Question
Why hasn't US broadband converged like Europe's
Everyone will tell you it's regulatory capture and monopolies, which is true but incomplete. The real problem is that the US already *won* the last infrastructure race, and that victory is now concrete.
Europe and Japan rebuilt their telecom infrastructure from scratch in the 90s and 2000s when fiber was economically viable at scale. They planned it regionally, made it publicly available wholesale, and moved on. The US had already spent the 1980s-90s pouring billions into cable plant and copper distribution that worked fine for what people wanted at the time. By the time fiber became obviously superior, that sunk cost was real money, tied to real assets, owned by companies with actual quarterly obligations. So Comcast in Denver services a neighborhood where they've already recouped their plant investment five times over. Ripping it out costs them money this year.
The regulatory piece matters, sure. But I've watched this happen in other domains. It's not primarily that politicians are captured (though they are). It's that once infrastructure gets to a certain state of entrenchment, even reasonable actors with good incentives get stuck. The gap between blocks isn't malice—it's the gap between "we upgraded this neighborhood in 2015" and "we haven't touched that one yet and the ROI still looks marginal." Then five years pass. Then another five.
The European advantage wasn't moral. It was timing.
3 comments
Log in to comment.
I'd push back a bit on the timing story, actually. Or rather, I'd say timing mattered, but not in the clean way this frames it.
I worked at a regional ISP for a couple years in the early 2010s, and the constraint we hit wasn't sunk costs on old plant—it was that the financial model for new deployment basically doesn't work outside dense urban areas, even with a blank slate. Europe got around this partly through subsidies and public ownership, which the post mentions, but also through something simpler: they're denser. A fiber drop in Amsterdam costs different money than a fiber drop in suburban Columbus. You can have perfect regulatory alignment and still not build to scattered exurban areas because the per-home cost is just brutal. We looked at it.
The other thing is, I'm not sure the "already recouped five times over" story fully explains Comcast's behavior. Utilities can be profitable on declining assets for decades. The real resistance I saw was organizational: cable companies weren't structured to transition their business model. That's not the same as sunk costs making it irrational. It's more that the leadership wasn't incentivized to cannibalize the video bundle before they had to. Regulatory capture actually does matter here—it lets them delay the inevitable transition because they don't have the pressure a competitive market would create.
So yeah, timing shapes things. But I'd hesitate to absolve the incentive structure itself. The entrenchment is real, but a lot of it's about what the market structure *allows* to be profitable, not what it requires.
I'd push back on the timing framing a bit. Yes, Europe had the advantage of rebuilding, but you're treating the sunk cost as more deterministic than it actually is. AT&T had massive sunk costs in copper too, and they've still deployed fiber to something like 70% of their footprint over the last decade. Comcast hasn't, despite being arguably in a stronger financial position.
The difference isn't just that the US "won early." It's that AT&T faces competition from cable and wireless in ways that force the upgrade, while cable companies in many markets face almost none. A company with sunk costs will absolutely rationalize inaction when the competitive pressure allows it. But that's not infrastructure being naturally sticky—that's a choice made legible by the lack of credible alternatives.
What I think you're underselling is how much of this *is* regulatory capture dressed up as economics. When Comcast can lobby municipalities into exclusive franchise agreements, they're not just operating with better ROI math on their existing plant. They're actively preventing the conditions that would make that sunk cost feel threatening. Europe's regional planning approach made bypass possible; the US structure made it illegal in a lot of places. That's not a natural outcome of timing. That's a design choice that became policy.
This tracks with what we saw at my last job when we were evaluating CDN providers for a client in the midwest. One of the options was this regional operator that literally couldn't upgrade their backbone because they'd financed their current setup with debt they'd still be paying off in 2028. The math on fiber expansion just didn't work for them—not because of greed, but because their previous investment was still generating enough revenue to service the debt. Meanwhile the big national players could absorb the capex across thousands of POPs. The regional operator wasn't being irrational, they were being mathematically realistic about their constraints.
I think you're right that this isn't mainly a regulatory capture story, though the regulations definitely *protect* the entrenchment once it exists. The harder problem is that infrastructure decisions from 20 years ago have these incredibly long tails. You can't just decide to be rational today; you're locked into decisions made by people who aren't even at the company anymore. Europe got lucky timing—they could greenfield it when the tech was ready and the economics aligned. The US got unlucky by succeeding too early with the wrong medium.
The question that actually interests me is whether this is inevitable or whether there are organizational structures that could break the cycle. Like, could a sufficiently large municipal broadband rollout in one region actually force the incumbents' hand the way competition did in a few cities? Or does the debt just... win every time?