MaxGhenis · Oct 24, 2026 · Oct 11, 2026 · Oct 11, 2026 · Oct 23, 2026 · Oct 23, 2026
diff --git a/changelog.d/obr-receipts-importer.added.md b/changelog.d/obr-receipts-importer.added.md
@@ -0,0 +1 @@
+- `import_obr_forecasts.py` now refreshes the OBR receipts growth series that project corporate SDLT and business rates (`yoy_growth.obr.stamp_duty_land_tax.*` and `.business_rates.*`) from a new EFO's receipts tables, with the previous EFO's tables (`--previous-receipts-file`) as the base for outturn growth, and refuses receipts tables from a different or unidentifiable EFO or without an unambiguous pound-billion unit declaration in every selected table in both releases.
diff --git a/policyengine_uk/parameters/gov/economic_assumptions/README.md b/policyengine_uk/parameters/gov/economic_assumptions/README.md
@@ -30,6 +30,19 @@ GDP and population assumptions remain supported.
 | `petrol_spending_litre_proxy` | Spending growth that preserves road-fuel litres    | HMRC + OBR              |
 | `diesel_spending_litre_proxy` | Same, for diesel                                   | HMRC + OBR              |
 | `finance_ni.domestic_rates` | Average Northern Ireland domestic rates bill (regional plus district poundage) | DoF rate poundages, LPS rates paid, NI draft Budget 2026-29 |
+| `stamp_duty_land_tax.residential` | SDLT receipts from residential property          | EFO receipts Table 3.7  |
+| `stamp_duty_land_tax.non_residential` | SDLT receipts from non-residential property  | EFO receipts Table 3.7  |
+| `business_rates.great_britain` | Business rates receipts, England, Scotland and Wales (cash) | EFO receipts, current receipts (cash basis): Table 3.8 in March 2026 |
+| `business_rates.northern_ireland` | Northern Ireland business rates receipts      | EFO receipts Table 3.2  |
+
+The four receipts series project the corporate SDLT and business rates
+statistics (`gov.hmrc.stamp_duty.statistics`,
+`gov.hmrc.business_rates.statistics`) from their latest outturn. Each is the
+ratio of consecutive published receipts levels, to six decimals: outturn
+growth for the EFO's outturn year (its outturn over the year before's outturn
+in an earlier EFO), the EFO's forecast years, and after them the
+`average_earnings` path. Years before the outturn year are kept, so each
+index keeps its base year.
 
 ## Time horizons
 
@@ -282,8 +295,19 @@ threshold by the same rises, up to the last year of
 
 1. Replace the 2025-2030 block in each series with values from the new EFO
    detailed forecast tables (note the EFO release month in the header comment).
+   The importer converts annual economy growth percentages to decimal rates
+   and rounds them to four decimal places. Review the economy-series diff
+   deliberately when running a full CLI import: with the real March 2026
+   economy workbook, the existing extraction changes cumulative 2030 indices
+   relative to the stored series by RPI −0.0880%, CPI −0.0584%, average
+   earnings −0.0585% and house prices +0.0194%. Receipts-only rendering can
+   reproduce the committed YAML byte for byte; a full import also refreshes
+   these economy series.
 2. Recompute the 2031-2073 convergence path so the first long-run year
-   continues smoothly from the new last forecast year (no jump).
+   continues smoothly from the new last forecast year (no jump). If the new
+   EFO forecasts a year that a series has no entry for, add that entry first:
+   the importer only replaces existing entries, and stops naming the series
+   and year it cannot update.
 3. Nothing needs regenerating: `create_economic_assumption_indices` rebuilds
    the cumulative `indices` parameters from these series whenever the
    tax-benefit system processes its parameters (`tax_benefit_system.py`).
@@ -293,9 +317,36 @@ threshold by the same rises, up to the last year of
    `yoy_growth.yaml`; pass the receipts tables with `--receipts-file` or
    `--receipts-url` so September CPI uses the OBR's September forecast. If
    you edit `yoy_growth.yaml` by hand instead, rerun it with `--gaps-only`.
-6. If the new EFO runs past `PROJECTION_END_YEAR`, move it on (see
+6. Refresh the receipts series (`stamp_duty_land_tax.*`,
+   `business_rates.*`). The importer does this in the same run when it also
+   gets the receipts tables of the EFO whose outturn year is one earlier
+   (March 2025 for March 2026), with `--previous-receipts-file` or
+   `--previous-receipts-url`; their outturn is the base for the outturn
+   year's growth. It finds each line by table title and row label, so table
+   renumbering does not matter, and reads only that table's rows. It takes
+   each workbook's release from its contents sheet, or else its file name or
+   URL, and refuses receipts tables from another EFO or of unknown release.
+   Every selected table in both releases must declare `£ billion` in its
+   header (`£bn` and `£ bn` are also accepted, ignoring case and whitespace).
+   Other units, including `£m`, and missing or ambiguous declarations stop
+   the import before any file is written, so different cash magnitudes
+   cannot be combined across releases.
+   Then update the published levels in
+   `policyengine_uk/tests/test_corporate_receipts_statistics.py`
+   (`OBR_RECEIPTS_BN`), which pin the series to the OBR's tables. Pass
+   `--skip-receipts-growth` only to leave them on the previous EFO.
+7. If the new EFO runs past `PROJECTION_END_YEAR`, move it on (see
    [Dataset projection horizon](#dataset-projection-horizon)).
 
