IFRIC 23 To Clarify Grey Areas In Income Tax

Unclear tax treatments are headaches for any executive, but for accountants they are especially annoying: this is because varying treatments mean varying accounting, and that leads to inconsistencies. (As accountants, we hate inconsistencies!) Consequentially, financial statements prepared under the International Financial Reporting Standards (IFRS) can lose much of their comparability.

A tax treatment can be considered "uncertain" when a taxpayer applies it without being sure that the tax authorities will accept it. Common examples would be the tax deductibility of certain expenses, tax-exemption of certain income, and transfer pricing rules to allocate income between jurisdictions.

New guidance from the IFRS Interpretations Committee

The current tax standard, IAS 12 Income Taxes, doesn't explicitly address the accounting issues that tax uncertainties bring. In reaction to this, the IFRS Interpretations Committee (IFRS IC) looked into the issue and confirmed that firms do use different accounting methods in cases of uncertainty.

Thus, in June 2017, the IFRS IC issued IFRIC 23, an interpretation applicable to any aspect of income tax accounting where there is uncertainty, notably taxable profit or loss as well as the tax basis of assets and liabilities, tax losses and credits, and tax rates.

IFRIC 23 clarifies that a company must assume that the taxing authorities have full knowledge of all the relevant information in assessing the proposed tax treatment. Thus, the probability of the tax uncertainty being detected is ignored.

How to treat uncertainty in financial statements

Under IFRIC 23, the key test is whether it is probable (i.e. more likely than not) that the taxing authority will accept the company's tax treatment as reported in the income tax filing. If the answer is yes, then you should record the same amount in the financial statements as what you submit (or plan to submit) in the income tax return.

If the answer is no, then the effect of the tax uncertainty should be reflected using the method that you expect will better predict its resolution. There are two methods:

the "most...