{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "565b601d",
   "metadata": {},
   "source": [
    "# Where Your Degree Takes You — 3. Geography, Affordability & the Debt Paradox\n",
    "\n",
    "Where a degree's jobs are, whether the pay covers the rent, and the headline finding: **debt and earnings are barely correlated across majors.**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "39ccd77a",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-19T18:49:48.679279Z",
     "iopub.status.busy": "2026-06-19T18:49:48.679279Z",
     "iopub.status.idle": "2026-06-19T18:49:49.267089Z",
     "shell.execute_reply": "2026-06-19T18:49:49.267089Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Reading committed data from: C:\\Users\\jelbe\\palavir-estate-audit\\repos\\data-portfolio\\public\\data\\degree\n"
     ]
    }
   ],
   "source": [
    "import json, os\n",
    "import numpy as np, pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "plt.style.use(\"dark_background\")\n",
    "plt.rcParams.update({\"figure.facecolor\": \"#0a0a0a\", \"axes.facecolor\": \"#0a0a0a\",\n",
    "                     \"savefig.facecolor\": \"#0a0a0a\", \"axes.edgecolor\": \"#444\",\n",
    "                     \"font.size\": 11, \"axes.grid\": True, \"grid.alpha\": 0.15})\n",
    "PURPLE, GREEN, ROSE = \"#a855f7\", \"#10b981\", \"#f43f5e\"\n",
    "DATA = os.environ.get(\"DEGREE_DATA\", os.path.join(\"..\", \"public\", \"data\", \"degree\"))\n",
    "load = lambda n: json.load(open(os.path.join(DATA, n), encoding=\"utf-8\"))\n",
    "print(\"Reading committed data from:\", os.path.abspath(DATA))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "42269ee6",
   "metadata": {},
   "source": [
    "## The debt paradox\n",
    "\n",
    "Across Bachelor's majors, does more debt buy more pay? We recompute the Pearson correlations from the major landscape."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "e54beeab",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-19T18:49:49.267089Z",
     "iopub.status.busy": "2026-06-19T18:49:49.267089Z",
     "iopub.status.idle": "2026-06-19T18:49:49.285644Z",
     "shell.execute_reply": "2026-06-19T18:49:49.285644Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "debt  vs earnings : r = +0.056\n",
      "AI    vs earnings : r = +0.181\n",
      "growth vs earnings: r = -0.128\n",
      "stored: {'debt_vs_earnings': 0.056, 'ai_exposure_vs_earnings': 0.181, 'growth_vs_earnings': -0.128, 'notes': \"Across Bachelor's-level majors. Pearson r. Observational.\"}\n"
     ]
    }
   ],
   "source": [
    "L = pd.DataFrame(load('major_landscape.json')['majors'])\n",
    "def r(a,b):\n",
    "    m = L[[a,b]].dropna(); return np.corrcoef(m[a], m[b])[0,1]\n",
    "print(f\"debt  vs earnings : r = {r('debt','earn_5yr'):+.3f}\")\n",
    "print(f\"AI    vs earnings : r = {r('ai_beta','earn_5yr'):+.3f}\")\n",
    "print(f\"growth vs earnings: r = {r('growth_pct','earn_5yr'):+.3f}\")\n",
    "print('stored:', load('national_overview.json')['correlations'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "3fb96844",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-19T18:49:49.287941Z",
     "iopub.status.busy": "2026-06-19T18:49:49.285644Z",
     "iopub.status.idle": "2026-06-19T18:49:49.410937Z",
     "shell.execute_reply": "2026-06-19T18:49:49.410937Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(8,5))\n",
    "m = L.dropna(subset=['debt','earn_5yr'])\n",
    "ax.scatter(m['debt'], m['earn_5yr'], s=24, c=PURPLE, alpha=0.6)\n",
    "ax.set_xlabel('Median debt ($)'); ax.set_ylabel('Median 5-yr earnings ($)')\n",
    "ax.set_title(f\"Debt buys little earnings (r = {r('debt','earn_5yr'):.2f})\")\n",
    "plt.tight_layout(); plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8d12bbed",
   "metadata": {},
   "source": [
    "An almost flat cloud (r ≈ 0.05): taking on more debt does **not** systematically buy more earning power. The lever that matters is *field*, not *spend*.\n",
    "\n",
    "## Where the jobs cluster\n",
    "\n",
    "From BLS OEWS we compute each state's **location quotient** — how over-represented a degree's occupations are vs. the national average (after dropping generic catch-all occupations like 'Managers' that the crosswalk attaches to every field). Petroleum Engineering is the classic test:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "8108a2f4",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-19T18:49:49.413276Z",
     "iopub.status.busy": "2026-06-19T18:49:49.413276Z",
     "iopub.status.idle": "2026-06-19T18:49:49.422641Z",
     "shell.execute_reply": "2026-06-19T18:49:49.422641Z"
    }
   },
   "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>state</th>\n",
       "      <th>concentration_x</th>\n",
       "      <th>jobs_share_%</th>\n",
       "      <th>wage</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Alaska</td>\n",
       "      <td>10.00</td>\n",
       "      <td>1.59</td>\n",
       "      <td>205380</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Oklahoma</td>\n",
       "      <td>8.30</td>\n",
       "      <td>7.45</td>\n",
       "      <td>155659</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Texas</td>\n",
       "      <td>7.01</td>\n",
       "      <td>65.34</td>\n",
       "      <td>157011</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Colorado</td>\n",
       "      <td>3.83</td>\n",
       "      <td>7.38</td>\n",
       "      <td>168740</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Louisiana</td>\n",
       "      <td>3.47</td>\n",
