PolicyEngine-UK 2.75.2__tar.gz → 2.75.4__tar.gz

{policyengine_uk-2.75.2 → policyengine_uk-2.75.4}/CHANGELOG.md

@@ -1,3 +1,17 @@
+## [2.75.4] - 2026-04-05
+
+### Fixed
+
+- - Ensure `uprate_rent` passes a NumPy array into vectorial parameter lookup for regional private rent growth.
+
+
+## [2.75.3] - 2026-03-26
+
+### Changed
+
+- Replace flat national land intensity ratio (0.673) with region-specific ratios from MHCLG 2023 land value estimates, so household_land_value reflects the much higher land share in London (0.85) vs the North East (0.42).
+
+
 ## [2.75.2] - 2026-03-17
 
 ### Changed

{policyengine_uk-2.75.2 → policyengine_uk-2.75.4}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: policyengine-uk
-Version: 2.75.2
+Version: 2.75.4
 Summary: PolicyEngine tax and benefit system for the UK.
 Project-URL: Homepage, https://github.com/PolicyEngine/policyengine-uk
 Project-URL: Repository, https://github.com/PolicyEngine/policyengine-uk
@@ -20,7 +20,7 @@ Classifier: Programming Language :: Python :: 3.12
 Classifier: Programming Language :: Python :: 3.13
 Classifier: Programming Language :: Python :: 3.14
 Classifier: Topic :: Scientific/Engineering :: Information Analysis
-Requires-Python: >=3.13
+Requires-Python: >=3.11
 Requires-Dist: microdf-python>=1.2.1
 Requires-Dist: policyengine-core>=3.23.6
 Requires-Dist: pydantic>=2.11.7
@@ -64,13 +64,12 @@ PolicyEngine UK enhances this dataset by fusing it to other surveys and reweight
 [^1]: Department for Work and Pensions, Office for National Statistics, NatCen Social Research. (2021). Family Resources Survey, 2019-2020. [data collection]. UK Data Service. SN: 8802, http://doi.org/10.5255/UKDA-SN-8802-1
 
 
-## Fast setup instructions
+## Fast setup instructions (Linux / macOS)
 
 1. Run `pip install policyengine-uk`
 
 2. Run `policyengine-uk` and go through the prompt to setup microdata.
 
-
 ## Contact
 
 The primary maintainer for PolicyEngine UK is Nikhil Woodruff, co-founder and CTO of PolicyEngine (nikhil@policyengine.org).

{policyengine_uk-2.75.2 → policyengine_uk-2.75.4}/README.md

@@ -20,13 +20,12 @@ PolicyEngine UK enhances this dataset by fusing it to other surveys and reweight
 [^1]: Department for Work and Pensions, Office for National Statistics, NatCen Social Research. (2021). Family Resources Survey, 2019-2020. [data collection]. UK Data Service. SN: 8802, http://doi.org/10.5255/UKDA-SN-8802-1
 
 
-## Fast setup instructions
+## Fast setup instructions (Linux / macOS)
 
 1. Run `pip install policyengine-uk`
 
 2. Run `policyengine-uk` and go through the prompt to setup microdata.
 
-
 ## Contact
 
 The primary maintainer for PolicyEngine UK is Nikhil Woodruff, co-founder and CTO of PolicyEngine (nikhil@policyengine.org).

{policyengine_uk-2.75.2 → policyengine_uk-2.75.4}/policyengine_uk/data/economic_assumptions.py

@@ -177,7 +177,7 @@ def uprate_rent(
     elif year < 2025:
         # We have regional growth rates for private rent.
         regional_growth_rate = growth.ons.private_rental_prices(year)[
-            region.values.astype(str)
+            np.array(region.values.astype(str))
         ]
         current_year.household["rent"] = np.where(
             is_private_rented,

{policyengine_uk-2.75.2 → policyengine_uk-2.75.4}/policyengine_uk/parameters/gov/hmrc/income_tax/rates/scotland/rates.yaml

