Lock policyengine-uk 2.151.0 and write the UK inputs it reads (#1095, #1278)

changelog.d/1095-uk-energy-price-basis.added.md

@@ -0,0 +1 @@
+Every UK H5 that stores domestic energy spending now declares its price period in the root `energy_price_basis` attribute, `{"period_type": "calendar_year", "period_value": 2024}`: the calendar-2024 prices `lcfs_consumption` sets every household spend column at, which the stage-time gate `uk_stage_lcfs_consumption_basis` holds (microcosm#1113). From policyengine-uk 2.148.1 (PolicyEngine/policyengine-uk#2571) the engine projects declared electricity, gas and domestic energy spending by its consumer price index and keeps undeclared spending at its stored prices in every later year, so a later year costed on a Microcosm release no longer prices energy at 2024 prices (PolicyEngine/policyengine-uk#2526, its remaining item 2; microcosm#1095). On a file that declares it, each household's electricity, gas and domestic energy spending moves to a later year by the engine's energy price index from 1 January 2024, every household alike, where an undeclared file keeps its stored spending; releases built before this change stay undeclared. An engine-tier test loads a written file through `Microsimulation` and checks 2025 and 2026 against the index, and a copy without the declaration against the stored spending (Vahid's review of microcosm#1383, item 3). The calibration's 2025 measurement of the energy rows moves with it.

changelog.d/1095-uk-held-rows-2026-11-26.changed.md

@@ -0,0 +1 @@
+The UK national rows held out until 2026-11-26 are settled (microcosm#1095, María's ruling of 2026-10-22). The zero-support floor (`uk/calibration_measure_exclusions.json`, schema 4) now holds its rows in scopes of one unit each, each with its own ceilings, because a share ceiling sums published values. A scope's unit is the one its targets' contract measurements give (a count's entity, or pounds), and the run refuses a scope labelled otherwise; the band-H scope counts households, the occupied dwellings its rows measure. A register at schema 3, the single-scope floor, is refused rather than read, so an old run directory's register no longer replays without its floor restated at schema 4. The scopes are the UC payment-distribution and benefit-cap amount bands, the HMRC SPI income-band counts, the HMRC SPI income-band amounts, and the band-H council-tax rows by English region. On #1333's policyengine-uk 2.145.0 spine every HMRC £1m-and-over cell and every band-H region cell has carriers. The three band-H region entries retire and the rows bind, with their figures withheld where fewer than ten source households carry them; Plan 5's liable stock, which the stage fits at design weights, binds too. The ten HMRC £1m-and-over cells were bound on a calibrate-only arm of that spine, which measured a distortion of their SPI-channel neighbours, so they stay held on that measurement under María's rule of 2026-10-21, re-signed to 2027-01-23. The OBR Universal Credit split at the welfare cap and the England postgraduate repayments leave the targets for `registry_parity.excluded`: the model marks no welfare-cap boundary, and the stage assigns no postgraduate plan. `obr.universal_credit` still binds the sum of the two cap cells.

changelog.d/1095-uk-held-rows-arm2.changed.md

@@ -0,0 +1 @@
+Arm 2 of the next policyengine-uk bump, one spine and one calibration on policyengine-uk 2.151.0, re-measures the UK rows still held out of the objective (microcosm#1095). Four of the twelve UC payment-distribution bands, which microcosm#1344 now draws by award band, come within 10 percent unbound, so their entries retire and the rows bind; the other eight keep their renewals of 2026-10-23. The four Housing Benefit rows and DWP's JSA claimants stay outside the 25 percent bound at design weights and are renewed, signed on 2026-10-24 to 2027-01-24, pending the Housing Benefit and JSA PR (microcosm#867, microcosm#869). No reason gives a new model figure of a row that rests in part on the households of the SPI support channel (SN 9422).

changelog.d/1095-uk-held-rows-renewed-arm1.changed.md

@@ -0,0 +1 @@
+The UK rows still held out of the objective are renewed on a calibrate-only arm of #1333's policyengine-uk 2.145.0 spine (microcosm#1095), signed on 2026-10-23 to 2027-01-23: the twelve UC payment-distribution bands (all outside 10 percent unbound; they wait on the award-band take-up of microcosm#1344), the two Plan 2 borrower stocks (the coverage gap stands), Plan 5's repayments (too few source households carry them for any weights to reach SLC's figure; its figures are withheld) and the any-tenure UC housing-element count (at the other/unknown-tenure gap itself). No reason gives a new model figure: the rows rest in large part on the households of the SPI support channel (SN 9422), and Plan 5's repayments on fewer than ten source households. Each adjudication keeps its history.

