ITAT Delhi remands bogus liability issue to CIT(A) for fresh adjudication; upholds 10% disallowance of administrative expenses

Background

The Delhi ‘F’ Bench of the Income Tax Appellate Tribunal has partly allowed two appeals filed by the Revenue against M/s Pearls Buildwell Infrastructure Ltd for Assessment Years 2012–13 and 2014–15, holding that the Commissioner of Income Tax (Appeals) erred in granting relief on the issue of alleged bogus liabilities without obtaining a remand report from the Assessing Officer. The appeals arose from search and seizure proceedings conducted under Section 132 of the Income Tax Act, 1961 in the PACL Group on 20 June 2013, wherein the assessee—engaged in real estate development—was covered as a contractor/land developer. Pursuant to notices issued under Section 153A, the Assessing Officer completed assessments making additions of ₹4.46 crore (AY 2012–13) and ₹1.17 crore (AY 2014–15) on account of alleged bogus trade liabilities reflected in the balance sheet, along with ad-hoc disallowances of 25% of administrative and employee expenses for want of supporting documents. On appeal, the CIT(A) partly deleted the additions relating to liabilities and restricted the disallowance of expenses to 10%. Aggrieved, the Revenue approached the Tribunal contending that the CIT(A) admitted additional evidence without confronting the same to the Assessing Officer, in violation of procedural requirements.

Outcome

Accepting the Revenue’s contention on this limited issue, the Bench held that the CIT(A) ought to have called for a remand report before adjudicating the matter. The Tribunal accordingly set aside the orders of the CIT(A) on the issue of bogus liabilities and remitted the matter back for fresh adjudication, after granting both sides a reasonable opportunity to present their cases. However, the Tribunal declined to interfere with the restriction of disallowance of administrative and employee expenses to 10%, holding the same to be reasonable. The Revenue’s appeals were thus partly allowed for statistical purposes.