Reason Foundation

Reason Foundation

Air Traffic Control Reform Newsletter #39

Topics include: Nav Canada's 10th anniversary; financing the NextGen ATC system; clear thinking on ATC User Fees; beefing up the controller workforce; and other news.

Robert Poole
November 1, 2006

In this issue:

Nav Canada at 10 Years

On Nov. 1, 1996, Canada's ATC system was transitioned from a government department to a not-for-profit, stakeholder-governed corporation called Nav Canada. Controllers and managers went home on Oct. 31 as civil servants and returned the next day as employees of the new company. A board of directors representing aviation stakeholders replaced the Canadian parliament as the primary governance mechanism, and Nav Canada began to be regulated for safety, at arms length, by the government. Over a two-year transition period, Canada's ticket tax was phased out and replaced by ATC user fees paid directly to Nav Canada. The company went to the capital markets and raised several billion dollars to fund its modernization program.

How has Nav Canada done over its first decade? Aviation groups and the Canadian media give the company high marks. So do several independent studies previously reported on in this newsletter, one from January 2006 by MBS Ottawa and three universities ( and another several months ago by Clint Oster and John Strong, published by the IBM Center for the Business of Government (

CBC News (Oct. 31) quotes Nav Canada CEO John Crichton on the organization's evolution since being a government entity. "I don't think a lot of people understand that government really is different. It runs completely differently than businesses do." He notes that about 1,000 employees (mostly middle managers), about 15% of the total, could not make the transition and retired, transferred to government jobs, or were laid off. But the company also increased the size of the controller workforce by 250, a 10% increase, to remedy long-standing shortfalls that had plagued Canadian ATC.

As an entrepreneurial company, Nav Canada has developed improved technologies and, in addition to using them in its own operations, has marketed them to other air navigation service providers (ANSPs). For example, its paperless flight data processing system and display for controllers (EXCDS) is now used in the control towers at three London airports (Gatwick, Stansted, and Luton, with Heathrow soon to follow) and at Copenhagen airport. The U.K.'s NATS is also implementing Nav Canada's oceanic control system, GAATS, on the North Atlantic.

Safety and cost are always big issues when ATC commercialization is discussed. Nav Canada's safety record is one of steady improvement over the decade. Losses of separation, as defined in Canada, declined from 1.36 per 100,000 aircraft movements in 1996 to 0.77 in 2006. Its charges for ATC services, on a per-passenger basis, are 20% lower than the ticket tax they replaced. And over the 10-year period, it has retained an investment-grade rating on its revenue bonds.

The Oster/Strong IBM report draws seven lessons from the Nav Canada experience. Among them are the following:


Financing the Next Generation System

Financing commercialized ANSPs' capital modernization programs is very straightforward: they issue bonds based on the revenue stream from fees and charges for ATC services. The markets readily buy the bonds, because these ANSPs generally have a statutory monopoly as the country's ATC provider. Moreover, aviation is a long-term growth industry. No wonder the bonds get investment-grade ratings.

But when you try to apply that straightforward financing model to FAA's Air Traffic Organization, you quickly run into a rat's nest of complications. This was brought out in testimony by the Congressional Budget Office at the Sept. 27th hearing (House Aviation Subcommittee) that I reported on last issue. The subcommittee itself provides a good summary of these complications (see They are embedded in well-established budget-scoring rules agreed to long ago by House and Senate Budget Committees, CBO, and OMB. These rules apply to the federal budget across the board.

To illustrate, let's consider a $10 billion modernization program�say, a major portion of the NextGen system. To pay for this up front (in cash), the ATO's budget would have to leap by $10 billion in the first year. Not very likely, even if making large commitments up-front were the most cost-effective way to do it.

OK, then, what about leasing, to spread the cost out over many years? Well, budget-scorers thought of that, too. Any long-term lease (or lease-purchase deal) is scored up-front, in the year of signing, at the present value of all future lease payments. The only way out of that box is an "operating lease," but to qualify as that, the asset must not be built to unique specifications of the government, and there must be a separate private market for the asset. "These conditions may effectively preclude the FAA from using leases as a significant part of financing [NextGen]," the report rightly concludes.

