Who relies on the wireless network — and why the affordable option matters.
The households on the rural wireless network are disproportionately older, lower-income, and without an affordable wired alternative. They're the ones who would feel a pause first — and who can least afford to replace an affordable local connection with a satellite bill. This map lays the county's median household income next to the network so you can see who depends on it. See poverty by tract layered on the network in 3D →
Who is "the ratepayer"?
Before you spend the ratepayers' money, meet the ratepayer.
When the District talks about protecting "the ratepayers' money," it's worth asking who the ratepayer actually is. Not an abstraction — a specific, mostly rural, modest-income, hard-working household. Here is that household, drawn entirely from public data.
The median Okanogan County household
62% rural
live outside the incorporated towns — among the highest rural shares of any WA county
72% own
own their home; about 1 in 6 of those is a manufactured home
$63,207
median household income — just 64% of the Washington median
age 43
median age; nearly 1 in 4 residents is 65 or older
At home
72%own their home (28% rent)
1 in 6homes is a mobile / manufactured home
2.3people in the average household
Family & age
29%of households have kids under 18 — most don't
23%of residents are 65 or older
43median age (Washington's is 38)
What they earn
$63Kmedian household income — 64% of the state's
$36,920the median worker's yearly earnings
4 in 10households draw Social Security income
Getting by
44%of households can't afford the basics (below the ALICE survival budget)
~1 in 5lives below the poverty line; 19% are on SNAP
1 in 6with a credit record has debt in collections (typically $1,774)
Work & land
#1employer is agriculture — about 1 in 4 jobs (orchards, ranch, farm)
56%labor-force participation; 4.5% unemployment
Health & longevity
2.7 yrsshorter life expectancy than the WA average (76.1 vs 78.8)
2×as likely to be uninsured as the state (16% vs 7%)
22%report only poor or fair health (state: 15%)
Getting around
70%of households keep 2 or more vehicles
4%have no vehicle — in a county you can't cross without one
Online
1 in 11households has no internet at all (current ACS) — more than double the state rate; the county's 2023 plan data ran near 1 in 5
~13%have no wired/fixed broadband; 12% lean on cell data only
↳this is the exact gap the PUD wireless network fills
And what's left over? Almost nothing.
"Owns a home" doesn't mean a big mortgage payment. More than half of Okanogan owners — about 56% — own free and clear, paying a median of just $491 a month in taxes, insurance, and utilities. That paid-off home is how a lot of modest-income households get by at all. For everyone else, it's tighter:
Median monthly housing cost by tenure — ACS 2020–2024. A third of renters pay 30%+ of their income on housing.
The bare minimum costs more than the median household makes. A no-frills "survival budget" for a family of four here — housing, food, child care, transportation, health care, taxes, nothing extra — runs $70,092 a year (United For ALICE, 2024). The county's median household income is $63,207. Even the leaner ALICE survival threshold ($59,106) leaves the median household only about $340 a month above that bare-minimum bar — to cover all savings, debt, and emergencies. The survival budget itself sets aside $0 for any of them.
And retirement here is mostly Social Security. Four in ten households draw it, and households led by someone 65 or older live on a median of $50,625 — about a fifth below the county median. County-level retirement savings aren't published anywhere — but at margins this thin, there isn't much room to build them.
It's not what they earn. It's what's left.
When a bare-minimum survival budget takes the whole paycheck, the number that decides everything isn't income — it's the margin. And here the margin is almost gone: after the basics, the median household has about $340 a month left, and 44% of households have nothing left at all.
So a price increase doesn't come out of "income" — it comes out of that sliver, or out of nothing. The roughly $45 a month more a household pays under a monopoly satellite than under the local network lands against that ~$340 cushion — and for the 44% below the line, against nothing. When there's nothing left over, $10 a month is real money.
That's why the cheapest, locally-owned way online isn't a luxury in this county — for most of it, it's the difference between getting by and not. Protecting the ratepayer and keeping this network are the same thing.
Sources: U.S. Census Bureau, American Community Survey 2020–2024 5-year estimates (tables B19013, B25003, B25024, B25044, B25010, B11005, B22003, B20002, B19055, B19049, B23025, C24050, S2801, B25081, B25088, B25064) and 2019–2023 5-year (DP05 age/race); United For ALICE, Washington 2026 Report (2024 data); Urban Institute, Debt in America (2025 credit-bureau data — "debt in collections" is measured among residents with a credit record); County Health Rankings & Roadmaps 2025 (life expectancy 2020–2022; uninsured & health status 2022); Washington OFM (rural share). Okanogan County, WA. Figures rounded; margins of error apply to county estimates.
A fair question — not an accusation
Two worthy projects, and an honest question about funding both.
