Financial Health

Margins, reserves, and revenue mix among the region’s larger nonprofits

Key finding The region’s larger nonprofits are, in aggregate, modestly healthy but thinly buffered: among 502 organizations with current Form 990 filings, the median operating margin is +5.0% and median reserves equal 10.6 months of expenses. Underneath those medians, one in three ran a deficit in its latest fiscal year and one in five holds less than three months of operating reserves — the level commonly considered the minimum safe cushion. Human services organizations hold thinner reserves: 27% sit below the three-month line.

GU = ({
  navy: "#002967",
  red: "#c8102e",
  s1: "#1c5cab",
  s2: "#1baf7a",
  s3: "#c8102e",
  s4: "#eda100",
  ink: "#1a1a1a",
  muted: "#898781",
  grid: "#e1e0d9",
  seq: ["#cde2fb", "#9ec5f4", "#6da7ec", "#3987e5", "#256abf", "#184f95", "#0d366b"]
})

fnum = d3.format(",")

fmoney = (x) => x == null || isNaN(x) ? "—"
  : Math.abs(x) >= 1e9 ? "$" + d3.format(".2f")(x / 1e9) + "B"
  : Math.abs(x) >= 1e6 ? "$" + d3.format(".1f")(x / 1e6) + "M"
  : "$" + d3.format(",.0f")(x)

fpct = (x) => x == null || isNaN(x) ? "—" : d3.format(".1f")(x) + "%"

fin = FileAttachment("data/fin_metrics.csv").csv({typed: true})

finByYear = FileAttachment("data/fin_by_year.csv").csv({typed: true})

Who this page covers

Financial-health analysis requires full Form 990 data, so this page covers active regional organizations with at least $500,000 in revenue (plus the 25 largest by assets) that have a 990 or 990-EZ filing from 2021 or later — 502 organizations. Metrics come from each organization’s most recent filing. Definitions and caveats (including why a true “overhead ratio” cannot be computed from extract data) are on the Methodology page.

Revenue distribution

Nonprofit revenue spans five orders of magnitude, so the axis is logarithmic. Each dot is one organization.

Plot.plot({
  height: 340,
  width: 900,
  marginLeft: 200,
  marginBottom: 45,
  style: {fontSize: "13px"},
  x: {type: "log", label: "Total revenue, most recent filing ($, log scale)", labelAnchor: "center", grid: true, tickFormat: d => d3.format("~s")(d).replace("G", "B")},
  y: {label: null},
  marks: [
    Plot.dot(fin.filter(d => d.totrevenue > 1000), {
      x: "totrevenue",
      y: "subsector_name",
      fill: GU.s1,
      fillOpacity: 0.45,
      r: 3.5,
      tip: true,
      title: d => `${d.name}\n${d.city} · FY${d.tax_prd_yr}\nRevenue: ${fmoney(d.totrevenue)}`
    }),
    Plot.tickX(
      d3.rollups(fin.filter(d => d.totrevenue > 1000), v => d3.median(v, d => d.totrevenue), d => d.subsector_name)
        .map(([subsector_name, med]) => ({subsector_name, med})),
      {x: "med", y: "subsector_name", stroke: GU.red, strokeWidth: 3}
    )
  ]
})

md`<span style="color:${GU.muted}; font-size:0.85rem;"><span style="color:${GU.s1}">●</span> organization &nbsp;&nbsp;<span style="color:${GU.red}">|</span> category median. Hover any dot for the organization.</span>`

Operating margins

Share of latest-year revenue left after expenses. Margins cluster near break-even by design — nonprofits price to mission, not profit — but persistent deficits erode reserves.

marginData = fin.filter(d => d.margin != null && isFinite(d.margin) && d.margin > -1 && d.margin < 1)

