Merge remote-tracking branch 'upstream/main' into codex/release-manifest

Author: MaxGhenis

CHANGELOG.md

@@ -1,3 +1,22 @@
+## [1.78.1] - 2026-04-12
+
+### Changed
+
+- Add SSTB QBI split inputs to `policyengine-us-data` by exposing
+  `sstb_self_employment_income`, `sstb_w2_wages_from_qualified_business`, and
+  `sstb_unadjusted_basis_qualified_property` from the existing PUF/calibration
+  pipeline. The current split follows the legacy all-or-nothing
+  `business_is_sstb` flag, so mixed SSTB/non-SSTB allocations remain approximate
+  until more granular source data or imputation is added.
+
+
+## [1.78.0] - 2026-04-12
+
+### Added
+
+- Add comparison-mode CTC diagnostics to `validate_national_h5`, including child-count and child-age drift reporting between national artifacts.
+
+
 ## [1.77.0] - 2026-04-10
 
 ### Added

docs/appendix.md

@@ -112,12 +112,19 @@ for iteration in range(5000):
 - w2_wages_from_qualified_business
 - unadjusted_basis_qualified_property
 - business_is_sstb
+- sstb_self_employment_income
+- sstb_w2_wages_from_qualified_business
+- sstb_unadjusted_basis_qualified_property
 - qualified_reit_and_ptp_income
 - qualified_bdc_income
 - farm_operations_income
 - estate_income_would_be_qualified
 - farm_operations_income_would_be_qualified
 - farm_rent_income_would_be_qualified
+
+The current PUF/calibration pipeline uses the legacy `business_is_sstb` flag to
+split these SSTB variables on an all-or-nothing basis. It does not yet infer
+mixed SSTB and non-SSTB allocations within the same record.
 - partnership_s_corp_income_would_be_qualified
 - rental_income_would_be_qualified
 - self_employment_income_would_be_qualified

policyengine_us_data/__init__.py

@@ -1,6 +1,9 @@
 from importlib import import_module
 
 from .geography import ZIP_CODE_DATASET
+from .utils.policyengine import ensure_policyengine_us_compat_variables
+
+ensure_policyengine_us_compat_variables()
 
 _LAZY_EXPORTS = {
     "CPS_2024": (
@@ -26,7 +29,16 @@
 
 def __getattr__(name: str):
     if name not in _LAZY_EXPORTS:
-        raise AttributeError(f"module {__name__!r} has no attribute {name!r}")
+        try:
+            value = import_module(f"{__name__}.{name}")
+        except ModuleNotFoundError as exc:
+            if exc.name == f"{__name__}.{name}":
+                raise AttributeError(
+                    f"module {__name__!r} has no attribute {name!r}"
+                ) from exc
+            raise
+        globals()[name] = value
+        return value
 
     module_name, attribute_name = _LAZY_EXPORTS[name]
     value = getattr(import_module(module_name), attribute_name)

policyengine_us_data/calibration/check_staging_sums.py

@@ -23,7 +23,7 @@
 VARIABLES = [
     "adjusted_gross_income",
     "employment_income",
-    "self_employment_income",
+    "total_self_employment_income",
     "tax_unit_partnership_s_corp_income",
     "taxable_pension_income",
     "dividend_income",

policyengine_us_data/calibration/puf_impute.py

@@ -50,9 +50,12 @@
     "pre_tax_contributions",
     "taxable_ira_distributions",
     "self_employment_income",
+    "sstb_self_employment_income",
     "w2_wages_from_qualified_business",
     "unadjusted_basis_qualified_property",
     "business_is_sstb",
+    "sstb_w2_wages_from_qualified_business",
+    "sstb_unadjusted_basis_qualified_property",
     "short_term_capital_gains",
     "qualified_dividend_income",
     "charitable_cash_donations",
@@ -122,6 +125,8 @@
     "w2_wages_from_qualified_business",
     "unadjusted_basis_qualified_property",
     "business_is_sstb",
+    "sstb_w2_wages_from_qualified_business",
+    "sstb_unadjusted_basis_qualified_property",
     "charitable_cash_donations",
     "self_employed_pension_contribution_ald",
     "unrecaptured_section_1250_gain",
@@ -693,6 +698,11 @@ def _impute_retirement_contributions(
             X_test[income_var] = puf_imputations[income_var]
         else:
             X_test[income_var] = cps_sim.calculate(income_var).values
+    if "sstb_self_employment_income" in puf_imputations:
+        X_test["self_employment_income"] = (
+            X_test["self_employment_income"]
+            + puf_imputations["sstb_self_employment_income"]
+        )
 
     del cps_sim
 
@@ -723,13 +733,13 @@ def _impute_retirement_contributions(
     catch_up_eligible = age >= 50
     limit_401k = limits["401k"] + catch_up_eligible * limits["401k_catch_up"]
     limit_ira = limits["ira"] + catch_up_eligible * limits["ira_catch_up"]
+    se_income = X_test["self_employment_income"].values
     se_pension_cap = np.minimum(
-        X_test["self_employment_income"].values * limits["se_pension_rate"],
+        se_income * limits["se_pension_rate"],
         limits["se_pension_dollar_limit"],
     )
 
     emp_income = X_test["employment_income"].values
-    se_income = X_test["self_employment_income"].values
 
     result = {}
     for var in CPS_RETIREMENT_VARIABLES:

policyengine_us_data/calibration/target_config.yaml

@@ -22,7 +22,7 @@ include:
     geo_level: district
   - variable: real_estate_taxes
     geo_level: district
-  - variable: self_employment_income
+  - variable: total_self_employment_income
     geo_level: district
   - variable: taxable_pension_income
     geo_level: district
@@ -163,9 +163,9 @@ include:
   - variable: non_refundable_ctc
     geo_level: national
     domain_variable: adjusted_gross_income,non_refundable_ctc
-  - variable: self_employment_income
+  - variable: total_self_employment_income
     geo_level: national
-    domain_variable: self_employment_income
+    domain_variable: total_self_employment_income
   - variable: tax_unit_partnership_s_corp_income
     geo_level: national
     domain_variable: tax_unit_partnership_s_corp_income
@@ -199,7 +199,7 @@ include:
   # Restore old loss.py's self-employment return-count target.
   - variable: tax_unit_count
     geo_level: national
-    domain_variable: self_employment_income
+    domain_variable: total_self_employment_income
 
   # === NATIONAL — identity / population count targets from old loss.py ===
   - variable: person_count