The next time you're soaring through the air at lightning speed, inside a metal tube, you might want to take the time to consider something of greater importance than the drink tray: the intricate web of human inte ligence that's keeping you safe.
That web is seeing some damage.
The Federal Aviation Administration is in the midst of a massive wave of air traffic controller retirements. These highly skilled workers guide your pilot, preventing him or her from crashing into a nearby jet. Controllers help with takeoffs and landings, and prevent your plane from running into, say, a snowplow in the middle of the runway.
As of early 2007, the FAA employed more than 14,000 controllers, who guide 50,000 aircraft each day. Over the next decade, 72 percent of those controllers will be eligible to retire, even as U.S. domestic flights are expected to increase by 27 percent.
In response, the FAA is conducting a mass hiring, shortening its years-long training process, and also looking at other measures, such as increased privatization and new technology. Though contacted more than a week in advance, FAA representatives would not reply to questions for this story.
Some government officials, including U.S. Rep. John Salazar, D-Colo., a pilot who sits on the House Aviation Subcommittee, assert the FAA is behind in its hiring process, and is cutting corners on safety to save money. And controllers, many of whom are now working overtime shifts and using decades-old equipment, put the situation in dire terms.
"Something has changed, that all of a sudden we're making mistakes that we never made before," says Mike Coulter, a recently retired Denver controller. "And I can tell you, it's fatigued controllers."
A little history
The controllers who are leaving the workforce now were hired after an infamous 1981 strike that led to 10,438 of their predecessors being fired.
President Ronald Reagan's mass dismissal left about 4,700 controllers on the job, and the controllers union (the Professional Air Traffic Controllers Organization or PATCO) in ruins.
From 1982 through 1991, the FAA hired an average of 2,655 controllers a year. But because most controllers are eligible to retire after 20 to 25 years with the FAA (and are required to retire at age 56), many of those controllers were destined to leave at roughly the same time.
That time is now.
The state of retirements
Bryan Sanford leans over the table at a Pueblo Starbucks, the lines of his face deepening.
"They're putting us in the type of situation," he says, "where you're probably going to have an accident."
After two decades with the FAA, Sanford is bitter. He's bitter about pay cuts, forced overtime, long shifts without breaks, and even the new, more formal, dress code. But mostly, he's bitter about safety.
Sanford currently works at Pueblo Memorial Airport's air traffic control tower. He also serves as a representative of the National Air Traffic Controllers Association (NATCA), the union that formed following the 1981 strike.
Controllers in Pueblo are facing a massive increase in traffic, now that the facility's been named a training ground for Air Force pilots. In addition to the traffic the airport already handled (much of which comes from military and private fliers, and other pilot training schools), the airport now manages traffic from 350 new students. That number's expected to increase to 1,000 in the next few months, with more to come.
All the while, controllers also are focusing their attention on training new hires, a process that can take years. Pueblo recently brought in new hires and controllers from other facilities, but all those recruits, regardless of experience, have to undergo training to learn the specifics of the facility. That means it will take a while for them to ease the staffing crisis.
Sanford started his career in 1987, after leaving his job as a dairy farmer. His new work served him well; he enjoyed the complexity of moving planes through the air, a process he compares to "a big video game that you're playing in your head."
Like many fellow controllers, Sanford is proud of having guided thousands of people safely through the air over the years. But he's worried. Short staffing, long hours and spotty use of safety equipment have put him and his coworkers in such a bind that there could easily be a mistake, he says.
"The FAA tries to downplay it, but they're getting to the point that they can't blow smoke," he says.
Sanford's just hoping he can bear the work until he reaches retirement age and can collect his pension. Even with that money coming in, though, Sanford would likely need to take another, lower-paying job to support his family, which nearly doubled recently, when he and his wife adopted five siblings.
How did this happen?
Sanford and other controllers say the troubles go back to the FAA's poor planning. After all, the FAA has known for decades that this retirement bubble was coming.
In reports, the FAA states that it does have a plan, and that it's keeping overtime to a minimum in most facilities while hiring plenty of new workers. But a person hired today will need to undergo an average of three and a half to nearly six years of training, including classes and extensive on-the-job education.
Even still, some new hires will find it impossible to learn the job. They'll move to an easier facility to start training all over again, or be fired.
"Not many people really can do this job," says Coulter. "It's just a different world we live in. I mean, we have no margin of error, and we live and breathe that."
The FAA says it has factored these challenges into its hirings, and it's adding the right amount of people each year. Yet its recent track record isn't full of accurate predictions; estimates for retirements have fell short several years in a row.
