Beta This part of GOV.UK is being rebuilt – find out what this means

HMRC internal manual

VAT Partial Exemption Guidance

HM Revenue & Customs
, see all updates

Partial Exemption principles: apportionment



Apportionment refers to dividing residual input tax between taxable and exempt supplies. All partial exemption methods will have one or more apportionment calculations.

Potential alternative methods of apportionment are discussed below. Where businesses wish to use a calculation other than one based on income they will need to demonstrate that it is fair and reasonable (PE24000). This includes the requirement that a calculation which uses a non income based proxy can be guaranteed to be more accurate than one using the default income based methodology.

Outputs based apportionment

This is apportioning between supplies in the same ratio as the value of taxable supplies bears to the value of all supplies.

This is the default measure of use, being the mechanism used for the standard method. It is most likely to give a fair and reasonable result where the supplies are made in the same tax year as the costs are incurred, and where the input tax bearing costs used in making a supply increase in proportion to the value of a supply, for instance where a £1,000 supply will use about ten times the costs of a £100 supply.

Where there is the likelihood of large, distortive supplies that nevertheless use significant amounts of residual input tax, a sensible approach is to ‘cap’ the value of supplies at a level so that the method still gives a fair and reasonable result. For instance, treating any supply exceeding £10,000 as though it was £10,000 for the purpose of the calculation.

A common problem encountered is when the input tax used in making supplies increases with the value of supplies made, but more slowly than the value of the supply itself. Often values and transactions may appear the most appropriate methods, but neither is ‘accurate’. In such cases, either an alternative calculation must be found, or a judgement made as to which gives a fair and reasonable result.

Transaction based apportionment

This is apportioning between supplies in the same ratio as the number of taxable transactions made bears to all transactions made.

Usually this involves counting the number of supplies made, rather than the value of those supplies. However, other ‘numbers of things’ might also be appropriate to count. For example, when considering residual input tax in respect of a landlord, the number of buildings where a taxable rental is charged divided by total number of buildings rented out.

Transaction counts tend to be appropriate where the supplies and the related input tax are incurred in the same tax year and where the costs involved are fairly constant whatever the value of a supply. It is a commonly used apportionment method in the finance sector. The costs involved are often fairly constant, whatever the value of shares traded, loans made, etc. Similar examples occur throughout the finance sector, and indeed in many other sectors.

When approving a transaction-based method, it is essential that both parties agree what will or will not be treated as a transaction, and that this is set out unambiguously in the special method approval letter. There are no simple answers as to what should be considered a transaction and what should not, it will depend on the supplies made, and how costs are used in making them.

It should be noted that transaction counts can be weighted, such that if, for example, supply A uses four times the input tax bearing costs of supply B, each occurrence of supply A will count as 4 transactions and each supply B will count as 1 transaction.

Staff time and headcount apportionment

These are calculations where the apportionment is made in the same ratio as the staff time spent making taxable supplies bears to all staff time.

Common residual input tax bearing costs are those that support staff, such as computers, lighting, heating, rental, telephones and training. Staff time spent on different activities is in theory a very good method of apportioning these types of costs.

However, staff time based methods are only appropriate where the staff record their time for purposes other than partial exemption calculations and the liability of the supplies they are working on can be identified. For example, solicitors and accountants will record time regularly on time sheets for fee accuracy, but most other businesses will not.

For these reasons, staff time is rarely a fair and reasonable method of attributing input tax, often being both difficult to operate and impossible to verify.

A headcount method is similar to the staff numbers allocation method and has a similar effect to the using of staff time. However, it can be used by a business that does not keep time sheets. It is most likely to give a result that reflects the use of the input tax in the same circumstances as for staff time, but only where a significant proportion of staff are employed in areas that are involved in making supplies of only one liability.

Input tax and inputs based apportionment

These are calculations where the apportionment is made in the same ratio as the costs directly attributable to taxable supplies bear to all directly attributable costs.

These calculations are most commonly used where costs involved in making supplies are normally incurred in a different tax year than the year the supplies are made. To be able to produce a fair and reasonable result:

there must be directly attributable costs incurred relating to both taxable and exempt supplies;

a significant proportion of the costs must be directly attributable; and

the directly attributable costs must be incurred in about the same proportion as the residual input tax is used to make taxable and exempt supplies.

A property developer is a good example of a business for which such a method may produce a fair and reasonable result. The majority of residual input tax is likely to be incurred on staff supporting costs relating to the management of the various projects. The greater the costs relating to a project, the more time is likely to be spent managing it and thus the more residual input tax bearing costs relate to it.