@@ -19,6 +19,7 @@ brackets:
         2023-04-06: 2_162
         2024-04-06: 2_306
         2025-04-06: 2_827
+        2026-04-06: 3_967
       metadata:
         uprating: gov.economic_assumptions.indices.obr.consumer_price_index
     rate:
@@ -32,6 +33,7 @@ brackets:
         2023-04-06: 13_118
         2024-04-06: 13_991
         2025-04-06: 14_921
+        2026-04-06: 16_956
       metadata:
         uprating: gov.economic_assumptions.indices.obr.consumer_price_index
     rate:

policyengine_uk-2.75.4/policyengine_uk/parameters/household/wealth/land/intensity.yaml

@@ -0,0 +1,63 @@
+description: Land intensity by asset type (where microsimulation data is not available)
+metadata:
+  unit: /1
+corporate_wealth:
+  2019-01-01: 0.349
+  2022-01-01: 0.321
+property_wealth_by_region:
+  description: >
+    Region-specific land intensity of property wealth, i.e. the fraction of
+    average house price attributable to land.  Derived via the residual method:
+    intensity = (mhclg_land_per_ha / dwellings_per_ha) / avg_house_price,
+    using MHCLG 2023 central-scenario densities and UK HPI Dec 2025 prices.
+    England regions use MHCLG local authority data aggregated to region.
+    Scotland and Wales are interpolated from the nearest English region
+    by house price (MHCLG covers England only): Scotland from North East/
+    North West, Wales from North West.
+  metadata:
+    label: Regional property wealth land intensity
+    unit: /1
+    reference:
+      - title: MHCLG Land Value Estimates for Policy Appraisal 2023
+        href: https://www.gov.uk/government/publications/land-value-estimates-for-policy-appraisal-2023
+      - title: UK House Price Index Dec 2025
+        href: https://www.gov.uk/government/statistics/uk-house-price-index-for-december-2025
+  LONDON:
+    values:
+      2023-01-01: 0.85
+  SOUTH_EAST:
+    values:
+      2023-01-01: 0.65
+  EAST_OF_ENGLAND:
+    values:
+      2023-01-01: 0.60
+  SOUTH_WEST:
+    values:
+      2023-01-01: 0.58
+  WEST_MIDLANDS:
+    values:
+      2023-01-01: 0.52
+  EAST_MIDLANDS:
+    values:
+      2023-01-01: 0.48
+  NORTH_WEST:
+    values:
+      2023-01-01: 0.47
+  YORKSHIRE:
+    values:
+      2023-01-01: 0.46
+  SCOTLAND:
+    values:
+      2023-01-01: 0.44
+  WALES:
+    values:
+      2023-01-01: 0.47
+  NORTH_EAST:
+    values:
+      2023-01-01: 0.42
+  NORTHERN_IRELAND:
+    values:
+      2023-01-01: 0.673
+  UNKNOWN:
+    values:
+      2023-01-01: 0.673

{policyengine_uk-2.75.2 → policyengine_uk-2.75.4}/policyengine_uk/scenarios/uc_reform.py

@@ -29,16 +29,8 @@ def add_universal_credit_reform(sim: Microsimulation):
         )  # Monthly amount * 12
         sim.set_input("uc_LCWRA_element", year, current_health_element)
 
-    # https://bills.parliament.uk/publications/62123/documents/6889#page=14
-
-    uc_uplift = rebalancing.standard_allowance_uplift
-
-    for year in range(2026, 2030):
-        if not rebalancing.active(year):
-            continue
-        previous_value = sim.calculate("uc_standard_allowance", year - 1)
-        new_value = previous_value * (1 + uc_uplift(year))
-        sim.set_input("uc_standard_allowance", year, new_value)
+    # Standard allowance uplift is handled in the formula itself so user
+    # reforms to the base amount are applied before the uplift.
 
 
 universal_credit_july_2025_reform = Scenario(

{policyengine_uk-2.75.2 → policyengine_uk-2.75.4}/policyengine_uk/simulation.py

@@ -144,7 +144,7 @@ class Simulation(CoreSimulation):
             self.build_from_multi_year_dataset(dataset)
         elif dataset is None:
             self.build_from_url(
-                "hf://policyengine/policyengine-uk-data/enhanced_frs_2023_24.h5"
+                "hf://policyengine/policyengine-uk-data-private/enhanced_frs_2023_24.h5"
             )
         else:
             raise ValueError(f"Unsupported dataset type: {dataset.__class__}")

