Add rail_usage variable derived from rail_subsidy_spending

changelog_entry.yaml

@@ -0,0 +1,4 @@
+- bump: minor
+  changes:
+    added:
+      - rail_usage variable derived from rail_subsidy_spending / fare_index at survey year, enabling fare reforms to modify prices independently of usage quantity

policyengine_uk_data/datasets/imputations/services/services.py

@@ -6,9 +6,31 @@
 """
 
 from policyengine_uk.data import UKSingleYearDataset
+from policyengine_uk.system import system
 from .nhs import impute_nhs_usage
 from .etb import impute_public_services, create_efrs_input_dataset
 
+# ETB survey year (most recent year in ETB data)
+ETB_SURVEY_YEAR = 2021
+
+# Fallback fare index for 2021 if parameter not yet available in policyengine-uk
+# This is the cumulative fare index from base year 2020 (+1.0% from 2020)
+FALLBACK_FARE_INDEX_2021 = 1.010
+
+
+def get_fare_index_survey_year() -> float:
+    """
+    Get the rail fare index for the ETB survey year.
+
+    Attempts to read from policyengine-uk parameters, falls back to
+    hardcoded value if parameter not yet available.
+    """
+    try:
+        return system.parameters.gov.dft.rail.fare_index(ETB_SURVEY_YEAR)
+    except AttributeError:
+        # Parameter not yet available in policyengine-uk
+        return FALLBACK_FARE_INDEX_2021
+
 
 def impute_services(
     dataset: UKSingleYearDataset,
@@ -44,6 +66,14 @@ def impute_services(
             .values
         )
 
+    # Derive rail_usage (quantity at base year prices) from rail_subsidy_spending
+    # rail_usage = rail_subsidy_spending / fare_index at survey year
+    # This allows reforms to modify fare_index independently of usage quantity
+    fare_index_survey_year = get_fare_index_survey_year()
+    dataset.household["rail_usage"] = (
+        dataset.household["rail_subsidy_spending"] / fare_index_survey_year
+    )
+
     visit_variables = [
         "a_and_e_visits",
         "admitted_patient_visits",