According to a report by the Department of Transportation, the higher-than-expected retirements have a lot to do with the FAA's recent actions.
In June 2006, the FAA pushed through its own work rules for controllers without ensuring NATCA's assent. According to a report by the Department of Transportation: "By September , when FAA began unilaterally implementing its own proposals for open Articles, actual retirements were nearly three times higher than FAA had projected (97 actual retirements compared to 39 projected)." The DOT report also warned that the FAA's pay cut to many controllers in early 2007 would likely cause more-than-projected retirements.
Coulter was one of the experienced controllers who left this year.
"I worked 25 years without having an operational error or operational deviation [a critical safety violation] to my name," he says. "I don't think I could work two more years and say that, simply because we're running out of people."
In Denver, Coulter also served as a NATCA representative. He says as old controllers retire and new controllers take their place, the system is losing the experience that has kept it safe.
The work environment has worsened, too.
"You can't even go get something to eat in Denver," he says. "You can't go for a walk on break to refresh. They just have you strapped in a working position forever, whether you're needed or not."
And Coulter, like the other controllers consulted for this story, says getting vacation time is difficult, if not impossible. At the same time, he and Sanford say, management obsesses over the dress code and buys new office furniture.
Quitting was a tough financial decision for Coulter; he'll still need to work to pay for his children's college education. But he feels the FAA was chasing him out the door.
During years where retirements exceed projections, FAA reports say, it simply hires more people to fill the empty slots. But even the FAA concedes it can't hire ceaselessly, because trainees can only account for about a third of staffing and the FAA says it needs to keep to a budget.
While the agency continues to hire aggressively in the short term, it does not plan to replace every controller it loses because doing so would flood the workforce with on-the-job trainees and cause financial strain.
To reconcile the hiring issues, the FAA states that it is increasing productivity, which will allow it to reduce staffing requirements by 10 percent, even as air traffic increases.
What do controllers think of recent policy? Well, a 2003 Employee Attitude Survey showed a few positive comments, but most were negative.
"To say that I am completely demoralized," wrote one, "is a gross understatement."
The problem with trainees
The FAA has changed the training process to make it quicker and cheaper. A nine-week screening, which weeded out nearly half of applicants, has been reduced to an eight-hour test that more than 90 percent pass.
Simulators have been brought into classrooms to help training. And the FAA is hoping other efficiencies will help reduce on-the-job training to just two to three years.
That's a cause of some ire: Some controllers, who are now training their replacements, say on-the-job training can't be accelerated.
One reason is that trainees are being used to counteract staffing shortages. Once they learn one of the (average of five to seven) jobs they're asked to master in the training process, they're eligible to work that job. Of course, this slows their training for the rest of the jobs, a fact that even the FAA concedes.
"It's kind of a "robbing Peter to pay Paul' kind of thing, where they need to kind of get through a day now because they don't have enough staffing," says Doug Church, national spokesman for NATCA.
The FAA saves money by paying trainees less, but maintains that doesn't affect its recruiting. Others tell a different story. According Church, a new trainee makes around $10,000 over the first six to eight months.
"Many of them have to sleep on couches in people's garages," Church says. "That is having an effect on their ability to hold on to these people."
The average salary for a new hire in on-the-job training is $31,700, according to a Department of Transportation report. While the FAA says a fully trained controller will make an average of $50,000 after one year and $94,000 after five, Church says the figures are exaggerated.
"Fifty thousand dollars is the highest you could possibly make; it's not the average after one year," he says. "Same with this nonsense about $90,000 after five years.
"We're talking about places like Fremont, Calif., where the cost of living is the highest in the country, or New York. We're not talking about Colorado Springs. It's propaganda to try to show that we're overpaid."
Surely, new hires won't make what controllers used to; in 2004, the average base salary of a fully certified controller was $113,000.
Tim Davis has seen trainees struggling. Davis is a controller at Colorado Springs' air traffic control tower and a representative for NATCA. His facility has been flooded with trainees, but he doesn't expect all to finish the program.
"I have two trainees who have to have a second job just to make ends meet," he says.
Some trainees in the Springs are military veterans who thought being a controller would give them a high enough salary to provide for their families. Others went to college for air traffic control and now have debt to pay off.
"We're sitting there as journeymen controllers, watching what these kids are going through, and we feel terrible for them," Davis says. "It's just wrong to be treated this way as what we call a "safety position' in the government."
Why the big need?
According to FAA documents, a substantial part of the staffing problem is not a lack of controllers, but rather that controllers are in the wrong place. When held to a union contract, the FAA had to have a certain amount of staff in each facility.