First, to be clear: we support the District's fiber work, the Conconully build included. Conconully is a small, isolated, unserved community — fiber there is a worthwhile investment, and we're glad it's happening. This was never fiber versus wireless.
Budgeted wireless reinvestment (bars) vs. subscribers (line), 2021–2026 — District capital budgets (public-records request) and monthly billing.
The honest question is about the other side of the ledger. The District is investing about $4.3 million in fiber to Conconully (367 homes), while the ~$1.2 million wireless upgrade — which would modernize the network serving ~2,700 households — is not in the 2026 budget. Budgeted reinvestment in that wireless network had already fallen about 92% (chart above), even as the segment banked a surplus every year — so the upgrade is affordable from the network's own earnings.
So we're not asking the Board to choose fiber or wireless, and we're not alleging anyone did anything improper. We're asking one question, in the open: the wireless network reaches the county's lower-income rural majority and pays its own way — what would it take to fund its upgrade too? At that same meeting the Board agreed to take it up in a workshop with the local providers. We're asking it to follow through → · Follow the money →
The pattern
Fiber follows the towns. Wireless covers everywhere else.
Start with where people actually live. About 3 in 5 Okanogan County residents — roughly 27,000 of ~43,000 — live in the unincorporated rural areas, not the incorporated towns (Washington OFM, 2024). And none of the towns is a growth engine: the largest, Omak, edged up about 3.6% since 2020, while Tonasket, Oroville, and Conconully stayed essentially flat (each within about 1%) from 2020 to 2024. So the wireless network isn't serving a shrinking fringe — it serves the majority of the county.
The PUD's fiber backbone and town distribution serve the seven incorporated towns, and the newest funded project (the Conconully middle-mile and fiber-to-the-home build) extends that reach further. That investment is real and worthwhile. But fiber is expensive to run to scattered rural homes — which is exactly why the wireless network exists: it blankets the valley floor and the hills between the towns, where running fiber to each property pencils out poorly.
Those in-between areas also tend to be the county's lower-income census tracts. Across the county, tract median household income runs from about $37,800 to $59,600 (ACS 2015–2019); the county-wide median is roughly $63,000 today. The households most dependent on PUD wireless are, on the whole, the ones with the least room in the monthly budget.
And the largest new fiber money is now being handed back. The District's $30.2 million federal ReConnect grant for the northeastern communities — Chesaw, Molson, Havillah, Nine Mile — was awarded in 2023 with a spring-2024 start, then sat unspent ($0 by the end of 2024). The District has now moved to rescind it — Resolution 1863, "Rescind RUS ReConnect Award WA 1707-A73," is on the June 22, 2026 board agenda (pending the final vote / posted minutes). It's the clearest sign of the real pattern: fiber to the dispersed rural sprawl is the part the District can't get out the door — while the wireless network already serves those same areas today, and that's the platform now being cut. (See it on the map below, struck through in red.)
~62%
of county residents live in the rural areas wireless serves (WA OFM, 2024)
$37.8K
lowest tract median household income (ACS)
~$63K
county-wide median household income (2024)
2–4%
of a $38K household's income to replace service with residential Starlink
$3K–$18K
per year for a farm or small business on Starlink Business
Explore the overlay
Income vs. PUD fiber investment
Hover a tract for its median household income. Toggle the fiber layers from the control on the map, and click a fiber line or project area for details. Darker tracts are higher-income; the rust and navy lines show existing and funded fiber. Two optional dot layers — “Approximate served area — fiber/wireless” — are off by default; switch them on from the map control for a rough reference of where fiber reaches versus where homes rely on wireless. They reflect just one retail provider's footprint (not the whole county), are offset for privacy, and don't represent customers or specific addresses.
Income: U.S. Census Bureau ACS 2015–2019 5-year (median household income, DP03). Boundaries: Census TIGER. Fiber layers are approximate/schematic, compiled from Okanogan PUD announcements — verify exact routes at okpudfiber.org. A fuller view of wireless coverage is on the coverage map.
Reach per dollar
Both builds are worthwhile — and the wireless upgrade stretches much further per dollar.
This isn't a knock on Conconully — fiber to an isolated, fire-exposed community is worth doing. It's simply a reach-per-dollar comparison the Board can weigh as it decides what to fund next. The two are different products and largely different money, but the contrast is worth seeing.
What each project actually connects
Cost per home — fiber: $4.3M ÷ 367 ≈ $11,700; wireless: $1.18M ÷ 878 ≈ $1,340.
Honest caveats: the $4.3M includes ~22 miles of middle-mile backbone that will serve future homes too, so the per-home cost falls as more connect; and fiber is a faster, longer-life product. The point is reach per dollar today. (Wireless: the $1.18M upgrade covers seven towers serving 878 customers — part of a network already serving ~2,700; the 367 are new fiber connections.) Full cost & funding breakdown →
The point isn't which homes by income — Conconully itself is a small, modest-income town, not a wealthy enclave. The point is simply that a self-funded wireless upgrade would modernize the service for thousands at a fraction of the per-home cost, even as the District takes on a major fiber commitment elsewhere. Both can be true; both can be funded.
