TCJA_perm_taxcalc.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "jewish-jumping",
   "metadata": {},
   "source": [
    "# TCJA Permanency Act\n",
    "To run this notebook, you need the following:\n",
    "1. Fork the repository https://github.com/OpenSourceEcon/TCJAperm and clone it to your local machine.\n",
    "2. Navigate to the `/TCJAperm/` directory in your terminal and create the `tcjaperm-dev` conda environment by typing: `conda env create -f environment.yml`\n",
    "3. Open the `TCJA_perm_taxcalc.ipynb` notebook and set the kernel in the notebook to the `tcjaperm-dev` conda environment.\n",
    "4. Execute the cells in the `TCJA_perm_taxcalc.ipynb` notebook.\n",
    "    * You will need to adjust the file path to the `puf.csv` file in code cell 2 and the path to the `TCJA_perm.json` file in code cell 3.\n",
    "    * If you do not have the IRS Public Use File, then you will need to use the CPS file included with Tax-Calculator. That code is commented out in cell 2."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "composite-while",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "from bokeh.plotting import figure, show\n",
    "from bokeh.io import output_notebook\n",
    "import taxcalc as tc\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "surprising-democracy",
   "metadata": {},
   "outputs": [],
   "source": [
    "puf_path = (\n",
    "    \"/Users/richardevans/Docs/Economics/OSE/Tax-Calculator/taxcalc/puf.csv\"\n",
    ")\n",
    "recs = tc.Records(data=puf_path)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d2f0e958",
   "metadata": {},
   "source": [
    "## 1. Solve for total effect of the policy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "eccd73c1",
   "metadata": {},
   "outputs": [],
   "source": [
    "# If you need to use the CPS data file included with Tax-Calculator, comment\n",
    "# out the previous lines in this cell and uncomment the following line.\n",
    "# recs = tc.Records.cps_constructor()\n",
    "pol = tc.Policy()\n",
    "calc_base = tc.Calculator(policy=pol, records=recs)\n",
    "CYR = 2024\n",
    "calc_base.advance_to_year(CYR)\n",
    "calc_base.calc_all()\n",
    "itax_rev_base = calc_base.weighted_total('iitax')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "cd7cb6a2",
   "metadata": {},
   "outputs": [],
   "source": [
    "reform_json_path = (\n",
    "    \"/Users/richardevans/Docs/Economics/OSE/TCJAperm/json/TCJA_perm.json\"\n",
    ")\n",
    "params_ref1 = pol.read_json_reform(reform_json_path)\n",
    "pol.implement_reform(params_ref1)\n",
    "calc_ref1 = tc.Calculator(policy=pol, records=recs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "supposed-magnitude",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Compute annual revenue for each year in a 10 year budget window\n",
    "# Do this for with and without the ARPA\n",
    "rev_base = {'IIT': {}, 'Payroll': {}}\n",
    "rev_ref1 = {'IIT': {}, 'Payroll': {}}\n",
    "for t in range(2024, 2034):\n",
    "    calc_base.advance_to_year(t)\n",
    "    calc_base.calc_all()\n",
    "    calc_ref1.advance_to_year(t)\n",
    "    calc_ref1.calc_all()\n",
    "    rev_base['IIT'][t] = calc_base.weighted_total('iitax')\n",
    "    rev_base['Payroll'][t] = calc_base.weighted_total('payrolltax')\n",
    "    rev_ref1['IIT'][t] = calc_ref1.weighted_total('iitax')\n",
    "    rev_ref1['Payroll'][t] = calc_ref1.weighted_total('payrolltax')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "abroad-samba",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>2024</th>\n",
       "      <th>2025</th>\n",
       "      <th>2026</th>\n",
       "      <th>2027</th>\n",
