Auditing of Income Tax Returns: Some Questions

Auditing is an essential tool for ensuring compliance and preventing tax evasion in modern societies. However, applying biased sampling instead of a random one during selection of the target population often proves to be counterproductive, as has been evident in some developing countries like Bangladesh. As is well-known, the country’s tax-GDP ratio is one of the lowest in the world, and it is certainly the lowest in South Asia. Policymakers, therefore, need to focus both on the tax-collection system and the methods and procedures applied for mobilising tax revenue. A recent example of adopting a deficient method has been the biased sampling employed by the National Board of Revenue (NBR) for selecting income tax (IT) returns in order to conduct audits on taxpayers across the country.
According to the Auditing Practices Board (APB) of the USA, “An audit of financial statements is an exercise whose objective is to enable auditors to express an opinion on whether the financial statements give a true and fair view (or equivalent) of the entity’s affairs at the period end and of its profit or loss (or income and expenditure) for the period that ended and have been properly prepared in accordance with the applicable reporting framework (for example, relevant legislation and applicable accounting standards) or where statutory or other specific requirements prescribe the term ‘present fairly’.”
The new ‘Income Tax Act, 2023’ was passed in Parliament in June 2023 after repealing the old ‘Income Tax Ordinance, 1984’. Section 180 of the new Act stipulates provisions related to ‘self-assessment’ of income tax return (ITR), while Section 182 elaborates on the procedural aspects of conducting audits. Based on these two articles of Act-2023, NBR formulated ‘Income Tax Return Audit Guidelines, 2025’ to facilitate the auditing role of zones, ranges and circles under its jurisdiction. The objectives of these audits, as detailed in Section 2 of these guidelines, are to prevent tax evasion, curb the tendency of not paying taxes, identify the grounds, methods and risks of tax evasion, and ensure proper compliance with and application of law and rules by the taxpayer and income tax officials, respectively.
However, deficiencies have been observed when Section 3 on ‘Selection of income tax returns for audit’ and Section 4 on ‘Guidelines to be followed in selecting income tax returns for audit in case of taxpayers other than individual taxpayers' are scrutinised. Section 3(A) of the guidelines stipulates that the tax authority would select IT returns for audit from among those submitted under Section 180 of the IT Act in accordance with procedures mentioned in Section 182. Besides, Section 3(B) of the guidelines specifies that the IT returns should be selected through a completely automated process based on ‘Risk-based Audit Selection Criteria’ for ensuring transparency and neutrality in the selection process. But contradicting these principles of neutrality and transparency, Section 4(L) states that among selected returns, at least one-third should be from among those who were never audited in the past. It clearly violates the principle of fairness and neutrality, as the sample becomes ‘biased’ instead of remaining ‘random’.
‘Random sampling’ is defined as a statistical technique whereby individuals are selected from a larger population so that every member has an equal and independent probability of getting chosen. On the other hand, a biased sampling occurs when the sampling method deliberately picks or leaves out certain members of a target population, making the sample unrepresentative of the whole. Consequently, when the precondition of including at least one-third of the sample from unaudited taxpayers is applied, it is no more random and becomes biased against this particular group of people. Moreover, the stipulation for the NBR guidelines (Section 4) was violated when individual taxpayers were also included in this group.
The question of fairness also comes, as it can be inferred that the taxmen had previously spared this group of unaudited people because they did not show anything incriminating in their returns. So, when they are chosen for auditing through biased sampling, it is akin to penalising them for fair practices in the past, similar to showing yellow or red cards to football players who have not committed any foul.
Now, what are the impacts of this biased sampling in the selection of IT returns for auditing? Firstly, choosing at least one-third of the sample from previously unaudited taxpayers allowed more likely defaulters to go scot-free, as their spots were taken by the former. This is likely to result in losses in revenue as well as moral high ground of NBR. Secondly, it came as a ‘bolt from the blue’ or an ‘unexpected shock’ to mostly innocent taxpayers, many of them in their sixties or seventies, who never had to face the ordeal of a completely unknown threat like auditing. There have been reports that even retired schoolteachers in remote villages and retired civil servants in their late 60s or 70s became innocent victims of this statistical game. The situation is probably more precarious for those senior citizens who are vulnerable due to health issues.  
According to media reports, NBR selected 15,494 returns for audit in the first phase in July 2025 and 72,341 returns from the 2023-24 tax year in the second phase this year. Each tax circle had to select a maximum of 200 and a minimum of 20 taxpayers for the purpose. The initiative was first taken by the interim government led by Professor Yunus, with the newly elected BNP government only continuing what was already started. But what surprises many is that an accomplished economist like Dr Salehuddin Ahmed was at the helm of the finance ministry when this deficient auditing initiative was rolled out. The current finance minister also has an excellent business-cum-accountancy background. Therefore, it is expected that he would look into this matter and prescribe remedial measures for ensuring neutrality and fair play in tax administration.
_____________________________________
The writer is a retired Additional Secretary, with a master’s degree in ‘Government Financial Management’ from the Ulster University, UK. He can be reached at [email protected]