The standard partial-exemption method at SI 1995/2518 regulation 101 apportions residual input tax (input VAT non-attributable to specific taxable or specific exempt supplies) by the ratio of taxable supply values to total supply values. For most mixed-portfolio landlords the standard method is administratively simple and approximates actual use closely enough. For some, it does not. Where the supply-value ratio materially differs from actual use of residual costs (typically because rental yields on the taxable and exempt sides of the portfolio do not track the cost-allocation pattern), three statutory levers move the landlord off standard-method default and toward a more accurate attribution.

This page is the depth on those three levers: the regulation 102 special-method approval route, the regulation 107A standard-method override, and the regulation 102B HMRC direction. Companion to our standard-method page, which covers the regulation 101 baseline. The special-method route is what comes after standard method when standard method does not work; standard-method override is the in-method correction when distortion is occasional rather than structural; HMRC direction is the route HMRC takes when the taxpayer has not moved voluntarily.

The standard-method baseline in 90 seconds

Regulation 101 of SI 1995/2518 sets the standard method. Inputs are classified as: directly attributable to taxable supplies (fully recoverable); directly attributable to exempt supplies (not recoverable); residual (non-attributable, apportioned). Residual input tax is apportioned by the ratio of the value of taxable supplies in the period to the value of total supplies in the period, expressed as a percentage rounded up to the nearest whole percentage point. The standard method is the default applied automatically; no HMRC approval is required to use it.

The standard method is statutory and well-suited to portfolios where supply values are a reasonable proxy for actual use. The Wave 5 standard-method page walks the mechanic in operational detail with the worked taxable / exempt ratio example.

The special-method route at regulation 102

Regulation 102 (verbatim heading 'Use of other methods') is the statutory route for a taxable person to move from the standard method to an HMRC-approved alternative tailored to the portfolio. Operative requirements:

  1. Prior HMRC approval. The taxable person cannot adopt a special method unilaterally. The Commissioners must have approved or directed the alternative method before it can be used. Applying a special method without approval produces invalid attribution; HMRC will recompute on the standard method and assess any under-recovered VAT with interest.
  2. Written method. The approved method must be set out in writing, identifying the supplies in respect of which it attributes input tax by reference to the relevant paragraph or paragraphs of VATA 1994 section 26(2). The written method is signed by HMRC and the taxable person and operates as the binding documentation of the methodology.
  3. Fair-and-reasonable declaration (from 1 April 2007). The taxable person must provide a written declaration that 'to the best of his knowledge and belief' the proposed method 'fairly and reasonably represents the extent to which goods or services are used by or are to be used by him in making taxable supplies'. The declaration is signed by the taxable person or representative and confirms reasonable steps have been taken to ensure all relevant information is held.
  4. Consistency requirement. Once approved, the special method must be used consistently until the taxable person requests a change (under a fresh regulation 102 application) or HMRC directs a change. Period-by-period switching between methods is not permitted.

The PESM (Proposed Special Method) document submitted with the regulation 102 application is the taxable person's analytical case for the proposed method. A well-prepared PESM contains: a description of the business and its property portfolio; identification of the supplies made (taxable, exempt, outside the scope); the proposed apportionment mechanic (e.g. floor area, headcount, sectorised); supporting evidence that the proposed method better reflects actual use than the standard method; numerical comparison of standard-method versus proposed-method residual recovery on a recent longer-period base; the fair-and-reasonable declaration. HMRC's Notice 706 paragraph 6 and the VAT Partial Exemption Manual at PE21000+ set out HMRC's expectations on PESM content.

Approval timelines vary. Simple PESMs (e.g. a floor-area-based sectorised method for a portfolio of clearly-distinguished commercial and residential elements) typically receive approval within 3 to 6 months. Complex PESMs (e.g. headcount methods with shared-overhead-attribution mechanics, or transaction-count methods on multi-channel rental portfolios) can take 6 to 12 months and frequently require negotiation with HMRC's Partial Exemption Team on the specific apportionment mechanics.

PESM typology: four methods typically approved for property portfolios

MethodMechanicBest fitCommon limitations
SectorisedPortfolio split into taxable-sector (commercial opted) and exempt-sector (residential exempt); separate apportionment in each; residual within-sector costs apportioned per sector; residual cross-sector costs apportioned by sector value or other agreed metricLarger portfolios with clearly-distinguishable commercial and residential elements; head-office overhead structure that can be sub-attributed by sectorCross-sector residual cost apportionment can itself distort; some PESM applications fail at this stage
Floor-areaResidual input tax apportioned by taxable-use floor area (m²) as percentage of total floor area; uses Land Registry plans or commercial lease plans to evidence floor areasMixed-use buildings with clean commercial / residential floor splits; portfolio with comparable yield per square metre across taxable and exempt elementsDistorts where rental yields per m² differ materially between sectors (high-yield commercial vs low-yield residential), which is common on London portfolios
HeadcountManagement overhead apportioned by headcount allocated to taxable-side versus exempt-side activities; useful where commercial and residential portfolios are managed by distinguishable teamsPortfolios with formal team structure where staffing is structurally split by tenant type; landlords with in-house management distinguishing commercial leasing from residential lettingsOutsourced or shared management invalidates the method; requires regular headcount-allocation reviews
Transaction-countApportionment by count of taxable-side transactions versus exempt-side transactions in the periodHigh-volume short-let portfolios with seasonal mix; multi-property letting platforms where the supply-value method distorts seasonallyLess common; HMRC scrutinises transaction-count methods carefully because transaction count is a weaker proxy for actual use than value or floor area on most property portfolios