{policyengine_uk-2.75.2 → policyengine_uk-2.75.4}/policyengine_uk/tax_benefit_system.py

@@ -37,7 +37,7 @@ from policyengine_uk.utils.parameters import (
 
 # Module constants
 COUNTRY_DIR = Path(__file__).parent
-ENHANCED_FRS = "hf://policyengine/policyengine-uk-data/enhanced_frs_2023_24.h5"
+ENHANCED_FRS = "hf://policyengine/policyengine-uk-data-private/enhanced_frs_2023_24.h5"
 
 # Cache for fully-processed parameter tree, so convert_to_fiscal_year_parameters
 # (22,538 param.update() calls) only runs once per process.

{policyengine_uk-2.75.2 → policyengine_uk-2.75.4}/policyengine_uk/tests/microsimulation/reforms_config.yaml

@@ -1,3 +1,7 @@
+# Regression fixtures for microsimulation reform impacts.
+# These expected impacts are derived from current model output and are used to
+# catch unintended behavior changes. They are not external fiscal estimates
+# unless a reform entry is explicitly annotated otherwise.
 reforms:
 - name: Raise basic rate by 1pp
   expected_impact: 7.8
@@ -17,7 +21,7 @@ reforms:
     gov.hmrc.child_benefit.amount.additional: 25
 - name: Reduce Universal Credit taper rate to 20%
   expected_impact: -41.9
-  tolerance: 1.5
+  tolerance: 3.0
   parameters:
     gov.dwp.universal_credit.means_test.reduction_rate: 0.2
 - name: Raise Class 1 main employee NICs rate to 10%
@@ -25,7 +29,7 @@ reforms:
   parameters:
     gov.hmrc.national_insurance.class_1.rates.employee.main: 0.1
 - name: Raise VAT standard rate by 2pp
-  expected_impact: 18.9
+  expected_impact: 21.3
   tolerance: 1.5
   parameters:
     gov.hmrc.vat.standard_rate: 0.22

{policyengine_uk-2.75.2 → policyengine_uk-2.75.4}/policyengine_uk/tests/microsimulation/test_reform_impacts.py

@@ -1,6 +1,8 @@
 """
 Test suite for PolicyEngine UK reform fiscal impacts.
 This file tests that model changes don't unexpectedly change reform impacts.
+Expected impacts come from the repo's current-model regression fixtures in
+reforms_config.yaml, not from external policy costings unless noted there.
 """
 
 import pytest
@@ -61,7 +63,7 @@ reform_names = [reform["name"] for reform in reforms_data]
 def test_reform_fiscal_impacts(
     baseline, reform, reform_name, expected_impact, tolerance
 ):
-    """Test that each reform produces the expected fiscal impact."""
+    """Test that each reform still matches its current-model regression target."""
     impact = get_fiscal_impact(baseline, reform)
 
     assert abs(impact - expected_impact) < tolerance, (

policyengine_uk-2.75.4/policyengine_uk/tests/microsimulation/test_state_pension_reform.py

@@ -0,0 +1,32 @@
+"""Test state pension parameter reforms affect microsimulation outputs."""
+
+import pytest
+
+from policyengine_uk import Microsimulation
+
+YEAR = 2025
+
+
+@pytest.fixture(scope="module")
+def baseline():
+    return Microsimulation()
+
+
+@pytest.fixture(scope="module")
+def reform_halved():
+    return Microsimulation(
+        reform={
+            "gov.dwp.state_pension.basic_state_pension.amount": {
+                "2025-01-01": 84.75,
+            }
+        }
+    )
+
+
+@pytest.mark.microsimulation
+def test_basic_state_pension_responds_to_reform(baseline, reform_halved):
+    baseline_total = baseline.calculate("basic_state_pension", YEAR).sum()
+    reform_total = reform_halved.calculate("basic_state_pension", YEAR).sum()
+
+    assert baseline_total > 0
+    assert reform_total < baseline_total * 0.9

policyengine_uk-2.75.4/policyengine_uk/tests/microsimulation/test_uc_standard_allowance_reform.py

