Airlines Confidential host and former Wall Street Journal airline reporter Scott McCartney had a fascinating discussion with FAA Administrator Bryan Bedford this week. I was enthusiastic about Bedford’s appointment to the role.
I am concerned with some of the rewriting of history he seems to do here – it seems far less than candid – and by his seeming desire to take greater government control over airline scheduling.
The First Trump Administration Wanted To Privatize Air Traffic Control, Now They Don’t, Why The Switch?
The first Trump administration worked on spinning off air traffic control into a stakeholder non-profit, along the lines of NavCanada. Here’s Bob Crandall (also on Airlines Confidential) blaming lobbying by private pilots for killing the effort that would have done more than anything else to improve performance and safety.
Bedford rewrites the narrative, leading off with an homage to President Trump (“in order to do big projects, you have to have strong leadership, strong vision”) and he says Trump was going to modernize air traffic control but that got hijacked by the airline industry and turned into a privatization debate. That’s a bizarre rewrite of history. Here’s President Trump announcing his air traffic control initiative,
At its core, our new plan will dramatically improve America’s air traffic control system by turning it over to a self-financing, non-profit organization.
He framed the FAA’s continuing role as safety oversight, not air traffic control operations:
The Federal Aviation Administration will focus firmly on what it does best — safety.
And he argued the separate entity would improve operations:
A separate non-profit entity would be charged with ensuring route efficiency, timely service, and a long-awaited reduction in delays.
A month later, he was clear his goal was to “transform our air traffic control services into an independent, non-profit organization.”
Trump’s Secretary of Transportation echoed this: “Air traffic control operations will be split, will be moved to a non-governmental independent, non-profit cooperative…the air traffic control system will be liberated from government constraints.”
The only possibilities I can see here in Bedford’s narrative about the history of air traffic control reform during the first Trump administration:
- Bedford doesn’t know what he’s talking about
- Bedford is being disingenuous
- Bedford thinks Trump didn’t actually want what he said he wanted, and is easily manipulated.
Number 1 and 2 reflect poorly on Bedford, number 3 is my Straussian take on his performative opening praise on the President. Making obsequious statements about the President creates room to implicitly criticize him.
The Problem With Air Traffic Control Wasn’t Just Who Was In Charge
Beford says the problem with air traffic control modernization became that “the Biden Buttigieg administration hated airlines.” But the problem is much more fundamental.
- Having the same entity regulating safety and providing service creates a conflict of interest. The FAA is its own regulator, which is a recipe for no accountability.
- The way they’ve done procurement for decades has been in long-term cycles, spending years specing projects instead of buying off-the-shelf software like the rest of the world. They depend on the vagaries of congressional appropriation cycles, rather than being able to issue bonds to make capital investments and charge fees to airlines for the service to cover that cost.
- And you get political interference in recruiting and training controllers.
That’s not just about spending money to replace aging equipment (and DOT hasn’t been great about maintaining the equipment they have). Indeed, Bedford acknowledges that it’s not just replacing obsolete equipment, but focuses on redesigning airspace and rethinking staffing and controller workload.
He blames the airlines, though, for air traffic congestion because they build their schedules around what customers want rather than what the status quo system can absorb.
The FAA is using AI tools to do a better job using airline schedules to forecast system overload. The solution, though, needs to be improving air traffic system throughput and staffing to travel patterns. And it certainly doesn’t work to start with current capacity as fixed and have the government parcel out flights to airlines based on that.
San Francisco Capacity Is Being Limited Because Of Palantir’s Systems
Bedford reveals that San Francisco parallel approach restrictions were the result of risks flagged by Palantir. The FAA now bars side-by-side visual approaches to the east-west parallel runways, 28L/28R (which are just 750 feet apart), even in clear weather and even when both crews report the other aircraft in sight. The new rule requires staggered approaches, one aircraft ahead of the other, and this limits capacity at the airport cutting maximum arrivals from 54 to 36 per hour.
What seems to have happened is a lot of false alerts about aircraft too close when there wasn’t any real safety risk. And parallel arrivals are being banned, but not parallel departures, because the same alerts weren’t coming up on departure (for technical reasons these are inhibited below 1,000 feet). Palantir may have flagged an issue that triggers a lot of alerts, and it’s low hanging fruit achieving metrics that correlate with safety rather than actual safety.
We Need More Controllers, But We Also Need Better Scheduling
Addressing the controversy over the FAA reducing its projected need for air traffic controllers, he says they have about 11,000 today and are accelerating hiring, but that they have a producitivity and scheduling problem.
Different facilities have different locally built schedules, controller transfer rules create staffing imbalances between faciltiies, and the agency does not have enough useful data on work spent on actual time-on-position. He frames this as hiring more, but ending the waste of controller capacity, rather than doing more with less. The problem is this gets back to having the FAA run air traffic control – it remains a political football, and Senator Chuck Schumer blocked controller assignment fixes for years.
He says we’re going to see facility consolidation at least as a test. And I’d argue we clearly need more remote towers to use controller staffing more efficiently.
The FAA Wants More Centralized Fixes To Air Traffic Control
Bedford’s view is that American air travel is delayed and made riskier because the FAA has been running a fragmented, analog-era system in a reactive fashion. He wants to move it toward a digital, centrally-optimized, more tightly-controlled airspace.
But when he says that the FAA lacked focus and accountability, producing five-year plans that no leader would ever be accountable for, he misses that centrally planning the airspace and continuing to have the FAA regulate itself is hardly a recipe for solving this.
Sounds to me like a proposal to bring back the Civil Aeronautics Board.
I think the article conflates for-profit privatization with the world of non-profit, stakeholder managed groups. The latter would include organizations like credit unions, Habitat for Humanity and community food banks which are mission driven and governed by consensus of the stakeholders. Bedford is distinguishing between a take over by a for-profit corporation versus what was actually proposed.