       "      <th>2028</th>\n",
       "      <th>2029</th>\n",
       "      <th>2030</th>\n",
       "      <th>2031</th>\n",
       "      <th>2032</th>\n",
       "      <th>2033</th>\n",
       "      <th>2024-2033</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>IIT</th>\n",
       "      <td>$0.000</td>\n",
       "      <td>$0.000</td>\n",
       "      <td>$-383.212</td>\n",
       "      <td>$-397.469</td>\n",
       "      <td>$-411.914</td>\n",
       "      <td>$-426.896</td>\n",
       "      <td>$-442.421</td>\n",
       "      <td>$-458.603</td>\n",
       "      <td>$-474.087</td>\n",
       "      <td>$-485.583</td>\n",
       "      <td>$-3480.186</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Payroll</th>\n",
       "      <td>$0.000</td>\n",
       "      <td>$0.000</td>\n",
       "      <td>$0.000</td>\n",
       "      <td>$0.000</td>\n",
       "      <td>$0.000</td>\n",
       "      <td>$0.000</td>\n",
       "      <td>$0.000</td>\n",
       "      <td>$0.000</td>\n",
       "      <td>$0.000</td>\n",
       "      <td>$0.000</td>\n",
       "      <td>$0.000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          2024   2025      2026      2027      2028      2029      2030  \\\n",
       "IIT     $0.000 $0.000 $-383.212 $-397.469 $-411.914 $-426.896 $-442.421   \n",
       "Payroll $0.000 $0.000    $0.000    $0.000    $0.000    $0.000    $0.000   \n",
       "\n",
       "             2031      2032      2033  2024-2033  \n",
       "IIT     $-458.603 $-474.087 $-485.583 $-3480.186  \n",
       "Payroll    $0.000    $0.000    $0.000     $0.000  "
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# make table of revenue estimates\n",
    "rev_base_df = pd.DataFrame.from_dict(rev_base).T\n",
    "rev_ref1_df = pd.DataFrame.from_dict(rev_ref1).T\n",
    "diff_df = (rev_ref1_df - rev_base_df) * 1e-9  # in $ billions\n",
    "pd.options.display.float_format = '${:.3f}'.format\n",
    "diff_df['2024-2033'] = diff_df.sum(axis=1)\n",
    "# print(diff_df.to_latex())\n",
    "diff_df"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "80fb0b01",
   "metadata": {},
   "source": [
    "## 2. Create decomposition of tax effects by category"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "e0d4cd77",
   "metadata": {},
   "outputs": [],
   "source": [
    "sec_ref1_list = [\n",
    "    (\n",
    "        \"Sec 1: (II_rt*, II_brk*, PT_rt*, PT_brk*, PT_qbid*)\",\n",
    "        {\n",
    "            \"II_rt2\": {\"2026\": 0.12},\n",
    "            \"II_rt3\": {\"2026\": 0.22},\n",
    "            \"II_rt4\": {\"2026\": 0.24},\n",
    "            \"II_rt5\": {\"2026\": 0.32},\n",
    "            \"II_rt7\": {\"2026\": 0.37},\n",
    "            \"II_brk1\": {\"2026\": [10275.0, 20550.0, 10275.0, 14650.0, 20550.0]},\n",
    "            \"II_brk2\": {\"2026\": [41775.0, 83550.0, 41775.0, 55900.0, 83550.0]},\n",
    "            \"II_brk3\": {\"2026\": [89075.0, 178150.0, 89075.0, 89050.0, 178150.0]},\n",
    "            \"II_brk4\": {\"2026\": [170050.0, 340100.0, 170050.0, 170050.0, 340100.0]},\n",
    "            \"II_brk5\": {\"2026\": [215950.0, 431900.0, 215950.0, 215950.0, 431900.0]},\n",
    "            \"II_brk6\": {\"2026\": [539900.0, 647850.0, 323925.0, 539900.0, 647850.0]},\n",
    "            \"PT_rt2\": {\"2026\": 0.12},\n",
    "            \"PT_rt3\": {\"2026\": 0.22},\n",
    "            \"PT_rt4\": {\"2026\": 0.24},\n",
    "            \"PT_rt5\": {\"2026\": 0.32},\n",
    "            \"PT_rt7\": {\"2026\": 0.37},\n",
    "            \"PT_brk1\": {\"2026\": [10275.0, 20550.0, 10275.0, 14650.0, 20550.0]},\n",
