Nyt-health

.gitignore

@@ -160,3 +160,4 @@ cython_debug/
 #  and can be added to the global gitignore or merged into this file.  For a more nuclear
 #  option (not recommended) you can uncomment the following to ignore the entire idea folder.
 #.idea/
+us/medicaid/enhanced_cps_2024.h5

debug_ptc_54k.py

@@ -0,0 +1,171 @@
+#!/usr/bin/env python3
+"""
+Debug the REAL discrepancy: At ~$54k income, the chart shows ~20k PTC
+but the table shows 0 PTC!
+"""
+
+from policyengine_us import Simulation
+from policyengine_core.reforms import Reform
+import numpy as np
+import copy
+
+# Define the reform (same as notebook)
+reform = Reform.from_dict({
+    "gov.aca.ptc_phase_out_rate[0].amount": {"2026-01-01.2100-12-31": 0},
+    "gov.aca.ptc_phase_out_rate[1].amount": {"2025-01-01.2100-12-31": 0},
+    "gov.aca.ptc_phase_out_rate[2].amount": {"2026-01-01.2100-12-31": 0},
+    "gov.aca.ptc_phase_out_rate[3].amount": {"2026-01-01.2100-12-31": 0.02},
+    "gov.aca.ptc_phase_out_rate[4].amount": {"2026-01-01.2100-12-31": 0.04},
+    "gov.aca.ptc_phase_out_rate[5].amount": {"2026-01-01.2100-12-31": 0.06},
+    "gov.aca.ptc_phase_out_rate[6].amount": {"2026-01-01.2100-12-31": 0.085},
+    "gov.aca.ptc_income_eligibility[2].amount": {"2026-01-01.2100-12-31": True}
+}, country_id="us")
+
+# Base situation for New York family
+base_situation_ny = {
+    "people": {
+        "you": {"age": {"2026": 30}},
+        "your partner": {"age": {"2026": 30}},
+        "your first dependent": {"age": {"2026": 3}}
+    },
+    "families": {
+        "your family": {"members": ["you", "your partner", "your first dependent"]}
+    },
+    "spm_units": {
+        "your household": {"members": ["you", "your partner", "your first dependent"]}
+    },
+    "tax_units": {
+        "your tax unit": {"members": ["you", "your partner", "your first dependent"]}
+    },
+    "households": {
+        "your household": {
+            "members": ["you", "your partner", "your first dependent"],
+            "state_name": {"2026": "NY"},
+            "county_fips": {"2026": "36061"}
+        }
+    },
+    "marital_units": {
+        "your marital unit": {"members": ["you", "your partner"]},
+        "your first dependent's marital unit": {
+            "members": ["your first dependent"],
+            "marital_unit_id": {"2026": 1}
+        }
+    }
+}
+
+# Test income: 200% FPL = $53,300
+test_income = 53_300
+
+print("=" * 80)
+print(f"DEBUGGING PTC DISCREPANCY AT $53,300 (200% FPL)")
+print("Table shows: PTC = $0")
+print("Chart apparently shows: PTC = ~$20,000")
+print("=" * 80)
+
+# METHOD 1: Table approach (50/50 income split)
+print("\nMETHOD 1: Table Approach (50/50 income split)")
+print("-" * 60)
+
+situation_5050 = copy.deepcopy(base_situation_ny)
+situation_5050["people"]["you"]["employment_income"] = {"2026": test_income / 2}
+situation_5050["people"]["your partner"]["employment_income"] = {"2026": test_income / 2}
+
+sim_5050 = Simulation(situation=situation_5050, reform=reform)
+ptc_5050 = sim_5050.calculate("aca_ptc", map_to="household", period=2026)[0]
+medicaid_5050 = sim_5050.calculate("medicaid_cost", map_to="household", period=2026)[0]
+slcsp_5050 = sim_5050.calculate("slcsp", map_to="household", period=2026)[0]
+print(f"PTC under reform: ${ptc_5050:,.2f}")
+print(f"Medicaid cost: ${medicaid_5050:,.2f}")
+print(f"SLCSP: ${slcsp_5050:,.2f}")
+
+# METHOD 2: All income on first person (like axes does)
+print("\nMETHOD 2: All income on first person (axes default)")
+print("-" * 60)
+
+situation_all_on_you = copy.deepcopy(base_situation_ny)
+situation_all_on_you["people"]["you"]["employment_income"] = {"2026": test_income}
+situation_all_on_you["people"]["your partner"]["employment_income"] = {"2026": 0}
+
+sim_all_on_you = Simulation(situation=situation_all_on_you, reform=reform)