changelog.d/1095-uk-legacy-closure-owners.fixed.md

@@ -0,0 +1 @@
+The Move to UC claim draw (`uc_legacy_closure_take_up`) now reads each benefit unit's legacy and Universal Credit reports from its claimant and partner only (`is_claimant_or_partner`; microcosm#1095). Since policyengine-uk 2.146.0 (pe-uk#2549) the engine closes a legacy award through its owners, so an award another member claims in their own right closes on its own. The draw counted every member's reports, so such an award could put the family in the draw, or keep it out as a Universal Credit unit.

changelog.d/1095-uk-local-thin-rows-withheld.changed.md

@@ -0,0 +1 @@
+Constituency and local-authority rows now follow the ten-household floor for their published figures (María's ruling of 2026-10-23 on the local line): a local row resting on fewer than ten source households keeps its place without its model figures, and a published local row that would give it back by difference is withheld beside it, as national, regional and BRMA rows are. The area-support rule still decides which local rows bind. `uk_withheld_row_sets` takes its span per block, since the local surface is too large for one dense span: the national, regional and BRMA rows together; each region's rows of every grain, so a region row less its published local cells, or a region's constituencies less its local authorities, is caught; and each local family with the higher-grain targets the cross-grain declarations or a nested measurement tie it to, so a national row less a family's published cells is caught too (Vahid's review of microcosm#1383, item 1). The blocks overlap, so a row one block withholds beside a thin one is protected in every other block that holds it, and the blocks run again until none withholds another row. Local rows need the frame's household region. The report guard matches a withheld local row by its own `family/area/metric` key, and a bare area code only for a withheld BRMA row (`uk_withheld_gate_areas`), so a published row of another family in the same area may still be named. BRMA rows publish from ten households under the national block's check, and beside Scotland's local rows in its region's block where a run carries them (María's rulings of 2026-10-17 and 2026-10-24).

changelog.d/1095-uk-self-employment-receipts-default.added.md

@@ -0,0 +1 @@
+`UK_ENGINE_DEFAULT_INPUTS` (`microcosm.frame.adapters.policyengine_uk_concepts`) signs the engine inputs the UK build leaves at policyengine-uk's default as approximations, each with the error it can make (microcosm#1095). Its first entry is `self_employment_gross_receipts`: no source of the self-employed's turnover is read yet, so `trading_allowance_deduction` relieves a profit of £1,000 or less in full and gives a larger profit no allowance, which puts taxable profit at most £1,000 a person from where receipts would put it. Its second is `housing_benefit_residential_capital_exception` (policyengine-uk 2.148.4): the FRS samples private households and records no assessment of the residential accommodation or saved circumstances the exception covers, so it stays False, and Housing Benefit is understated only for a qualifying working-age resident, who takes tariff income on capital above £6,000 instead of £10,000 (Vahid's review of microcosm#1383, item 4). Tests check that each entry is an engine input the mapping never binds and that no spine column writes it.

changelog.d/1095-uk-spi-public-use-tape-marker.added.md

@@ -0,0 +1 @@
+UK builds mark the outputs that rest on the SPI 2022-23 Public Use Tape (SN 9422), whose Special Conditions add two obligations to the End User Licence: no attempt to attribute a record to a named individual, and analyses of the tape go to HMRC at least ten working days before they are distributed or published (microcosm#1095). `microcosm.build.uk_runtime.spi_dependency` reads which of a run's stages declare the tape among their artifacts. The spine's build record carries the marker (`spi_public_use_tape`), the spine provenance in the calibrated candidate and its signed diagnostics carry it on, and both release assemblers restate it in `release_manifest.json`: on the microdata's own artifact entry (`spi_public_use_tape_dependent`) and, with the conditions, under `licensed_data`. The assemblers rederive it from the stages the signed provenance names and refuse a recorded marker that disagrees. How SPI-dependent outputs are released stays María's ruling; the marker records the dependency only.

changelog.d/1095-uk-uc-take-up-esa-kept.fixed.md

@@ -0,0 +1 @@
+The UC take-up redraw keeps a reported income-related ESA award off a Universal Credit claim. From policyengine-uk 2.149.0 a claim for Universal Credit ends a claimant's or partner's income-related ESA (pe-uk#2563), and the earlier single-rate draw sets many units that report the award, and no UC, to claim it, so the engine would end their ESA at the calibration year. A unit whose claimant or partner reports income-related ESA, and that neither reports Universal Credit nor is moved off its legacy benefits, now claims none, as a reported receipt is a fact; the closure draw reads the same flag, and the stage receipt bands the units it keeps off Universal Credit (microcosm#1095). Housing Benefit, which the engine also ends on a claim, keeps the stored flag pending the Housing Benefit and JSA PR (microcosm#867, microcosm#869).