    "            \"PT_brk2\": {\"2026\": [41775.0, 83550.0, 41775.0, 55900.0, 83550.0]},\n",
    "            \"PT_brk3\": {\"2026\": [89075.0, 178150.0, 89075.0, 89050.0, 178150.0]},\n",
    "            \"PT_brk4\": {\"2026\": [170050.0, 340100.0, 170050.0, 170050.0, 340100.0]},\n",
    "            \"PT_brk5\": {\"2026\": [215950.0, 431900.0, 215950.0, 215950.0, 431900.0]},\n",
    "            \"PT_brk6\": {\"2026\": [539900.0, 647850.0, 323925.0, 539900.0, 647850.0]},\n",
    "            \"PT_qbid_rt\": {\"2026\": 0.2},\n",
    "            \"PT_qbid_taxinc_thd\": {\"2026\": [170050.0, 340100.0, 170050.0, 170050.0, 340100.0]},\n",
    "            \"PT_qbid_taxinc_gap\": {\"2026\": [50000.0, 100000.0, 50000.0, 50000.0, 100000.0]},\n",
    "            \"PT_qbid_w2_wages_rt\": {\"2026\": 0.5},\n",
    "            \"PT_qbid_alt_w2_wages_rt\": {\"2026\": 0.25},\n",
    "            \"PT_qbid_alt_property_rt\": {\"2026\": 0.025}\n",
    "        }\n",
    "    ),\n",
    "    (\n",
    "        \"Sec 2: (STD, ID_*)\",\n",
    "        {\n",
    "            \"STD\": {\"2026\": [12950.0, 25900.0, 12950.0, 19400.0, 25900.0]}\n",
    "        }\n",
    "    ),\n",
    "    (\n",
    "        \"Sec 3: (CTC_*, ACTC_*, ODC_c)\",\n",
    "        {\n",
    "            \"CTC_c\": {\"2026\": 2000.0},\n",
    "            \"CTC_ps\": {\"2026\": [200000.0, 400000.0, 200000.0, 200000.0, 400000.0]},\n",
    "            \"ACTC_c\": {\"2026\": 1600.0},\n",
    "            \"ODC_c\": {\"2026\": 500.0},\n",
    "            \"ACTC_Income_thd\": {\"2026\": 2500.0}\n",
    "        }\n",
    "    ),\n",
    "    (\n",
    "        \"Sec 4: (AMT_*)\",\n",
    "        {\n",
    "            \"AMT_em\": {\"2026\": [75900.0, 118100.0, 59050.0, 75900.0, 118100.0]},\n",
    "            \"AMT_em_ps\": {\"2026\": [539900.0, 1079800.0, 539900.0, 539900.0, 1079800.0]},\n",
    "            \"AMT_em_pe\": {\"2026\": 776100.0},\n",
    "        }\n",
    "    )\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "0b4bebd8",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Simulating baseline\n",
      "\n",
      "Simulating Sec 1: (II_rt*, II_brk*, PT_rt*, PT_brk*, PT_qbid*)\n",
      "      2024   2025      2026      2027      2028      2029      2030      2031  \\\n",
      "IIT $0.000 $0.000 $-143.121 $-149.832 $-156.407 $-162.981 $-169.666 $-176.234   \n",
      "\n",
      "         2032      2033  2024-2033  \n",
      "IIT $-181.282 $-189.343 $-1328.866  \n",
      "\n",
      "Simulating Sec 2: (STD, ID_*)\n",
      "      2024   2025     2026     2027     2028     2029     2030     2031  \\\n",
      "IIT $0.000 $0.000 $-70.694 $-72.944 $-75.368 $-77.979 $-80.766 $-83.784   \n",
      "\n",
      "        2032     2033  2024-2033  \n",
      "IIT $-86.502 $-88.491  $-636.529  \n",
      "\n",
      "Simulating Sec 3: (CTC_*, ACTC_*, ODC_c)\n",
      "      2024   2025     2026     2027     2028     2029     2030     2031  \\\n",
      "IIT $0.000 $0.000 $-79.175 $-79.926 $-80.646 $-81.479 $-82.144 $-82.753   \n",
      "\n",
      "        2032     2033  2024-2033  \n",
      "IIT $-83.272 $-83.764  $-653.159  \n",
      "\n",
      "Simulating Sec 4: (AMT_*)\n",
      "      2024   2025     2026     2027     2028      2029      2030      2031  \\\n",
      "IIT $0.000 $0.000 $-90.221 $-94.766 $-99.494 $-104.456 $-109.844 $-115.833   \n",
      "\n",
      "         2032      2033  2024-2033  \n",
      "IIT $-123.031 $-123.985  $-861.631  \n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Provisions</th>\n",
       "      <th>2024</th>\n",
       "      <th>2025</th>\n",
       "      <th>2026</th>\n",
       "      <th>2027</th>\n",
