[This is a Guest Post by Dhruva Gandhi and Sahil Raveen.]
Last year in Hindustan Construction Company Limited v Union of India (“Hindustan Construction”), the Supreme Court of India delivered a widely welcomed decision in so far as arbitral proceedings in India were concerned. Once an award is passed by an arbitral tribunal under the Arbitration and Conciliation Act, 1996 (“Act”), it can be challenged by any aggrieved party under Section 34 of the Act before a principal civil court having jurisdiction vis-à-vis the seat of arbitral proceedings. The award is executed as per Section 36 of the Act as if it were a decree of a court. For nearly two decades, there has been confusion prevailing with respect to whether the mere filing of an application under Section 34 amounts to an automatic stay of the execution proceedings.
In Hindustan Construction, the Supreme Court clarified this proposition. It said that there mere filing of a Section 34 application does not lead to a stay on execution. However, in the process, it struck down two statutory provisions using the doctrine of ‘manifest arbitrariness’. In this post, we critique the application of this doctrine. Last year, commenting on the IBC decision in Committee of Creditors in Essar v Satish Kumar Gupta, we had remarked how ‘manifest arbitrariness’ is a powerful tool and could lead the Judiciary into unchartered territory. In this post, we show how in Hindustan Construction, the Court has effectively substituted the Legislature’s preferences with its own under the garb of this doctrine.
Section 36 and the conundrum of automatic stay on enforcement of arbitral awards
Section 36 of the Act, prior to the Arbitration and Conciliation (Amendment) Act, 2015 (“2015 Amendment”) said,
36. Enforcement-Where the time for making an application to set aside the arbitral award under Section 34 has expired, or such application having been made, it has been refused, the award shall be enforced under the Code of Civil Procedure, 1908 (5 of 1908) in the same manner as if it were decree of the Court.
The Supreme Court in National Aluminum Company Ltd. (NALCO) v. Pressteel & Fabrications (P) Ltd. and Anr (“NALCO”) interpreted Section 36 to mean that the moment an application challenging an arbitral award is filed under Section 34 of the Act, there will be an automatic stay on the execution of the award. In Fiza Developers and Inter-trade Private Limited v. AMCI (India) Private Limited and Anr, (“Fiza Developers”), the Supreme Court reiterated this position. This position was undesirable for an award holder because an award debtor simply had to lodge an application to evade payment/compliance under the award. With the pendency in Indian legal system, it meant that an award debtor could hold up execution for at least a few years.
Subsequently, by the 2015 Amendment, the Legislature amended Section 36 of the Act to read,
36. (1) Where the time for making an application to set aside the arbitral award under section 34 has expired, then, subject to the provisions of sub-section (2), such award shall be enforced in accordance with the provisions of the Code of Civil Procedure, 1908, in the same manner as if it were a decree of the court.
(2) Where an application to set aside the arbitral award has been filed in the Court under section 34, the filing of such an application shall not by itself render that award unenforceable, unless the Court grants an order of stay of the operation of the said arbitral award in accordance with the provisions of sub-section (3), on a separate application made for that purpose.
(3) Upon filing of an application under sub-section (2) for stay of the operation of the arbitral award, the Court may, subject to such conditions as it may deem fit, grant stay of the operation of such award for reasons to be recorded in writing:
Provided that the Court shall, while considering the application for grant of stay in the case of an arbitral award for payment of money, have due regard to the provisions for grant of stay of a money decree under the provisions of the Code of Civil Procedure, 1908.
Therefore, the 2015 Amendment clarified in no uncertain terms that the mere filing of a Section 34 application would not lead to an automatic stay on execution.
What is crucial for our purposes though, is Section 26 of the 2015 Amendment. Section 26 said,
26. Nothing contained in this Act shall apply to the arbitral proceedings commenced, in accordance with the provisions of section 21 of the principal Act, before the commencement of this Act unless the parties otherwise agree but this Act shall apply in relation to arbitral proceedings commenced on or after the date of commencement of this Act.