changelog.d/1095-uk-was-property-draw.added.md

@@ -0,0 +1 @@
+The WAS wealth stage now draws the share split and the debt secured on other property, the inputs policyengine-uk reads inside `corporate_wealth` (pe-uk#1974) and in valuing Universal Credit property capital (UC Regs 2013 reg 49(1)(b)); microcosm#1095, María's ruling of 2026-10-23. Three draws are appended after consumer debt, each fitted on the donors and drawn for the recipients holding the column it rests on, so the first eight segments draw as before. The first is the directly held fraction of the share-like holdings outside ISAs: `directly_held_shares` (WAS UK shares plus employee shares and options) and `unit_and_investment_trusts` (WAS collective investments) are written on the final `corporate_wealth` less the stocks-and-shares ISA, so the three sum to `corporate_wealth` and the engine's `share_like_capital` does not move. The other two are `other_residential_property_secured_debt` (the debt on other houses and buy-to-let property) and `non_residential_property_secured_debt`: each is zero where its asset is zero, and the two are parts of the mortgages on other property, scaled down together where they would pass them; the donor cleaning refuses a donor whose debts pass its other mortgages, and the wealth coherence gate holds the split and the debts as identities. The recorded Universal Credit and Pension Credit capitals (`uc_capital_coherence`, and the reporter redraw's screen) now net each class's secured debt after the costs of sale and floor each class at zero, as the engine values UC property capital and as SPC Regs 2002 reg 19(b) values Pension Credit capital; owned land carries no debt, as the engine has no input for it (pe-uk#1969). The four columns are reviewed extras of the export surface.

changelog.d/1222-uk-widowed-parents-allowance.changed.md

@@ -0,0 +1 @@
+FRS benefit code 6, Bereavement Support Payment and its predecessors, now splits by rate (microcosm#1222, María's ruling of 2026-10-22). A weekly amount within 1% of a Bereavement Support Payment rate's weekly equivalent (GBP 350 or GBP 100 a month, frozen since April 2017, at 12/52 or at four weeks) stays `bsp_reported`; an amount at Widowed Parent's Allowance's 2024-25 rate of GBP 148.40 a week (within 1%), or below it as a reduced award, goes to `widowed_parents_allowance_reported`, the input policyengine-uk 2.149.0 adds (pe-uk#2572); any other amount, above the allowance's rate and at no rate (a lump sum, a part month or a misreport), stays `bsp_reported` (Vahid's review of microcosm#1383, item 12). Unlike Bereavement Support Payment, the allowance is taxable, Universal Credit unearned income, a welfare benefit for the benefit cap and an overlapping benefit for Carer's Allowance and Carer Support Payment, and Housing Benefit counts it less £15 a week. The column goes wherever War Widow's Pension goes: the SPI channel restores it from the FRS twin, and it joins the export allow-list, since the enhanced FRS keeps the whole code in `bsp_reported`.

changelog.d/1278-uk-heating-fuels.added.md

@@ -0,0 +1 @@
+The LCFS stage now draws household spending on liquid fuels (central-heating oil, paraffin and bottled gas, COICOP 04.5.3) and solid fuels (coal, coke, wood and peat, 04.5.4) into `liquid_fuel_consumption` and `solid_fuel_consumption`, the inputs policyengine-uk 2.149.0 reads (pe-uk#2572; microcosm#1278; María's ruling of 2026-10-22: LPG with liquid, heat in the remainder). They are the last chain targets, so every earlier draw is unchanged; they are uprated to calendar-2024 prices as every diary spend column is; and the housing total nets their draws and adds them back, so it always holds them, while hot water, steam and ice stay in its remainder. The engine counts them inside housing, so `consumption` does not move. The recomposed-totals receipt's `uncarried` block becomes `published_fuels`: ONS 04.5.3 and 04.5.4 beside each fuel's donor total at diary prices and the frame's draw, which the stage-time gate recomputes from the vendored rows. COICOP puts liquefied hydrocarbons under 04.5.2 (gas), so ONS's 04.5.3 leaves out the bottled gas the liquid column holds, and the receipt gives the donor's liquid fuels without it too (Vahid's review of microcosm#1383, item 8). Neither column is levelled to ONS, unlike electricity and gas: the donor's diary totals are about twice ONS's, and the frame's draws reach neither those nor ONS's, so a costing on either needs levelling to ONS first; no engine formula reads them yet. `ons.household_energy_expenditure`, which the contract keeps for registry parity and the reference surface signs out as superseded by its fuel rows, is measured on electricity, gas, liquid and solid fuel spend, and its signed rationale no longer says the frame lacks the last two. The consumption drift block reports the two fuels where the frame carries them, and `uk_consumption_drift_flags` gives warn-only flags at #1278's R1 thresholds: a total moving more than 2 percent, or a share more than half a point, between the design and the final weights. The calibration diagnostics carry each flag as a warn-only `diagnostic_warnings` entry, naming the total or share and the threshold (Vahid's review of microcosm#1383, item 11).