+For example, for March 2026:
+
+```bash
+uv run python -m policyengine_uk.utils.import_obr_forecasts \
+  --file economy_tables_march_2026.xlsx \
+  --receipts-file receipts_tables_march_2026.xlsx \
+  --previous-receipts-file receipts_tables_march_2025.xlsx
+```
+
 The EFO refresh does not touch `finance_ni.domestic_rates`; OBR does not
 publish a Northern Ireland domestic rates bill series (see below).
 

diff --git a/policyengine_uk/parameters/gov/economic_assumptions/yoy_growth.yaml b/policyengine_uk/parameters/gov/economic_assumptions/yoy_growth.yaml
@@ -713,19 +713,22 @@ obr:
             href: https://obr.uk/efo/economic-and-fiscal-outlook-march-2026/
 
   # Receipts used to project the statistics behind corporate incidence.
-  # Each is the OBR's growth in a receipts line: 2024 is outturn growth
-  # (2024-25 outturn in the March 2026 EFO over 2023-24 outturn in the March
-  # 2025 EFO), 2025-2030 the March 2026 EFO forecast, and 2031 onwards the
-  # average_earnings long-run path (the OBR forecasts receipts only to
-  # 2030-31). Values are ratios of the published levels, to six decimals.
-  # The 2024 value never enters a projection: indices start at 1.0 in a
-  # series' first year, so it only anchors the index at 2024, where the
+  # Each is the OBR's growth in a receipts line, the ratio of consecutive
+  # published levels to six decimals. The latest EFO supplies outturn growth
+  # for its outturn year (its outturn over the previous year's outturn in an
+  # earlier EFO) and growth for each forecast year. After the OBR's receipts
+  # forecast, each series follows the average_earnings long-run path. The
+  # comments in each series name the EFOs and tables behind its values.
+  # policyengine_uk/utils/import_obr_forecasts.py rewrites these series from
+  # a new EFO's receipts tables and keeps the years before its outturn year.
+  # The first value, 2024, never enters a projection: indices start at 1.0
+  # in a series' first year, so it only anchors the index at 2024, where the
   # statistics with 2024-25 as their latest outturn are uprated from.
   stamp_duty_land_tax:
     residential:
       description: Growth in stamp duty land tax receipts from residential property (England and Northern Ireland).
       values:
-        # Outturn (2024-25 over 2023-24), Table 3.7 "SDLT from residential property"
+        # Outturn (2024-25 over 2023-24) from OBR EFO March 2026 over March 2025, receipts Table 3.7 "SDLT from residential property"
         2024-01-01: 0.211202
         # Forecast (2025-2030) from OBR EFO March 2026, receipts Table 3.7
         2025-01-01: 0.1302
@@ -752,7 +755,7 @@ obr:
     non_residential:
       description: Growth in stamp duty land tax receipts from non-residential property (England and Northern Ireland).
       values:
-        # Outturn (2024-25 over 2023-24), Table 3.7 "SDLT from non-residential property"
+        # Outturn (2024-25 over 2023-24) from OBR EFO March 2026 over March 2025, receipts Table 3.7 "SDLT from non-residential property"
         2024-01-01: 0.151067
         # Forecast (2025-2030) from OBR EFO March 2026, receipts Table 3.7
         2025-01-01: 0.048515
@@ -780,7 +783,7 @@ obr:
     great_britain:
       description: Growth in business rates receipts in England, Scotland and Wales (the OBR "Business rates" line, cash basis).
       values:
-        # Outturn (2024-25 over 2023-24), Table 3.8 (March 2025: Table 3.9) "Business rates"
+        # Outturn (2024-25 over 2023-24) from OBR EFO March 2026 over March 2025, receipts Table 3.8 (March 2025: Table 3.9) "Business rates"
         2024-01-01: 0.093818
         # Forecast (2025-2030) from OBR EFO March 2026, receipts Table 3.8
         2025-01-01: 0.051739
@@ -807,7 +810,7 @@ obr:
     northern_ireland:
       description: Growth in Northern Ireland business rates receipts (the OBR "Northern Ireland business rates" line in other taxes).
       values:
-        # Outturn (2024-25 over 2023-24), Table 3.2 "Northern Ireland business rates"
+        # Outturn (2024-25 over 2023-24) from OBR EFO March 2026 over March 2025, receipts Table 3.2 "Northern Ireland business rates"
         2024-01-01: 0.032513
         # Forecast (2025-2030) from OBR EFO March 2026, receipts Table 3.2
         2025-01-01: 0.032388

