Air Traffic Control Reform Newsletter #22

Topics include: FAA's "perfect storm" budget crisis, operational error reporting, expanding ATC capacity, and the controller workforce issue.
In this issue:

Perfect Storm Hitting New Air Traffic Organization

Although former American Airlines executive Russ Chew knew the job would be a big challenge, it's hard to believe he had any idea how much of squeeze he'd find himself in, as the first Chief Operating Officer of the FAA's new Air Traffic Organization (ATO). After a year on the job, Chew is being hit with a perfect-storm budget crisis that is structural, not cyclical, in nature. Fixing it is going to require even more fundamental reform than most people had expected.

In a nutshell, ATO costs are rising at a much faster pace than revenues. In a series of recent internal and external briefings, Chew has been educating ATO staff, customers, and owners about the harsh new realities. On the revenue side, the recession and 9/11 led to a several-year slump in air travel that depressed all aviation user-tax revenues. But as air travel has recovered, the single largest revenue source — the 7.5% tax on the value of airline tickets — has not. That is due to the huge growth in low-cost carriers (LCCs), which now compete on about three-quarters of all routes. That has driven down fare levels and hence ticket tax revenues. And there are no signs that this is merely a temporary aberration.

Unlike the airlines, which have had to make major cuts in expenses over the past four years, the FAA budget has kept growing, from $12.1 billion in FY2001 to $13.8 billion in FY2004 (up 14% over three years). To close the most recent gap between aviation tax revenues and FAA spending, Congress in 2004 used $3 billion in general-fund revenue and drew down another $1 billion from the balance in the Aviation Trust Fund. But as Chew's charts make clear, there is now only $2.7 billion left unobligated in the Trust Fund, and it's likely to be all used up by FY2007 or 2008. Moreover, at a time of severe pressure on all federal agencies to cut spending due to the federal budget deficit, the odds of continuing to get $3 billion a year in general-fund money are very low. (General-fund support for FAA has averaged $2.2 billion/year over the past decade.) So with no relief in sight from either the general fund or the Trust Fund, the only other current revenue source is aviation user taxes. And with the airlines already losing billions and complaining bitterly about the level of those taxes, the odds of increasing them are close to nil.

Short of some deus-ex-machina on the revenue side, that leaves Chew with cutting costs as the only alternative. And that will not be easy. As Inspector General Kenneth Mead has been pointing out for several years, the Operations budget has appeared to be on autopilot, thanks to a very expensive contract with the controllers agreed to by former Administrator Jane Garvey (and extended for two years by current Administrator Marion Blakey). Blakey recently set the stage for upcoming contract negotiations by warning that FAA "cannot and will not sign something we can't afford."

Chew is saying all the right things about measuring (and delivering) value to the customers, and about reducing unit costs. Unless there are major changes, Aviation Week reports Chew projecting a $5 billion shortfall in operations and a $3.2 billion shortfall in capital spending over the next five years. Given the relative sizes of the capital and operating budgets, it looks as if the biggest hit would be to the capital budget. That's bad news for the customers, especially the airlines. With airline delays poised to exceed those of the summer of 2000, the urgent need is for advanced technology that expands airspace capacity. That vital modernization is the most likely victim of this emerging perfect storm.

Operational Errors: Garbage In/Garbage Out

Inspector General Ken Mead has performed another real service with his Sept. 20th audit of the process for recording operational errors (violations of the FAA's standards for keeping planes safely separated). FAA claims to have made modest progress in 2004 in reducing these errors, but the IG's audit points out that accurate data only exist for 20 of the 524 ATC facilities. Only the 20 en-route centers have automated systems for identifying impending loss of separation, alerting controllers, and recording that fact. All towers and TRACONs, by contrast, rely on self-reporting of operational errors. And according to the IG's report, such reporting is, shall we say, suspect.

First, consider the raw data. Although en-route centers handle about 25% of all operations, their automated systems reported 684 operational errors in FY2003. By contrast, for the 75% of operations handled by towers and TRACONs, only 501 total errors were self-reported. And of those 501 reported tower and TRACON errors, 22% came from an outside source (i.e., were not self-reported by a controller or manager). By contrast, at en-route centers, only 4% of the errors were identified by outside sources.

To reinforce your intuition that not everything is getting self-reported, the IG staff followed up on a whistle-blower complaint regarding the Dallas/Ft. Worth TRACON. Though this investigation is still in process, the preliminary findings are troubling. That facility had self-reported only two operational errors between January and June 2004. But digging into the data for just May and June, the IG's sleuths found five more (unreported) operational errors.

