The FAA ground stop and the brittle state

The Federal Aviation Administration issued a nationwide ground stop on Thursday, halting all departing flights across the United States after a telecommunications outage disrupted communications between air traffic controllers and aircraft. Al Jazeera reported that the halt affected every major airport in the country, while NPR confirmed that the agency declared the stop after controllers lost reliable voice communications. The disruption triggered cascading delays and cancellations: the Guardian reported that nearly 11,000 flights were delayed or cancelled before operations resumed. NPR further noted that Homeland Security Secretary Markwayne Mullin publicly stated there was no known threat, an intervention that itself communicated the severity of the anxiety inside the administration. Passengers sat on tarmacs without information for hours, trapped in an experience of institutional helplessness that has become, for air travellers in the United States, something disturbingly close to routine.

The received wisdom

The standard interpretation of a disruption like this treats it as essentially a technical event — an unfortunate malfunction in a complex system, to be resolved by engineers, logged by regulators, and addressed through the standard federal procurement and IT-modernisation process. Aviation safety in the United States is, by global standards, impressively good. The FAA oversees thousands of daily flights with a fatality rate that is statistically negligible. The ground stop worked as it was designed to: when controllers lost reliable communications, they grounded the planes rather than risk miscoordination. That is the system functioning, not failing.

The progressive reading would add a layer of infrastructure investment argument: the federal government has underfunded the FAA’s technology base for decades, and what is needed is a large, sustained public commitment to modernising the air traffic control system. This argument has real merit. The FAA’s core systems run on technology that predates the smartphone. Congress has chronically failed to provide the sustained, multi-year funding required to replace legacy communications infrastructure. The Biden administration’s infrastructure law made some provision for this, but implementation has been slow. If you want someone to blame, the standard frame suggests, look to the appropriations committees that keep short-changing a safety-critical federal agency.

A different read

The infrastructure-funding argument is not wrong, but it is incomplete. It treats the FAA’s fragility as primarily a resource problem — a gap between what the agency needs and what Congress has provided. The harder truth is that the United States has not simply underfunded its aviation infrastructure; it has designed a governance structure almost perfectly calibrated to prevent the kind of sustained, competent institutional effort that reliable modernisation requires.

The FAA has not had a permanent, Senate-confirmed administrator for significant stretches of the past decade. The agency is simultaneously a safety regulator and an aviation industry advocate — a dual mandate that creates structural tensions. Its procurement processes are governed by federal contracting rules designed, ostensibly, for accountability but which in practice produce cost overruns, decade-long timelines, and systems that arrive obsolete. The NextGen air traffic modernisation programme, begun in the 2000s, has cost billions and delivered a fraction of its promised capability. This is not a story of underfunding. It is a story of institutional design that cannot spend money effectively even when money is provided.

There is also a question the mainstream frame is reluctant to ask: how much of this fragility is the product of risk aversion dressed up as prudence? The communications system that failed on Thursday has known vulnerabilities that have been documented in inspector-general reports for years. The Guardian noted the scale of the disruption — nearly 11,000 flights — without dwelling on the institutional archaeology of how the vulnerability came to exist. The answer, in brief, is that replacing a live system is genuinely difficult, that every attempted modernisation has faced political opposition from stakeholders who benefit from the status quo, and that the federal workforce capable of executing complex technology transitions has been hollowing out for two decades.

Historically, the United States built critical infrastructure — the interstate highway system, air traffic control itself, GPS — through concentrated executive attention during periods when the federal government retained the technical competence and institutional credibility to deliver large projects. That era ended somewhere around the 1980s, and the country has been living on the institutional capital accumulated during those years ever since. The result is infrastructure that works, until it does not, and then fails catastrophically and visibly.

The ground stop also arrived in the context of a nearly simultaneous and unrelated outage at Microsoft, which took down Outlook and Teams globally for close to twenty-four hours. That the two events coincided is coincidence. But the coincidence is instructive: it illustrates how thoroughly critical social infrastructure — both physical and digital — has been allowed to consolidate into systems with single points of failure, operated by organisations whose incentive structures do not reward resilience. The FAA’s telecommunications vendor and Microsoft’s Azure infrastructure share a common feature: they are very large, they are very hard to replace, and when they fail, the costs are borne not by them but by millions of people who had no meaningful choice in the matter.

The right-of-centre case for taking this seriously is not about federal spending increases. It is about the state’s obligation to maintain the prerequisites of a functioning civil society — an obligation that fiscal conservatives sometimes wave away with talk of private-sector dynamism, but which admits of no private-sector substitute when the market in question is natural monopoly infrastructure operated under federal licence. A government that cannot keep its air traffic control communications system operational is a government that has failed in its most basic functions, regardless of what the discretionary budget looks like.

What to watch

Watch whether Congress uses this episode as the catalyst for genuine FAA reauthorisation reform — specifically, provisions that allow longer-term funding commitments and streamlined procurement for communications infrastructure. Watch also for the inspector-general investigation that will follow, and whether it surfaces the same documentation of known vulnerabilities that previous reviews have produced. If the FAA administrator’s response focuses primarily on the specific vendor whose equipment failed rather than on the structural governance deficits that made the failure consequential, that is a signal that the institutional lessons will not be learned. And watch Microsoft’s concurrent outage proceedings: if two of the most visible critical infrastructure failures in recent memory happen within hours of each other, the question of systemic concentration risk in American infrastructure deserves a serious policy hearing rather than separate, siloed investigations.

— J