Plot.plot({
  height: 320,
  width: 900,
  marginLeft: 55,
  marginBottom: 45,
  style: {fontSize: "13px"},
  x: {label: "Operating margin (latest fiscal year)", labelAnchor: "center", tickFormat: ".0%", domain: [-0.6, 0.6]},
  y: {label: "Organizations", grid: true},
  marks: [
    Plot.rectY(marginData, Plot.binX(
      {y: "count"},
      {
        x: "margin",
        interval: 0.05,
        fill: d => d.margin < 0 ? GU.red : GU.s1,
        insetLeft: 1,
        insetRight: 1,
        rx: 2,
        tip: true
      }
    )),
    Plot.ruleX([0], {stroke: GU.ink, strokeWidth: 1.5}),
    Plot.ruleY([0], {stroke: "#c3c2b7"})
  ]
})

md`<span style="color:${GU.muted}; font-size:0.85rem;"><span style="color:${GU.red}">&#9632;</span> deficit (${fpct(d3.mean(marginData, d => d.margin < 0) * 100)} of organizations) &nbsp;&nbsp;<span style="color:${GU.s1}">&#9632;</span> surplus. Organizations with margins beyond ±60% (24, mostly foundations with lumpy grant income) fall outside the displayed range.</span>`

Operating reserves

Months of expenses covered by net assets — the sector’s standard solvency gauge. Three months is the commonly cited minimum; the red band marks organizations below it.

reserveData = fin.filter(d => d.months_reserves != null && isFinite(d.months_reserves))

reserveBuckets = {
  const buckets = [
    {label: "Negative", test: d => d.months_reserves < 0, atRisk: true},
    {label: "0–3 months", test: d => d.months_reserves >= 0 && d.months_reserves < 3, atRisk: true},
    {label: "3–6 months", test: d => d.months_reserves >= 3 && d.months_reserves < 6, atRisk: false},
    {label: "6–12 months", test: d => d.months_reserves >= 6 && d.months_reserves < 12, atRisk: false},
    {label: "1–3 years", test: d => d.months_reserves >= 12 && d.months_reserves < 36, atRisk: false},
    {label: "3+ years", test: d => d.months_reserves >= 36, atRisk: false}
  ];
  return buckets.map(b => ({bucket: b.label, atRisk: b.atRisk, n: reserveData.filter(b.test).length}));
}

Plot.plot({
  height: 300,
  width: 900,
  marginLeft: 55,
  marginBottom: 45,
  style: {fontSize: "13px"},
  x: {label: "Net assets as months of expenses", labelAnchor: "center", domain: reserveBuckets.map(d => d.bucket)},
  y: {label: "Organizations", grid: true},
  marks: [
    Plot.barY(reserveBuckets, {
      x: "bucket",
      y: "n",
      fill: d => d.atRisk ? GU.red : GU.s1,
      rx: 3,
      insetLeft: 5,
      insetRight: 5,
      tip: true,
      title: d => `${d.bucket}: ${fnum(d.n)} organizations (${fpct(d.n / reserveData.length * 100)})`
    }),
    Plot.textY(reserveBuckets, {
      x: "bucket", y: "n", text: d => fnum(d.n), dy: -8, fill: GU.ink, fontSize: 12
    }),
    Plot.ruleY([0], {stroke: "#c3c2b7"})
  ]
})

md`<span style="color:${GU.muted}; font-size:0.85rem;"><span style="color:${GU.red}">&#9632;</span> below the three-month minimum cushion (${fpct(d3.mean(reserveData, d => d.months_reserves < 3) * 100)} of organizations).</span>`

Size vs. cushion

Do bigger organizations hold bigger buffers? Only weakly — thin reserves appear at every size. Deficit-running organizations (red) with low reserves (lower half of the chart) are the region’s financially fragile institutions.

scatterData = fin.filter(d =>
  d.totrevenue > 10000 && d.months_reserves != null && isFinite(d.months_reserves) &&
  d.months_reserves > -12 && d.months_reserves < 120 && d.margin != null
)