While Section 36 was amended to usher in much needed clarity, Section 26 of the 2015 Amendment led to a confusion of its own. It was unclear whether or not Section 26 implied that the amended Section 36 would also apply to pending court proceedings that had arisen out of arbitrations initiated before 23.10.2015 (the date of enactment of the 2015 Amendment).
Different High Courts offered conflicting opinions. Finally, the Supreme Court in Board of Control for Cricket in India v. Kochi Cricket Pvt. Ltd. (“BCCI”) said that the amendments to Section 36 would apply to pending court proceedings, even if they arose out of arbitrations initiated before 23.10.2015. It said that Section 36 was procedural in nature and therefore, amendments made to it could be applied retrospectively. However, the rest of the 2015 Amendment would apply prospectively.
Notably, in our opinion, (even so far as Section 36 of the Act was concerned) Section 26 could equally fairly have been interpreted to apply only to arbitrations commenced after 23.10.2015 and court proceedings arising thereof. The text of the statute allowed for a literal interpretation of that nature.
Thereafter, the Legislature enacted the Arbitration and Conciliation (Amendment) Act, 2019 (“2019 Arbitration”) and further introduced Section 87 in the Act. Section 87 replaced Section 26 of the 2015 Amendment, which was also deleted. Section 87 said,
87. Unless the parties otherwise agree, the amendments made to this Act by the Arbitration and Conciliation (Amendment) Act, 2015 shall—
(a) not apply to––
(i) arbitral proceedings commenced before the commencement of the Arbitration and Conciliation (Amendment) Act, 2015;
(ii) court proceedings arising out of or in relation to such arbitral proceedings irrespective of whether such court proceedings are commenced prior to or after the commencement of the Arbitration and Conciliation (Amendment) Act, 2015;
(b) apply only to arbitral proceedings commenced on or after the commencement of the Arbitration and Conciliation (Amendment) Act, 2015 and to court proceedings arising out of or in relation to such arbitral proceedings.
Evidently then, the Legislature differed with the view espoused in BCCI and sought to take away the basis for that decision.
Constitutional Challenge to Section 87 of the 2019 Amendment
In Hindustan Construction, the constitutionality of Section 87 was challenged. The Supreme Court in an opinion delivered by Nariman J. upheld the constitutionality of the section in so far as it took away the basis for a previous Apex Court decision. However, it struck down the provision for being manifestly arbitrary and thus, in breach of Article 14 of the Constitution of India.
What piques our interest in this post are the reasons offered by Nariman J. to hold Section 87 to be manifestly arbitrary. They were,
- The 2019 Amendment was based on a Report by the Srikrishna Committee. This Committee had said that there were conflicting High Court decisions on Section 26 of the 2015 Amendment and there was a need to clarify the position. However, this Report was delivered in 2017 whereas the BCCI judgement was delivered in 2018. Therefore, according to Nariman J., the Legislature ought not to have relied on this Report once the Supreme Court had decided the issue.
- The Report opined that to make provisions of the 2015 Amendment applicable retrospectively may create confusion and uncertainty in that parties may have to be heard afresh. According to Nariman J. there was little substance to this observation as fresh applications for lifting of stay could be decided forthwith.
- The benefits of Order XLI, Rule 5 of the Civil Procedure Code were available in case of an appeal from a decree but not in case of a challenge to an arbitral award under Section 34 of the Act. Order XLI, Rule 5 says that there shall be no automatic stay on proceedings under a decree merely on filing of an appeal. Section 34 of the Act though, is not an appeal.
- Section 87 does not consider implications under the Insolvency and Bankruptcy Code, 2016. Award holders unable to recover operational debts under an award may themselves have to bear the brunt of an insolvency application from creditors downstream. Therefore, an automatic stay on the execution of an arbitral award would create unwarranted economic hardship.
In our opinion, none of these reasons hold much merit and we shall deal with them in seriatim.
