The Andhra Pradesh High Court's Section 147A Strike-Down: Courageous Constitutionalism or Overreach of Its Own?
The Hook: When Parliament Rewrites History, Who Gets the Last Word? There is a particular species of legislative mischief that Indian constitutional lawyers have watched with growing unease for decades: the retrospective validating amendment. Parliam...
The Hook: When Parliament Rewrites History, Who Gets the Last Word?
There is a particular species of legislative mischief that Indian constitutional lawyers have watched with growing unease for decades: the retrospective validating amendment. Parliament loses in court, waits for the judgment to settle, and then rewrites the statute with retrospective effect — effectively telling the judiciary, "nice try, but we were always right." The Andhra Pradesh High Court's judgment in Bala Sreenivas Rao Somisetty v. Income Tax Officer (WP No.28520 of 2023, decided 12.10.2026) is the latest — and one of the most consequential — judicial pushbacks against this practice. The Division Bench of Justices Ninala Jayasurya and T.C.D. Sekhar struck down Section 147A of the Income Tax Act, 1961 as unconstitutional, holding that Parliament cannot nullify judicial decisions without first removing the statutory basis on which those decisions rest.
This is not a small case. It involves over 150 writ petitions, implicates reassessment proceedings worth thousands of crores, and lands squarely in the middle of a nationwide judicial conflict over the faceless assessment regime. The Punjab and Haryana High Court had already struck down Section 147A on 10.09.2026 — a judgment whose operation the Supreme Court stayed within eight days. The Andhra Pradesh High Court knew this and proceeded anyway, stating pointedly that it had "dealt with the matter independently." That takes institutional nerve. Whether the reasoning holds up under Supreme Court scrutiny is a different question altogether.
The Facts: A Scheme, a Notice, and a Legislative Counter-Punch
The factual matrix is deceptively simple. The Income Tax Department's Jurisdictional Assessing Officers (JAOs) issued notices under Sections 148A and 148 of the Income Tax Act, 1961 to hundreds of assessees, asserting that income had escaped assessment. The assessees challenged these notices on a straightforward ground: the e-Assessment of Income Escaping Assessment Scheme, 2022, notified under Section 151A on 29.03.2022, required that proceedings under Sections 147 and 148 be conducted in a faceless manner through the National Faceless Assessment Centre (NFAC). The JAOs — physical, territorial officers — had no business issuing these notices.
A co-ordinate bench of the Andhra Pradesh High Court agreed, quashing the notices on 28.10.2025. Similar judgments followed from multiple High Courts across the country. The Revenue appealed to the Supreme Court. While those SLPs were pending, Parliament enacted the Finance Act, 2026 (Act No.4 of 2026), inserting Section 147A with retrospective effect from 01.04.2021. This new provision declared — with a non-obstante clause overriding "any judgment, order or decree of any Court" — that the Assessing Officer for Sections 148 and 148A "shall mean and shall always be deemed to have meant" a JAO, not the NFAC.
The Supreme Court then disposed of the SLPs, set aside the High Court judgments on the limited ground that their foundation had been altered by the amendment, and remitted the matters to the respective High Courts for fresh consideration, expressly leaving open the question of Section 147A's constitutional validity.
There is an additional wrinkle that proved decisive: the Income Tax Act, 1961 was itself repealed on 31.03.2026, with the Income Tax Act, 2025 coming into force from 01.04.2026. The Finance Act, 2026 — which inserted Section 147A into the 1961 Act — also came into force on 01.04.2026. The petitioners argued, and the Court ultimately accepted, that Parliament had amended a dead statute.
The Arguments: Two Very Different Visions of Legislative Power
The petitioners mounted a multi-front assault. Their core contention was that Section 147A did not cure any defect in the law — it simply declared, by legislative fiat, that the High Courts were wrong. The scheme under Section 151A remained unamended; Section 148 remained within the scheme's faceless mandate; yet Parliament purported to "clarify" that JAOs always had jurisdiction. This, they argued, was not validation but judicial overreach by the legislature — a violation of the separation of powers doctrine. They also raised the repeal argument: the Finance Act, 2026 came into force on 01.04.2026, by which time the Income Tax Act, 1961 had already been repealed at midnight on 31.03.2026.
The Revenue, represented by no less than the Additional Solicitor General, invoked the well-established twin test for validating legislation from Shri Prithvi Cotton Mills Ltd. v. Broach Borough Municipality, (1969) 2 SCC 283: first, the legislature must remove the cause of invalidity identified by the court; second, it must do so through a subsequent enactment with retrospective effect. The Revenue contended that Section 147A satisfied both limbs — it removed the "defect" (the ambiguity over who constitutes the Assessing Officer) and applied retrospectively. They also argued the presumption of constitutionality, the wide latitude afforded to fiscal legislation, and the staggering practical consequences of invalidation — reassessment proceedings against 8,82,146 assessees involving Rs.21.74 lakh crores in assessed income.
On the repeal point, the Revenue argued that the Finance Act, 2026 received Presidential assent on 30.03.2026 — one day before the 1961 Act's repeal — and was therefore validly enacted.
The Judgment: Two Independent Grounds, One Fatal Conclusion
The Bench struck down Section 147A on two independent grounds, each sufficient on its own.