The county's own Digital Equity plan says the same. The Broadband Action Team's June 2023 Broadband & Digital Equity Action Plan is built around affordability and the rural digital divide — and its own list of obstacles names the "high cost of construction of fiber networks in low population density areas." It also shows why the public network matters for affordability: the District's network lists residential fiber at $35/mo, with its wireless tiers cheaper still, against commercial plans at $80–$120/mo and satellite higher again. For a county the plan calls "historically and economically distressed," the cheapest, furthest-reaching layer is the one that matters most — and it's the one being cut. (Okanogan County / CCT Broadband Action Team, June 14, 2023 — read it here.)
The need, neighborhood by neighborhood
Where the county is least connected.
This is the data behind the county's own 2023 broadband map, made interactive. Each census tract is shaded by the share of households with no internet at all — countywide, about 1 in 5 (≈20%) in the plan's tract data (ACS 2015–2019; the newest county-wide ACS runs near 1 in 11), and far higher in the rural and reservation tracts. Toggle the other measures to see how speed, computer access, and poverty line up the same way.
Shade tracts by
Tract data: NoaNet / Washington broadband planning layer — ACS adoption & poverty and Ookla median speeds, the dataset behind the 2023 county plan's map (tap a tract for details). Tract estimates; actual coverage and speeds vary.
The deepest gap
The county's least-connected ground is the Colville Reservation — and its government is the one proving the case.
The map above makes it plain: the two reservation census tracts are among the worst-served in the county, with roughly 3 in 10 households having no internet at all. That isn't our characterization — the 2023 broadband plan, which the Confederated Tribes of the Colville Reservation (CCT) co-authored, documents it directly.
In the plan's own words, "the Colville Reservation is one of the areas that lack broadband services," the southeastern county where the reservation sits "does not show many ISPs," and that area "is poorer than the rest of the county… fewer people using the internet, and fewer people that own a computer." It names "a large tribal community that has historically not been served." The reservation — which spans Okanogan and Ferry counties and is among the largest in Washington — is where the private market thins out the most.
This isn't a county-versus-Tribes story; it's a partnership. Since 2019 the county and the Tribes have run a joint Broadband Action Team — the body that produced the 2023 plan, on the premise that "broadband knows no boundaries." And the plan credits the public builders together: the infrastructure "provided by the Okanogan PUD District No. 1, Electric co-ops, and the Colville Tribe" is what the smaller local ISPs lean on.
Here's the part that should settle the technology debate. The Tribes are investing a $48.4 million federal award in their own network — and it is a deliberate fiber-and-wireless hybrid: fiber where it reaches, fixed wireless where it can't, built expressly for telehealth and distance learning. A neighboring government, planning for the same rural terrain, chose wireless as part of the answer — at the very moment the PUD is treating its own wireless as expendable. (NTIA Tribal Broadband Connectivity Program; the Tribes also hold a $19.1M USDA ReConnect award in Okanogan County. The TBCP project serves the reservation across Okanogan and Ferry counties.)
A pause doesn't fall evenly. The lowest-income, most rural, and reservation ground is exactly where the wireless layer is the only affordable option — so it's where a pause would be felt first. Reservation residents sit squarely inside the federal Digital Equity Act's "covered populations" — as rural residents, as a racial or ethnic minority, and as low-income households. Letting the public wireless layer age out falls first, and hardest, exactly where the map is darkest.
The households with the least are the ones a public network reaches when no one else will. That is the whole reason a public utility exists — and it is the ground most exposed the moment it steps back.
The fiber affordability gap
Even where fiber exists, almost no one can afford the drop.
The PUD doesn't pay for the fiber drop to a home — the customer does, at about $1 per foot from the nearest splice point, with a $500 minimum. In a county where homes sit far apart, a household even a mile from the line faces a $5,000+ bill just to get connected. So fiber reaches the people who can afford the install — and in the lower-income, spread-out areas, almost everyone else stays on wireless. One local provider's service records show it plainly:
15%
of one provider's 274 active connections are on fiber
85%
connect by wireless instead
$1/ft
customer-paid fiber drop, $500 minimum
12–17%
fiber adoption even in Tonasket & Oroville, where fiber is being built
Fiber vs. wireless, by community
On fiberOn wireless
Tonasket
7 fiber / 57 total (12%)
Pateros
12 fiber / 54 total (22%)
Oroville
8 fiber / 48 total (17%)
Brewster
3 fiber / 41 total (7%)
Loomis
5 fiber / 34 total (15%)
Omak
3 fiber / 17 total (18%)
Okanogan
1 fiber / 10 total (10%)
Source: Will Connect active service records (a PUD wireless retail provider), 274 active connections. In the smallest communities — Riverside, Malott, Conconully, Carlton — fiber adoption is effectively zero; nearly every connection is wireless. Cutting wireless investment removes the one option these households can actually afford.