Finally, what about bonding? Treasury has anticipated that, and has made sure the scoring rules provide that "debt issued by a federal entity such as FAA would be scored similar to direct, up-front appropriations. Therefore, debt-financing would not provide FAA with a budget scoring advantage relative to up-front appropriations."

It looks as if the only way out of this box would be to privatize the ATO, taking it completely out of the federal government so that neither its revenues nor its spending were in the budget. That way, the budget-scoring rules would not apply, and normal revenue-bond financing of capital modernization could take place. But why dream of that, when Democrats have just taken control of both House and Senate, and are not likely to favor anything called "privatization"?

At this point of despair, I turned to the actual CBO testimony (as opposed to the subcommittee's summary) and discovered one paragraph that had not been summarized. Under the heading of "Alternative Forms of Federal Borrowing," it described the Tennessee Valley Authority as "an example of an agency that has direct spending authority in the form of borrowing authority. TVA has the authority to issue bonds up to a specified ceiling (now $30 billion) and a source of income (the sale of electricity) that can be used to service its debt." And note that despite the Treasury's insistence that it alone should issue any and all such debt because, after all, it's the government and has the lowest borrowing cost, "The costs of [TVA's] borrowing�which are typically 30 to 40 basis points higher than the costs of comparable Treasury securities�are thus borne by the users of the electricity it generates." Testimony that same day from Goldman Sachs pointed out that TVA gets something for those slightly higher costs: the flexibility to refinance or call in the bonds if and when it makes sense to do so.

Just to be sure about the budget treatment accorded TVA, I queried staff at CBO. They replied that TVA's spending does show up in the federal budget (but those monies are not appropriated by Congress, since customers pay TVA directly�this is simply a matter of record-keeping). But�of great importance�the proceeds of TVA bonds are not�repeat not�counted as a receipt in the budget.

So let's suppose that a Democratic Congress concluded that the $20 billion or so in capital costs of implementing NextGen needed to be financed�but without busting the federal budget due to budget-scoring rules. They could spin off the ATO as a TVA-like federal corporation, selling its services directly to its customers and issuing its own revenue bonds just like TVA does. No $10-20 billion spike in revenue would show up in the federal budget when the bonds were issued, and the expenditures of the proceeds�the new facilities and equipment needed for NextGen�would show up only in the ATO's annual budget over many years, reported nominally as part of the federal budget, like TVA's.

The TVA was created long before the current budget-scoring rules, and my impression from talking with people at Treasury, OMB, and CBO is that they are very wedded to those rules and really don't want to see exceptions made. But it's up to Congress to decide these matters. If they decide that implementing a very expensive, technologically challenging Next Generation ATC system can best be done by a TVA-like government corporation, they are certainly free to legislate accordingly.


Let's Have a Serious Discussion on ATC User Fees

The general aviation campaign against ATC user fees is getting very tiresome. It is deliberately distorting the issue in order to (1) stir up every private pilot in the land with unrealistic fears, in hopes of mobilizing an NRA-like army at the grass-roots, and (2) create such an overwhelming negative spin around the issue that the FAA and Congress will decide it's too hot to handle.

AOPA, GAMA, and NBAA are getting the greatest mileage out of pretending that the airlines' plan is synonymous with user fees. The ATA has correctly pointed out the large disparity in use of the IFR system by business aviation in relation to how much they pay in aviation taxes (2.3% of the user taxes but 15.6% of the usage)�an important point in any factual discussion of how to pay for the system. But just because the airlines would ideally like to achieve a system in which all "blips" pay the same rate per mile or per hour or per landing does not mean that's the only alternative. We haven't even seen the much-discussed (but not yet released) FAA proposal, and there are dozens of other user-fee systems in use today�in Canada, Europe, Australia, etc. But because the ATA approach would lead to the biggest cost-shift between bizjets and airlines, it obviously has the most scare value, so that's the only kind of user fee these guys talk about.