According to a February 2007 report from the Department of Transportation, the FAA has contracted with MITRE Corp., a national scientific and technical nonprofit, to develop realistic staffing ranges for its air traffic facilities. The report states that the FAA has had staffing models since the 1970s, but the models relied more on a national staffing number than the amount of employees needed at each facility.
MITRE's report, expected to take into account the full complexity of each facility, is scheduled for completion at the end of 2008.
In the meantime, the FAA has established staffing ranges for all facilities. For Pueblo, this means working with 11 to 13 controllers. The union contract specified 12 in 1999 when daily traffic accounted for about a third of what it is today.
Understaffed facilities might hope for relief from experienced controllers, who will take much less time than new hires to learn a new facility's complexities. Pueblo has seen some of these controllers lately; some will stay for about a year to ease staffing problems.
But many controllers say it's not realistic to assume that staff will relocate where they're needed most, because there's little incentive to move. Controllers can maintain their current salary if they stay put. Moving might mean their salary is dictated by the new pay scale, so they actually could be paid less even if they move to a busier facility.
That effect has strained markets, Sanford says.
"There are a lot of people at lower-activity towers that would be interested in moving up, if they could move under the old pay scale," he says.
Faith in the future
Forget hiring more controllers, or redistributing them. Why not eliminate the need for so many of them?
The FAA says it's hoping to automate most air traffic control by 2025. The new system is called Next Gen, and to the FAA, it's the Holy Grail.
"They're talking about a 10-year-old kid with one hour worth of lessons can actually take off on a Cessna from one airport and land it in another one," says Rep. Salazar. "So, that's how automated this thing could be."
It's a long way off, but the FAA is enthusiastic.
"The future system will rely heavily on automation for routine tasks, and the role of pilots, controllers, flow managers and dispatchers will progress from routine tasks to managing exceptions and the unexpected," an FAA report states.
That's quite a vision, considering that today in Pueblo, controllers are using 1960s technology, according to Sanford. Davis says in Colorado Springs, it's from the 1980s. Even in Denver, controllers still record vital information on paper strips.
"It's ludicrous, the things that I can work on here at home, and I go to work and it's like, "Why am I working on this antiquated crap?'" Davis says.
Controllers do a lot of work in their head, and some use windows to view the path of planes.
Just a slight bump in technology would mean a lot to many facilities. In Pueblo, for instance, controllers aren't worrying about automation. They'd be happy with new radar screens.
"It's like the difference between a black-and-white TV and a flat-screen plasma," Sanford says.
Where's the money?
In 2003, the FAA formed the Air Traffic Organization, a separate organization within the FAA that helps it "run more like a business." Basically, the ATO helps the FAA achieve one of its biggest goals: saving money.
The FAA says it needs to cut costs. After all, it will need to pay to hire and train all those replacements an expected $18.2 million cost in fiscal year 2007 alone.
There's also the problem of fewer dollars coming from airline ticket taxes. Since 9/11, more airlines have been offering cheaper fares, which translates to less money for the FAA. Compounding the problem: Airlines have also been flying smaller planes, which translates to more traffic.
The FAA says it's saving money out of necessity, but not everyone agrees that should be a top priority.
"I understand that we need to be fiscally responsible, but I mean, there's needs out there that are not being met," Rep. Salazar says.
In an ironic twist, Salazar says Congress is willing to give more money to the FAA but FAA leaders don't want it.
"The FAA could have the budget," he says. "We had authorized $3 billion, they only asked for $2.5 billion, and we know that there are a lot of needs that are not being met. And we can't understand why they're doing this, and we couldn't get the answer this morning."
Salazar says one reason Congress is offering extra money is to update facilities invaded by mold and asbestos. But instead of accepting those funds, the FAA has found a number of ways to further cut costs, including:
having controllers go to fewer meetings, which means they offer less input and spend less time conducting union business on the clock;
scrutinizing injured and disabled claims by workers, and getting them back to work sooner even if their claims are found to be legitimate;
reducing workforce, particularly "nonsafety" personnel;
hiring more part-time controllers and imposing more split shifts;
reducing separation standards;
and privatizing more of its operations.
Multiple FAA reports make no mention of how their new policies could affect employee morale.
And then there's the impact on safety. For instance, according to Salazar and several controllers, the FAA is considering tasking Denver employees with running Colorado Springs and Pueblo's radar. Pueblo air traffic manager Dave Zimmers says it's a good idea that would save money, give his facility more space and reduce the amount of training needed for new hires.