@@ -0,0 +1,27 @@
+"""Test UC standard allowance reforms are respected (#1472)."""
+
+from policyengine_uk import Simulation
+
+YEAR = 2026
+
+SITUATION = {
+    "people": {"person": {"age": {YEAR: 30}}},
+    "benunits": {"benunit": {"members": ["person"]}},
+    "households": {"household": {"members": ["person"]}},
+}
+
+REFORM = {
+    "gov.dwp.universal_credit.standard_allowance.amount.SINGLE_OLD": {
+        "2025-01-01.2100-12-31": 800,
+    },
+}
+
+
+def test_uc_standard_allowance_responds_to_reform():
+    baseline = Simulation(situation=SITUATION)
+    baseline_amount = baseline.calculate("uc_standard_allowance", YEAR)[0]
+
+    reformed = Simulation(situation=SITUATION, reform=REFORM)
+    reform_amount = reformed.calculate("uc_standard_allowance", YEAR)[0]
+
+    assert reform_amount / baseline_amount > 1.5

policyengine_uk-2.75.4/policyengine_uk/tests/policy/baseline/household/wealth/household_land_value.yaml

@@ -0,0 +1,89 @@
+- name: London household - high land intensity (0.85)
+  period: 2023
+  absolute_error_margin: 1
+  input:
+    region: LONDON
+    residential_property_value: 500000
+    owned_land: 0
+  output:
+    household_land_value: 425000
+
+- name: North East household - low land intensity (0.42)
+  period: 2023
+  absolute_error_margin: 1
+  input:
+    region: NORTH_EAST
+    residential_property_value: 500000
+    owned_land: 0
+  output:
+    household_land_value: 210000
+
+- name: South East household with owned land
+  period: 2023
+  absolute_error_margin: 1
+  input:
+    region: SOUTH_EAST
+    residential_property_value: 400000
+    owned_land: 50000
+  output:
+    household_land_value: 310000
+
+- name: Scotland household
+  period: 2023
+  absolute_error_margin: 1
+  input:
+    region: SCOTLAND
+    residential_property_value: 200000
+    owned_land: 0
+  output:
+    household_land_value: 88000
+
+- name: Wales household
+  period: 2023
+  absolute_error_margin: 1
+  input:
+    region: WALES
+    residential_property_value: 200000
+    owned_land: 10000
+  output:
+    household_land_value: 104000
+
+- name: Zero property wealth still counts owned land
+  period: 2023
+  absolute_error_margin: 0
+  input:
+    region: LONDON
+    residential_property_value: 0
+    owned_land: 25000
+  output:
+    household_land_value: 25000
+
+- name: Unknown region uses national average (0.673)
+  period: 2023
+  absolute_error_margin: 1
+  input:
+    region: UNKNOWN
+    residential_property_value: 300000
+    owned_land: 0
+  output:
+    household_land_value: 201900
+
+- name: East Midlands household
+  period: 2023
+  absolute_error_margin: 1
+  input:
+    region: EAST_MIDLANDS
+    residential_property_value: 250000
+    owned_land: 0
+  output:
+    household_land_value: 120000
+
+- name: Northern Ireland household
+  period: 2023
+  absolute_error_margin: 1
+  input:
+    region: NORTHERN_IRELAND
+    residential_property_value: 200000
+    owned_land: 0
+  output:
+    household_land_value: 134600

{policyengine_uk-2.75.2 → policyengine_uk-2.75.4}/policyengine_uk/tests/test_deterministic_variables.py

@@ -154,6 +154,29 @@ class TestExplicitOverrides:
         assert result[0] == False
 
 
+class TestDefaultDatasetUrl:
+    """Test that the default dataset URL points at the private HF repo."""
+
+    def test_simulation_defaults_to_private_hf_repo(self, monkeypatch):
+        captured = {}
+
+        class _StopDefaultDatasetLoad(Exception):
+            pass
+
+        def fake_build_from_url(self, url):
+            captured["url"] = url
+            raise _StopDefaultDatasetLoad
+
+        monkeypatch.setattr(Simulation, "build_from_url", fake_build_from_url)
+
+        with pytest.raises(_StopDefaultDatasetLoad):
+            Simulation()
+
+        assert captured["url"] == (
+            "hf://policyengine/policyengine-uk-data-private/enhanced_frs_2023_24.h5"
+        )
+
+
 class TestIsHigherEarner:
     """Test deterministic tie-breaking for is_higher_earner."""
 

