Bind local-authority energy and regional bus journeys before the Budget (#1207)

changelog.d/1207-bus-engine-parity-and-vat-base.added.md

@@ -0,0 +1 @@
+Two report-only additions for microcosm#1207. The UK measure resolver's receipt (carried in every build's measures receipt) gains `bus_engine_parity`: policyengine-uk's `gov.dft.bus.boardings_per_trip` and `yield_per_boarding` beside the boardings per trip and fiscal-year yield the `lcfs_consumption` stage priced each area at, with each relative gap marked against a 0.1% tolerance. The engine prices bus fares itself from the dataset's trips, so this is the check that its pricing still matches the stage's; an engine without those parameters records why it has none. The calibration diagnostics' `consumption_drift` block gains the full-rate VAT base (consumption times the ETB VAT stage's full-rate expenditure rate) at design and final weights, since no bound row holds it.

changelog.d/1207-bus-journeys-by-region.added.md

@@ -0,0 +1 @@
+Nine national target rows bind DfT BUS01a local bus passenger journeys for FY2024-25 in London and the eight other English regions (family `dft_bus_journeys`, microcosm#1207). The frame carries NTS trips, not boardings, so the resolver's new household measures `bus_in_london_boardings` and `other_local_bus_boardings` price each household's trips at the boardings per trip that the `lcfs_consumption` stage priced each series at. A persisted column of either name is refused. London's row counts every resident's bus-in-London trips. Each other region counts its residents' other local bus trips. The rows constrain the frame's trip level through London and England outside London, and add the split between the eight regions.

changelog.d/1207-family-loss-share-floors.added.md

@@ -0,0 +1 @@
+A check-only loss-share floor per target family for the UK local dense solve (microcosm#1207). `UK_LOCAL_FAMILY_LOSS_SHARE_FLOORS` in `uk_runtime/local_doctrine.py` sets the minimum share of the loss weight that DESNZ's local energy family (2.5%) and the BUS01 region journeys (1.25%) must carry. `assert_uk_family_loss_share_floors` refuses a weight vector that leaves either family below its floor. It runs on the row-metadata rules' weights in `prepare_uk_full_solve`, and the problem's bindings record it beside the loss-weight receipt. The weights themselves do not change.

changelog.d/1207-lcfs-donor-capture-receipts.added.md

@@ -0,0 +1 @@
+The `lcfs_consumption` stage's report-only `division_capture` block now records, for each COICOP division and the total, the donor's weighted total (the annualised donor the chain drew from, carried to the base year) and the frame's over it, in all and per household, so an over-draw of a division against its own donor shows in every build's evidence (microcosm#1207's furnishings and health item). The `recompose_from_remainder` receipt's `uncarried` block, the published ONS 04.5.3 and 04.5.4 spend that stays in the housing total, gains the donor's own other fuels at diary prices and their ratio to the published spend (microcosm#1207's liquid and solid fuels item). Nothing is gated on either; no column moves.

changelog.d/1207-local-energy-appliers.added.md

@@ -0,0 +1 @@
+Two UK uprating appliers for microcosm#1207's local-authority energy rows, in `uk_runtime/domestic_energy_targets.py`. `desnz.energy_trends.gb_domestic_level` multiplies each authority's DESNZ sub-national consumption by one factor per fuel, so that Great Britain's authorities sum to the `lcfs_consumption` stage's Energy Trends level (the rows keep only DESNZ's split between areas). Northern Ireland's part of that UK level is taken out at Great Britain's share of UK households (ONS LFS, the level's year) until Chronicle carries Northern Ireland's consumption (PolicyEngine/chronicle#356). `desnz.subnational.gas_meter_share` multiplies an authority's census households by its gas meters over its electricity meters. A missing meter row, or a share outside (0, 1], is refused. Both stamp `uprating_hold`, so the holds register declares the rows. `household_energy_kwh` reads each household's kWh back from its stored spend at the stage's declared prices. With it, the calibration diagnostics' `consumption_drift` block gains electricity and gas kWh at design and final weights wherever the frame carries the household's region.