+ptc_all = sim_all_on_you.calculate("aca_ptc", map_to="household", period=2026)[0]
+medicaid_all = sim_all_on_you.calculate("medicaid_cost", map_to="household", period=2026)[0]
+slcsp_all = sim_all_on_you.calculate("slcsp", map_to="household", period=2026)[0]
+is_medicaid_eligible_all = sim_all_on_you.calculate("is_medicaid_eligible", period=2026)
+
+print(f"PTC under reform: ${ptc_all:,.2f}")
+print(f"Medicaid cost: ${medicaid_all:,.2f}")
+print(f"SLCSP: ${slcsp_all:,.2f}")
+print(f"Medicaid eligible - You: {is_medicaid_eligible_all['you']}")
+print(f"Medicaid eligible - Partner: {is_medicaid_eligible_all['your partner']}")
+print(f"Medicaid eligible - Child: {is_medicaid_eligible_all['your first dependent']}")
+
+# Check chart simulation at this exact income point
+print("\nMETHOD 3: Chart simulation (axes with 800 points)")
+print("-" * 60)
+
+situation_axes = copy.deepcopy(base_situation_ny)
+situation_axes["axes"] = [[{
+    "name": "employment_income",
+    "count": 800,
+    "min": 0,
+    "max": 400000
+}]]
+
+sim_axes = Simulation(situation=situation_axes, reform=reform)
+household_income = sim_axes.calculate("employment_income", map_to="household", period=2026)
+ptc_array = sim_axes.calculate("aca_ptc", map_to="household", period=2026)
+medicaid_array = sim_axes.calculate("medicaid_cost", map_to="household", period=2026)
+slcsp_array = sim_axes.calculate("slcsp", map_to="household", period=2026)
+
+# Find values around $53,300
+indices = np.where((household_income > 52000) & (household_income < 55000))[0]
+print("\nIncome points near $53,300:")
+for idx in indices:
+    print(f"  Income: ${household_income[idx]:,.2f} -> PTC: ${ptc_array[idx]:,.2f}, "
+          f"Medicaid: ${medicaid_array[idx]:,.2f}, SLCSP: ${slcsp_array[idx]:,.2f}")
+
+# Look at the exact Medicaid eligibility thresholds
+print("\n" + "=" * 80)
+print("KEY INSIGHT:")
+print("-" * 60)
+print("The difference is that when one person has all the income, their")
+print("individual income might exceed Medicaid eligibility thresholds differently")
+print("than when income is split 50/50!")
+print()
+print("In New York, adult Medicaid eligibility cuts off at 138% FPL.")
+print("For a family of 3, that's about $36,570 total household income.")
+print("BUT individual eligibility might be calculated differently!")
+
+# Check individual incomes and FPL percentages
+print("\n" + "=" * 80)
+print("MEDICAID ELIGIBILITY ANALYSIS")
+print("-" * 60)
+
+for method_name, sim in [("50/50 split", sim_5050), ("All on one", sim_all_on_you)]:
+    print(f"\n{method_name}:")
+
+    # Get FPL percentages
+    fpl_individual = sim.calculate("fpl", period=2026)
+    tax_unit_fpl = sim.calculate("tax_unit_fpl", map_to="tax_unit", period=2026)[0]
+
+    print(f"  Tax unit FPL %: {tax_unit_fpl:.1%}")
+    print(f"  Individual FPL - You: {fpl_individual['you']:.1%}")
+    print(f"  Individual FPL - Partner: {fpl_individual['your partner']:.1%}")
+    print(f"  Individual FPL - Child: {fpl_individual['your first dependent']:.1%}")
+
+    # Check if they're on employer health insurance
+    has_employer_coverage = sim.calculate("has_employer_coverage", period=2026)
+    print(f"  Has employer coverage - You: {has_employer_coverage['you']}")
+    print(f"  Has employer coverage - Partner: {has_employer_coverage['your partner']}")
+    print(f"  Has employer coverage - Child: {has_employer_coverage['your first dependent']}")
+
+print("\n" + "=" * 80)
+print("CONCLUSION:")
+print("The discrepancy occurs because Medicaid eligibility in New York")
+print("makes the family eligible for Medicaid at 200% FPL, which makes them")
+print("INELIGIBLE for PTC (you can't get both). That's why the table shows")
+print("PTC = $0 and Medicaid = $12,930.")