       "      <td>4.34</td>\n",
       "      <td>145098</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Utah</td>\n",
       "      <td>1.24</td>\n",
       "      <td>1.45</td>\n",
       "      <td>169820</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>California</td>\n",
       "      <td>0.54</td>\n",
       "      <td>7.02</td>\n",
       "      <td>136361</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Pennsylvania</td>\n",
       "      <td>0.41</td>\n",
       "      <td>1.88</td>\n",
       "      <td>103618</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          state  concentration_x  jobs_share_%    wage\n",
       "0        Alaska            10.00          1.59  205380\n",
       "1      Oklahoma             8.30          7.45  155659\n",
       "2         Texas             7.01         65.34  157011\n",
       "3      Colorado             3.83          7.38  168740\n",
       "4     Louisiana             3.47          4.34  145098\n",
       "5          Utah             1.24          1.45  169820\n",
       "6    California             0.54          7.02  136361\n",
       "7  Pennsylvania             0.41          1.88  103618"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pet = load('by_cip_geo/1425.json')  # Petroleum Engineering\n",
    "top = sorted([(k,v) for k,v in pet.items() if v.get('concentration')],\n",
    "             key=lambda kv:-kv[1]['concentration'])[:8]\n",
    "pd.DataFrame([{'state': v['name'], 'concentration_x': v['concentration'],\n",
    "               'jobs_share_%': v.get('jobs_share'), 'wage': v.get('wage')} for _,v in top])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0fcb45d8",
   "metadata": {},
   "source": [
    "Oklahoma, Texas, Alaska, Colorado, Louisiana — oil country, exactly right.\n",
    "\n",
    "## Can the paycheck cover the rent?\n",
    "\n",
    "We pit each major's graduate-weighted metro pay (OEWS) against market rent (Zillow), as a share of income. The classic guidance: **keep rent under 30%**. For Computer Science, how many metros clear that line?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "70018efb",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-19T18:49:49.424685Z",
     "iopub.status.busy": "2026-06-19T18:49:49.424685Z",
     "iopub.status.idle": "2026-06-19T18:49:49.436746Z",
     "shell.execute_reply": "2026-06-19T18:49:49.436746Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "368 of 370 metros keep CS rent under 30%\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>metro</th>\n",
       "      <th>wage</th>\n",
       "      <th>rent</th>\n",
       "      <th>burden_%</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>222</th>\n",
       "      <td>Weirton, WV</td>\n",
       "      <td>110790</td>\n",
       "      <td>842</td>\n",
       "      <td>9.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>209</th>\n",
       "      <td>Decatur, IL</td>\n",
       "      <td>110658</td>\n",
       "      <td>928</td>\n",
       "      <td>10.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>221</th>\n",
       "      <td>Johnstown, PA</td>\n",
       "      <td>98542</td>\n",
       "      <td>873</td>\n",
       "      <td>10.6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>273</th>\n",
       "      <td>Wheeling, WV</td>\n",
       "      <td>105787</td>\n",
       "      <td>938</td>\n",
       "      <td>10.6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>178</th>\n",
       "      <td>Ames, IA</td>\n",
       "      <td>114533</td>\n",
       "      <td>1024</td>\n",
       "      <td>10.7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>340</th>\n",
       "      <td>Muncie, IN</td>\n",
       "      <td>103308</td>\n",
       "      <td>961</td>\n",
       "      <td>11.2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             metro    wage  rent  burden_%\n",
       "222    Weirton, WV  110790   842       9.1\n",
       "209    Decatur, IL  110658   928      10.1\n",
       "221  Johnstown, PA   98542   873      10.6\n",
       "273   Wheeling, WV  105787   938      10.6\n",
       "178       Ames, IA  114533  1024      10.7\n",
       "340     Muncie, IN  103308   961      11.2"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "metros = {m['cbsa']: m for m in load('affordability_metros.json')['metros']}\n",
    "wage = load('by_cip_afford/1107.json')  # Computer Science\n",
    "rows = []\n",
    "for cbsa, w in wage.items():\n",
    "    m = metros.get(int(cbsa))\n",
    "    if m: rows.append({'metro': m['name'], 'wage': w, 'rent': m['zori_monthly'],\n",
    "                       'burden_%': round(m['zori_monthly']*12/w*100,1)})\n",
    "aff = pd.DataFrame(rows).sort_values('burden_%')\n",
    "print(f\"{(aff['burden_%']<=30).sum()} of {len(aff)} metros keep CS rent under 30%\")\n",
    "aff.head(6)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "340ea607",
   "metadata": {},
   "source": [
    "*Rent data: Data Provided by Zillow Group. OEWS covers fewer metros than Zillow, so unmatched metros are omitted — never estimated.*\n",
    "\n",
    "---\n",
    "\n",
    "### Sources & reproducibility\n",
    "Every figure above is recomputed from the committed `public/data/degree/*.json`, which the pipeline in `scripts/degree/` derives from College Scorecard, BLS OEWS, NCES CIP→SOC, O\\*NET, the Eloundou & AIOE AI-exposure datasets, and Zillow ZORI. See `sources.json` for full citations and licenses."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.11.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}