While the Statement of Objects and Reasons of the Arbitration and Conciliation (Amendment) Bill, 2019 does refer to the Srikrishna Committee Report, it primarily does so in the context of institutional arbitration. On Section 87, it simply says that the Legislature intends to clarify the manner in which Section 26 of the 2015 Amendment would apply. Moreover, the tools of statutory interpretation lead us to imply that the Legislature would be presumed to be aware of the BCCI judgement and simply wished to remove a prop of that decision.
Moving on, the observation of the Srikrishna Committee too was not entirely baseless. In some cases where the Section 34 application was at an advanced stage, the court would now have to consider whether or not to issue a stay on the award under Section 36. However, the decision which a court would actually make is whether or not to lift the stay already in place. Considerations that play on a judge’s mind when lifting a stay are quite different from the ones for the issuance of a stay. Even the mere pendency of a Section 34 application would be a consideration. Therefore, there was both a rational and a plausible basis for the Committee to say that retrospective operation of Section 36 of the Act could potentially lead to inconsistent outcomes.
With respect to the point that no automatic stay operates in case of civil decrees, the Legislature was entitled to create different regimes for arbitral awards as opposed to a decree of a court and then to change the regime with the 2015 Amendment. If at all, one could consider whether this met the twin tests of ‘intelligible differentia’ and ‘rational nexus’ under Article 14. These tests though find no mention in the Supreme Court’s deliberation.
Lastly, the implications under insolvency law that Nariman J. mentions are not unique. It is commonplace to see companies and individuals be rendered insolvent owing to forward payments not having been cleared. Courts expect debtors to factor such recoveries into their transactions. While this is also a point to be dealt with by insolvency law, there is no reasoning offered to show how a one-off economic hardship may lead to a violation of the ‘Right to Equality’ under Article 14 of the Constitution.
Therefore, in our opinion, none of the reasons offered to say that Section 87 is manifestly arbitrary really pass the muster.
The real reason to employ ‘Manifest Arbitrariness’
More than the reasons, what is noteworthy is the language employed by the Court. In our opinion, it is the language which seems to be indicative of the unspoken premise of the decision.
To refer to the Srikrishna Committee Report (without at all referring to this Court’s judgement) even after the judgement has pointed out the pitfalls of following such provision, would render Section 87 and the deletion of Section 26 of the 2015 Amendment Act manifestly arbitration, having been enacted unreasonably without adequate determining principle and contrary to the public interest sought to be subserved by the Arbitration Act, 1996 and the 2015 Amendment Act….
…. the anomaly, therefore, of Order XLI Rule 5 of the CPC applying in the case of full-blown appeals, and not being applicable by reason of Section 36 of the Arbitration Act, 1996…., is itself a circumstance which militates against the enactment of Section 87, placing the amendments made in the 2015 Amendment Act, in particular Section 36, on a backburner. For this reason also, Section 87 must be struck down as manifestly arbitrary under Article 14….
…The retrospective resurrection of an automatic stay not only turns the clock backwards contrary to the object of the Arbitration Act, 1996 and the 2015 Amendment Act, but also results in payments already made under the amended Section 36 to award-holders in a situation of no-stay or conditional-stay now being reversed.” (emphasis supplied)
There is a distinct preference that comes to the fore in the language employed by the Court. The objects of the 2015 Amendment are worthy, those of the 2019 Amendment are not. The former must be prioritised and preferred over the latter. The 2019 Amendment to the extent that it militates against the 2015 Amendment must be reversed.
However, this cannot be the basis to strike down a provision of law as unconstitutional. A provision cannot be held unconstitutional only because it purportedly reversed the clock on an earlier amendment. During the evolution of the doctrine of ‘manifest arbitrariness’ one of the objections put forth before the Supreme Court was that it would simply result in a replacement of policy choices of the Legislature with those of the Judiciary. The decision in Hindustan Constructions is a prime example of these fears coming to life. With time, what may happen is that the outcome of a case may even boil down to the court in question. Some judges may agree with the policy preferences of the Legislatures, others may not. The doctrine of ‘manifest arbitrariness’ thus carries a lurking danger of being arbitrary. .