Where fiber is built, adoption still stays low — because the drop is the customer's to pay. Even in Tonasket and Oroville, where fiber is going in, only about 12–17% of homes take it. At ~$1 a foot with a $500 minimum, the install cost is the barrier — so in the spread-out, lower-income areas, nearly everyone stays on wireless. Toggle the Approximate served area — fiber (navy) and wireless (orange) layers on the map above to see it.
And the drop is the second cost barrier — there's a bigger one before it. Even the expensive customer drop only matters where fiber gets built in the first place. Across Okanogan's terrain, building fiber runs about 9× more per home than on flat ground (~$40,000 vs. ~$4,300) — so for the dispersed, lower-income rural majority, fiber may never economically arrive at all. That's the equity heart of it: “just wait for fiber” isn't a plan for these households — it's leaving them offline. The affordable wireless network is the equitable option precisely because it's the one that actually reaches them. Why fiber costs ~9× more here →
What this network actually provides
For these households, broadband was never about speed. It's about service.
The decision in front of the Board frames this as a question of megabits and subscriber counts. But for the people this network serves — disproportionately rural, lower-income, and older (about 1 in 4 residents is 65+, vs ~1 in 6 nationwide — Census) — getting online was never mainly about raw speed. It's about service: a local provider who answers the phone, drives out to the property, and helps a household actually use what the connection can do. Assistance, context, and a relationship — delivered affordably, up front, with no premium. That human layer is what brings a rural community online and keeps it there, and it's exactly what the local retail providers on this open-access network provide.
The "competitive marketplace" the District names as a replacement doesn't offer that — and the national record shows the commercial market often does the opposite to communities like this one:
9 of 10
cities where the worst broadband deals — the same monthly price for far slower service — landed in lower-income neighborhoods (The Markup, 2022)
~5×
the share of income lower-income communities pay for broadband — 2.43% vs 0.51% — more money for worse service (Federal Reserve Bank of NY, 2025)
77%
of households on a low-income internet subsidy said they'd drop or downgrade service when it ended — they cut back, they don't trade up (FCC, 2024)
That's the pattern — clearest in the city data above, but starker in the country: the private market under-serves and over-charges exactly the rural, lower-income households this network was built for — and where it can't turn a profit, it doesn't show up at all. The federal government's own watchdog puts it plainly: "the market does not support private broadband investment in low-density areas" (GAO) — which is the whole reason a public utility exists to do it. A replacement built on out-of-area satellite and national carriers doesn't serve these communities; it extracts from them. A public, open-access network with local providers is the answer to that — the model that serves and educates these households instead of billing them more for less.
And we know what happens when the affordable option disappears. When the federal Affordable Connectivity Program — a $30/month subsidy reaching 23 million households — ran out of funding in 2024, the FCC's own survey found more than three-quarters would have to drop or change their service, and low-income families reported cutting back on other essentials to keep the connection. People at this income level don't trade up to something pricier — they cut back, and many go without.
Sources: The Markup, "Dollars to Megabits" (Oct 2022), analysis of more than 800,000 service offers across 38 cities; Federal Reserve Bank of New York (Sept 2025); U.S. Government Accountability Office (GAO-22-104611); FCC Affordable Connectivity Program Consumer Survey (Dec 2023). These are national patterns describing disparate outcomes, not a finding of intent against any one provider.
Why it matters
When the affordable option fades, the lowest-income households feel it first.
When a public network is allowed to fall behind, the people who feel it first are those who can't simply buy their way out. A residential Starlink subscription runs $660–$1,560 a year before equipment — for a household earning $38,000, that's roughly 2–4% of gross income just to stay online. For a farm or small business, the Business plans run $3,000–$18,000+ a year. Every one of those dollars leaves the county.
The District's own broadband manager has made the case that rural broadband needs public stewardship precisely because ordinary private-market economics fall short out here — the same logic that justified the $30 million USDA ReConnect grant the District accepted in 2023 — the grant it is now moving to give back, unspent (Res. 1863, June 2026). That logic doesn't stop at the edge of the fiber footprint. If anything, the rescission proves it: when fiber to the rural sprawl proved too costly to build, the wireless platform is what still reaches those addresses — and it's the one being cut.
None of this means fiber investment is wrong, or that the Board must reach any particular outcome. It means the equity stakes belong in the open — weighed in public, by the elected commissioners, alongside the financial case, as part of revisiting the wireless decision.