policyengine_uk-2.75.4/policyengine_uk/tests/test_scottish_income_tax_rates.py

@@ -0,0 +1,7 @@
+from policyengine_uk.system import system
+
+
+def test_scottish_income_tax_thresholds_updated_for_2026():
+    rates = system.parameters.gov.hmrc.income_tax.rates.scotland.rates("2026-04-06")
+
+    assert rates.thresholds == [0.0, 3967, 16956, 31092, 62430, 112570]

{policyengine_uk-2.75.2 → policyengine_uk-2.75.4}/policyengine_uk/variables/gov/dwp/additional_state_pension.py

@@ -18,12 +18,17 @@ class additional_state_pension(Variable):
             data_year = period.start.year
         reported = person("state_pension_reported", data_year) / WEEKS_IN_YEAR
         type = person("state_pension_type", data_year)
-        maximum_basic_sp = parameters(
-            data_year
-        ).gov.dwp.state_pension.basic_state_pension.amount
+        sp_amount = parameters.gov.dwp.state_pension.basic_state_pension.amount
+        max_sp_data_year = sp_amount(data_year)
+        max_sp_period = sp_amount(period)
         amount_in_data_year = where(
             type == type.possible_values.BASIC,
-            max_(reported - maximum_basic_sp, 0),
+            max_(reported - max_sp_data_year, 0),
             0,
         )
-        return amount_in_data_year * WEEKS_IN_YEAR
+        uprating = where(
+            max_sp_data_year > 0,
+            max_sp_period / max_sp_data_year,
+            1,
+        )
+        return amount_in_data_year * uprating * WEEKS_IN_YEAR

{policyengine_uk-2.75.2 → policyengine_uk-2.75.4}/policyengine_uk/variables/gov/dwp/basic_state_pension.py

@@ -23,23 +23,22 @@ class basic_state_pension(Variable):
 
         reported = person("state_pension_reported", data_year) / WEEKS_IN_YEAR
         pension_type = person("state_pension_type", period)
-        maximum_basic_sp = parameters(
-            data_year
-        ).gov.dwp.state_pension.basic_state_pension.amount
-
-        # For BASIC pension type, cap at the maximum; otherwise return 0
-        amount_in_data_year = where(
-            pension_type == pension_type.possible_values.BASIC,
-            min_(reported, maximum_basic_sp),
+        sp_amount = parameters.gov.dwp.state_pension.basic_state_pension.amount
+        max_sp_data_year = sp_amount(data_year)
+        max_sp_period = sp_amount(period)
+
+        # Compute the person's share of the data-year maximum so reforms can
+        # scale the current-period amount while preserving the original cap.
+        share = where(
+            max_sp_data_year > 0,
+            min_(reported, max_sp_data_year) / max_sp_data_year,
             0,
         )
-
-        # Apply triple lock uprating only when using dataset
-        # (i.e., when data year differs from simulation period)
-        if has_dataset:
-            triple_lock = parameters.gov.economic_assumptions.indices.triple_lock
-            uprating = triple_lock(period) / triple_lock(data_year)
-        else:
-            uprating = 1
-
-        return amount_in_data_year * uprating * WEEKS_IN_YEAR
+        return (
+            where(
+                pension_type == pension_type.possible_values.BASIC,
+                share * max_sp_period,
+                0,
+            )
+            * WEEKS_IN_YEAR
+        )

{policyengine_uk-2.75.2 → policyengine_uk-2.75.4}/policyengine_uk/variables/gov/dwp/universal_credit/standard_allowance/uc_standard_allowance.py