Conclusion
Like in Essar, in Hindustan Construction too, in our opinion, Nariman J. could have arrived at the same conclusion through alternate means.
In Hindustan Construction, Nariman J. held both NALCO and Fiza Developers to be per incuriam. It is now a settled position [(1988 2 SCC 602, Paras 40, 183] that once a decision has been held to be per incuriam, it loses precedent value. As a consequence, the interpretation proposed by NALCO and Fiza Developers was never the law. Therefore, Nariman J. could have stated that there has never been an automatic stay of execution proceedings on the filing of a Section 34 application in India. This has always been the law and there has been no change in the interpretation of Section 36. After that, the 2015 Amendment in so far as Section 36 was concerned would have become merely declaratory e- clarifying the position of law by means of abundant caution. Extending that logic, Section 87 would not have to be applied to Section 36 at all because there would be no question of its retrospective operation. Section 87 only clarified when and how the other amendments introduced by the 2015 Amendment ought to apply.
It is our opinion that laws must not be struck down as unconstitutional unhesitatingly, especially when there are alternatives available. Although slightly unconventional, Nariman J. could have explored this route of interpretation to arrive at the same conclusion and thereby, minimised the use of ‘manifest arbitrariness’, the increased usage of which is matter grave concern.
Constitutional Law and Philosophy 
Dhruv, Sahil – disagreements with your post:
1. The declaration on the 2019 amendment being ‘manifestly arbitrary’ was not, as you have contended, based on the Court exercising its subjective opinion on which amendment, i.e. w/c of 2015 or 2019, it ‘preferred’. It was based on the premise that the 2019 amendment directly militated against a fundamental objective of the 1996 Act and the Model Law (and w/c lead to the 2015 amendment): for arbitration proceedings to be time-efficient and suffer from minimal court interference. While you can contest that the 2019 amendment was not counter to these objectives (difficult), it would be incorrect – and contrary to decades of precedent – to say that a court could not have struck down a provision as being arbitrary on that premise.
2. Your contest to Nariman, J.’s dismissal of the Government’s reliance on the Srikrishna report is somewhat misconceived. To clarify – the Court said that the Srikrishna report could not have been used as a basis to bring in the 2019 amendment because that report had already been discussed, and the position of law clarified as a result, in BCCI v. Kochi Cricket. What the government did was to then re-apply (verbatim) the reasoning of the Srikrishna report, in bringing the 2019 amendment, despite the court having been alive to it in an earlier decision. What this demonstrated to the court was a non-application of mind, with the fundamental premise and reasoning of the BCCI decision not having been dislodged by the introduction of Section 87 (or the removal of Section 26). The merits of the Srikrishna report could not, therefore, have been the subject of another analysis.
3. Having declared NALCO and Fiza per incuriam, Hindustan is explicit in stating that Section 36, even prior to amendment, did not have an ‘automatic stay’ mechanism built-in. What this implies is that the law from 1996 onwards was steady, i.e. that a party could seek a deposit of the decree amount from a court irrespective of when the arbitration had commenced (see Para 32 of the SC version). So it is entirely plausible to suggest that the court could have declared that Section 87 would just apply to the remaining amendments made in 2015. What remains, however, is the court’s understanding of the arbitrariness of the 2019 amendment, its lack of determining principle etc. It was incumbent on the court, in my opinion, to then declare the constitutional invalidity of the 2019 amendment, and also to prevent the interpretative inconsistency that has typically plagued the arbitration act.
But there is a strong case to be made that the arbitrariness doctrine itself needs more in-built conditions/limitations to prevent the judiciary from substituting legislative wisdom for their own. On that, we agree.
