Scottish Building Safety Levy: indicative rates

Annex A – Methodology note

Levy base

New build completions

The main component of the levy base is new build completions. This is a wider measure than new build sales, since not all new build completions are sold (for example, Build to Rent and self build). As a result of significant variation in new residential completions from year to year, and given that the levy is intended to be in place for a period of 15 years, a 6-year average of new build completions^[5] has been used as the base. Since more recent data has been published since the Financial Memorandum for the Act was prepared, this 6-year window has been updated by one year and now covers the period 2020-2025.

The calculation starts with this 6-year average of total new build completions (including both private and social sector)^[6] and deducts a corresponding 6-year average of new build units (for social rent, affordable rent and affordable home ownership) which have received funding from the Scottish Government’s Affordable Housing Supply Programme.^[7]

This figure is then further reduced by new build completions in the island areas. A 6-year (2020-2025) average of private sector^[8] new build completions in the three island local authorities (Na h-Eileanan Siar, Orkney Islands and Shetland Islands) is used as a proxy for this.^[9]

Conversions

Data on conversions is collected from local authorities in a separate return to new build completions, and only once per fiscal year. The six-year average is therefore based on fiscal years covering the period from 2019-20 to 2024-25 (the latest data at the time of modelling). The components used are new residential units which have been created from non-residential buildings, as well as new residential units created from integration and subdivision of residential properties.^[10]

Purpose build student accommodation (PBSA)

PBSA is not included in the new build data returns to the Scottish Government, and there is no official dataset on this sector. Therefore, a commercial dataset (CoStar) was used to obtain data on developments classified as student accommodation in that dataset. The year built or refurbished was used to determine the likely date of completion, and a 6-year average from 2020-2025 was calculated.^[11] Since the levy will be charged on an area basis, the field “rentable building area” in the dataset was used as the closest match to the definition of area in the Act, 15% was deducted from this measure of area to make allowance for common areas in the building. This deduction is a general assumption on the proportion of these developments relating to communal areas.

Other components of calculation

New build price

New build prices are taken from the Registers of Scotland data. Again, a 6-year average (2020-2025) of (mean) house prices in each local authority is used.^[12]

Levy-free Allowance

The first 29 units completed by a developer will be Levy-free. To estimate the impact of this allowance, a bespoke 3-year dataset (covering fiscal years 2022-23 to 2024-25) was obtained from Registers of Scotland. Linked developers were identified using Companies House data, and this linking exercise was quality assured by an independent contractor. The share of sales for each developer which was up to 29 units comprised 23% on average over 3 years. This figure was used as a proxy for the share of new build, conversions and PBSA completions which could be eligible for the Allowance.

Greenfield: Brownfield split

To estimate the Greenfield:Brownfield split, an internal Scottish Government dataset was used containing Housing Land Audit data from 18 local authorities.^[13] For those local authorities not covered by this dataset, the split was proxied by that of their nearest neighbour in terms of the urban:rural make-up of local authorities. It should be noted that the greenfield:brownfield split only affects the total magnitude of the levy, and not the variation of the levy across local authorities (see further below). On average, the share of residential units on greenfield site is estimated to be around two-thirds.

Area

The Energy Performance Certificate database was used to obtain an estimate of the average area of residential units in different local authorities. Once again, a 6-year average was used (2020-2025). New build was identified by the type of assessment undertaken (new build SAP). Potential duplicates were removed, and dwellings were allocated to local authorities using a postcode to local authority look-up, and where that did not result in a match,^[14] the assessor’s identification of the local authority was used.

Another field available in the dataset is the tenure of the dwelling. There are some issues with this field, in particular the EPC assessor may not know what the tenure will be (as evidenced by a significant share of records were the tenure is marked as “unknown”). The data show that the average area of dwellings where the assessor marked the tenure as being “social” was smaller than other categories. Since AHSP-funded units are exempt, these units were excluded. However, the number of units marked as “social” is smaller than would be expected given the number of social completions; thus it is likely that some of the unknown tenure are social units. The data limitations could therefore have resulted in an underestimate of the average size of non-AHSP funded units, but analysis suggest this is likely to be small.

Calculation of levy

The levy is set so that its value expressed as a percentage of the average purchase price is the same in each local authority. To obtain the levy on per m2 basis, the levy is divided by the average size of a unit in that local authority, so an equivalent statement is that the levy is set in such a way that it comprises the same percentage of average house price per m2 in each local authority.

While the variation in the levy between local authorities reflects differences in house prices per m2, the overall level of the levy is set so that, given the number of completions in scope, and the fact that completions on brownfield are subject to a 50% relief, the revenue target of £30 million per year is achieved.