle) is inconsistent with
four Supreme Court cases: Videocon, Yograj, Dozco and Reliance. It is inconsistent with Reliance, Videocon and Dozco because
there the Supreme Court held that the parties had impliedly excluded Part I by
choosing, in addition to a foreign
seat, a foreign law (English and Korean respectively) to govern the arbitration
agreement. It is inconsistent with Yograj
because the Supreme Court in that case held that the parties had impliedly
excluded Part I by choosing, in addition to
a foreign seat (Singapore), the SIAC Rules 2007 to govern the arbitration, which
provided that the lex arbitri would
be the Singapore International Arbitration Act 2002. Thirdly, the Bombay High
Court does not distinguish between the law governing the arbitration and the law governing the arbitration agreement: for jurisdictional questions
(such as implied exclusion), it is the former that is likely to be decisive.
These points do not mean that the
Bombay High Court’s analysis is wrong: indeed, it is submitted that the view it
has expressed is preferable as a matter of principle, although, given the
current state of the authorities, it is one that is not open at any level below
a Constitution Bench of the Supreme Court. The reason (in principle) the mere
choice of a foreign seat should be taken to exclude Part I is that it is a
powerful indication of which courts the parties intended should supervise the arbitration. This, in a
different context, is the point that Hamblen J makes in his recent judgment in Shagang, which we will consider in a
separate post.
Bombay High Court’s analysis is wrong: indeed, it is submitted that the view it
has expressed is preferable as a matter of principle, although, given the
current state of the authorities, it is one that is not open at any level below
a Constitution Bench of the Supreme Court. The reason (in principle) the mere
choice of a foreign seat should be taken to exclude Part I is that it is a
powerful indication of which courts the parties intended should supervise the arbitration. This, in a
different context, is the point that Hamblen J makes in his recent judgment in Shagang, which we will consider in a
separate post.
* The distinction between express
and implied choice is not as straightforward as it is often thought to be. In Yograj, the so-called ‘express’ choice
of the Singapore Act was in fact a result of the parties having chosen the SIAC
Rules, which provided that that Act would apply. It is hard to see how this is
any more ‘express’ than the choice of a foreign seat.
and implied choice is not as straightforward as it is often thought to be. In Yograj, the so-called ‘express’ choice
of the Singapore Act was in fact a result of the parties having chosen the SIAC
Rules, which provided that that Act would apply. It is hard to see how this is
any more ‘express’ than the choice of a foreign seat.