The Procedure as Punishment: Why Nfra’s Retrospective Reach is Unconstitutional

Introduction

Following large-scale financial frauds such as the IL&FS debacle, India created the National Financial Reporting Authority (‘NFRA’) as a strong, autonomous regulator to revive public confidence in the profession of accountancy. Though its establishment was a much-needed change, there has been a grave legal confrontation. NFRA’s functions are constitutional, but its process, when retrospectively applied, is a challenge to the provisions of natural justice. The question was raised in the Delhi High Court’s judgment in Deloitte Haskins & Sells LLP v. Union of India. The court upheld NFRA’s substantive powers, including its retrospective jurisdiction, on the premise that it was a simple ‘change of forum’ from the ICAI. But it set aside the disciplinary notices due to the fact that NFRA did not demarcate its investigative and judicial functions, and hence there was a possibility of bias. This essay maintains that the High Court’s decision regarding retrospective application is flawed.

The paper argues that the shift from the ICAI to NFRA was not a mere ‘change of forum’ but an overhaul of the disciplinary process. This overhaul greatly erodes an auditor’s right to a fair defence as well as introduces new obligations on audit firms. Hence, the implementation of this new, more restrictive procedure which imposes new obligations to past conduct is unconstitutional. This paper will initially determine the quasi-criminal character of NFRA’s proceedings. It will then contrast the procedural protections of the ICAI with the restrictions of the NFRA framework, contending that this transition undermines substantive rights and renders its retrospective application unfair. Finally, it will determine that NFRA has usurped powers similar to the Public Company Accounting Oversight Board of America (hereinafter ‘PCAOB’) without adequate procedural protections as envisaged under the PCAOB rules.

The Quasi-Criminal Nature of NFRA Proceedings

Before examining NFRA’s particular procedures, it is necessary to define the standard of fairness they need to satisfy. The regulator’s disciplinary actions are not at all civil or administrative issues; they are essentially quasi-criminal in character. This characterisation is important as it requires a significantly greater degree of procedural protection to the accused. The Supreme Court in An Advocate v. Bar Council of India ruled that where a professional is liable to suffer a heavy penalty such as debarment, impacting his right to practice the profession, the proceedings are quasi-criminal in nature. It characterized the stripping of a professional off the rolls as a penalty of gigantic size, akin to a death penalty in penal law. The Court observed that such a penalty deprives an individual of their means of livelihood and has the potential to make them a social apartheid. This is a principle directly applicable to NFRA, which, according to Section 132(4) of the Companies Act, can disbar auditors for a period of ten years. This quasi-criminal nature directly affects the burden of proof to be established.  The Supreme Court in L.D. Jaisinghani v. Naraindas N. Punjabi ruled that in cases of this gravity, the evidence should be of such a nature that it leaves no reasonable doubt regarding guilt. This implies that the auditor has to be given the benefit of doubt. This standard is far higher than the standard of “preponderance of probabilities” which was held to be applicable in the Deloitte case. Hence, any evaluation of NFRA’s procedures should be subjected to this high threshold of quasi-criminal fairness.

(A) From an Adversarial Trial to an Inquisitorial Review

The processes undertaken by the ICAI and NFRA are essentially different in nature. The change is towards a regulator-led, inquisitorial model from a quasi-judicial, adversarial one. The model of the ICAI was set up as a conventional disciplinary system that deals with complaints, establishing a clear-cut adversarial structure involving a complainant and an accused auditor. It had the features of a trial. As the comparative tables in the Deloitte judgment show, the process under Rule 18 of the Misconduct Rules, 2007, permitted the ‘examination of witnesses.’ Importantly, it also provided for the Disciplinary Committee to exercise its discretion to allow cross-examination. This enabled the auditor to aggressively challenge the facts, evidence, and witnesses produced against them, which is a foundation stone of an adversarial system. By sharp contrast, NFRA’s process under Rule 11 of the NFRA Rules, 2018, is inquisitorial. The Delhi High Court recognized this difference but employed it to justify the lack of certain safeguards. The Court argued that as NFRA’s process depends not upon the oral testimony of a third-party complainant whose credibility must be tested, the right of cross-examination is not vital. However, this argument is premised on a partial knowledge of the ICAI’s own procedural powers. It was assumed that the ICAI could only proceed on formal, third-party complaints. Rule 7 of the Misconduct Rules, 2007, provided for the disciplinary committee to proceed on ‘any written information’ it received, which acted as a suo motu power. Even if the provider of this information refused to lodge a formal complaint, the Director (Discipline) still had the authority to review the case, make a prima facie opinion, and initiate an investigation.