The other easy scare factor is to raise the specter of the slippery slope. Even though neither FAA nor ATA is proposing user fees for recreational general aviation, the type of flying done by most AOPA members, AOPA and now even the Experimental Aircraft Association are scaring their members by saying it's inevitable that fees will spread to everyone, even to fees for safety-related items like weather briefings. But no user fees for anybody will be implemented without Congress giving the OK. I have no doubt that AOPA will retain its lobbying clout as far into the future as we can imagine, so whatever Congress may decide about airline and bizjet user fees next year, no way are they going to approve user fees for recreational and VFR flying.

Instead of a propaganda war, it's high time we had a serious discussion of who should pay what, as Aviation Week called for last week (Nov. 6, editorial). I take seriously NBAA's point that a great many bizjets don't use the crowded airspace at ORD or JFK or LAX, so it's not clear to me that every blip in the entire system should pay the same terminal-area ATC charge. That's especially relevant to the thousands of very light jets (VLJs) that will enter service over the next decade. If they are mostly serving small airports out of congested traffic patterns, I'm all in favor of their paying lower terminal-airspace charges. However, we can't really work this out until the ATO releases accurate cost-of-service data. For example, if it costs the ATC system the same to guide a bizjet through NY/NJ terminal airspace to a landing at Teterboro as it does to get it to Newark, then the same charge probably should apply.

While we're on the subject of terminal-area charges, I've taken some heat from my economist friends for saying we might need to include weight in some of these formulas, as most other countries do. There is no relationship of weight to cost, they say, chiding me for being too pragmatic. But consider this factor: the rate at which you can land planes at busy airports depends on how much space you need to leave between them, because of tip vortices that pose a hazard to the following plane. And which planes produce the biggest and longest-lasting vortices? The "heavies": the 767s, 777s, A-330s, etc., not the Citations and Learjets. So there may be a legitimate cost-based reason for taking weight into account, to some extent, in devising terminal-area charges. (And please note that few if any ATC charging systems use unvarnished weight in the formula; they typically use the square root or some other root, which means that a plane twice the weight doesn't pay anywhere near twice as much.)

I'm also not enamored of the idea of basing en-route charges on time in the system, as both the ATA and (reportedly) the FAA formulas would do. That approach means the ANSP profits from delays�the longer a flight drags on, the higher its revenue. Talk about perverse incentives! It also adds to the complexity of record-keeping and billing�another grossly exaggerated scare point used by NBAA and its allies. The charging formula should be simple, based on a few parameters that can be taken directly off a plane's flight plan (e.g., great-circle miles for the route from origin to destination; if weight is used, something standard like gross take-off weight).

The FAA should not release its "cost-based" user-fee proposal until it is prepared to share the ATC cost-of-providing services data on which it is based. And then we should have a serious dialog about a fair, simple approach that bears some reasonable relationship to the underlying ATC costs and provides incentives for wise use of the system. The annual total revenue from the fees should be enough to provide the total annual budget of the ATO, both capital and operating. A bondable revenue stream of this sort will make it possible to finance the large capital costs of the NextGen system. (If military and other government users balk at paying fees for their use of the system, then the general fund will have to provide their share of the costs.)

By the way, if you would like to see a debate on ATC user fees for bizjets between me and Ed Bolen, president of NBAA, check out the November issue of Professional Pilot, which invited us to hold forth on this important subject in its pages. My hat's off to publisher Murray Smith for wanting to have his readers hear both sides on this key issue. (See


Beefing Up the Controller Workforce

A reader of this newsletter who spent 20 years as a US Air Force air traffic controller offers the following suggestion to help the Air Traffic Organization cope with the coming retirement bubble (as large numbers of controllers hired in the early 1980s after the PATCO strike reach retirement age). Here are his comments, verbatim.