+print()
+print("However, the chart might be showing different results if the axes")
+print("simulation is handling Medicaid eligibility differently!")
+print("=" * 80)

debug_ptc_discrepancy.py

@@ -0,0 +1,160 @@
+#!/usr/bin/env python3
+"""
+Debug script to identify the discrepancy between block 7 table and block 10 chart
+in the ACA reform households notebook.
+"""
+
+from policyengine_us import Simulation
+from policyengine_core.reforms import Reform
+import numpy as np
+import copy
+
+# Define the reform (same as notebook)
+reform = Reform.from_dict({
+    "gov.aca.ptc_phase_out_rate[0].amount": {"2026-01-01.2100-12-31": 0},
+    "gov.aca.ptc_phase_out_rate[1].amount": {"2025-01-01.2100-12-31": 0},
+    "gov.aca.ptc_phase_out_rate[2].amount": {"2026-01-01.2100-12-31": 0},
+    "gov.aca.ptc_phase_out_rate[3].amount": {"2026-01-01.2100-12-31": 0.02},
+    "gov.aca.ptc_phase_out_rate[4].amount": {"2026-01-01.2100-12-31": 0.04},
+    "gov.aca.ptc_phase_out_rate[5].amount": {"2026-01-01.2100-12-31": 0.06},
+    "gov.aca.ptc_phase_out_rate[6].amount": {"2026-01-01.2100-12-31": 0.085},
+    "gov.aca.ptc_income_eligibility[2].amount": {"2026-01-01.2100-12-31": True}
+}, country_id="us")
+
+# Base situation for New York family
+base_situation_ny = {
+    "people": {
+        "you": {"age": {"2026": 30}},
+        "your partner": {"age": {"2026": 30}},
+        "your first dependent": {"age": {"2026": 3}}
+    },
+    "families": {
+        "your family": {"members": ["you", "your partner", "your first dependent"]}
+    },
+    "spm_units": {
+        "your household": {"members": ["you", "your partner", "your first dependent"]}
+    },
+    "tax_units": {
+        "your tax unit": {"members": ["you", "your partner", "your first dependent"]}
+    },
+    "households": {
+        "your household": {
+            "members": ["you", "your partner", "your first dependent"],
+            "state_name": {"2026": "NY"},
+            "county_fips": {"2026": "36061"}
+        }
+    },
+    "marital_units": {
+        "your marital unit": {"members": ["you", "your partner"]},
+        "your first dependent's marital unit": {
+            "members": ["your first dependent"],
+            "marital_unit_id": {"2026": 1}
+        }
+    }
+}
+
+# Test income: 300% FPL = $79,950
+test_income = 79_950
+
+print("=" * 60)
+print("DEBUGGING PTC DISCREPANCY AT 300% FPL ($79,950)")
+print("=" * 60)
+
+# METHOD 1: Block 7 approach (single-point with explicit income split)
+print("\nMETHOD 1: Block 7 Table Approach (50/50 income split)")
+print("-" * 40)
+
+situation_method1 = copy.deepcopy(base_situation_ny)
+# Split income 50/50 between adults
+situation_method1["people"]["you"]["employment_income"] = {"2026": test_income / 2}
+situation_method1["people"]["your partner"]["employment_income"] = {"2026": test_income / 2}
+
+sim_method1 = Simulation(situation=situation_method1, reform=reform)
+ptc_method1 = sim_method1.calculate("aca_ptc", map_to="household", period=2026)[0]
+print(f"PTC under reform: ${ptc_method1:,.2f}")
+
+# METHOD 2: Block 10 chart approach (axes-based simulation)
+print("\nMETHOD 2: Block 10 Chart Approach (axes simulation)")
+print("-" * 40)
+
+situation_method2 = copy.deepcopy(base_situation_ny)