Plot.plot({
  height: 420,
  width: 900,
  marginLeft: 55,
  marginBottom: 45,
  style: {fontSize: "13px"},
  x: {type: "log", label: "Total revenue ($, log scale)", labelAnchor: "center", tickFormat: d => d3.format("~s")(d).replace("G", "B"), grid: true},
  y: {label: "Months of reserves", grid: true},
  color: {
    legend: true,
    domain: ["Surplus", "Deficit"],
    range: [GU.s1, GU.red]
  },
  marks: [
    Plot.ruleY([3], {stroke: GU.muted, strokeDasharray: "4,4"}),
    Plot.text([{x: 15000, y: 3}], {x: "x", y: "y", text: ["3-month minimum"], dy: -8, textAnchor: "start", fill: GU.muted, fontSize: 11}),
    Plot.dot(scatterData, {
      x: "totrevenue",
      y: "months_reserves",
      fill: d => d.margin < 0 ? "Deficit" : "Surplus",
      fillOpacity: 0.55,
      r: 4,
      tip: true,
      title: d => `${d.name}\n${d.city} · ${d.subsector_name} · FY${d.tax_prd_yr}\nRevenue: ${fmoney(d.totrevenue)}\nMargin: ${fpct(d.margin * 100)}\nReserves: ${d3.format(".1f")(d.months_reserves)} months`
    }),
    Plot.ruleY([0], {stroke: "#c3c2b7"})
  ]
})

Reserves above 10 years and below −1 year are clipped from the chart (mostly endowed foundations and a few organizations in wind-down). Hover any dot for the organization.

Revenue mix: earned vs. contributed

Contribution reliance — the share of revenue from donations and grants — shapes risk. Organizations funded mostly by contributions ride donor and grant cycles; fee-funded organizations ride demand.

mixData = fin.filter(d => d.contrib_share != null && d.contrib_share >= 0 && d.contrib_share <= 1)

Plot.plot({
  height: 300,
  width: 900,
  marginLeft: 55,
  marginBottom: 45,
  style: {fontSize: "13px"},
  x: {label: "Contributions & grants as share of revenue", labelAnchor: "center", tickFormat: ".0%"},
  y: {label: "Organizations", grid: true},
  marks: [
    Plot.rectY(mixData, Plot.binX(
      {y: "count"},
      {x: "contrib_share", interval: 0.1, fill: GU.s1, insetLeft: 1, insetRight: 1, rx: 2, tip: true}
    )),
    Plot.ruleY([0], {stroke: "#c3c2b7"})
  ]
})

The distribution is U-shaped: the region’s nonprofits are mostly either donation-funded charities (right) or fee/program-funded operations such as health systems, housing, and credit-counseling services (left), with fewer blended models between.

Financial detail: every covered organization

viewof finCounty = Inputs.select(
  ["All counties"].concat(Array.from(new Set(fin.map(d => d.county))).sort()),
  {label: "County"}
)

viewof finSearch = Inputs.search(
  fin.filter(d => finCounty === "All counties" || d.county === finCounty),
  {placeholder: "Search names…", columns: ["name", "city"]}
)

Inputs.table(
  finSearch
    .sort((a, b) => d3.descending(a.totrevenue, b.totrevenue))
    .map(d => ({
      Name: d.name,
      County: d.county,
      FY: String(d.tax_prd_yr),
      Revenue: fmoney(d.totrevenue),
      Expenses: fmoney(d.totfuncexpns),
      "Margin": d.margin != null ? fpct(d.margin * 100) : "—",
      "Months reserves": d.months_reserves != null && isFinite(d.months_reserves) ? d3.format(".1f")(d.months_reserves) : "—",
      "Contrib. share": d.contrib_share != null ? fpct(Math.min(Math.max(d.contrib_share, 0), 1) * 100) : "—"
    })),
  {rows: 15, layout: "auto"}
)

Source: IRS Form 990/990-EZ extracts via ProPublica Nonprofit Explorer API, latest filing per organization (2021 or later).