Bedford has always been a chump. From his days at Business Express, Mesaba, Chautauqua, Frontier, Midwest Express (and their pilot suicides caused by him), Republic. He has been totally self serving and every possible way. Of course he will rewrite history to benefit him or his cronies. On top of all this BS he thumps his Bible. Wow! What an A-Hole!!!!!
Airlines want to bank flights and have flights leave at optimal times. Few want to arrive in London at 4AM or arrive in NYC at 7PM. So that means stress on larger airports which translates into lack of gates, long taxis, multiple holds and incoming flights being slowed down in speed because of landing and take off congestion. All of that translates into passenger frustration. Landing and taxing to the gate at ORD and taxing can be a real adventure in itself.
It was only discussed briefly but there was also mention of SFO and the restriction on parallel approaches.
Although there has been talk of permanent damage to UA’s SFO hub efficiency, Bedford said new offset approach procedures would bring the capacity back.
The kakistocracy is alive! The FAA doesn’t regulate itself, Gary; Congress has oversight, should they grow a spine and choose to use it. I recommend we elect better people, say, in about 158 days. (Privatization benefits no one but the AA-DL-UA oligopoly.)
@1990 – Congress is not a regulator. And no one is suggesting privatization, Canadian air traffic control was not privatized. I do think that an entity that can issue bonds to pay for technology upgrades rather than relying on annual appropriations, and then charge user fees to pay for it, makes a lot of sense. NavCanada is far more efficient than the FAA ATO. But even just putting the Air Traffic Organization into a separate government entity other than FAA would be an improvement.
I’m not sure how remote towers would use controller staffing more efficiently. It would require the same amount of staffing, just at a remote location. It isn’t like someone can be certified in three different towers at the same time.
And if there ever is a consolidation of facilities on a large scale, you will see more controllers retiring or quitting than you could imagine. You have worked in your hometown for 20 years and they are going to move you 250 miles away, are you going to retire or move?
Not sure this is contradictory. Trump was for spinning off ATC, got flack, and retreated because there was no clear path to getting it done. This is how it works, and Gary knows how it works.
@jack the ladd – that’s not the story bedford is telling, he says trump didn’t actually want to do a spinoff, just to invest in ATC modernization but the process got hijacked by others. i offered the receipts that trump himself pushed the spinoff.
Trump would bring back Zeppelins if you paid him enough and called them Trump Balloons, I doubt he cares about any of this
Amazing that this article ignores the problems that the FAA created when they moved portions of Newark Approach to PHL apprach, including total radar and communication loss.
@JK – Well, the problems pre-existed. NY TRACON had refused to add controllers in order to protect overtime, it was the heaviest overtime TRACON in the country by a lot. And fixes were blocked by Schumer ofr years. But FAA initially mangled the fix. It was a brilliant workaround, really, but they’re bad at managing tech projects of all kinds so…
Since the entire industry is unified against privatization, it seems that sense got knocked into those in charge. But, also, what has happened since the first Trump administration? Covid.
NavCanada — as it is user-funded — was forced to lay off a large number of controllers, and they still have years to go until they’ll recover from that. Their airspace handles a fraction of the traffic of ours, and their headcount is also much smaller. Can you imagine what would’ve happened if we’d pursued it back then? The challenges faced today would be exponentially amplified.
Both Canada and the UK have also prioritized airlines over general aviation — and our entire system is based on equal access. That’s engrained in how the FAA operates, how laws are written — and, frankly, that’s what makes this such a great place to learn to fly. None of us should want the US to be more like Europe or other parts of the world where general aviation faces massive restrictions and much higher costs — it’s already expensive enough that many can’t pursue this as a career.
Plus, it took Canada and the UK several years to complete the transition to privatization. US airspace, again, is far busier and far more complex. How long would that take here? 10 years? 15? 20? And can you imagine how many billions of dollars would be wasted on this effort?
It’s fair to point out policy flops without reasonable explanations, but pretending like privatization should have been pursued because it was proposed originally is the wrong answer.
@Gary Leff — I know this topic is one of your rotating go-to’s, but, Canada’s airspace is simply different, less complicated than the US. I still think public infrastructure should remain under direct democratic control, rather than being handed over to a board dominated by corporate interests. User fees are regressive and will isolate rural/small airports. See you again when you repost next time.
@1990 – the claim about less complicated airspace in Canada completely misses that NavCanada handles the traffic across the North Atlantic. In any case, NavCanada’s far better management of tech upgrades positions them to handle complex airspace far better than the FAA.
And as for ‘user fees are regressive’ on the contrary the reason an ATC spinoff was killed 9 years ago is because *private aircraft owners* fought having to incur the fees.
@Gary Leff — Apples and oranges. Managing the North Atlantic verses the Northeast Corridor are two entirely different operational realities. Gander oceanic airspace is a massive block of high-altitude, en-route tracking. The complexity of stacking, sequencing, and routing thousands of flights an hour through overlapping, often weather-delayed airspace like NYC, Chicago, or Atlanta. No, we not could easily scale Canada’s model to the density of the U.S. system.
Not to mention, major carriers would want seats on that non-profit board because it would give them power over capacity allocation and fee structures. Who do you think gets prioritized during a weather delay? It won’t be a regional flight to a small hub… it will be profitable international routes connecting major hubs on AA, DL, or UA.
Sure, the general aviation lobby protects its billionaires and their private jets alike, but a privatized board would more likely allow major airlines to protect their oligopoly. The FAA has procurement and technology issues that Congress very much needs to fix, but handing over control of public infrastructure to the users with the deepest pockets is not the way to improve the system.