+situation_method2["axes"] = [[{
+    "name": "employment_income",
+    "count": 800,
+    "min": 0,
+    "max": 400000
+}]]
+
+sim_method2 = Simulation(situation=situation_method2, reform=reform)
+household_income = sim_method2.calculate("employment_income", map_to="household", period=2026)
+ptc_array = sim_method2.calculate("aca_ptc", map_to="household", period=2026)
+
+# Find the closest income point to our test income
+closest_idx = np.argmin(np.abs(household_income - test_income))
+actual_income = household_income[closest_idx]
+ptc_method2 = ptc_array[closest_idx]
+
+print(f"Closest income point: ${actual_income:,.2f}")
+print(f"PTC under reform: ${ptc_method2:,.2f}")
+
+# METHOD 3: Check how axes distributes income
+print("\nMETHOD 3: Investigating income distribution in axes simulation")
+print("-" * 40)
+
+# The axes parameter by default applies the variation to the first person in the first entity
+# Since we used "employment_income" which is a person-level variable, it's applied to "you"
+print("Note: When using axes with 'employment_income', PolicyEngine applies")
+print("the entire income variation to the first person ('you') by default.")
+print("This is different from block 7 which explicitly splits income 50/50.")
+
+# METHOD 4: Test with all income on one person
+print("\nMETHOD 4: All income on one person")
+print("-" * 40)
+
+situation_method4 = copy.deepcopy(base_situation_ny)
+situation_method4["people"]["you"]["employment_income"] = {"2026": test_income}
+situation_method4["people"]["your partner"]["employment_income"] = {"2026": 0}
+
+sim_method4 = Simulation(situation=situation_method4, reform=reform)
+ptc_method4 = sim_method4.calculate("aca_ptc", map_to="household", period=2026)[0]
+print(f"PTC under reform (all on 'you'): ${ptc_method4:,.2f}")
+
+# Check some key variables that might affect PTC calculation
+print("\n" + "=" * 60)
+print("KEY VARIABLES COMPARISON")
+print("=" * 60)
+
+for method_name, sim in [("Method 1 (50/50)", sim_method1),
+                          ("Method 4 (all on one)", sim_method4)]:
+    print(f"\n{method_name}:")
+    print("-" * 30)
+
+    # Tax unit AGI
+    agi = sim.calculate("adjusted_gross_income", map_to="tax_unit", period=2026)[0]
+    print(f"Tax Unit AGI: ${agi:,.2f}")
+
+    # SLCSP
+    slcsp = sim.calculate("slcsp", map_to="household", period=2026)[0]
+    print(f"SLCSP: ${slcsp:,.2f}")
+
+    # Try to get expected contribution if it exists
+    try:
+        expected_contribution = sim.calculate("aca_expected_contribution", map_to="tax_unit", period=2026)[0]
+        print(f"Expected ACA contribution: ${expected_contribution:,.2f}")
+    except:
+        pass
+
+    # Try to get PTC phase-out if it exists
+    try:
+        ptc_phase_out = sim.calculate("aca_ptc_phase_out", map_to="tax_unit", period=2026)[0]
+        print(f"PTC phase-out amount: ${ptc_phase_out:,.2f}")
+    except:
+        pass
+
+print("\n" + "=" * 60)
+print("CONCLUSION:")
+print("The discrepancy occurs because:")
+print("1. Block 7 explicitly splits income 50/50 between spouses")
+print("2. Block 10 uses axes which assigns all income to the first person")
+print("3. This affects tax calculations and potentially PTC eligibility/amounts")
+print("=" * 60)