FAA claims that because of the complexity of traffic at towers and TRACONs, an automated system like that used at centers is not feasible. But the IG's report suggests two feasible alternatives. First, data playback systems retain detailed operational history for 45 days; those records could be regularly reviewed to identify operational errors not previously reported. Second, for those errors that occur on the ground, the busiest airports are in the process of getting collision-avoidance systems. Some 34 airports have recently received Airport Movement Area Safety Systems (AMASS,) and another 25 are due to get the more-advanced Airport Surface Detection Equipment (ASDE-X), both of which can provide automated identification of this type of operational error—if FAA chooses to use it that way. Unfortunately, budget cuts are pushing back the installation dates for ASDE-X till 2008-09 (a delay which FAA ludicrously declares will not affect safety).

Not mentioned in the IG's report is the other serious flaw in operational error reporting: its lack of international comparability. It's high time FAA worked with the International Civil Aviation Organization (ICAO) and Eurocontrol to standardize the definition of "operational error." Only then will it be possible to make meaningful comparisons of the safety performance of different air navigation service providers.

Is Expanding ATC Capacity Threefold Really Do-Able?

For me, the most exciting session at the recent annual meeting of the Air Traffic Control Association was Tuesday morning's plenary session on ATC 20 years hence. While several speakers talked about technology advances, two presenters stood out because their visions made such a stark contrast. The first of these was Doug Fralick, Director of Safety & Technology for the controllers' union, NATCA. Having heard the previous speaker talk about the needed transformation in the role of the controller (analogous to the change in the pilot's role over the past 20 years, thanks to much better technology), Fralick was having none of it. "Human-centric" ATC is still our best bet for the next 20 years, he maintained. Besides, weather will remain a major source of problems. And anyway, the real need is for a lot more runways (yet we all know expanding airports is becoming exceedingly difficult). In short, we'll just have to muddle through with the system more or less as it is today.

It was the perfect set-up for Mike Lewis, Director of Future Systems at Boeing. Lewis took on the challenge of expanding ATC capacity in the terminal area, and made the startling claim that we can increase this capacity threefold (3X) in the next 20 years. How? Contra Fralick, it's by making ATC "network-centric," not human-centric. That means we need to create and fully exploit "shared precision information." Translated into everyday English, that means recognizing that today's standards for how far apart planes must be kept by ATC are not laws of nature. Rather, they are a function of how good our information is about where each plane is (and will be one second later), where bad weather is (precisely), where wake vortexes are spinning behind a plane (exactly), etc. If we can know these things much more precisely, in real time, we can fly planes much closer together, safely.

Lewis then laid out seven key steps which, used together, would deliver 3X today's airport capacity (as assessed by simulation modeling at Boeing). They are:

  1. Remove visibility as an issue (via tools like NASA's synthetic vision);
  2. Provide arrival time precision within +/- 2 seconds (as Boeing recently tested in Sydney and Melbourne);
  3. Make use of all nonhazardous air space via much better real-time weather information;
  4. Reduce final-approach spacing to 2 miles, from the 3-6 miles used now due to wake turbulence;
  5. "Pave down the middle"—i.e., add new runways between existing ones (Raytheon is working on a system for landing an echelon of planes on three closely spaced parallel runways.);
  6. Enable safe multi-aircraft operations on runways (e.g., a regional jet landing closely behind a 777 taking off near the end of a two-mile runway).
  7. Use all available capacity of all airports (e.g., better-coordinated use of the airports in each metro area).

He told us that the big-payoff items in this list were numbers 4 and 5, accounting for about three-quarters of the total capacity gain. And he also noted that although much of the technology is nearly in hand, breakthroughs are needed in (1) short-term weather forecast accuracy and (2) wake vortex detection and prediction.

The business case for revamping ATC along these lines, he said, is overwhelmingly positive. Three times the capacity means three times the revenue, but he maintains that this technology-intensive system would require little or no increase in staff and less expensive infrastructure than today's costly-to-maintain ground-based infrastructure.

This is the kind of vision that results from viewing ATC as a high-tech service business. It's this kind of outside-the-box thinking that aviation needs, to escape the perfect storm now engulfing the fledgling ATO.