changelog.d/1207-local-energy-targets.added.md

@@ -0,0 +1 @@
+Three local-authority target rows from DESNZ's sub-national energy statistics bind in the UK local dense solve (microcosm#1207). They are `desnz.subnational.electricity_kwh`, `desnz.subnational.gas_kwh` and `desnz.subnational.gas_connected_households`, census family `domestic_energy`, for England, Wales and Scotland. The kWh rows keep DESNZ's split between authorities at the `lcfs_consumption` stage's Energy Trends level, and DESNZ's region rows stay unbound (`no_higher_control`). The gas-connected rows are each authority's census households times its gas meters over its electricity meters. They join the census household partition (`census_gas_connected_share`) and take the census-household uprating. The local metrics `energy/electricity_kwh`, `energy/gas_kwh` and `energy/gas_connected_households` read each household's stored spend at the dataset period and turn it back into kWh at the stage's prices. Orkney and Shetland, with no mains gas, are signed deferrals. A new vendored resource, `desnz_local_energy_facts.json`, carries the authority meter counts and the LFS household rows the appliers read.

changelog.d/1207-ni-bus-fares-own-yields.changed.md

@@ -0,0 +1 @@
+Northern Ireland's bus fares are re-priced from its fiscal-year yield to calendar 2024 with its own published yields (microcosm#1207). The `price_bus_journeys` area declares `price_index: {basis: own_fiscal_yields}`: the yield `R / (B - C)` of the fiscal years opening in 2023 and 2024, from the area's own DfI receipts, journeys and full-fare-concession journeys, weighted three months and nine as `calendar_year_window` weights them, over the 2024-25 yield. It replaces England's local bus fares index, a declared proxy that carried the January 2025 rise in the English fare cap Northern Ireland did not have. The factor moves from 0.982 to 0.989, so Northern Ireland's fare per trip rises by about 0.7%. `sources.schema.json` admits the new basis beside a published index.

packages/microcosm-build/src/microcosm/build/spec_engine/schema/sources.schema.json

@@ -11180,33 +11180,49 @@
           "type": "string"
          },
          "price_index": {
-          "type": "object",
-          "additionalProperties": false,
-          "required": [
-           "resource",
-           "concept",
-           "frequency"
-          ],
-          "properties": {
-           "resource": {
-            "type": "string"
-           },
-           "concept": {
-            "type": "string"
-           },
-           "geography_id": {
-            "type": "string"
-           },
-           "frequency": {
-            "enum": [
-             "quarterly",
-             "annual"
-            ]
+          "oneOf": [
+           {
+            "type": "object",
+            "additionalProperties": false,
+            "required": [
+             "resource",
+             "concept",
+             "frequency"
+            ],
+            "properties": {
+             "resource": {
+              "type": "string"
+             },
+             "concept": {
+              "type": "string"
+             },
+             "geography_id": {
+              "type": "string"
+             },
+             "frequency": {
+              "enum": [
+               "quarterly",
+               "annual"
+              ]
+             },
+             "basis": {
+              "type": "string"
+             }
+            }
            },
-           "basis": {
-            "type": "string"
+           {
+            "type": "object",
+            "additionalProperties": false,
+            "required": [
+             "basis"
+            ],
+            "properties": {
+             "basis": {
+              "const": "own_fiscal_yields"
+             }
+            }
            }
-          }
+          ]
          },
          "trips_basis": {
           "enum": [

packages/microcosm-build/src/microcosm/build/uk/country_package.json

@@ -477,6 +477,11 @@
       "kind": "legacy_json",
       "schema_id": "legacy_json"
     },
+    {
+      "path": "desnz_local_energy_facts.json",
+      "kind": "legacy_json",
+      "schema_id": "legacy_json"
+    },
     {
       "path": "local_small_cell_rule.json",
       "kind": "legacy_json",

