Complete analysis of household 4428 itemization

final_analysis.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Household 4428 Itemization Analysis\n",
    "\n",
    "This notebook clarifies the itemization behavior and apparent calculation issues."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from policyengine_us import Microsimulation\n",
    "from policyengine_us.model_api import *\n",
    "import numpy as np\n",
    "\n",
    "# Setup\n",
    "DATASET = \"hf://policyengine/policyengine-us-data/enhanced_cps_2024.h5\"\n",
    "YEAR = 2026\n",
    "\n",
    "# Reform stack\n",
    "reform = Reform.from_dict({\n",
    "    \"gov.irs.income.bracket.rates.2\": {\"2026-01-01\": 0.15},\n",
    "    \"gov.irs.income.bracket.rates.3\": {\"2026-01-01\": 0.25},\n",
    "    \"gov.irs.deductions.standard.amount.JOINT\": {\"2026-01-01\": 48900},\n",
    "    \"gov.irs.income.exemption.amount\": {\"2026-01-01\": 0},\n",
    "    \"gov.contrib.reconciliation.ctc.in_effect\": {\"2026-01-01\": True},\n",
    "    \"gov.irs.credits.ctc.amount.base[0].amount\": {\"2026-01-01\": 2200},\n",
    "}, country_id=\"us\")\n",
    "\n",
    "sim = Microsimulation(reform=reform, dataset=DATASET)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Key Finding: Household 4428 Has Two Tax Units"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Get data for both tax units\n",
    "tax_unit_ids = sim.calculate(\"tax_unit_id\", period=YEAR).values\n",
    "tu1_idx = np.where(tax_unit_ids == 442801)[0][0]\n",
    "tu2_idx = np.where(tax_unit_ids == 442802)[0][0]\n",
    "\n",
    "def get_value(var, idx):\n",
    "    return sim.calculate(var, period=YEAR).values[idx]\n",
    "\n",
    "print(\"TAX UNIT 442801 (Parents + 4 kids):\")\n",
    "print(f\"  AGI: ${get_value('adjusted_gross_income', tu1_idx):,.2f}\")\n",
    "print(f\"  Itemizes: {bool(get_value('tax_unit_itemizes', tu1_idx))}\")\n",
    "print(f\"  Deductions: ${get_value('taxable_income_deductions', tu1_idx):,.2f}\")\n",
    "print(f\"  Tax: ${get_value('income_tax', tu1_idx):,.2f}\")\n",
    "print(f\"  Tax if itemizing: ${get_value('tax_liability_if_itemizing', tu1_idx):,.2f}\")\n",
    "print(f\"  Tax if standard: ${get_value('tax_liability_if_not_itemizing', tu1_idx):,.2f}\")\n",
    "\n",
    "print(\"\\nTAX UNIT 442802 (23-year-old):\")\n",
    "print(f\"  AGI: ${get_value('adjusted_gross_income', tu2_idx):,.2f}\")\n",
    "print(f\"  Itemizes: {bool(get_value('tax_unit_itemizes', tu2_idx))}\")\n",
    "print(f\"  Deductions: ${get_value('taxable_income_deductions', tu2_idx):,.2f}\")\n",
    "print(f\"  Tax: ${get_value('income_tax', tu2_idx):,.2f}\")\n",
    "\n",
    "print(\"\\n✓ Both tax units make rational decisions\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## The Aggregation Issue\n",
    "\n",
    "When values are mapped to household level, confusing things happen:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Get household-level values\n",
    "household_ids = sim.calculate(\"household_id\", map_to=\"household\", period=YEAR).values\n",
    "h_idx = np.where(household_ids == 4428)[0][0]\n",
    "\n",
    "h_itemizes = sim.calculate(\"tax_unit_itemizes\", map_to=\"household\", period=YEAR).values[h_idx]\n",
    "h_deductions = sim.calculate(\"taxable_income_deductions\", map_to=\"household\", period=YEAR).values[h_idx]\n",
    "h_tax_if_item = sim.calculate(\"tax_liability_if_itemizing\", map_to=\"household\", period=YEAR).values[h_idx]\n",
    "h_tax_if_std = sim.calculate(\"tax_liability_if_not_itemizing\", map_to=\"household\", period=YEAR).values[h_idx]\n",
    "\n",
    "print(\"HOUSEHOLD-LEVEL AGGREGATION:\")\n",
    "print(f\"  Itemizes: {bool(h_itemizes)} (from first tax unit)\")\n",
    "print(f\"  Total deductions: ${h_deductions:,.2f}\")\n",
    "print(f\"  Tax if itemizing: ${h_tax_if_item:,.2f}\")\n",
    "print(f\"  Tax if standard: ${h_tax_if_std:,.2f}\")\n",
    "\n",
    "print(\"\\nThis creates the false appearance that the household itemizes\")
",
    "print(\"despite worse tax outcomes, but it's really two separate decisions.\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## The Deduction Mystery\n",
    "\n",
    "The $18,906.70 in household deductions comes from a later reform (likely senior deduction)\n",
    "that adds benefits to itemizers beyond just itemized deductions."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}