The Controller Workforce Issue, Part 2

I got a boatload of responses to the item last issue about controller "Jane Doe" and her charges that many facilities are over-staffed. Let me share a few of those comments, before adding a few other thoughts on coping with the impending controller retirement bulge.

Of the controllers and managers responding, most agreed with Jane Doe that lower-level facilities are over-staffed, with controller time-on-position running four hours or less out of an eight-hour shift. Several pointed to an FAA affirmative action program called "Train to Succeed" under which, to quote one controller, "unqualified people were hired, trained until their trainers couldn't take it any more, and then sent to lower-activity facilities. The agency now has a cadre of sub-standard controllers... who could never move up to the busy, under-staffed facilities because they never should have been hired to begin with." One manager wrote that FAA cannot move controllers from over-staffed to under-staffed facilities because it gave away this right in the last contract. But another said that even when the agency had that right, it never used it. Yet another one reported that his combined tower/TRACON facility has gone from 23 controllers and 4 supervisors five years ago to 18 controllers and 2 supervisors—"And for the first time in years, I would consider our staffing optimal." He added, "If 20 people are doing the same work now that 27 people did five years ago, it follows that we were overstaffed by 26%."

One of the most interesting suggestions for how to cope with the impending retirement bulge also came from one of these controllers. He wrote that he knows many recently retired controllers who would love to work part-time again, for hourly pay with no benefits, for 3-6 hours a day several days a week. Since operations at most facilities have significant peaks and valleys, that would be a lot more efficient than paying everybody for 8-hour shifts, and it would make use of experienced people who are good at controlling traffic. (I note in passing that Delta recently negotiated contract changes that will let recently retired pilots return part-time to help address pilot shortages.)

Meanwhile, the ATO should soon start getting a better handle on how controllers are spending their time at work. The long-awaited labor distribution system, a part of the agency's new accounting system, is finally starting to enter service. Version 1 has completed testing at New Orleans and Van Nuys towers, and is being installed in five other facilities. For the first time ever, it provides a system by which controllers sign on when beginning their shifts, and record which activities they are engaged in. NATCA president John Carr claims the system (CRU-X) can't handle larger facilities. That may be literally true, since the system is being implemented in several versions, with the simplest version going into the least-busy facilities. I'm sure the Inspector General will keep us posted on CRU-X, since his office has been leaning on the FAA to get a handle on who's doing what, and where.

Finally, I'm heartened to see that the House Transportation Appropriations subcommittee, while approving funds to hire and train 120 more controllers, made clear in its report that it expects the FAA to replace retiring controllers on less than a one-for-one basis, since no staff reductions would imply no productivity gains.

News Notes

ATC Rate Reduction in UK. Now that air traffic on the North Atlantic has recovered, the part-privatized UK ATC provider, NATS, has announced that ATC fees and charges will be reduced by 3.8% next year. (That contrasts with an 11% increase the previous year.) NATS still hopes to be able to cut rates by 40% over the next 10 years, thanks to productivity increases. NATS is owned 46% by airlines and airport operator BAA, 49% by the UK government, and 5% by NATS staff.

Now 37 ATC Corporations. With the addition of Bulgarian ATC provider ATSA, global membership in the Civil Air Navigation Services Organization (CANSO) has reached 37. The 37 full members are all self-supporting ATC corporations, regulated at arms-length for safety. Well-known members include Airservices Australia , DFS ( Germany ), NATS (UK), and Nav Canada.

Corporatized Europe Taking Lead in Modernization? Aviation Week quotes several US ATC executives as concluding that Europe is moving ahead faster with ATC modernization than the United States . "They're out-investing us," said Michael Harrison of Aviation Management Associates. "They're outthinking us, said consultant Gerald Thompson, a member of the board of the Air Traffic Control Association.

New Player in Contract Towers. Airservices Australia has won a five-year contract to operate control towers in Hawaii (Molokai, Lihue, Kona, and Kalealoa), Guam, and Saipan . That contract was previously held by Serco, one of the big three in US contract tower operations. Airservices hopes to use the contract as a springboard for further overseas operations.

Should Private Sector Take the Lead? Jaws dropped at the ATCA convention when Lockheed-Martin's Don Antonucci questioned whether the FAA could modernize itself. As Aviation Week paraphrased his comments, "The time might have arrived for industry to replace government as the moving force behind modernization and global standardization of air traffic control."

Robert Poole is Searle Freedom Trust Transportation Fellow and Director of Transportation Policy