packages/microcosm-build/src/microcosm/build/uk/cross_grain_declarations.json

@@ -155,6 +155,15 @@
       ],
       "kind": "share_of_parent",
       "reason": "DWP's constituency UC households by number of children divide each constituency's UC caseload; the published unknown-children category is not bound, so the four bands cover part of the parent and move by its reconciliation factor, keeping their published shares. The local metrics count benefit units on UC, as the parent does. Until the area series counts paid claims, the bands carry all-claims shares, which run richer in childless households than the paid caseload does."
+    },
+    {
+      "partition_id": "census_gas_connected_share",
+      "parent_target_id": "ons.census.households",
+      "member_target_ids": [
+        "desnz.subnational.gas_connected_households"
+      ],
+      "kind": "share_of_parent",
+      "reason": "Each authority's gas-connected households are its census households times its DESNZ gas meters over its electricity meters, a share of the parent; they take the census household uprating and then the authority's household reconciliation factor, keeping DESNZ's share. The households off the gas network are not bound (microcosm#1207)."
     }
   ],
   "band_bridges": [
@@ -576,6 +585,14 @@
     "microcosm.high_value_homes.over_2m": {
       "route": "no_higher_control",
       "reason": "No publisher counts owner-occupied homes over £2m nationally, so no national row binds; the constituency and local-authority cells are Microcosm's derived Land Registry stock (microcosm#1240, fix 2), both grains summed from the same weighted dwellings, so they agree by construction. The OBR costing note's Table 1.1 counts every residential property in scope and is a diagnostic beside the family (uk_high_value_homes_benchmarks), never a control."
+    },
+    "desnz.subnational.electricity_kwh": {
+      "route": "no_higher_control",
+      "reason": "DESNZ's region and country rows are the sums of their authorities, so a region control adds no information, and a gas bridge would put the three authorities DESNZ publishes no gas for (the Isles of Scilly, Orkney and Shetland) in controlled legs. The electricity cells are levelled onto the lcfs stage's Energy Trends level by their uprating index instead (desnz.energy_trends.gb_domestic_level; microcosm#1207, local-authority grain only)."
+    },
+    "desnz.subnational.gas_kwh": {
+      "route": "no_higher_control",
+      "reason": "DESNZ's region and country rows are the sums of their authorities, so a region control adds no information, and a gas bridge would put the three authorities DESNZ publishes no gas for (the Isles of Scilly, Orkney and Shetland) in controlled legs. The gas cells are levelled onto the lcfs stage's Energy Trends level by their uprating index instead (desnz.energy_trends.gb_domestic_level; microcosm#1207, local-authority grain only)."
     }
   },
   "relations": [

packages/microcosm-build/src/microcosm/build/uk/desnz_domestic_energy_facts.json

@@ -3,7 +3,7 @@
   "country": "uk",
   "kind": "uk_vendored_ledger_facts",
   "resource": "desnz_domestic_energy_facts.json",
-  "purpose": "DESNZ domestic energy volumes: Energy Trends domestic electricity (table 5.5) and gas (table 4.1) consumption at actual temperature, annual and quarterly, United Kingdom, 2022 to the latest quarter (the published level the LCFS kWh rake scales to over FY2024-25), and the subnational electricity and gas statistics' domestic consumption and meter counts for Great Britain, England and Wales, the countries and the English regions, 2023 and 2024 (the published gas-connected share by region: gas meters over electricity meters).",
+  "purpose": "DESNZ domestic energy volumes: Energy Trends domestic electricity (table 5.5) and gas (table 4.1) consumption at actual temperature, annual and quarterly, United Kingdom, 2022 to the latest quarter (the published level the LCFS kWh rake scales to, summed over the four quarters of calendar 2024; microcosm#1113), and the subnational electricity and gas statistics' domestic consumption and meter counts for Great Britain, England and Wales, the countries and the English regions, 2023 and 2024 (the published gas-connected share by region: gas meters over electricity meters; and the Great Britain totals that level the local-authority energy rows, microcosm#1207).",
   "consumers": [
     "uk_runtime.energy_pricing"
   ],