diff --git a/policyengine_uk/parameters/gov/hmrc/business_rates/statistics.yaml b/policyengine_uk/parameters/gov/hmrc/business_rates/statistics.yaml
@@ -49,7 +49,7 @@ revenue:
         - title: MHCLG, National non-domestic rates collected by councils in England 2025 to 2026, Table 1 "Non-domestic rating income from rates retention scheme" (2021-22 to 2025-26; 2025-26 provisional)
           href: https://www.gov.uk/government/statistics/national-non-domestic-rates-collected-by-councils-in-england-2025-to-2026
         - title: MHCLG, NNDR3 2019 to 2020 local authority data, November 2025 update, Datasheet1 "nndrincome", England
-          href: https://www.gov.uk/government/statistics/national-non-domestic-rates-collected-by-councils-in-england-2019-to-2020
+          href: https://www.gov.uk/government/statistics/national-non-domestic-rates-collected-by-councils-in-england-2019-to-2020-provisional-data
         - title: MHCLG, NNDR3 2020 to 2021 consolidated data, November 2025 update, "nndrincome", England
           href: https://www.gov.uk/government/statistics/national-non-domestic-rates-collected-by-councils-in-england-2020-to-2021
   ENGLAND_CENTRAL_LIST:

diff --git a/policyengine_uk/tests/fixtures/obr_receipts/README.md b/policyengine_uk/tests/fixtures/obr_receipts/README.md
@@ -0,0 +1,22 @@
+# Real OBR receipts table fixtures
+
+These parser fixtures retain the original worksheet XML for Contents, Other
+taxes (3.2), Property transactions taxes: receipts by sector (3.7), and Current
+receipts (on a cash basis: 3.9 in 2025, 3.8 in 2026), plus the original shared
+strings. Only the workbook's sheet list and relationships were reduced to those
+entries. The published unit cells, fiscal-year headers, markers, row labels and
+numeric cells are unchanged. These are minimal archives for the importer's XLSX
+reader, rather than complete Excel documents with styles and charts.
+
+Sources: OBR [March 2025 EFO](https://obr.uk/efo/economic-and-fiscal-outlook-march-2025/)
+and [March 2026 EFO](https://obr.uk/efo/economic-and-fiscal-outlook-march-2026/),
+receipts detailed forecast tables. The original files used for extraction are
+`obr_efo_march_2025_receipts.xlsx` and `obr_efo_march_2026_receipts.xlsx`.
+Their SHA-256 hashes are:
+
+- March 2025: `6b20e05e93c50e3ccae1d3111fd84b8fa326b75c38c8ab055fad694648c75905`
+- March 2026: `f65b6cb7f96931d4abd85940794eb60f6f65afca8bbf16c74a4548bf92766a28`
+
+Tests mutate the unit cell and selected receipt levels inside these real tables;
+the unchanged control must reproduce the committed YAML byte for byte. No test
+downloads a workbook or depends on a developer's local source directory.
diff --git a/policyengine_uk/tests/fixtures/obr_receipts/obr_efo_march_2025_receipts.xlsx b/policyengine_uk/tests/fixtures/obr_receipts/obr_efo_march_2025_receipts.xlsx
diff --git a/policyengine_uk/tests/fixtures/obr_receipts/obr_efo_march_2026_receipts.xlsx b/policyengine_uk/tests/fixtures/obr_receipts/obr_efo_march_2026_receipts.xlsx
Original file line number	Diff line number	Diff line change
		@@ -0,0 +1 @@
		- `import_obr_forecasts.py` now refreshes the OBR receipts growth series that project corporate SDLT and business rates (`yoy_growth.obr.stamp_duty_land_tax.` and `.business_rates.`) from a new EFO's receipts tables, with the previous EFO's tables (`--previous-receipts-file`) as the base for outturn growth, and refuses receipts tables from a different or unidentifiable EFO or without an unambiguous pound-billion unit declaration in every selected table in both releases.