FRS 102 deferred tax recognition works on a full-provision basis for liabilities and a probability test for assets. Section 29 requires an entity to recognise a deferred tax liability for every timing difference that exists at the reporting date, and a deferred tax asset only to the extent that recovery against future taxable profits is more likely than not. A handful of carve-outs sit on top of that rule, and the measurement and disclosure choices that follow flow from the recognition decision.1Paragraph 29.6, FRS 102
How Timing Differences Trigger Recognition
A timing difference exists whenever revenue or an expense enters the accounts in one period but the tax computation in another. Capital allowances are the standard illustration. A company that buys machinery for £100,000 and claims full first-year allowances takes £100,000 of tax relief immediately, while depreciation of £10,000 a year over ten years puts only £10,000 through the profit and loss account in year one. The £90,000 gap between relief already taken and depreciation still to come is a timing difference that reverses as depreciation continues without further allowances.
Direction matters. Where tax relief runs ahead of the accounting charge, the difference is taxable and produces a liability. Where the tax deduction lags behind the accounting expense, the difference is deductible and produces an asset. Every deferred tax balance recognised under FRS 102 traces back to one of those two patterns.
When a Deferred Tax Liability Must Be Recognised
Paragraph 29.6 takes an unqualified line: if a taxable timing difference exists at the reporting date, the liability is recognised. There is no materiality escape and no likelihood-of-reversal filter. The recognition test for liabilities is simply whether the timing difference is there.1Paragraph 29.6, FRS 102
Accelerated capital allowances are the most common trigger for UK companies. Any entity claiming annual investment allowances or first-year allowances on plant and machinery will depreciate the same asset over a longer period, generating a taxable timing difference that persists until the asset is fully written down.
Revaluations create liabilities too. An investment property carried at fair value under Section 16 generates a timing difference when its value rises, because no tax is due until disposal. FRS 102 requires a deferred tax provision on that unrealised gain, measured using the rates that would apply on sale. A £35,000 fair value gain on an investment property produces a deferred tax liability of £8,750 at the 25% main rate of corporation tax.
When a Deferred Tax Asset Can Be Recognised
Assets face a tougher hurdle. Paragraph 29.7 permits recognition only to the extent that recovery is probable, which the FRS 102 glossary defines as more likely than not.2Paragraph 29.7, FRS 102 The existence of unused tax losses is itself treated as evidence that future taxable profits may not arrive, so the entity carries the burden of showing they will.
In practice this means working through profit forecasts, order books, and past performance. A company with a run of taxable profits and a credible forecast of more will meet the threshold; one with several consecutive loss-making years will struggle. The assessment is redone at each reporting date. An asset held back in one year can be picked up in the next if the outlook improves, and an asset previously recognised may need to come off if it worsens.
Pension deficits and share-based payment charges are the everyday sources of deferred tax assets: the accounting expense hits first, the tax deduction follows. A £50,000 pension liability supports a deferred tax asset of £12,500 at 25%, but only if the entity expects sufficient future profits to absorb the eventual deduction. Without that expectation the asset stays off the balance sheet entirely, which keeps the accounts from reflecting tax savings that may never materialise.
Timing Differences That Are Not Recognised
Several situations sit outside the general rule. Applied mechanically, Section 29 would sweep them in; the standard specifically prevents that.
- Goodwill. No deferred tax is recognised on goodwill arising in a business combination. Any provision would change the goodwill figure, which would change the provision, and the standard cuts the loop by prohibiting the calculation.
- Initial recognition outside a business combination. Where an asset or liability is first recognised in a transaction that is not a business combination and affects neither accounting nor taxable profit at the time, no deferred tax is recognised. A lease liability recorded at inception with no immediate impact on either profit measure is the standard example.
- Unremitted earnings of subsidiaries and associates. A parent is not required to provide for deferred tax on undistributed profits of a subsidiary where it controls the timing of distributions and does not expect remittance in the foreseeable future. This matters most for groups whose overseas subsidiaries reinvest locally rather than repatriate.
