TCS Daily

How to Relieve Those Air Traffic Delays

By Robert W. Poole - November 6, 2000 12:00 AM

Something is wrong in our skies, and you don`t have to be a gold-level frequent flyer to know it. This year is setting new records for airline delays, but the problem has been growing for a long time. Between 1995 and 1999 flight delays increased 58 percent, taxi-out delays of one hour or more soared 130 percent, and cancellations were up by 68 percent.

What we`re seeing is the inevitable result of trying to run a 21st Century aviation sector with mid-20th Century technology: long-range radars powered by vacuum tubes, mainframe computers running patched-together code written in the 1960s, and "airways" in the sky defined by ground-based beacons on hilltops. Those are the maintstays of our air traffic control system. The agency in charge -- the Federal Aviation Administration (FAA) -- has spent tens of billions of dollars over the past two decades with very little to show for it.

At its root, air-traffic management is an information-processing problem. There is no shortage of air space. It`s simply a question of having and using enough information, in real time, to keep fast-moving planes safely separated while they get from point A to point B in three dimensions.

The FAA`s obsolete approach is the "mother, may I?" model. Nobody can move without a controller on the ground giving the order (and trying to keep track, in his or her head, of an uncomfortably large amount of data). This anachronistic system forces the FAA to define a small number of "airways," separated from each other by huge buffer zones, both vertically and horizontally. Pilots actually have very little information in their cockpits about such basics as weather and other planes - it`s all centralized on the ground, on controllers` screens.

The alternative model is called "free flight." It lets pilots select their own paths. It`s based on technologies such as the GPS satellite network, which provides highly accurate, three-dimensional position-location in real time; digital datalinks that send information back and forth much faster and more reliably than voice, and clever software tools that predict potential conflicts far in advance.

With this technology, far more information is provided to each pilot in the cockpit, and the role of the controller on the ground becomes more that of an assistant, who`s there to help spot and resolve conflicts. The whole sky becomes available for flight, as long as all the planes in that batch of sky are properly equipped. Once they are (and the costs of the on-board gear are dropping rapidly), it becomes possible to save billions of dollars in costs on the ground by consolidating facilities and reducing the workforce.

And that gives you a clue as to why the FAA has been so resistant to change. It`s a lot like railroads making the transition from labor-intensive steam locomotives to far more efficient diesels. That transition happened rapidly in this country, where railroads were investor-owned and cost-saving mattered more than preserving obsolete jobs. But where railroads were government-owned, steam persisted a decade or two longer. (Indeed, it`s still a major factor on China`s railroads today.)

Ironically, when it comes to air traffic control, the rest of the world is far ahead of us. Over the past decade, 17 countries have shifted this vital service out of government agencies like the FAA, setting it up instead as a commercial enterprise. The pioneer was tiny New Zealand, but other countries include Australia, Canada, Belgium, Germany, Switzerland, and Britain. In every case, out the window went civil service rules, cumbersome procurement regulations, and the micromanagement that comes with government funding. These ATC corporations proved easy to finance. With a growing revenue stream from aviation user payments, they can sell long-term revenue bonds to finance high-tech modernization.

What about safety? Defenders of the status quo argue that a commercial approach somehow conflicts with the security of passengers -- as if investor-owned airlines were indifferent to crashes. But, in fact, safety is improved by ATC commercialization. First, getting the ATC service provider out of government puts it at arms-length from the safety regulator. In our case, the FAA would no longer be regulating (and covering up for) itself.

Second, the key to improving air safety is better technology - and that`s what these new ATC corporations have been investing in, more successfully than the FAA. And third, these corporations have to buy liability insurance (something government never does), giving them another layer of safety oversight -- the insurers` money is on the line if the company screws up, so the insurers are tough enforcers too.

Sounds great in theory, you might say, but how`s it actually working out? When Germany commercialized air traffic control, it cut airline delays by 25 percent in the very first year. Both Canada and New Zealand have seen costs (and user fees) come down by about one-third since their systems were commercialized. (Neither has resorted to layoffs, either; they`ve downsized by attrition.) And in Canada, air-safety-as measured by "operational errors"-has improved by 20 percent since 1996 (while the U.S. trend is in the opposite direction).

If the technology and the organizational models are out there, when can U.S. frequent flyers expect relief? The Clinton administration tried and failed in 1995 to shift air traffic control to a commercialized government corporation. Several Republican members of Congress have drafted bills to create a non-profit, private ATC corporation on the order of the successful Nav Canada. But so far, opposition from controllers (who fear downsizing), private pilots (who fear higher fees), and other members of Congress (who fear loss of control) has stymied reform.

A new president and a new Congress will have a window of opportunity next year. But they are unlikely to act unless airlines and their passengers demand real change. Both Gore and Bush claim to be reformers. Fixing air traffic control will be an acid test for whichever one becomes president.

TCS Daily Archives