No prudent utility calls a hedge a waste — until, apparently, it's broadband.

Public power was never handed to Okanogan. It was won — slowly, and against the odds.

The right to build a public utility didn't come from the legislature — it came from the people, over the objections of the private power industry. In 1930, a Grange-led citizens' initiative authorized Washington's public utility districts at the ballot box, passing 152,487 to 130,901 against a far better-funded opposition that fed prewritten stories, editorials, and cartoons to newspapers across the state. But authorizing a PUD and having one were a decade apart — and Okanogan's own took years of court fights to make real.

Nobody handed the county this network — not the power lines won in 1945, and not the broadband towers built on the same public-stewardship bet decades later. It was voted for, litigated for, and paid for, by people who took a real risk that it might not work. What takes years to win can be given away in a single budget cycle — which is exactly what a 25-year broadband legacy is now up against.

You don't have to predict the future to keep a hedge that already pays you to hold it.

Start with what isn't in dispute: the upgrade is a one-time ~$1.2 million, and the wireless line earns money — the telecom segment posted an audited $808K surplus in 2024, and wireless's share of that surplus, allocated by its third of the revenue, is roughly $270,000 a year. A hedge whose premium is negative — one that pays a dividend the whole time you hold it — is one a prudent utility keeps regardless of what comes next, because keeping it never depended on guessing the future right. That, by itself, settles the decision.

The dollars below don't carry that argument — they just picture the downside the hedge also covers. The network reaches about 13,400 households (~32,000 residents at ~2.4 per home — see the map). If the local option goes and the monopoly fallback does what monopolies do, here is the rough scale of a modest increase on a typical ~$80/month bill — an illustration, not a forecast:

In that picture, a one-time ~$1.2 million upgrade sits to the left of every bar but the smallest: a 10% increase would run on the order of ~$1.3 million a year, a 20% increase ~$2.6 million — each, in a single year, larger than the whole upgrade the District called unaffordable. Over the gear's ten-year life that range is $13–26 million, and the share paid to out-of-county providers leaves Okanogan and never comes back. None of those figures has to be exact for the shape to hold: the cost of losing the option dwarfs the cost of keeping it — and the option earns money while you keep it.

The assumptions, in the open: ~13,400 covered households, a ~$80 average bill, the increase applied across them — an illustration, not a forecast (not every bill rises the same). And the two figures are different scopes, on purpose: the $1.2M is the District's first-phase upgrade — seven towers serving 878 customers, per its own Tarana workbooks (see The Case) — not the price of rebuilding the whole network. The price protection, by contrast, is the network's: it accrues to the ~13,400 households in its footprint, because a local option holds prices across the market it reaches, and the District's pause puts that whole network at risk. So the point isn't that $1.2M modernizes 13,400 homes — it's that the District called a $1.2M investment unaffordable while a single year of a modest increase across the network's reach costs more. ("Leaves the county" = the margin paid to out-of-county providers; a local, at-cost network keeps those dollars home.) Move any input you like — the conclusion holds: the upgrade is a rounding error against what unchecked prices cost.

Counting providers doesn't measure competition. Counting incentives does.

The District's letter calls this “a highly competitive marketplace,” and points at the provider list. But a market can carry several names and still behave like one company — if those companies all answer to the same kind of owner, with the same cost structure and the same pressure to return cash. That isn't our theory. It's the framework the federal antitrust agencies use.

The government's own merger framework says aligned incentives are the problem. Under the 2023 Merger Guidelines (DOJ / FTC), coordination between rivals doesn't require a meeting or an agreement — it happens tacitly, through observation and response to rivals. Guideline 3 treats a merger as unlawful when it raises that risk, and §2.3.A names the loss of a “maverick” — a firm whose different incentives disrupt the comfortable equilibrium — as a primary factor. “Aligned incentives” sits among the secondary factors. Read that against the letter: antitrust enforcers treat the disappearance of the differently-motivated firm as a harm in itself, no matter how many competitors remain on the list.

And the price data says two isn't enough. On FCC availability data, roughly 35% of Americans sit in a monopoly and another 37% in a duopoly. As one summary of that record puts it: people with one choice “pay monopoly prices, and people with only two… pay the higher prices typically charged by duopolies. People with three or more choices typically pay less.” Where the count reaches three or more, 300 Mbps runs about $58 against roughly $82 in monopoly markets.

So the question isn't how many companies serve you. It's how many different answers they owe. In this county the retail names sit on top of genuinely different owners:

Ownership isn't a moral scorecard — investor-owned firms bring capital and engineering this county genuinely needs. It's that a cost-based operator and a return-seeking one respond differently to the same opportunity to raise a price, and that difference is what a captive rural customer is actually buying. The pattern shows up wherever it's measured: consumer-owned electric utilities charge residential rates averaging about 13% less than investor-owned utilities (American Public Power Association). (That's electricity, not broadband — evidence about how ownership shapes pricing, not a forecast of anyone's internet bill.)

Sources: 2023 Merger Guidelines, DOJ / FTC (Guideline 3; §2.3.A on mavericks and coordination) — the “maverick” and “parallel accommodating conduct” concepts date to the 2010 Horizontal Merger Guidelines §7. Provider-count and price figures: FCC availability data as summarized by the Benton Institute for Broadband & Society. Longmont: Institute for Local Self-Reliance. Public-vs-investor-owned electric rates: American Public Power Association. The underlying theory is Shleifer, “A Theory of Yardstick Competition,” RAND J. Econ. 16(3), 1985 — a cost-based benchmark disciplines a monopolist. Broadband price-comparison studies cited above are drawn from wireline markets; we use them for direction, not as a claim about what a wireless tower does to a fiber price in Omak.

A local bill mostly stays here. A monopoly bill mostly leaves.

And it isn't only the size of the bill that matters — it's where it goes. Price increases on a rural household aren't rare; for a for-profit provider, raising the price is the plan.

Here's the part the "recoup" lens misses. When a rural household pays a national ISP or a satellite multinational, most of that money leaves the county — the margin is repatriated to a distant headquarters and doesn't come back. When the same household pays one of the seven local providers on the District's network, the money largely recirculates here: local payroll, local taxes, local spending. So the protection works twice — it holds the household's price down, and it keeps the dollars that would otherwise leak out of Okanogan in local hands. (To be precise: a bill paid out of county isn't 100% "lost" — the provider still pays some local costs. The dollars that truly leave are the margin a distant owner extracts, and on a monopoly bill that margin is large.)