Repository navigation
Households' council tax on second homes and other property (engine and data) #1423
Description
Context. OBR's council tax receipts cover every chargeable dwelling, including second homes and empty homes. Microcosm's frame carries one council tax bill per household: the bill on its main residence. So draft PR #1422 (#1095, ruling D14f) binds obr.council_tax at the main-residence share only. That share uses MHCLG's October 2025 taxbase for England: chargeable dwellings less second homes and empty homes, about 0.968. Scotland and Wales take England's share, as a stated approximation.
What is missing.
- Engine (PolicyEngine/policyengine-uk):
council_taxis the bill on the household's main residence. A household that also owns a second home pays council tax on it, and councils may charge premiums on second homes and long-term empty homes. The engine has no input or formula for it. - Data (Microcosm): the frame carries the value of other residential property (WAS), but no count or band of other dwellings, and no flag for one that is a second home rather than let.
- High Value Council Tax Surcharge: Microcosm measures the surcharge on main residences only (UK: data fixes to model the High Value Council Tax Surcharge (mansion tax) correctly, and the related housing and council-tax gaps #1240). Coverage beyond main residences also waits on this.
Proposed scope.
- Engine: an input for council tax on other dwellings (or a count and band of them), added to household taxes.
- Data: draw it for households with other residential property that is not let, from the WAS and the second-home counts in Chronicle: MHCLG's England taxbase, StatsWales's chargeable second homes, and the Scottish Government's vacant private dwellings and second homes (one combined count).
- Then bind
obr.council_taxin full and retire the main-residence share.
Not in scope. Empty homes that no household holds stay outside a household file. Their share of the receipts stays out of the binding, as now.
🤖 Generated with Claude Code
Activity
vahid-ahmadi commentedon Oct 29, 2026
A check on the interim share, against MHCLG's Council Taxbase 2025, Tables 1 to 5 (revised 21 January 2026).
The 0.968 reproduces on dwelling counts. Chargeable dwellings 25,056,421; second homes 267,894 (Table 4); chargeable empty homes 542,260 (Table 3a). (25,056,421 − 267,894 − 542,260) / 25,056,421 = 0.9676. Only the chargeable empty count gives this: Table 3a's total of 754,264 empty homes includes 212,004 exempt ones and would give 0.959. Worth confirming that the Chronicle fact
mhclg.council_taxbase.empty_dwellingsis the chargeable count.On liability, the share is about 0.952. #1422's note says premiums make the true share "a little smaller". Sizing it from Table 2 by band, at Band D ratios (6/9 for A up to 18/9 for H):
That gives main residences 0.952 of Band D equivalents, second homes 0.020 and empty homes 0.028. On the £48.77bn row 15 window, the 1.6-point gap is about £0.8bn that the binding still asks main-residence bills to carry.
Caveats. Table 3a's premium tiers aren't split by band, so I spread them by the band shares of empty homes with a premium. The second homes discount is taken as 25% (6,861 homes; it barely matters). This is liability before council tax reduction, which falls on main residences only, so on the net receipts the binding uses the main-residence share would be a little lower again. Band A- is left out.
For step 3 of this issue, Table 2's band split of second homes, and of second homes with a premium, gives the band distribution to draw against.
🤖 Generated with Claude Code