“Governments never learn. Only people learn.” — Milton Friedman
Lately there has been much in the news about the misconduct of air traffic controllers. There have been reports of controllers sleeping on the job and watching movies in the control tower. This is only the surface of a corrupt, union controlled, bureaucratic air traffic control system operated by the Federal Aviation Administration.
Federal involvement in air traffic control has a long history. The 1926 Air Commerce Act tasked the Department of Commerce with issuing and enforcing air traffic rules, licensing pilots, certifying aircraft, establishing airways, and operating aids for air navigation. In the mid-1930s, the federal Bureau of Air Commerce took over the initial air traffic control centers for en route tracking created by the airlines, which complemented the operation of local control towers by municipal authorities. During the 1940s, the federal government began taking over the operation of local control towers, and following World War II all aspects of air traffic control became federal.
Airport development took a different path. In the early years of commercial aviation, some private airports (e.g., Burbank, California) existed alongside those established by state and local governments. Today, virtually all U.S. commercial airports are owned by state and local governments. The federal government's role has been to regulate and subsidize those facilities. Federal financial aid to airports began with work relief programs in the 1930s, and was followed by the Federal Airport Act of 1946, which provided $500 million in airport grants to state and local governments over seven years. These airports are not to be confused with the hundreds of General Aviation airports and airfields across the United States. These air fields are privately owned and operated and service the owners of non-commercial aircraft.
General aviation airports normally have one runway, hanger facilities and a tower manned on an irregular basis by private air traffic controllers. They are subject to all safety regulations of the FAA and ICAO and usually have stated hours of operation.
The coming of jet aircraft and the large number of aviation accidents spurred Congress to pass the Federal Aviation Act of 1958, which created the Federal Aviation Administration (FAA). The new administration replaced previous federal agencies involved in air traffic control and airport development.
Congress started taxing the commercial aviation industry soon after it was established. It passed an excise tax on gasoline and aviation fuels in 1932 and an excise on airline passenger tickets in 1941. The revenue from these levies went into the federal government's general fund. That changed in 1970 when Congress passed legislation creating an Airport and Airway Trust Fund, which had dedicated streams of revenue to be used for air traffic control and federal aid to airports. Trust Fund revenue sources included ticket taxes on domestic and international flights, taxes on fuels, and various fees.
The Airport and Airway Trust Fund currently raises more than $12 billion annually from a 7.5 percent tax on domestic airline tickets, a tax on each segment flown, taxes on gasoline and jet fuel, international departure and arrival taxes, and half a dozen other fees. Trust Fund revenues pay for almost four-fifths of the FAA's $16.4 billion budget, with the balance coming from general federal funds covering the FAA's safety regulatory and miscellaneous other activities.
While air traffic control is an increasingly technology-intensive industry, labor union issues have long played an important role in the ATC system. A period of labor unrest began in the late 1960s as FAA controllers pushed for job improvements and official status as an employee union. In 1969, about 500 members of the Professional Air Traffic Controllers Organization stayed home "sick" causing air service interruptions. The following year, 3,000 PATCO members took part in another "sickout" or illegal strike, which caused chaos for the nation's air traffic.3
Labor problems continued during the 1970s, with various work slowdowns and union protests over contract issues. Then in 1981, PATCO declared a major system-wide illegal strike after negotiations on a new contract broke down. That prompted President Ronald Reagan to order controllers to return to work within 48 hours or else face termination. More than 11,000 controllers refused to return to work and were fired by Reagan and initially banned from federal service. PATCO was dissolved and a new controllers union was created in 1987, the National Air Traffic Controllers Association (NATCA).
Today, an important aspect of the federal ATC system is the high labor costs. In 2010, the operations portion of FAA had about 43,000 workers who earned a total of $6.5 billion in wages and benefits, or about $151,000 per worker.4 Just looking at controllers, a 2005 FAA study found that compensation packages averaged $166,000 annually.5 Labor costs account for two-thirds of the cost of FAA operations.
While organized labor has created management challenges for the FAA, so has the implementation of new technologies. Delays and cost overruns on major technology projects have been common. For example, the Advanced Automation System project was launched in the early 1980s and was originally expected to cost $2.5 billion and be completed by 1996. But by 1994, estimated project costs had soared to $7.6 billion and the project was seven years behind schedule.7 The FAA terminated some parts of the AAS program and restructured others, but $1.5 billion of spending ended up being completely wasted.
