Trust, Proportionality, and the Future of Tax Enforcement in India

The income-tax enforcement policy has been decades-old and based on the concept of deterrence. Even small violations of compliance, simply ignored, may open the door to criminal prosecution. The logic was clear: tax evasion hurts public finance, and only strong penalties could discourage bad actors. However, this strategy has resulted in a culture of excessive compliance, intimidation and lawsuits as well. A wrong record or a deadline that has been missed may not only lead to reprimand, but prosecution.

The working paper of NITI Aayog on the Income-tax Act, 2025 this year is a pivotal change in the line of thinking. The piece does not simply critique the text of the new law- it challenges the very structural form of the Indian tax system, in general, with regard to criminal liability. It is a basic concept, law must distinguish between mistake and will, and punishment must be equal to the damage.

A Reform Decades in the Making

The new Income-tax Act, 2025 was aimed at substituting the old-fashioned 1961 system. While it does streamline certain aspects and removes thirteen outdated offences, it still lists 35 actions as criminal, many of which are administrative in nature. Twenty-five of these carry mandatory minimum imprisonment. Each of them can bring jail, as well as fines.

NITI Aayog argues that this is excessive. Their stand is not rooted in ideology but on a simple principle, which states that criminal sanctions are to be applied in case of serious, fraudulent activities. Even the administrative lapses that are inconvenient, or non-compliant do not necessarily warrant custodial sentences.

Reframing the Question: When Should Tax Defaults Be Crimes?

One of the paper’s most valuable contributions is its framework for evaluating when criminal law should apply. It sets out five criteria: the offence must protect a core societal value; involve real and significant harm; be best addressed through criminal law; be proportionate; and be clearly drafted with scope for periodic review.

Through these principles, the paper makes three major conclusions. To begin with, 12 out of the 35 crimes ought to be decriminalised. Second, another 17 should attract criminal liability only if fraudulent or malicious intent is clearly established. And third, only 6 offences—those involving deliberate evasion or tampering—should remain fully criminal.

This is not decriminalisation for its own sake. It is an argument for fairness, legal clarity, and institutional credibility.

Procedural Defaults and Criminal Overreach

The paper also identifies areas in the new Act that are most overreaching. These are not hypotheticals but actual instances of how the law as it stands does not leave much to discretion:

• Section 473 criminalises the violation of seizure or attachment orders, even if the violation is accidental or unintentional.
• Section 474 makes it a crime to deny access to digital devices or refuse to provide passwords to tax authorities. While aimed at transparency, the paper notes that this could clash with constitutional protections against self-incrimination.
• Section 478 casts a wide net by criminalising any act that “facilitates” tax evasion, including omissions and errors that may have no intent behind them.
• Section 482 punishes false statements in verifications, but currently does not differentiate between intentional misstatements and honest mistakes.
• Section 494 makes confidentiality breaches by tax officials a criminal offence, even if the breach was a procedural error rather than an intentional leak.

In each case, the criticism is not that these actions should go unpunished, but that criminal punishment should be a last resort, not the only option available.

Restoring Balance: The Case for a Trust-Based Regime

The language of “trust-based taxation” has gained traction over the last few years, especially through initiatives like Transparent Taxation – Honouring the Honest . Rhetoric in itself cannot be used to establish trust. It has to be embodied in law, procedure and institutional behaviour.

The recommendations of NITI Aayog include the repealing of the mandatory minimum sentences, resetting of judicial discretion and putting back the burden of proving afresh on the state. These are not cosmetic changes. They symbolize a reconsideration of the relationship of the law with the tax-payer.

The paper also demands improved design, which is also very crucial. The legislation should be clear and should contain language which is not difficult to comprehend by a non specialist. Crimes and offences should be defined in a narrow sense. And when a new criminal measure is proposed, the legislators must not only consider the effect on the revenues, but also the workload it will impose on the courts, enforcement, and the rights of the citizens.

A System Worth Building

The shift to a compliance first regime does not mean the decrease of the barrier to tax evasion. What it does mean is that the legal system should act with precision. It must know when to punish, when to prosecute and when to merely correct.

This change also necessitates the need of tools that can assist the enforcement agencies to differentiate conduct according to intent, procedural fairness in documents, and lifecycle transparency of a dispute. The spirit of this reform can be achieved by having digital systems, which instantiate values of due process, e.g., organized records of decisions, auditable workflows, and proportional sanction logic. These are not abstract ideals; they are capabilities that justice tech platforms like Jupitice are already building into real-world systems.

If adopted thoughtfully, the approach proposed by NITI Aayog could mark a shift in the taxpayer state relationship. Less fear. More predictability. And a legal regime that recognises the difference between a mistake and a crime.

That would be a system worth building.