       "      <th>2028</th>\n",
       "      <th>2029</th>\n",
       "      <th>2030</th>\n",
       "      <th>2031</th>\n",
       "      <th>2032</th>\n",
       "      <th>2033</th>\n",
       "      <th>2024-2033</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Sec 1: (II_rt*, II_brk*, PT_rt*, PT_brk*, PT_q...</td>\n",
       "      <td>$0.000</td>\n",
       "      <td>$0.000</td>\n",
       "      <td>$-143.121</td>\n",
       "      <td>$-149.832</td>\n",
       "      <td>$-156.407</td>\n",
       "      <td>$-162.981</td>\n",
       "      <td>$-169.666</td>\n",
       "      <td>$-176.234</td>\n",
       "      <td>$-181.282</td>\n",
       "      <td>$-189.343</td>\n",
       "      <td>$-1328.866</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Sec 2: (STD, ID_*)</td>\n",
       "      <td>$0.000</td>\n",
       "      <td>$0.000</td>\n",
       "      <td>$-70.694</td>\n",
       "      <td>$-72.944</td>\n",
       "      <td>$-75.368</td>\n",
       "      <td>$-77.979</td>\n",
       "      <td>$-80.766</td>\n",
       "      <td>$-83.784</td>\n",
       "      <td>$-86.502</td>\n",
       "      <td>$-88.491</td>\n",
       "      <td>$-636.529</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Sec 3: (CTC_*, ACTC_*, ODC_c)</td>\n",
       "      <td>$0.000</td>\n",
       "      <td>$0.000</td>\n",
       "      <td>$-79.175</td>\n",
       "      <td>$-79.926</td>\n",
       "      <td>$-80.646</td>\n",
       "      <td>$-81.479</td>\n",
       "      <td>$-82.144</td>\n",
       "      <td>$-82.753</td>\n",
       "      <td>$-83.272</td>\n",
       "      <td>$-83.764</td>\n",
       "      <td>$-653.159</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Sec 4: (AMT_*)</td>\n",
       "      <td>$0.000</td>\n",
       "      <td>$0.000</td>\n",
       "      <td>$-90.221</td>\n",
       "      <td>$-94.766</td>\n",
       "      <td>$-99.494</td>\n",
       "      <td>$-104.456</td>\n",
       "      <td>$-109.844</td>\n",
       "      <td>$-115.833</td>\n",
       "      <td>$-123.031</td>\n",
       "      <td>$-123.985</td>\n",
       "      <td>$-861.631</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                          Provisions   2024   2025      2026  \\\n",
       "0  Sec 1: (II_rt*, II_brk*, PT_rt*, PT_brk*, PT_q... $0.000 $0.000 $-143.121   \n",
       "1                                 Sec 2: (STD, ID_*) $0.000 $0.000  $-70.694   \n",
       "2                      Sec 3: (CTC_*, ACTC_*, ODC_c) $0.000 $0.000  $-79.175   \n",
       "3                                     Sec 4: (AMT_*) $0.000 $0.000  $-90.221   \n",
       "\n",
       "       2027      2028      2029      2030      2031      2032      2033  \\\n",
       "0 $-149.832 $-156.407 $-162.981 $-169.666 $-176.234 $-181.282 $-189.343   \n",
       "1  $-72.944  $-75.368  $-77.979  $-80.766  $-83.784  $-86.502  $-88.491   \n",
       "2  $-79.926  $-80.646  $-81.479  $-82.144  $-82.753  $-83.272  $-83.764   \n",
       "3  $-94.766  $-99.494 $-104.456 $-109.844 $-115.833 $-123.031 $-123.985   \n",
       "\n",
       "   2024-2033  \n",
       "0 $-1328.866  \n",
       "1  $-636.529  \n",
       "2  $-653.159  \n",
       "3  $-861.631  "
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Create blank pandas dataframe with columns [\"Provisions\", \"2024\", \"2025\",..., \"2033\", \"2024-2033\"]\n",
    "diff_seq_df = pd.DataFrame(\n",
    "    columns=[\n",
    "        \"Provisions\", \"2024\", \"2025\", \"2026\", \"2027\", \"2028\", \"2029\", \"2030\",\n",
    "        \"2031\", \"2032\", \"2033\", \"2024-2033\"\n",
    "    ]\n",
    ")\n",
    "\n",
    "print(\"\")\n",