"I was always appalled by the 31-year age limit to get into the FAA as a controller. There is a pool of qualified replacements who are fully certified professional controllers, but many of them are not allowed to transition to the FAA ranks due to age restrictions�and that at an age when they are most experienced. The PC-20 program has been less than successful, if you ask me. I applied for several years, never getting a call. I have both tower and radar certifications at very busy USAF locations.

"I think one solution might be to keep the deployable military controller force young. (There will always be those dedicated military men and women that decide to make a career of it, and they would provide supervision of facilities.) This would allow a direct pipeline from the military to the FAA to supplement the controller workforce."

I'd be interested in responses from those at the ATO working on the controller staffing problem.


News Notes

German ATC Privatization on Hold. Despite the German parliament having passed legislation to permit the sale of 74.9% of Deutsche Flugsicherung (DFS), the country's ANSP, the country's president vetoed the sale. His legal advisors contend that the law runs counter to constitutional provisions requiring ATC to be carried out by a state organization. DFS is currently a commercialized government corporation (like TVA), consistent with those provisions. Some government and DFS sources say the decision will postpone, rather than stop, the privatization.

Nav Canada Award. Canada's privatized ANSP has received a nice 10th anniversary present in the form of yet another award. It is the 2006 winner of the J.A.D. Curdy Award, given each year by the Air Force Association of Canada for outstanding achievements in civil aviation in Canada. Ted Mahood, the AFAC's president, said "You have delivered real value to the airlines and to business and general aviation, all the while reducing the costs of air navigation services for the flying public. Nav Canada is a privatization success story, and all Canadians should be very proud of our achievements in the field of civil aviation."

NATS Sale Off. The consortium of seven British airlines that owns 42% of public-private National Air Traffic Services, the commercialized ANSP of the United Kingdom, has decided not to proceed with its proposed sale of their stake to outside investors. The sale had been opposed by company unions and some members of Parliament.

Tunisia Commercializes ATC. The 45th member of the Civil Air Navigation Services Organization is Tunisia's Civil Aviation and Airports Authority (OACA). CANSO announced its new member on Nov. 10th. OACA is a financially self-supporting ANSP housed within the country's Ministry of Transport, similar to the situation in France.

Your Editor Honored. As editor of this newsletter, and a long-time researcher on ATC reform issues, I was pleasantly surprised to open the third-quarter 2006 issue of Air Traffic Management and see my name and photo on p. 41, as one of a number of people to be singled out as "ATM's Finest." While calling the choice "controversial," the write-up went on to say that Reason Foundation "has been hugely influential in arguing the merits of privatization." It adds, "As study after study is presently showing �and what Poole has been advocating since the early 1980s � commercialized or privatized air traffic management approaches look set to be the way the industry is heading."

Air Traffic Visualized. I'll admit to not being a regular YouTube visitor, but I commend to you the nicely done video of U.S. air traffic, based on data from FAA's ETMS. You can find it at:


Quotable Quote

"Although rarely discussed in public, pushing the air transportation system to handle substantially more capacity with current technology has serious potential safety implications. Since those implications are unthinkable, severe constraints will have to be imposed on the system to protect public safety. Besides restricting traffic, these actions could make the price of an airline ticket exorbitantly high and effectively put air travel out of reach for a whole segment of society."

—George L. Donohue, George Mason University (and former FAA Associate Administrator for Acquisitions and Research)

"[I propose that] ICAO take a bold position with regard to the so-called protection dates for the terrestrial navaids. ICAO should be pro-active in encouraging the decommissioning of all of these aids (particularly VOR, DME, and all forms of NDB) so as to relieve states of the burdens of maintenance, push users into upgrading their avionics with the attendant safety benefits, free aeronautical spectrum for new uses and/or reallocation for other safety radio services . . . . The really big deal is for ICAO and our ANSPs to act to curtail the seemingly endless lives of the generation after generation of navaids that populate our landscapes."

—Frank Frisbie, Apptis, in The Journal of Air Traffic Control, April-June 2006

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

Print This