Rather than costs, directly attributable input tax may be a more appropriate calculation, although this will not be the case if there are substantial exempt and / or zero-rated supplies received, or where one or both liabilities of supply is labour intensive.

Floor space apportionment

This is apportionment in the same ratio as the floor space occupied by income generating staff involved exclusively in making taxable supplies bears to the total floor space of income generating staff.

This is most likely to produce a fair and reasonable result where the residual input tax bearing costs relate primarily to buildings, such as taxable rent incurred, heating, lighting, fixtures and fittings. In order to reflect use, it requires that specific areas be used for supplies of a particular liability and that these represent the predominant proportion of the floor area.

Where staff are spread relatively evenly, such as in an office, a floor space calculation will normally produce a similar result to a headcount calculation. However, where the distribution of staff is uneven, a floor space calculation is more likely to reflect use than a headcount one. For instance, consider a business with a warehouse making exclusively taxable supplies and an office above making exclusively exempt supplies. The warehouse will have few staff in it, but use a significant amount of residual costs. Thought will need to be given as to whether input tax incurred on non-building related costs should be apportioned in the same ratio, since it is likely that the office would incur a greater proportion of, for example, telephone and computer related costs.

When considering a floor space calculation, you should satisfy yourself that the business has correctly identified directly attributable floor space. For instance, shops that sell taxable goods and exempt insurance on those goods often claim the display areas relate only to the taxable supplies. In fact, these areas are also used to make exempt supplies, since the intention in displaying the goods is to make both types of supply. The exempt insurance could not be sold if the goods themselves were not displayed. This means that very little floor space is directly attributable and the method is unlikely to produce a meaningful result. Accordingly floor space calculations are rarely fair and reasonable for the retail sector.

This was illustrated in the Tribunal case of Optika Ltd [VTD 18627] and confirmed in Banbury Visionplus & others [VTD 19266 & STC 1568]. Optika, a retail optician appealed against the Commissioners’ rejection of their proposed floor area based special method, and the Tribunal agreed that it did not give a fair and reasonable result. The following features from the Optika case demonstrate this.

Notional Splits of ‘Mixed Use’ Areas - Few areas of Optika’s premises were exclusively used for taxable or exempt activities, and the proposals called for notional splits of mixed use floor areas to create a workable calculation. This is unacceptable due to the inherent uncertainties in making and checking them.

Weighting of Floor areas - The proposed method required a weighting of the floor areas with those at the front of the shop having greater impact on the amount of tax deductible. Optika argued that this reflected a series of weightings, known as zoning used by landlords to calculate rents. The Tribunal held that each rental is a single supply covering the entire premises.

Use of Retail Areas - Optika assumed that the use made of the retail area and the shop front related to their taxable products. However the shop front attracted purchasers to purchase all of their products, and much of the selling of the exempt products took place in the retail area. Because the proposals were based on a mistaken assumption, they were unacceptable.

Similar findings were made in the Banbury Visionplus case.

You must not approve any method that includes one or more of the features confirmed as unacceptable in the Optika case. If you discover that a retailer is using such a method, you should contact your local PESO or TAPE officer for advice regarding withdrawal or amendment of the method.

The value to be used for the denominator

With all apportionment calculations, it is worth considering whether a fair and reasonable result is achieved by putting the taxable plus exempt figure or the total figure in the denominator (bottom line of the calculation). It is often argued that use of a denominator that includes the residual amount being apportioned is mathematically flawed.

We are concerned with achieving a fair and reasonable result, not with the mathematical purity of calculations. The purpose of a method is to reflect the use of input tax bearing costs in making taxable supplies. If that is best achieved by the inclusion of the residual amount within the denominator then it should be so included.

For example, imagine a situation where 5% of the input tax bearing cost of an organisation is directly attributable to taxable supplies and 20% of the input tax bearing cost is directly attributable to exempt supplies. This means 75% of the input tax is used to support both taxable and exempt supplies although the vast majority of it is used in support of exempt activity (say for example to support the provision of member’s facilities at a golf club where some food and alcoholic drinks are sold; this tax is residual following the decision in Bridgnorth Golf Club, [2009] UKFTT 126).

In such a situation a formula of taxable input tax over total input tax is more likely to produce a fair and reasonable result than taxable input tax over taxable plus exempt input tax.

This is because 5/100 or 5% is likely to be closer to the actual use of the costs than 5/25 (5+20) or 20%.