    "print(\"Simulating baseline\")\n",
    "pol = tc.Policy()\n",
    "calc_base = tc.Calculator(policy=pol, records=recs)\n",
    "CYR = 2024\n",
    "calc_base.advance_to_year(CYR)\n",
    "calc_base.calc_all()\n",
    "rev_base = {'IIT': {}}\n",
    "for t in range(2024, 2034):\n",
    "    calc_base.advance_to_year(t)\n",
    "    calc_base.calc_all()\n",
    "    rev_base['IIT'][str(t)] = calc_base.weighted_total('iitax')\n",
    "\n",
    "rev_base_df = pd.DataFrame.from_dict(rev_base).T\n",
    "\n",
    "# Simulate each reform\n",
    "for i,tpl in enumerate(sec_ref1_list):\n",
    "    print(\"\")\n",
    "    print(\"Simulating \" + tpl[0])\n",
    "    params_ref = pol.read_json_reform(tpl[1])\n",
    "    pol.implement_reform(params_ref)\n",
    "    calc_ref = tc.Calculator(policy=pol, records=recs)\n",
    "\n",
    "    rev_ref = {'IIT': {}}\n",
    "    for t in range(2024, 2034):\n",
    "        calc_ref.advance_to_year(t)\n",
    "        calc_ref.calc_all()\n",
    "        rev_ref['IIT'][str(t)] = calc_ref.weighted_total('iitax')\n",
    "\n",
    "    rev_ref_df = pd.DataFrame.from_dict(rev_ref).T\n",
    "    diff_sec_df = (rev_ref_df - rev_base_df) * 1e-9  # in $ billions\n",
    "    pd.options.display.float_format = '${:.3f}'.format\n",
    "    diff_sec_df['2024-2033'] = diff_sec_df.sum(axis=1)\n",
    "    print(diff_sec_df)\n",
    "    diff_seq_df.loc[i, \"Provisions\"] = tpl[0]\n",
    "    numeric_columns = [\n",
    "        \"2024\", \"2025\", \"2026\", \"2027\", \"2028\", \"2029\", \"2030\", \"2031\", \"2032\",\n",
    "        \"2033\", \"2024-2033\"\n",
    "    ]\n",
    "    for col in numeric_columns:\n",
    "        diff_seq_df.loc[i, col] = diff_sec_df.loc[\"IIT\", col]\n",
    "\n",
    "    rev_base_df = rev_ref_df.copy()\n",
    "\n",
    "diff_seq_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "dd97c302",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Provisions    Sec 1: (II_rt*, II_brk*, PT_rt*, PT_brk*, PT_q...\n",
       "2024                                                     $0.000\n",
       "2025                                                     $0.000\n",
       "2026                                                  $-383.212\n",
       "2027                                                  $-397.469\n",
       "2028                                                  $-411.914\n",
       "2029                                                  $-426.896\n",
       "2030                                                  $-442.421\n",
       "2031                                                  $-458.603\n",
       "2032                                                  $-474.087\n",
       "2033                                                  $-485.583\n",
       "2024-2033                                            $-3480.186\n",
       "dtype: object"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "diff_seq_df.sum(axis=0)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "523d6c17",
   "metadata": {},
   "source": [
    "## 3. Create average effective tax rates plot for 2026"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "3677d8bb",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Restart calculation to get back to 2026\n",
    "pol_2026 = tc.Policy()\n",
    "calc_base_2026 = tc.Calculator(policy=pol_2026, records=recs)\n",
    "pol_2026.implement_reform(params_ref1)\n",
    "calc_ref1_2026 = tc.Calculator(policy=pol_2026, records=recs)\n",
    "calc_base_2026.advance_to_year(2026)\n",
    "calc_ref1_2026.advance_to_year(2026)\n",
    "calc_base_2026.calc_all()\n",
    "calc_ref1_2026.calc_all()\n",
    "\n",
