Trump’s Death Spiral Story of Air Traffic Control

Earlier this week President Trump, perhaps in desperate search of some good press, announced plans to privatize the air traffic control arm of the FAA. Under almost any other circumstances, this announcement would have been greeted by some reservations, but generally by tepid acknowledgement that the US air traffic control system is consistently on the cusp of obsolescence and is perhaps in need of a good shaking up. However, it was President Trump who said it, and President Trump’s team who planned it, so the idea is being taken with an enormous grain of salt by almost everyone involved. Nonetheless, as NPR pointed out, it’s hardly a new concept.

There’s a deeper story here, however, that highlights some of the issues that come from a fairly clueless President trying to “drain the swamp.” It seems, based on reporting by NPR and Politico, that the primary impetus for this decision was based on the perception and reality that the FAA was consistently behind schedule and over budget when it came to the delicate business of revolutionizing American air travel. I say “perception” and “reality” as two distinct items, dear reader, because this is one of those stories where there is truth… and then again there’s a different sort of truth.

It is an undeniable fact that the FAA has bungled key aspects of the comprehensive modernization program for US air traffic control, called NextGen. It is an equally undeniable fact that this bungling is at least in part the result of the dysfunction of Washington, DC, and therefore beyond the FAA’s control. And it is a further undeniable fact that, even if the FAA did do a bit of bungling, it’s unclear that private industry can do much better.

A lot of people don’t know this, but the average citizen is perfectly capable of looking up the various high-value IT-related projects (such as NextGen) under development by the Federal Government. These projects are collected, recorded, and updated on a very good resource called the Federal IT Dashboard. Using this tool, you too can take a look at the various projects being worked by the Department of Transportation (or any other non-classified Agency), and you can easily filter down to FAA. And once on the FAA summary page, with a little digging, just about anyone can see more or less what the ongoing projects are with respect to air traffic control.

In short, this view paints a picture of several key projects that are over budget, over schedule, and generally showing a slow rollout of key functionality. This, then, along with reports by the DOT Inspector General, is what Trump’s advisers saw when they coached him to say, “After billions and billions of tax dollars spent and the many years of delays we’re still stuck with an ancient, broken, antiquated, horrible system that doesn’t work.”

More on that in a moment. Remember our little aside earlier about perception and reality? This is one of those cases where although something looks bad, all is not lost. The simple fact of the matter is that the FAA was and is trying to do something very ambitious in the midst of one of the most desperate periods of political dysfunction in American history. In short, if you read the project descriptions (you did, didn’t you? I knew I could count on you!) you would notice that the FAA is already deploying solutions to the problem Trump outlined.

Basically, the FAA is replacing several key systems that are, as President Trump mentioned, antiquated (although they are not horrible — the US has an incredibly safe air transit system, as anyone with a shred of ability to access the internet can find out). These include:

• Upgrading all aircraft to use GPS navigation and, critically, report their positions via satellite. Prior to NextGen American air traffic controllers relied on flight plans, limited radar coverage, and radio communications to track aircraft; the new system would add GPS tracking and broadcasting capabilities, enabling controllers to maintain a constant view of aircraft across the US airspace and over oceans. No, they didn’t have this before. I know, I thought they did too. They didn’t. Now you know.
• Upgrading all aircraft and ground stations with new radar and radio equipment.
• Upgrading the handoff protocols and facilities to allow a smaller number of controllers to manage a plane’s ascent and descent. You didn’t even know that was a thing, did you? It totally is! Different controllers handle planes on ascent, descent, and in flight. You learned something today.
• Upgrading runways to incorporate automated status lights to help guide planes in on approach, in coordination with all other aspects of the landing process.

So to recap, the FAA is already replacing radars and radios, is working to outfit aircraft with a brand-new(ish, other countries already do this) GPS application, and is working to improve takeoff and landing procedures. That sounds like pretty much everything Trump might be talking about, right? So what’s the problem?

Well, part of the problem has been, undeniably, that the FAA has not had a good success rate with IT development. However, and this is a big “however,” a canny reader should appreciate that no one has a good record of IT development. Projects are measured as “succeeding” or “failing” based on whether they deliver the right product, on time, on budget. Based on this metric, it might shock you to know that in industry, better than 60% of IT projects fail, and 25% are canceled outright. In government, that number rises to 70%.