The unremitted earnings exception is not permanent. It falls away as soon as the parent loses control over the distribution decision or changes its intention, and most groups reassess the judgement annually and disclose the cumulative amount on which no deferred tax has been provided.
A separate carve-out applies to Pillar Two top-up taxes. The OECD rules impose a 15% global minimum effective tax rate on multinational groups with consolidated revenue above €750 million, and applying normal deferred tax mechanics to those top-up taxes would produce complex figures of limited use because the tax operates on a jurisdictional blending basis. The FRC responded with a mandatory temporary exception: entities within scope must not recognise or disclose deferred tax assets and liabilities related to Pillar Two top-up taxes. Targeted disclosures replace the measurement work. An entity that is, or expects to be, within scope must say so; where the legislation has been enacted or substantively enacted but is not yet in effect, qualitative and quantitative information about exposure is required, which can take the form of an indicative range. Where even a range is not reasonably estimable, the entity says so and describes its progress in assessing the exposure.
Measuring What Gets Recognised
Once a balance qualifies for recognition, FRS 102 requires measurement using tax rates and laws that have been enacted or substantively enacted by the reporting date and that are expected to apply when the timing difference reverses. In the UK, substantive enactment occurs when a Finance Bill receives its second reading in the House of Commons.
The current UK corporation tax structure has a 25% main rate for companies with profits above £250,000 and a 19% small profits rate for those below £50,000, with marginal relief bridging the two thresholds. Selecting the rate is a forecasting exercise: the entity has to estimate the level of taxable profits it expects in the period the difference reverses, not simply apply today’s headline rate.
FRS 102 prohibits discounting. A deferred tax balance is carried at its full nominal value regardless of when the underlying difference is expected to reverse, so a liability that will not crystallise for fifteen years sits on the balance sheet at the same figure as one reversing next year. This keeps the calculation simple, but the balance sheet number overstates the economic burden in present-value terms, which readers of the accounts should keep in mind.
How the Difference Is Defined Under FRS 102
FRS 102 defines the deferred tax base through timing differences, focusing on the income statement: are income or expenses recognised for tax and accounting purposes in different periods? IAS 12 under IFRS and ASC 740 under US GAAP use a temporary difference model based on the balance sheet, comparing the carrying amount of an asset or liability with its tax base.
The two models usually produce the same answer. The temporary difference approach catches certain items the timing difference approach misses, most notably a revaluation of a non-depreciable asset such as land: under the balance sheet approach, the gap between revalued carrying amount and tax base generates deferred tax automatically. Under a pure timing difference approach, no income statement entry has occurred, so no timing difference arises unless the standard specifically requires one. FRS 102 does require one for fair-value investment property under Section 16, but the requirement is not universal.
Uncertain Tax Treatments and Other Changes From January 2026
The September 2024 edition of FRS 102 introduces amendments that take effect for accounting periods beginning on or after 1 January 2026, with early adoption permitted provided all the changes are applied together. The most significant addition for recognition purposes is a new framework for uncertain tax treatments. Where an entity has taken a position in its tax return that HMRC might challenge, it must assume the tax authority has full knowledge of all relevant information. If, on that assumption, the entity concludes it is not probable the authority would accept the treatment, the effect of the uncertainty is reflected in the current and deferred tax balances.
This aligns FRS 102 more closely with IFRIC 23 under IFRS. Any change in facts, circumstances, or the entity’s judgement about an uncertain treatment is treated as a change in estimate rather than a prior-period adjustment, so it runs through the current period’s tax charge. In practice, aggressive tax positions will attract more scrutiny, and some entities will end up recognising additional liabilities that previously stayed off the balance sheet.
The 2026 amendments also expand small-entity disclosure. Small entities must now disclose the net reversal of deferred tax expected in the next year, balances by type of timing difference, and expiry dates for any unused losses and credits. Those disclosures were previously required only from entities applying the full Section 29 requirements.