    "# get data for plots\n",
    "mars = 'ALL'\n",
    "mtr_wrt_full_compen = True\n",
    "record_variables = ['s006', 'XTOT']\n",
    "if mars != 'ALL':\n",
    "    record_variables.append('MARS')\n",
    "record_variables.append('expanded_income')\n",
    "record_variables.append('e00200')\n",
    "vdf = calc_base_2026.dataframe(record_variables)\n",
    "# create 'tax1' and 'tax2' columns given specified atr_measure\n",
    "vdf['tax1'] = calc_base_2026.array('iitax')\n",
    "vdf['tax2'] = calc_ref1_2026.array('iitax')\n",
    "# compute mtrs for mtr plot data\n",
    "(mtr_base_ptax, mtr_base_itax, mtr_base_combined) = calc_base_2026.mtr(\n",
    "    variable_str='e00200p', wrt_full_compensation=mtr_wrt_full_compen\n",
    ")\n",
    "(mtr_ref1_ptax, mtr_ref1_itax, mtr_ref1_combined) = calc_ref1_2026.mtr(\n",
    "    variable_str='e00200p', wrt_full_compensation=mtr_wrt_full_compen\n",
    ")\n",
    "vdf['mtr1'] = mtr_base_itax\n",
    "vdf['mtr2'] = mtr_ref1_itax\n",
    "\n",
    "mtr_data = tc.calculator.mtr_graph_data(vdf,\n",
    "                              year=2026,\n",
    "                              mars=mars,\n",
    "                              mtr_measure='itax',\n",
    "                              alt_e00200p_text='This text',\n",
    "                              mtr_wrt_full_compen=True,\n",
    "                              income_measure='expanded_income',\n",
    "                              pop_quantiles=True,\n",
    "                              dollar_weighting=False)\n",
    "\n",
    "atr_data = tc.calculator.atr_graph_data(vdf,\n",
    "                              year=2026,\n",
    "                              mars=mars,\n",
    "                              atr_measure='itax',\n",
    "                              pop_quantiles=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "3d58fbca",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(mtr_data['lines'].index, mtr_data['lines'].base.values, label='Expiring TCJA')\n",
    "plt.plot(mtr_data['lines'].index, mtr_data['lines'].reform.values, label='TCJA made permanent')\n",
    "plt.xlabel('Income Percentile')\n",
    "plt.ylabel('Marginal Tax Rate on Employment Income')\n",
    "plt.legend(loc=9, bbox_to_anchor=(0.5, -0.15), ncol=2)\n",
    "plt.savefig('images/tcja_perm_mtr_plot.png', bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "e0dcc730",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(atr_data['lines'].index, atr_data['lines'].base.values, label='Expiring TCJA')\n",
    "plt.plot(atr_data['lines'].index, atr_data['lines'].reform.values, label='TCJA made permanent')\n",
    "plt.grid(which=\"major\", color=\"lightgray\", linestyle=\"--\")\n",
    "plt.xlabel('Income Percentile')\n",
    "plt.ylabel('Average Tax Rate on Employment Income')\n",
    "plt.legend(loc=9, bbox_to_anchor=(0.5, -0.15), ncol=2)\n",
    "plt.savefig('images/tcja_perm_atr_plot.png', bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "atomic-throat",
   "metadata": {},
   "source": [
    "# Distributional results\n",
    "\n",
    "See below + table by Cody"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "dedicated-navigator",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Restart calculation to get back to first year\n",
    "pol = taxcalc.policy.Policy()\n",
    "tc_base = taxcalc.calculator.Calculator(policy=pol, records=rec)\n",
    "pol.implement_reform(ref)\n",
    "tc_arpa = taxcalc.calculator.Calculator(policy=pol, records=rec)\n",
    "tc_base.advance_to_year(2021)\n",
    "tc_arpa.advance_to_year(2021)\n",
    "tc_base.calc_all()\n",
    "tc_arpa.calc_all()\n",
    "\n",
    "# get data for plots\n",