Hi Praharsh,
Thank you for reading our post and for having taken the time to provide some feedback. We really appreciate it. Here is what we have to say,
1. As you may well know, the doctrine of manifest arbitrariness is of recent import and we do not have much to go by in terms of when a court may strike down a ‘statute’ as arbitrary. Therefore, the observation about ‘decades of precedent’ appears to be somewhat confusing. However, we do feel your point misses a basic premise of what we seek to argue. Even if the 2019 Amendment militated against the objectives of the 1996 Act r/w the Model Law, in our opinion, it does not matter. This fact alone does not make the law unconstitutional. It is true that a premise of the 1996 Act r/w the 2015 Amendment was to make arbitrations time efficient and to minimise the interference of the courts. Even then, a Legislature of one time cannot bind a Legislature of a subsequent time unless a law has been deemed to be entrenched. Put otherwise, our point is this, even if Parliament wanted to make arbitrations time-inefficient and unwieldy, it was entitled to do so. The fact that the Parliament made this choice does not mean that the law is in breach of Article 14.
2. As to point 2, we think this response too misses the premise of what we sought to say. Perhaps, we did not make it amply clear. Firstly, the Statement of Objects and Reasons of the 2019 Bill does not trace Section 87 back to the Srikrishna Committee Report. This is a blank that Nariman J fills. Secondly, the Supreme Court had no business discussing the Srikrishna Committee Report in BCCI and saying that a provision akin to Section 87 (which was then a draft amendment) ought not be introduced. Those portions of the SC judgement were mere obiter. Parliament was under no burden to dislodge an obiter. Given that these observations were made though and assuming that Parliament relied on the Srikrishna Committee Report to introduce Section 87, it was well within its rights to do so.
As we argue (and again, maybe you missed this or we were not amply clear), the reasoning of the Srikrishna Committee was not entirely baseless. They had a rational/reasonable point, i.e. inconsistency in pending cases. Parliament wanted to simply avoid the sheer confusion that may be caused (and was in fact caused in some cases) by the retrospective application of the 2015 Amendment. In BCCI, the SC said that this confusion might not actually come to be. The Parliament felt differently. At the end of the day, the policy call rests with the Legislature. Lastly, ‘non-application of mind’ has been a reason to strike down executive actions (in most cases). We would be happy to be corrected, but we have not come across a case where an ‘Act of Parliament’ has been struck down for non-application of mind. We would go so far as to say that this might set a dangerous precedent if it were the case.
3. We are of the opinion (and you may differ) that a court must be hesitant to strike down a law when an alternate route is available. Para 32 of the SC opinion is of no application here. Your reasoning proceeds along the lines of – once the court found there to be no determinative principle, it ought to have struck down the law. In our conclusion, we said – the court could have avoided the discussion in the first place.
We have addressed points 1 and 2 that you make in our post. Our point as to preference is our reading of the language employed by the court.
Once again, thank you for having taken out the time to read our post. We would be happy to address any further doubts/comments that you may have.
Hi Dhruva – thanks a bunch for the reply. I’m having to dig pretty deep to continue a defence, but here goes:
1. It is certainly within the competence of the Government to enact amendments that may have differing objectives to that of the parent statute. However, the SC has declared consistently – that arbitrariness in bringing amendments must be tested against the ‘provisions of the statutes and the objectives thereof’ (Indian Express, 1985; Khoday, 1996; Otis, 2003; Bombay Dyeing, 2006). To contend then that the Parliament’s enactment of the 2019 amendment is immune from judicial scrutiny even if it is overtly contrary – in the Court’s assessment – to the objectives of the 1996 statute (w/c is for arbitrations to be time-efficient and with minimal court interference) is incorrect.
2. Necessary, at the outset, to concede to the following –
(i) It is correct that the Statement of Objects and Reasons to S.87 makes no reference to the Srikrishna Report.
(ii) It is also correct to say that the reasoning of the Srikrishna Report is sound – his fear of divergent interpretations by High Courts of Section 26 of the 2015 amendment was quickly realised.
(iii) Also correct, in my opinion, to contend that the Srikrishna Report’s recommendations was not strictly before the court in BCCI.