First, on the separation of powers: The Court held that a validating law must remove the basis of the judicial decision it seeks to overcome. Here, the High Courts had quashed the JAO notices because Section 148 was included in the faceless scheme under Section 151A. That scheme remained — and remains — unamended. Section 148 is still in the scheme. Without amending the scheme or Section 151A, Parliament cannot simply declare by legislative ipse dixit that the scheme means something different from what the courts held it to mean. The Bench relied on State of Tamil Nadu v. M. Rayappaiah Gounder, (1971) 3 SCC 1, and Municipal Corporation of Ahmedabad v. New Shrock Spg. & Wvg. Co., AIR 1970 SC 1292, for the proposition that while the legislature can remove the basis of a court's decision, it cannot simply direct instrumentalities to disregard judicial decisions.
The Court was particularly scathing about the Revenue's conduct, observing that it was "blowing hot and cold" — arguing on one hand that the scheme required faceless proceedings, and on the other that JAOs retained concurrent jurisdiction. The Bench noted that if the Revenue truly believed Section 148 was never intended to be part of the faceless scheme, it could have removed it before the 31.03.2022 deadline prescribed in the proviso to Section 151A(2). It did not.
Second, on the repeal argument: This is where the judgment becomes genuinely interesting. The Court held that the Income Tax Act, 1961 stood repealed at midnight on 31.03.2026. The Finance Act, 2026, by its own Section 2(a), came into force on 01.04.2026. Section 5 of the General Clauses Act — which provides that an Act comes into force on the date of Presidential assent — applies only when the Act itself does not specify a commencement date. Since the Finance Act expressly specified 01.04.2026, Presidential assent on 30.03.2026 was irrelevant. The Court relied on Common Cause v. Union of India, (2003) 8 SCC 250, to hold that when the legislature itself provides a commencement date, that date controls.
The conclusion: Parliament amended a repealed statute. Section 147A was stillborn.
The Critique: Strong Instincts, But Some Reasoning Needs Shoring Up
Let me be direct: I think the Court reached the right result, but through a route that is more vulnerable than it needed to be.
The separation of powers analysis is the stronger ground, and the Bench handled it competently. The point that Section 148 remains in the scheme, unamended, is devastating to the Revenue's validation theory. You cannot validate your way out of a judicial decision by leaving intact the very provision the court relied upon. The twin test from Prithvi Cotton Mills requires removal of the defect — and here, the "defect" (from the Revenue's perspective) was the inclusion of Section 148 in the faceless scheme. That inclusion was never removed. The amendment simply asserted a contrary interpretation. That is not validation; it is legislative overruling, and the Constitution does not permit it.
But the repeal argument — while clever — is where I have reservations. The Court's technical analysis of the commencement provisions is sound as far as it goes. Section 2(a) of the Finance Act, 2026 does specify 01.04.2026. And Common Cause does support the proposition that an express commencement date displaces Section 5 of the General Clauses Act. However, the argument proves too much. If accepted literally, it would mean that Parliament can never amend a repealed statute with retrospective effect, even when the repealing statute contains a savings clause — like Section 536(2)(c) of the Income Tax Act, 2025 — that expressly preserves pending proceedings under the repealed Act. The savings clause keeps the 1961 Act alive for pending matters. If it is alive for pending matters, it is alive for amendment in respect of those matters. The Bench addressed Section 536 but did not fully grapple with the argument that the savings clause creates a continuing legislative framework within which retrospective amendments can operate.
I suspect the Supreme Court — which has already stayed the Punjab and Haryana High Court's similar judgment — will find this ground attractive as a way to reverse without engaging with the more fundamental separation of powers question. That would be unfortunate, because the separation of powers ground is where the real constitutional principle lies.
There is also a missed opportunity. The petitioners argued that the amendment violated their "vested right" to faceless assessment, invoking promissory estoppel and legitimate expectation. The Court did not engage with this argument at all. It should have. Whether an assessee has a vested right to a particular procedure — as opposed to a substantive right — is a question with significant implications for tax litigation. The conventional answer is that procedural rights do not vest, but the faceless assessment regime arguably creates something more than a mere procedural entitlement: it is a structural safeguard against arbitrary exercise of power. The Bench could have added real jurisprudential value here.
The Takeaway: What Practitioners Should Do Now
First, do not treat this as the final word. The Supreme Court has already stayed the Punjab and Haryana High Court's identical conclusion. This judgment will almost certainly be appealed and stayed. Advise clients accordingly — the notices remain in limbo, not dead.
Second, preserve the separation of powers argument. If you are litigating this issue before any High Court, the strongest ground is not the repeal argument (which is technically clever but jurisprudentially fragile) but the failure to amend the scheme under Section 151A. That is the argument that goes to the heart of legislative competence and judicial independence.
Third, watch the scheme amendment. If the Revenue is smart — and it usually is, eventually — it will now amend the scheme under Section 151A to remove Section 148, and then re-enact Section 147A. If that happens, the constitutional calculus changes entirely. The twin test would then be satisfied, and the challenge would need to shift to whether retrospective withdrawal of a procedural safeguard constitutes an unreasonable restriction under Article 19(1)(g).
Fourth, the faceless assessment regime needs structural reform. This litigation has exposed a fundamental design flaw: Parliament created a faceless scheme but left the trigger mechanism — the Section 148 notice — in ambiguous territory. The result has been years of litigation, wasted judicial resources, and uncertainty for lakhs of assessees. Whatever the Supreme Court ultimately decides, the legislature needs to stop governing by retrospective amendment and start drafting with precision in the first place.
The Andhra Pradesh High Court has drawn a line. Whether it holds will depend on whether the Supreme Court is willing to enforce the principle that judicial decisions cannot be legislatively erased without removing their foundation. The stakes — for tax administration, for judicial independence, and for the rule of law — could not be higher.
Related Cases
Tags
Published by AnrakLegal AI