    "mars = 'ALL'\n",
    "mtr_wrt_full_compen = True\n",
    "record_variables = ['s006', 'XTOT']\n",
    "if mars != 'ALL':\n",
    "    record_variables.append('MARS')\n",
    "record_variables.append('expanded_income')\n",
    "record_variables.append('e00200')\n",
    "vdf = tc_base.dataframe(record_variables)\n",
    "# create 'tax1' and 'tax2' columns given specified atr_measure\n",
    "vdf['tax1'] = tc_base.array('iitax')\n",
    "vdf['tax2'] = tc_arpa.array('iitax')\n",
    "# compute mtrs for mtr plot data\n",
    "(mtr1_ptax, mtr1_itax, mtr1_combined) = tc_base.mtr(\n",
    "    variable_str='e00200p', wrt_full_compensation=mtr_wrt_full_compen)\n",
    "(mtr2_ptax, mtr2_itax, mtr2_combined) = tc_arpa.mtr(\n",
    "    variable_str='e00200p', wrt_full_compensation=mtr_wrt_full_compen)\n",
    "vdf['mtr1'] = mtr1_itax\n",
    "vdf['mtr2'] = mtr2_itax\n",
    "\n",
    "mtr_data = taxcalc.calculator.mtr_graph_data(vdf,\n",
    "                              year=2021,\n",
    "                              mars=mars,\n",
    "                              mtr_measure='itax',\n",
    "                              alt_e00200p_text='This text',\n",
    "                              mtr_wrt_full_compen=True,\n",
    "                              income_measure='expanded_income',\n",
    "                              pop_quantiles=True,\n",
    "                              dollar_weighting=False)\n",
    "atr_data = taxcalc.calculator.atr_graph_data(vdf,\n",
    "                              year=2021,\n",
    "                              mars=mars,\n",
    "                              atr_measure='itax',\n",
    "                              pop_quantiles=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "healthy-newton",
   "metadata": {},
   "outputs": [],
   "source": [
    "plt.plot(mtr_data['lines'].index, mtr_data['lines'].base.values, label='Pre-ARPA')\n",
    "plt.plot(mtr_data['lines'].index, mtr_data['lines'].reform.values, label='Post-ARPA')\n",
    "plt.xlabel('Income Percentile')\n",
    "plt.ylabel('Marginal Tax Rate on Wage Income')\n",
    "plt.legend(loc=9, bbox_to_anchor=(0.5, -0.15), ncol=2)\n",
    "plt.savefig('arpa_mtr_plot.png', bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "incorrect-session",
   "metadata": {},
   "outputs": [],
   "source": [
    "plt.plot(atr_data['lines'].index, atr_data['lines'].base.values, label='Pre-ARPA')\n",
    "plt.plot(atr_data['lines'].index, atr_data['lines'].reform.values, label='Post-ARPA')\n",
    "plt.xlabel('Income Percentile')\n",
    "plt.ylabel('Average Tax Rate')\n",
    "plt.legend(loc=9, bbox_to_anchor=(0.5, -0.15), ncol=2)\n",
    "plt.savefig('arpa_atr_plot.png', bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "brave-gravity",
   "metadata": {},
   "source": [
    "# Tax benefits by family size"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "numerical-subject",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = tc_base.dataframe(['nu18', 'XTOT', 'iitax', 'c00100', 's006', 'aftertax_income', 'expanded_income'])\n",
    "df['ARPA_iitax'] = tc_arpa.array('iitax')\n",
    "df['tax_change'] = df['ARPA_iitax'] - df['iitax']\n",
    "df['aftertax_income_ARPA'] = tc_arpa.array('aftertax_income')\n",
    "df.groupby('nu18')[['s006', 'XTOT']].sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "focused-nicaragua",
   "metadata": {},
   "outputs": [],
   "source": [
    "mdf.weighted_mean(df, 'nu18', 's006')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "australian-protein",
   "metadata": {},
   "outputs": [],
   "source": [
    "decile_bounds = np.arange(0, 1.1, 0.1)\n",