This two-way ability of the ICAI is important because it indicates the extent of the procedural change. Even if the ICAI was behaving in a regulator-led capacity on the basis of information, it was still subject to Rule 18’s adversarial protection. This still allowed the Disciplinary Committee to summon the ‘examination of witnesses’ and have the discretion to allow ‘cross-examination’ to maintain a fair hearing. This establishes that the ICAI structure esteemed these trial-like protections as core to natural justice, no matter how a case was instituted. NFRA, when performing an equivalent regulator-led function, does not offer these protections. Its whole procedure is limited to the audit file and its AQRR. There is no provision to cross-examine the NFRA officials who wrote the very report that acts as the charge sheet. Therefore, the High Court’s justification for the absence of cross-examination rests on a false dichotomy. The real difference is that the ICAI’s framework, even when acting inquisitorially, recognized the need for adversarial safeguards to ensure a fair hearing. NFRA’s framework removes them entirely, marking a fundamental shift in the system of justice itself. Also, the right to cross-examination is a cornerstone of a fair defence. The Supreme Court in State of M.P. v. Chintaman Sadashiva Waishampayan directed that this right is a severe breach of natural justice. In addition, the right to legal representation was also watered down. Although the ICAI framework assured representation through an advocate, the relative table in the Deloitte judgment emphasises that NFRA initially construed an in-person hearing as denying this right. While the High Court observes NFRA subsequently made a course correction, the original vagueness is a substantive erosion of an assured right. Where there is a complex matter, the Supreme Court has held in Board of Trustees v. Dilipkumar that withholding legal representation amounts to non-compliance with natural justice. These lost protections are not minor technical modifications; they are the foundation of a just trial, and their elimination markedly predisposes the accused auditor.

(B) The Waiver of Substantive Defence Rights

It is well settled as a principle of law that a procedural legislation could not be given retrospective effect where it would prejudice vested substantive rights. As the Supreme Court held in Hitendra Vishnu Thakur v. State of Maharashtra, a change in procedure could not be retrospective if it imposes ‘new disabilities or obligations’ on an individual for past acts. The NFRA framework fails this test on two essential counts. Firstly, it creates wholly new liabilities and far stricter penalties. The NFRA has the power to impose significantly greater monetary fines than the ICAI. Whereas the ICAI was able to impose a maximum financial penalty of ₹5 lakh on a person, Section 132(4) of the Companies Act gives NFRA the authority to impose penalties between five times or ten times the audit fee collected, which can reach crores. This is a very big increase in the penalty. In the Deloitte case, the High Court met this challenge by accepting an undertaking from NFRA. The regulator promised that for any audit done before October 2018, it would not issue a fine that would be more than the old limit of ₹5 lakh. By agreeing to this assurance, the court effectively nullified the argument on stricter monetary penalties. But this logic is constitutionally indefensible. The constitutionality of a law should be measured by what its language allows, rather than by the manner in which an authority vows to enforce it at a given instant. An administrative initiative is not an irreversible solution to a potentially unconstitutional measure. A subsequent leadership of NFRA would not be obligated by this commitment and would be legally permitted to invoke the more severe penalties envisioned by the statute. The statute itself does not change and continues to vest the power to impose a higher penalty retrospectively. The constitutionality of a law cannot be preserved by the discretionary self-restraint of its administrator. Hence, imposition of these fresh liabilities and increased penalties, regardless of any interim undertaking, is a new disability which fails the Hitendra Vishnu Thakur test.

More significantly, the express inclusion of audit firms within the disciplinary jurisdiction, making them vicariously liable for their partners’ wrongdoing, represents a fundamental departure from the Chartered Accountants Act, 1949, which was largely concerned with individual members. The retrospective application of such increased liabilities and this kind of liability is against the settled legal principle that penal statutes must not have a retrospective operation. The High Court, in its ruling, explained that firm-level liability was not novel. It referenced Section 147 of the Companies Act, 2013, to contend that the law already ‘contemplates a firm incurring a liability as a result of the act of its Engagement Partners.’ Accordingly, the court held that Section 132 of the Companies Act did not impose a foreign or uncontemplated liability on the pre-existing statutory framework. But this argument does not seem convincing due to High Court’s wrong analogy. The ‘change of forum’ at the centre of the case before us is the transfer of disciplinary jurisdiction from the ICAI mechanism under the Chartered Accountants Act, 1949 to NFRA’s new mechanism. Hence, the proper analysis is not to compare NFRA’s disciplinary regulations with the general liability provisions of the Companies Act, but to compare them with the specific disciplinary provisions that were in place under the CA Act and the Misconduct Rules, 2007. The Companies Act provides for general liabilities for violating its provisions, while the CA Act established the particular self-contained procedure for professional misconduct, which NFRA has now substituted for Public Interest Entities.