And this takes us straight into the discussion of Congressional dysfunction. Part of the problem with trying to judge the Federal Government in comparison to industrial project development is that the government projects operate on a variable budget that changes annually, or worse yet, doesn’t change at all. It is an acknowledged fact that the worst things you can do with a project are mess with the budget and mess with the people. Over the last few years, Congress has done both.

Some exposition may be necessary here. Under normal circumstances, the budget is set by Congress, approved by the President, and delivered to the various Agencies sometime between February and April. This budget may change the funding for a large project, and in fact should do so based on the needs of that project. In reality, the budget changes based on the whims of politicians who really don’t care, or even know that much, about the budgetary needs of most projects. This leaves shortfalls and overages, both of which count as inefficiencies and both of which can result in delays in project performance.

On the other hand, over the past few years the government has subsisted on what are called “continuing resolutions,” which is another way of saying that Congress can’t come to a decision on budgetary levels and just passes the previous year’s spending levels. This is bad news for any large-scale development project that does not have even levels of spending over the entire project lifecycle (in other words, all large-scale IT projects). And this, too, creates inefficiencies. Congress passed continuing resolutions at some point in the budgetary cycle every year for the last nine years. There is no way this did not result in inefficiencies and project delays.

Here, then, is the perception gap. To President Trump and those who support him, the US air traffic control modernization program (NextGen) looks like it’s in a death spiral. It looks like we’re pouring money into a black hole and getting nothing out of it. And it looks like the solution is to privatize it all and let an industrial solution handle it, since obviously government is incapable of taking the bit between its teeth and running.

But. There are problems with this assessment. Three big ones, in fact. The first is the allocation of blame. Trump and his advisers assume that the FAA is incompetent. However, they neglect to consider the fact that the FAA might be totally competent, but Congress might simply be dysfunctional enough to generate enough budget slippages to negatively impact FAA project management. Actually, that’s the generous version of this assessment. The less generous, but possibly more honest version is that Republican strategists are using this as another front in their ongoing battle to “Starve the Beast” — sabotaging federal programs, complaining federal programs are running poorly, and then spinning them off to private interests.

The second problem is the assumption that a private corporation can do better. They neglect to consider that, even in Silicon Valley, a private firm would statistically only have a 10% better chance to succeed where the FAA failed. That’s before one factors in the disruptions that will come along with handing over a bunch of in-progress projects to a new organization; it’s not for nothing that we have the saying “don’t change horses in the middle of a race.” In almost any other setting, this would be viewed as an incomprehensible risk, especially given the fact that the FAA is finally showing progress with NextGen. To put this in project management terms, you do not torpedo forward progress because you are upset with sunk costs. A real-estate developer in particular should know better… if efficiency was his actual goal, at any rate.

And the third problem is, of course, about money. Trump’s team seems to have grasped that federal funding has been inconsistent with project needs in this instance. Their solution is to allow the proposed new air traffic control entity to charge fees for service. This will likely result in a more dependable funding stream. It will also, almost inevitably, result in passing the costs on to consumers. We are, in essence, moving from a public highway to a toll road in the sky. This is undoubtedly fine for billionaires, but the regular Joe will likely see an increase in his ticket price.

If this is a microcosm of the Trump approach to infrastructure investment, then it is a troubling one. The idea of “fixing” the problem by outsourcing it may work here — there are large risks, but other countries have implemented a similar system. The broader implications are troubling, though. The “fix” represent a near-total disregard of actual project realities, does not identify or address the underlying problems of legislative dysfunction, and does not offer any particular plan to improve overall Federal project management.

Though I am loathe to use the slippery-slope argument, at what point will it stop looking like a good idea to do away with government instead of fixing it? Will we wake up in a few mornings to discover that all of our national highways are now toll roads? Will we outsource the FBI? Or the Army (ok, to be fair, we’re already well down the path of mercenaries, but still)? The simple fact of the matter is that there are things that government does because private industry cannot or will not do them in a way that benefits all members of society equally. We need to ask some hard questions, starting with this one:

Just because government does things slowly, does that mean government should not do them at all?