    "deciles = mdf.weighted_quantile(df, 'c00100', 's006', decile_bounds)\n",
    "quintile_bounds = np.arange(0, 1.2, 0.2)\n",
    "quintiles = mdf.weighted_quantile(df, 'c00100', 's006', quintile_bounds)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "subject-virgin",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_group = mdf.weighted_mean(df, 'tax_change', w='s006', groupby=['nu18'])\n",
    "df_group.plot.bar()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "empty-parameter",
   "metadata": {},
   "outputs": [],
   "source": [
    "from cycler import cycler\n",
    "from cycler import cycler\n",
    "N = 7\n",
    "plt.rcParams[\"axes.prop_cycle\"] = plt.cycler(\"color\", plt.cm.Blues(np.linspace(0.2,1,N)))\n",
    "\n",
    "\n",
    "df['quintiles'] = pd.cut(df['expanded_income'], bins=quintiles, labels=np.arange(1,6).tolist())\n",
    "df_group = pd.DataFrame(mdf.weighted_mean(df, 'tax_change', w='s006', groupby=['nu18', 'quintiles']))\n",
    "# df_group.plot.bar()\n",
    "df_group.reset_index(inplace=True)\n",
    "df_group.rename(columns={0: 'Average Change in Taxes', 'nu18': 'Number of Children', 'quintiles': 'Income Quintiles'}, inplace=True)\n",
    "df_group['Number of Children'] = df_group['Number of Children'].astype('int64')\n",
    "ax1 = df_group.pivot(\"Income Quintiles\", \"Number of Children\", \"Average Change in Taxes\").plot(kind='bar', alpha=1.0)\n",
    "plt.ylabel('Mean Tax Change')\n",
    "plt.xticks(rotation = 0)\n",
    "plt.legend(title='Number of Children', loc=9, bbox_to_anchor=(0.5, -0.15), ncol=3)\n",
    "plt.savefig('bar_plot_by_kids.png', bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "large-apache",
   "metadata": {},
   "outputs": [],
   "source": [
    "df.drop(df[df['expanded_income'] < 0].index, inplace=True)\n",
    "percentile_bounds = np.arange(0, 1.01, 0.01)\n",
    "percentiles = mdf.weighted_quantile(df, 'expanded_income', 's006', percentile_bounds)\n",
    "df['Percentile'] = pd.cut(df['expanded_income'], bins=percentiles, labels=np.arange(1,101).tolist())\n",
    "df_group = pd.DataFrame(mdf.weighted_mean(\n",
    "    df, ['aftertax_income', 'aftertax_income_ARPA'], w='s006', groupby='Percentile'))\n",
    "df_group.reset_index(inplace=True)\n",
    "df_group.plot(kind='line', alpha=1.0)\n",
    "plt.ylim([0, 50000])\n",
    "# plt.ylabel('Mean Tax Change')\n",
    "# plt.xticks(rotation = 0)\n",
    "# plt.legend(title='Number of Children', loc=9, bbox_to_anchor=(0.5, -0.15), ncol=3)\n",
    "# plt.savefig('bar_plot_by_kids.png', bbox_inches='tight')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "electrical-offering",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_group"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "square-following",
   "metadata": {},
   "outputs": [],
   "source": [
    "decile_bounds = np.arange(0, 1.1, 0.1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "german-squad",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "connected-cassette",
   "metadata": {},
   "outputs": [],
   "source": [
    "pct_bounds = np.array([0, 0.2, 0.4, 0.6, 0.8, 0.9, 0.99, 1.0])\n",
    "percentiles = mdf.weighted_quantile(df, 'cexpanded_income', 's006', pct_bounds)\n",
    "df['Percentile'] = pd.cut(df['expanded_income'], bins=percentiles)\n",
    "df_group = pd.DataFrame(mdf.weighted_mean(\n",
    "    df, ['aftertax_income', 'aftertax_income_ARPA'], w='s006', groupby='Percentile'))\n",
    "df_group.reset_index(inplace=True)\n",
    "df_group.plot(kind='line', alpha=1.0)\n",
    "# plt.ylim([0, 50000])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "abroad-funeral",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}