The right method depends on portfolio shape, cost structure, and the specific distortion the standard method produces. A floor-area method works where the value-based standard method over-credits high-yield commercial space; a headcount method works where management time is the relevant residual cost driver; a sectorised method works where the portfolio decomposes cleanly into separate operational books. PESM preparation typically involves modelling each candidate method against the existing portfolio data to identify which produces the closest match to actual-use evidence.

The standard-method override at regulation 107A

Regulation 107A (verbatim heading 'Adjustment of attribution') is the in-method correction provision, inserted by SI 2002/1074 (The Value Added Tax (Amendment) Regulations 2002) with effect from 18 April 2002. (House position §29.6 originally locked citing SI 2002/1142 as the inserting instrument; the verified citation against legislation.gov.uk on 2026-05-25 is SI 2002/1074. The position has been corrected at Stage 2b.)

The override operates within the standard method. The taxable person continues to use the standard method under regulation 101 but corrects the standard-method attribution to actual use where the attribution 'differs substantially' from actual use of the goods or services in making taxable supplies. The correction is calculated at the longer-period reconciliation and reported in the prescribed accounting period following the longer period.

'Differs substantially' is not numerically defined in regulation 107A. HMRC guidance (Notice 706 paragraph 7) indicates that differences of a few percentage points or more on residual input tax of material size trigger the override; smaller differences or differences on small residual cost bases may not. The override is taxpayer-initiated and self-assessed; HMRC does not need to direct it. The taxpayer who fails to apply the override when the conditions are met is exposed to HMRC assessment with interest at the next enquiry; the taxpayer who over-applies the override produces inaccurate VAT returns and may face an inaccuracy penalty.

The override is materially useful as a stop-gap in two scenarios: while applying for and awaiting regulation 102 approval (the override corrects standard-method distortion during the interim period); and where distortion is occasional rather than structural (a one-off project, an unusual tenant mix in a single longer period). For persistent distortion, the regulation 102 special method is the cleaner solution.

The HMRC direction route at regulation 102B

Regulation 102B authorises HMRC to direct a taxable person to use a particular partial-exemption method. The direction route is the HMRC-led counterpart to the taxpayer-led regulation 102 application. The direction is binding from the date specified.

HMRC issues directions sparingly because they require HMRC analytical work (whereas regulation 102 applications come with the taxpayer's analysis built in). Direction territory typically opens after HMRC enquiry concludes the taxable person's current method (usually the standard method) is materially over-recovering input tax, and the taxable person has not voluntarily applied for a special method. The direction sets out the HMRC-preferred alternative method.

The defensive move where HMRC has indicated direction is contemplated is to pre-empt with a regulation 102 application: the taxpayer prepares the PESM, submits it for approval, and HMRC's review of the application essentially substitutes for the direction process. The pre-emption keeps the methodology choice in the taxpayer's hands rather than HMRC's; the same special method may emerge either way, but the route shapes which side prepares the analysis.

The annual longer-period adjustment under regulation 107

Partial-exemption attributions are provisional at each quarterly (or monthly) VAT return; the annual longer-period reconciliation under regulation 107 corrects the cumulative position to the annual ratio. The longer period typically aligns with the taxable person's annual accounting period (often 1 April to 31 March for property-business VAT groups; alternative year-ends require HMRC agreement).

At the end of the longer period, the taxable person calculates the annual taxable-supplies ratio (under standard method) or annual approved-method ratio (under special method) and compares it with the cumulative quarterly attributions. The difference (positive or negative) is reported on the VAT return for the period in which the longer period ends. Where the regulation 107A standard-method override is being applied, it is calculated against the longer-period figures, not quarterly.

The longer-period adjustment is one of the most-common HMRC enquiry triggers because errors crystallise late and tend to be material. A landlord who has been over-recovering on quarterly returns through the year produces a large negative adjustment at year-end; sessions should reconcile quarterly to longer-period at least once mid-year to catch the position early.

The de minimis test at regulation 106

Regulation 106 sets the de minimis threshold below which the taxable person is treated as fully taxable (recovering all input tax including the exempt-related portion). Two conditions, both required:

  • Total exempt input tax does not exceed £625 per month on average (£7,500 per year).
  • Total exempt input tax is not more than 50% of total input tax.

The test is applied annually at the longer-period reconciliation. Where the test passes, no partial-exemption mechanic is required for the year; all input tax is recoverable. Small mixed-portfolio landlords with limited exempt residential activity routinely pass the de minimis test annually and avoid partial-exemption complexity entirely. As portfolio mix shifts year to year, the de minimis position can change; the test must be re-run each longer period.

Want this checked against your specific situation?

Drop your email and a one-line summary. We reply within 24 hours, no phone call needed.