When we take a look at the appropriate structure, it’s obvious that the CA Act was designed based on the personal misconduct of the single member, rather than the firm. The only scenario where a firm was liable under the previous disciplinary rules was under Sections 21A(6) and 21B(6) of the CA Act. As the High Court itself has pointed out, this was only where a partner was ‘repeatedly found guilty of misconduct’. This was not a general principle of vicarious liability, but a narrow, specific, and exceptional provision for recidivist offenders. Had the legislature wished to impose vicarious liability upon firms for one isolated example of a partner’s delict at the outset, it would not have drafted such a limited and qualified provision. The reason that the CA Act was amended in 2022 to specifically introduce wider firm-level liability further establishes its earlier absence. Hence, NFRA’s jurisdiction to make a firm vicariously liable for a partner’s act of default on the first occasion is a new liability. The retroactive application of this new type of liability is an absolute contravention of the principles established in Hitendra Vishnu Thakur.

Adopting PCAOB’s Powers Without Its Protections

A central rationale for NFRA’s design is that it conforms to international best practices, specifically with international institutions such as the US Public Company Accounting Oversight Board (‘PCAOB’). But this analogy is fundamentally flawed as closer inspection discloses that NFRA has borrowed the strict powers of international organizations without importing their essential procedural mechanisms for a just hearing. This has created an imperfect, one-way model of regulation that favours enforcement at the expense of justice. The most important distinction is the separation of powers. This is the cornerstone of PCAOB’s structure and serves the purpose of forestalling the very bias to which NFRA’s first hearing fell victim. As quoted in the Deloitte judgment, PCAOB Rule 5200(d) (Separation of Functions) imposes an absolute separation between its investigatory and adjudicatory branches. It creates a Division of Enforcement and Investigations to serve as the prosecutor, with an independent Hearing Officer to preside over the hearing.  The rule plainly prohibits the investigative staff from serving or advising on the ultimate decision. It was quite different in the Delhi High Court, which held that NFRA’s Executive Body did both, writing the investigative report (the AQRR) and subsequently ordering disciplinary action based on its own findings. This blurring of roles is a clear departure from the international standard for an objective regulator.

 In addition, the PCAOB design ensures a robust, trial-like hearing. PCAOB Rule 5444 specifically gives the accused the right to make a case of their own through oral and documentary evidence and to carry out such cross-examination as is appropriate for a full and true disclosure of facts. The right to appear by counsel is also specifically safeguarded under Rule 5109(b).  Such are basic precautions that guarantee a fair adjudicatory process. NFRA’s undefined summary process, being document-based and without these express rights, falls well below this international benchmark. Hence, the contention that NFRA is simply conforming to world standards is indefensible. In practice, its procedural model exposes auditors to a regime that is less equitable and offers less safeguards than the preceding domestic regulation under the ICAI and indeed the very global standard it was meant to replicate.

Conclusion

It goes without saying, then, that the Delhi High Court’s characterisation of the transition to NFRA as a simple ‘change of forum’ does not do justice to the magnitude of the procedural change. Throughout this paper, it was contended that the new system amounts to a departure in substance from the quasi-judicial procedure of the ICAI. The disciplinary process of the ICAI, with its scope for cross-examination and a trial-like hearing, provided a material right to a fair defence. NFRA’s inquisitorial, document-based summary procedure does away with these essential protections. The stripping away of basic defensive rights generates a fresh disability for the accused, as delineated by the Supreme Court in Hitendra Vishnu Thakur. The retrospective operation of Section 132 is thus rendered unconstitutional. It offends Article 14 by making auditors confront an arbitrary and less fair process for past conduct. It also offends Article 21 since the right to a fair and just trial is an integral component of personal freedom. For NFRA to effectively play its crucial role and acquire legitimacy, its powers have to be counterpoised with procedural fairness. Maintaining the retrospective application of a paradigm that violates the right to a fair defence creates a dangerous precedent, eroding the very rule of law that the regulator is trying to maintain.

Author:– Ritik Kumar , in case of any queries please contact/write back to us atsupport@ipandlegalfilings.com or   IP & Legal Filing.