However, let’s look at the implications of the BCCI decision on the relevance of the Srikrishna Report:
(I) The Report identified the problem of inconsistency of interpretation of the 2015 amendment, w/c was premised on certain reasons. On this basis, it recommended the enactment of a provision that would solve these problems.
(II) The Court in BCCI is simultaneously seized of the identical problem: that of inconsistency in the application of the 2015 amendments.
(III) The Court in BCCI ‘solves’ this problem – before an amendment is brought – by declaring the manner in which the 2015 amendments were to be applied, creating a bifurcation premised on when court proceedings in relation to arbitration commenced. It does so by also taking into account the reasons given/fears expressed by the Srikrishna Report.
(IV) In 2019, however, using the same problems identified in the Srikrishna Report, i.e. inconsistency in the application of the 2015 amendment (and explicitly referring to the Report as the basis of the enactment, para 48 of the SC version), S.87 is enacted.
The implication of HCC is to say that the Report could not have been used as the sole basis to bring the 2019 amendment – BCCI having already given quietus to the fears expressed in that report. Therefore, the enactment of Section 87 would have had to have been premised on a reference to the need to do away with the regime created by BCCI (i.e. bifurcation on the basis of court proceedings), and the difficulties of its interpretation of Section 26. This is why the 2019 amendment was declared to have been enacted without adequate determining principle – which, as also repeatedly held, is a facet of the inquiry under Article 14 arbitrariness (Shyam Sunder, 2011; A.P Dairy, 2011).
[Personally, I would think that if the government had done so – to state, for e.g. that bifurcation on the basis of when court proceedings began was never the intent of Section 26, or that the bifurcation had resulted in unforeseen complications not desired by the Government, and as a result, it was necessary to remove Section 26 and bring in Section 87, it would have allowed significantly less wiggle room to the Court to examine the validity of Section 87.]
3. I agree with you on principle – Courts should exercise restraint when striking down legislation. As I also admitted in my first reply, there is a conversation to be had in the SC on the nature of judicial review it is permitted to undertake under Article 14. I just think that this was one of those instances, for the reasons above and in my first reply, where the exercise of this power was justified.
Thanks again for replying as quickly as you did!
Very well written. It is very difficult to accept the view of Nariman J. Striking off legislation on the grounds of Manifest Arbitrariness is something which is unheard of and constitutionally unacceptable. It is a recent evolution of the Court. Unless the law is beyond the competence of the parliament or Legislature or manifestly violative of fundamental rights, it cannot be wished away. Sadly, the court has observed that not taking into consideration the Judgment of a Supreme Court and placing reliance upon the report of the committee, is a good ground for challenge. let us assume a case of non-reference to both. Could it be a reason for striking down the law? It is a firm “N0”. the recent pronouncement in INSAF, dated 6th March 2020, 2004 9 SCC 580, 19773 SCC 592, are guidelines as to when a STatute could be struck down as Unconstitutional.
Dhurva and his partner have done well to expose the fallacy in the reasoning of the Court.
[…] Several posts on this blog have considered some of the recent judgments of the Supreme Court; but an enunciation of the actual standard remains elusive. This series of essays argues that although the recent cases are labelled as accepting an “arbitrariness” challenge to legislation, they ought not to be taken as referring to the ‘arbitrariness’ of administrative law. When one is thinking through the lens of administrative law, ‘arbitrariness’ is a ground for review of administrative actions. But when one speaks of ‘arbitrariness’ as a matter of constitutional law, one is not speaking of the same thing. ‘Arbitrariness’ in constitutional law is distinct from the ‘arbitrariness’ of administrative law. The constitutional law test is of ‘manifest arbitrariness’; and this series of essays will suggest that “manifest arbitrariness” is not simply “an extreme form of administrative law arbitrariness”: the difference is not merely of degree. […]
[…] along with Sahil Raveen (here and here) I have critiqued the use of manifest arbitrariness by Nariman J. in Hindustan […]