Lansdowne Estates worked example: floor-area PESM uplift

Lansdowne Estates Limited (anonymised mixed-portfolio landlord) owns three opted commercial buildings and seven residential blocks managed in-house. Annual rental income: commercial £4.8m plus 20% VAT; residential £3.2m exempt. Annual residual input tax (management costs, professional fees, software, insurance, head-office overheads): £400,000 input VAT.

Under the standard method (regulation 101): ratio = £4.8m taxable / £8m total = 60%. Recovery on residuals = 60% × £400,000 = £240,000.

Lansdowne's analysis of actual use of residual costs (a 6-month exercise reviewing management time logs, building services breakdowns, professional fee invoices) shows actual use is 72% taxable. The residential blocks consume disproportionately less management time per rental £ because tenant management is largely automated through letting agents whereas commercial leases involve substantial ongoing landlord administration (rent reviews, dilapidations, insurance claims, planning correspondence, business rates appeals).

Under a floor-area-based PESM (regulation 102): recovery = 72% × £400,000 = £288,000. Uplift of £48,000 per year.

Lansdowne applies for the PESM under regulation 102. HMRC review takes 6 months. The PESM is approved with effect from the following longer-period start. For the longer period preceding approval, Lansdowne applies the regulation 107A standard-method override: an adjustment of £48,000 (the difference between standard-method 60% and actual-use 72% on the £400,000 residual base) is reported on the VAT return for the period in which that longer period ends.

Implementation cost: approximately £15,000 to £25,000 of advisory time preparing the PESM, supporting analysis (floor-area survey, management-time studies), and the HMRC application. Ongoing compliance cost is minimal once the PESM is approved. Two-year payback on the initial advisory cost; recurring £48,000 annual uplift thereafter for as long as the portfolio shape supports the floor-area approach.

The de minimis transition and the partial-exemption cliff

The de minimis test under regulation 106 is annual at the longer-period reconciliation. A portfolio that passes the test in year 1 (full input-tax recovery, no partial exemption) and fails in year 2 (because the exempt residential element has grown) sees a sharp transition: input-tax recovery on the residual cost base drops from 100% to the standard-method ratio overnight, with no graduated phase-in. The cliff effect is most acute on small portfolios near the £625-per-month threshold; a single new residential acquisition can push the portfolio across the line.

The defensive operational move where the portfolio is near the de minimis threshold is to plan acquisitions to either keep the portfolio safely inside the test (acquisitions structured to maintain a low exempt-input-tax ratio) or to push the portfolio decisively outside the test with a structurally appropriate special method ready to file (so the post-cliff position is a properly-designed PESM rather than a defaulted standard method that may itself distort). Sessions advising on portfolio growth planning should run the de minimis test annually against the projected portfolio shape, with PESM scoping work commissioned 6 to 9 months ahead of the projected cliff if the structural distortion is foreseeable.

Capital Goods Scheme interaction with the partial-exemption method

The Capital Goods Scheme runs alongside the partial-exemption method, not separately. For each capital item within the CGS (£250,000+ capital expenditure on land or buildings, VAT-exclusive), the initial recovery rate at acquisition or first interval is calculated under the partial-exemption method in operation at that point (standard method, special method, or override-adjusted standard method). At each subsequent CGS interval the recovery is re-tested against the prevailing partial-exemption method.

The structural implication: a change in partial-exemption method (e.g. switching from standard to a floor-area PESM under regulation 102) cascades into the CGS adjustment on every capital item then under adjustment. The CGS interval following the method switch will produce a different recovery proportion under the new method, with the difference (positive or negative) flowing through as the standard CGS interval adjustment. Where the method switch produces a substantial uplift on the standard-method baseline (the Lansdowne 12-percentage-point example above), each subsequent CGS interval recovers more of the original capital-item input VAT until the 10-interval period expires. The cumulative effect over the full CGS adjustment period can be materially larger than the annual residual-cost uplift alone.

The decision tree: matching portfolio shape to method choice

  1. De minimis check first. Does the annual longer-period de minimis test pass under regulation 106? If yes, no partial exemption applies; recover all input tax.
  2. Pure-commercial portfolios. If the portfolio is entirely commercial-opted with no residential exempt activity, no partial-exemption mechanic is required. The option to tax handles the entire input recovery.
  3. Standard-method-vs-actual-use comparison. Calculate the standard-method ratio and an actual-use indicator (floor-area, headcount, or transaction-count proxy). If within 1 to 2 percentage points, stay on standard method.
  4. One-off distortion. If the distortion is occasional, apply the regulation 107A standard-method override at the longer-period reconciliation and stay on standard method going forward.
  5. Structural distortion. If the distortion is persistent year-on-year, apply for a special method under regulation 102 with a tailored PESM. Floor-area is the most common starting point for mixed-property portfolios; sectorised methods suit larger structurally-split portfolios; headcount and transaction-count methods are niche.
  6. HMRC pressure. If HMRC has indicated dissatisfaction with the current method, pre-empt with a regulation 102 application before a regulation 102B direction is issued. The taxpayer-led PESM preserves methodology choice; the HMRC direction does not.

Authorities