More recently, a 2005 study by the Department of Transportation's Office of Inspector General looked at 16 major air traffic control upgrade projects and found that the combined costs had risen from $8.9 billion to $14.5 billion.8 The cost of the Standard Terminal Automation Replacement System project had jumped 194 percent to $2.7 billion and was seven years behind schedule. The OIG said that the STARS project was "facing obsolescence" even before it was completed.9 Meanwhile, the cost of the Wide Area Augmentation System project had jumped 274 percent to $3.3 billion and was 12 years behind schedule. A Government Accountability Office analysis in 2005 found similar cost overruns and delays in these projects.
Delays and cost overruns have not been uncommon in federally subsidized airport projects either. For example, Denver's new international airport finally opened in 1995 after many delays and huge cost overruns. The project was originally supposed to cost $1.7 billion but ended up costing almost three times as much at $4.9 billion, with $685 million coming from federal taxpayers.
In sum, federal funding of airports and the operation of the nation's ATC system have not been models of efficiency over the decades. There is large room for improvement in the management of the nation's aviation infrastructure.
The way to address all three of these organizational problems is to take the ATC system out of the federal budget process and make it a self-supporting entity, funded directly by its customers. Variants of this commercialization approach have been recommended by a series of federal studies and commissions over the past 15 years.
As part of Vice President Al Gore's efforts at "reinventing government" in the 1990s, for example, the Clinton administration proposed turning the ATC system into a separate, self-funded, nonprofit government corporation within the Department of Transportation. The 1997 National Civil Aviation Review Commission, which was chaired by Norman Mineta, similarly proposed moving toward a self-supporting air traffic control organization.
Commercialization would entail shifting from aviation-related taxes paid to the U.S. Treasury to fees for ATC services paid directly by customers to a new self-supporting Air Traffic Organization. This change would allow fees to grow in proportion to the growth of flight activity, rather than being tied to a less-stable variable, such as fuel prices or airline ticket prices. Moreover, a predictable revenue stream that was not subject to the federal budget process would provide the basis for the ATO to issue long-term bonds for funding capital investments.
Commercialization would also address the management problems that have plagued the FAA's efforts to modernize. A non-civil-service ATO could attract the best private-sector managers and engineers skilled at implementing complex technology projects. Such an ATO could hire, fire, and compensate its employees as other high-tech businesses do. Private sector managers would have an incentive to ask tough questions about whether new investments offered real value for the money, a process that often doesn't occur at the FAA or in Congress.
In addition, a separate, self-supporting ATO—no longer part of the FAA—would be overseen at arm's length for aviation safety by the remaining FAA. Numerous studies have pointed out that the FAA's air-safety role is compromised when it comes to the ATC system, since that system is operated "in-house" by a different branch of the same FAA. All other players in aviation—pilots, mechanics, aircraft manufacturers, airlines, and so forth—are regulated at arm's length for safety by the FAA. This separation of ATC operations from safety regulation is especially critical given the major changes entailed by shifting to the semi-automated next generation (NextGen) and Free Flight, where numerous safety versus capacity questions will need to be addressed in a rigorous and transparent manner.
Finally, a self-supporting ATO would address the political obstacles to improving system efficiency, such as making decisions to close facilities. By passing the enabling legislation for ATC reform, Congress would delegate such contentious issues to the customer-oriented ATO organization.
During the past two decades, nearly 50 governments have commercialized their air traffic control systems. That means they have separated their ATC activities from their transport ministries, removed them from the civil service, and made them self-supporting from fees charged to aircraft operators. These new Air Navigation Service Providers (ANSPs) are usually regulated at arm's length by their government's aviation safety agency.
Britain's ATC system has been commercialized by means of a "public-private partnership." National Air Traffic Services is a jointly owned company, with British airlines owning 42 percent, the airport company BAA owning 4 percent, employees owning 5 percent, and the government owning the remaining minority stake. NATS is operated on a not-for-profit basis.
Canada's ATC system has been fully commercialized. In 1996, Canada set up a private, nonprofit ATC corporation, Nav Canada, which is self-supporting from charges on aviation users. The Canadian system has been widely praised for its sound finances, solid management, and its investment in new technologies. The Canadian system is a very good reform model for the United States to consider.