changelog.d/1278-uk-lcfs-division-excess.changed.md

@@ -0,0 +1 @@
+The LCFS stage now levels the COICOP divisions' relative over-draw before it holds each recomposed total to the donor's maximum (microcosm#1278; María's rulings of 2026-10-23 and 2026-10-24). A division's ratio is the spend the chain drew for it per frame household over the donor's mean of the same spend per donor household: for housing and transport, the remainder against the donor's total less the same parts, since the levelled electricity, gas, heating fuels, road fuel and priced bus fares keep their published levels and are no over-draw (Vahid's review of microcosm#1383, item 9); for any other division, its whole draw. A division above the divisions' shared ratio, measured once before any levelling, comes down to it, and no other division moves: measured again after each cut, the shared ratio would fall until every division sat at the lowest one's, close to levelling to the donor totals (María's ruling of 2026-10-24). The recompose operation declares the rule as `level_division_excess`, the recomposed-totals receipt records each division's ratio and factor, the shared ratio and each total's `levelling_factor`, and the stage-time gate refuses a missing or undeclared levelling, a factor outside [0, 1], a division that moves though it is not above the shared ratio, and a total whose remainder moves by other than its division's factor.

changelog.d/1278-uk-obr-vat-held.changed.md

@@ -0,0 +1 @@
+`obr.vat` stays held out of the objective, measured on every evaluation as a diagnostic (microcosm#1113, microcosm#1278). María's ruling D3 of 2026-10-08 held it out until a locked policyengine-uk replaced the constant survey-coverage factor in `vat`; policyengine-uk#2187, released in 2.135.7, does that, but the factor it uses was derived on a certified Microcosm UK release so that its VAT meets OBR's receipts, so binding the row would mostly pull the consumption weights back toward that release (Vahid's review of microcosm#1383, item 10; María's ruling of 2026-10-24). Its entry is re-signed to 2027-01-24, and the row rebinds when policyengine-uk re-derives the factor on a file that does not bind it. The notes declare the residual the row keeps: OBR's receipts include VAT on private school fees from 1 January 2025, which policyengine-uk's current-law `vat` leaves out until policyengine-uk#2319.

changelog.d/1278-uk-orr-rail-held-arm2c.changed.md

@@ -0,0 +1 @@
+The ORR rail-support row stays held out of the objective (microcosm#1278). Bound on ORR table 7271's net operational government support from all sources, the default the plan of 2026-10-22 set on the open 7270-versus-7271 ruling (microcosm#890 G), it reached its target on a calibrate-only arm of the next policyengine-uk bump, but the weights concentrated past the release bound (`uk_weight_ratio` failed) and the effective sample size fell by about a tenth. Its entry keeps table 7270's basis and its 2026-12-11 window, and its reason records the measurement.

changelog.d/1308-uk-new-stage-health-gates.added.md

@@ -0,0 +1 @@
+The three stages #1308 left ungated have stage-health gates (María's ruling of 2026-10-17, taken up with the next policyengine-uk bump), each release-blocking and blocking on an absent receipt. `uk_stage_property_purchases_fit` holds each purchase kind to HMRC's Stamp Tax transactions within 5% and to its value less a recorded shortfall of at most one fifth, the floor the stage itself refuses below. `uk_stage_regional_home_values_ranking` requires rows re-ranked in every English region and Wales on the 2024 knots, at the weighted midpoint within each FRS region, with no source at the value cap. `uk_stage_uc_legacy_closure_take_up_contract` checks what the Move to UC receipt prints, since it withholds what a thin combination realizes: the design-weight Great Britain basis, a probability and a declared source for every combination's rate, a printed support band, and a well-formed unreachable flag. The property purchases and regional home-values gates check the frame against published facts, so a synthetic smoke build records their failures without blocking on them (`population_fact_check`), as it does for the other fact gates; the legacy closure gate blocks every posture. The mirrored battery digests move.

changelog.d/1319-uk-carers-allowance-scotland.changed.md

@@ -0,0 +1 @@
+The Carer's Allowance take-up draw (`carers_allowance_take_up`) now draws Scotland too (María's ruling of 2026-10-23; microcosm#1319). In Scotland the claim flag is Carer Support Payment's, and since policyengine-uk 2.146.1 (PolicyEngine/policyengine-uk#2473) the engine pays it only under its own earnings limit. So a Scottish carer whom the 2024-25 limit kept from claiming is drawn at the newly entitled rate as elsewhere, and is paid only where Carer Support Payment's rules allow. The receipt reports Scotland apart (`scotland_eligible_support`, `scotland_eligible_weighted_persons`, `scotland_drawn_support`) in place of the closed-region check.