Islamabad: The Federal Government has proposed a new data-driven tax compliance mechanism through the insertion of Section 165AB in the Income Tax Ordinance, 2001, under the Finance Bill 2026. The proposed law would require banks and Electronic Money Institutions (EMIs) to share high-value transaction data with the Federal Board of Revenue (FBR) for automated cross-matching against taxpayer declarations.
The measure is aimed at identifying significant discrepancies between banking transactions and reported income while reducing direct interaction between taxpayers and tax officials through the use of algorithmic risk assessment.
Banks and EMIs to Report High-Value Transactions
Under the proposed section, every banking company and Electronic Money Institution will be required to electronically upload financial transaction data to a centralized data hub maintained by the tax authorities.
The reporting obligation will apply to account holders whose total deposits or withdrawals exceed Rs. 100 million during a six-month reporting period across one or multiple accounts.
Financial institutions will be required to provide details including:
- Total deposits and withdrawals.
- Opening and closing account balances.
- Peak credit balance recorded on any single day during the reporting period.
- Aggregate credits received during the period.
The proposed provision overrides confidentiality restrictions contained in other laws, including the Banking Companies Ordinance, 1962, and the Protection of Economic Reforms Act, 1992.
Automated Processing to Protect Taxpayer Privacy
The Finance Bill emphasizes that the submitted banking information will initially be processed through an automated digital system.
According to the proposal, the shared data will not be directly visible to income tax officers during the initial cross-matching stage. Instead, an algorithm will compare banking activity with information already available in tax returns and wealth statements.
The FBR will be legally bound to maintain strict confidentiality of the information and ensure that the data is not misused.
Gross Mismatches to Trigger Risk-Based Investigations
Where the automated system identifies what the law describes as a “gross mismatch” between a taxpayer’s banking transactions and declared income or assets, the information will be transferred to the Compliance Risk Management (CRM) system.
The CRM platform will evaluate potential risks related to under-reporting of income, undisclosed assets, or other tax compliance concerns.
Any further inquiry or enforcement action arising from these mismatches will be handled exclusively through the proposed National Faceless Centre, a move intended to minimize physical interaction between taxpayers and tax authorities.
Twice-Yearly Reporting Schedule Introduced
The proposed law establishes two reporting periods each year:
- Transactions from July 1 to December 31 must be reported by January 31.
- Transactions from January 1 to June 30 must be reported by July 31.
Banks and EMIs will be required to ensure timely submission of the prescribed information through integrated digital systems.
Heavy Penalties for Non-Compliance
The Finance Bill also introduces penalties for institutions that fail to integrate their IT systems with the Central Data Hub or do not provide the required information.
Under the proposal, the Principal Officer of the institution may face a personal penalty of Rs. 500,000 for the first default and Rs. 1 million for each subsequent default.
Significant Expansion of FBR’s Data Analytics Framework
Tax experts view Section 165AB as one of the most significant data-collection and compliance-monitoring measures proposed in the Finance Bill 2026. The initiative forms part of the government’s broader strategy to enhance tax documentation through technology, risk-based enforcement, and automated compliance systems.
If approved by Parliament, the provision will substantially expand the FBR’s ability to detect undeclared income by electronically comparing high-value banking activity with taxpayer disclosures while relying on algorithmic analysis rather than manual scrutiny.