"You can run a TRACON [radar] from anywhere," he says. "I mean, you can theoretically move that thing to Guam."
Coulter, however, says the move doesn't make sense because the traffic at those three airports is so different. With Pueblo largely used for training, for instance, it sees the same planes constantly landing and taking off; that's quite different from in Denver and Colorado Springs, with their large passenger jets just making a pit stop.
"People in Pueblo do things that, if I took a controller from Denver TRACON down there, they would sit there and say, "You people are nuts.' But it's local knowledge," Coulter says.
Consolidations make sense in areas like New York or Washington, D.C., Coulter says, because traffic is similar and cooperation is needed.
Controllers see other problems with the FAA's new cost-cutting measures, not least of which is loosening safety standards. According to an FAA report, airplanes once required to have 2,000 feet of vertical space between them now must have only 1,000 feet of separation. The move is expected to save air carriers $5 billion in fuel costs through 2016.
If it sounds like the FAA is turning into a business, consider: Many people are concerned that it actually wants to sell out to a business.
The privatizing option
Actually, the FAA started to privatize in 1982, when Congress authorized the FAA to begin putting certain towers that had been closed after the mass controller firing under the auspices of private companies.
The program grew, and in fiscal year 1999, Congress allowed the FAA to expand the program further. By February 2007, there were 210 contract towers.
The FAA also negotiated a deal with Lockheed Martin in 2005, after the largest non-military, federal outsourcing competition in history. Lockheed took over the FAA's 58 automated flight-service stations, which provide weather and aeronautical information to pilots of small aircraft. Lockheed merged those centers, reducing the number to fewer than 20. The FAA says that move will save $2.2 billion over 13 years.
According to Salazar's office, the House Oversight and Investigations subcommittee will conduct a hearing on the flight-service stations in September, and Lockheed Martin's safety practices are expected to be examined.
Already, some are voicing concern.
"They're trying to privatize the FAA, and that's what we've been hearing," Salazar says. "That's what the administration's main focus has been since they've been here, is to privatize many branches of government because they say they can save money.
"But in the long term, if you look at it, we can use the other argument. If you look at what happened, for example, in Iraq, we have private contractors who are earning $200,000 for doing the same kind of work in security as a soldier who is only earning $30,000."
From Coulter's point of view, the idea of privatizing the FAA is criminal.
"For the most part, [the FAA] has served the U.S. taxpayer very, very well," he says. "To see that sold off, as a taxpayer, I'm livid. That's too valuable of a resource to sell off."
Coulter even wonders if the issue boils down to a conflict of interest. Some FAA executives have left to work in the aviation business.
Russell Chew, former chief operating officer for the Air Traffic Organization, went to work for JetBlue in March but not before he helped implement new FAA air-space and traffic policies expected to save air carriers billions. Monte Belger, a former FAA senior executive and acting deputy administrator, now works for Lockheed Martin.
How it hits home
The FAA has reached a critical point. If it mismanages the replacement of retirees, America's skies could become a lot less friendly. That means an increase in airport delays, and, perhaps, plane crashes.
From 2012 to 2018, the FAA projects a 27 percent increase in domestic air traffic. If nothing changes, that means a 62 percent increase in delay, since controllers can only handle so much traffic at one time. If they're too busy, planes just have to wait, even if runway space is available.
In May 2007, fewer than 78 percent of flights were on time, according to the Bureau of Transportation Statistics. Denver was slightly above average, and much better than, say, New York's LaGuardia Airport, where just over 63 percent of flights were on time.
Then there are the accidents. The National Transportation Safety Board has a long list of accident and near-accident investigations.
Of course, they're not all caused by controllers. Last year in Lexington, Ky., an airplane used a runway that was too short and crashed, killing 49 people. That was blamed on pilot error. In February at DIA, a snowplow driver almost got himself killed by crossing a runway without clearance, as a plane was landing. But in May in San Francisco, a seasoned controller caused a near-miss between two planes on one runway.
Air carriers had nine accidents in June alone, though most weren't fatal.
While some types of errors made by controllers have gone down over the years, operational errors (errors made by controllers in which two planes get too close to each other) have risen from 894 in 1998 to 1,304 in 2006, according to FAA data.
The odds of an accident on a commercial jet are still a fraction of a decimal. In fact, over the past three years, accident trends have been at an all-time low. It is safe to fly.
But with decades of experience walking out the doors of America's air traffic control centers, and the FAA struggling to adjust to changing times, the question is: Will it remain safe? email@example.com