Nav Canada's corporate board is composed largely of aviation stakeholders. It has 4 seats for the airlines, 3 for the government, 2 for employees, and 1 for the non-commercial aviation industry. Those 10 stakeholders select 4 directors from outside aviation, and then those 14 select the company president, who becomes the 15th board member. To further strengthen governance, neither elected officials nor anyone connected with suppliers to Nav Canada can serve on the board. Nav Canada also has a 20-member outside Advisory Committee.
A number of studies have found that ATC commercialization has generally resulted in improvements to service quality, better management, and reduced costs. At the same time, air safety has remained the same or improved in the countries that have pursued reforms to set up independent ANSP organizations.
A thorough 2009 report by Glen McDougall and Alasdair Roberts (Commercializing Air Traffic Control: Have the Reforms Worked?") compared the performance of 10 commercialized ATC systems and the FAA during the 1997 to 2004 period. They looked at large amounts of performance and safety data from the systems in the various countries and conducted over 200 interviews with managers, workers, and users of the different systems. The researchers found:
“ANSP commercialization has generally achieved its objectives. Service quality has improved in most cases. Several ANSPs have successfully modernized workplace technologies. The safety records of ANSPs are not adversely affected by commercialization and in some cases safety is improved. Costs are generally reduced, sometimes significantly. Other risks of commercialization—such as erosion of accountability to government, deterioration of labor relations, or worsened relationships between civil and military air traffic controllers—have not materialized.”
Other nations and regions that have privatized their ATC systems are: EUROCONTROL, COCESNA and Deutsche Flugsicherung (DFS). EUROCONTROL and COCESNA are regional organizations for Europe and Central America respectively while DFS is the German state owned ATC company. None of these organizations use union or civil service air traffic controllers. COCESNA is responsible for all ATC above 17,000 feet over Guatemala, El Salvador, Honduras, Nicaragua, Costa Rica and Belize. All airports in the countries have their own ATC controllers to deal with approach and take off actions.
I have worked as a contractor for the FAA providing Safety of Navigation airfield surveys and for airports in Guatemala, El Salvador, Honduras, Nicaragua, Costa Rica, Belize and Venezuela providing airport surveys to develop Global Navigation Satellite System (GNSS) approach and departure procedures. I have also given several seminars on this subject in Port-a Prince, Trinidad-Tobago, San Salvador, Quito, Ecuador, Caracas, Venezuela Miami, Jeju Island, South Korea and at the FAA headquarters in Washington, D.C. Click here for a PDF file of the presentation slides.
I recall sitting in the conference room at the FAA with the Jane Garvey, the then administrator, and the assistant administrator for international projects, my client, having a discussion about the implantation the GNSS for approach and departure procedures. I was a bit shocked when I heard Garvey’s response that no such implementation could be made at commercial airports without the approval of the ATC controller’s union. He said any automation technology that would have an influence of the jobs of the controllers has to be signed off by the union. In essence the NATCA runs the FAA when it comes to cost saving or safety issues — they want to protect the jobs of the controllers.
On October 26, 2009 US Department of Transportation Inspector General Calvin Scovel and US Government Accountability Office Director of Civil Aviation Issues Gerald Dillingham told Congress that the FAA faces considerable challenges in implementing a satellite-based NextGen ATC system, ranging from delays in approving new procedures and technology to skepticism among airlines regarding investment in new equipment. Testifying before the House of Representatives aviation subcommittee, Scovel warned that "the cost, schedule and benefits for NextGen are uncertain". Dillingham added that the "FAA faces cultural and organizational challenges in implementing NextGen capabilities".
Keep in mind that this year the FAA got $700 million dollars for ATC controller’s salaries. This, like the TSA, is an area where the civil service unions are not only dictating the terms of employment and compensation, but also impacting the efficiency and safety of our air traffic control system. 50 other nations are not taking this path.
For the United States, a commercialized ATC organization would be more likely than the FAA to efficiently implement the major aviation infrastructure advances that the nation desperately needs. Air traffic control is more complex and dynamic than ever, and it needs to be managed in the sort of efficient and flexible manner that only a commercialized environment can offer. Countries like Canada have shown the way forward for air traffic control, and U.S. policymakers should adopt the proven organizational reforms that have been implemented abroad.