@@ -11,4 +11,8 @@ class uc_standard_allowance(Variable):
     def formula(benunit, period, parameters):
         p = parameters(period).gov.dwp.universal_credit.standard_allowance
         claimant_type = benunit("uc_standard_allowance_claimant_type", period)
-        return p.amount[claimant_type] * MONTHS_IN_YEAR
+        value = p.amount[claimant_type] * MONTHS_IN_YEAR
+        rebalancing = parameters(period).gov.dwp.universal_credit.rebalancing
+        if rebalancing.active:
+            value = value * (1 + rebalancing.standard_allowance_uplift)
+        return value

policyengine_uk-2.75.4/policyengine_uk/variables/household/wealth/household_land_value.py

@@ -0,0 +1,56 @@
+from policyengine_uk.model_api import *
+import numpy as np
+
+
+class household_land_value(Variable):
+    entity = Household
+    label = "Land value"
+    documentation = "Estimated total land value directly owned by the household"
+    unit = GBP
+    definition_period = YEAR
+    value_type = float
+    quantity_type = STOCK
+
+    def formula(household, period, parameters):
+        region = household("region", period)
+        Region = region.possible_values
+        property_wealth = household("property_wealth", period)
+        owned_land = household("owned_land", period)
+
+        regional = parameters(
+            period
+        ).household.wealth.land.intensity.property_wealth_by_region
+
+        property_wealth_intensity = np.select(
+            [
+                region == Region.LONDON,
+                region == Region.SOUTH_EAST,
+                region == Region.EAST_OF_ENGLAND,
+                region == Region.SOUTH_WEST,
+                region == Region.WEST_MIDLANDS,
+                region == Region.EAST_MIDLANDS,
+                region == Region.NORTH_WEST,
+                region == Region.YORKSHIRE,
+                region == Region.SCOTLAND,
+                region == Region.WALES,
+                region == Region.NORTH_EAST,
+                region == Region.NORTHERN_IRELAND,
+            ],
+            [
+                regional.LONDON,
+                regional.SOUTH_EAST,
+                regional.EAST_OF_ENGLAND,
+                regional.SOUTH_WEST,
+                regional.WEST_MIDLANDS,
+                regional.EAST_MIDLANDS,
+                regional.NORTH_WEST,
+                regional.YORKSHIRE,
+                regional.SCOTLAND,
+                regional.WALES,
+                regional.NORTH_EAST,
+                regional.NORTHERN_IRELAND,
+            ],
+            default=regional.UNKNOWN,
+        )
+
+        return property_wealth * property_wealth_intensity + owned_land

{policyengine_uk-2.75.2 → policyengine_uk-2.75.4}/pyproject.toml

@@ -1,6 +1,6 @@
 [project] 
 name = "policyengine-uk"
-version = "2.75.2"
+version = "2.75.4"
 description = "PolicyEngine tax and benefit system for the UK."
 readme = "README.md"
 authors = [
@@ -22,7 +22,7 @@ classifiers = [
     "Programming Language :: Python :: 3.14",
     "Topic :: Scientific/Engineering :: Information Analysis",
 ]
-requires-python = ">=3.13"
+requires-python = ">=3.11"
 dependencies = [
     "policyengine-core>=3.23.6",
     "microdf-python>=1.2.1",

policyengine_uk-2.75.2/policyengine_uk/parameters/household/wealth/land/intensity.yaml

@@ -1,9 +0,0 @@
-description: Land intensity by asset type (where microsimulation data is not available)
-property_wealth:
-  2019-01-01: 0.635
-  2022-01-01: 0.673
-corporate_wealth:
-  2019-01-01: 0.349
-  2022-01-01: 0.321
-metadata:
-  unit: /1

policyengine_uk-2.75.2/policyengine_uk/variables/household/wealth/household_land_value.py

@@ -1,20 +0,0 @@
-from policyengine_uk.model_api import *
-
-
-class household_land_value(Variable):
-    entity = Household
-    label = "Land value"
-    documentation = "Estimated total land value directly owned by the household"
-    unit = GBP
-    definition_period = YEAR
-    value_type = float
-    quantity_type = STOCK
-
-    def formula(household, period, parameters):
-        wealth = parameters(period).household.wealth
-        property_wealth_intensity = (
-            wealth.land.value.aggregate_household_land_value / wealth.property_wealth
-        )
-        property_wealth = household("property_wealth", period)
-        owned_land = household("owned_land", period)
-        return property_wealth * property_wealth_intensity + owned_land