Full Text: GAO Official's Testimony at House Transportation and Infrastructure Aviation Subcommittee Hearing

UNITED STATES GENERAL ACCOUNTING OFFICE

TESTIMONY

AIRPORT AND AIRWAY TRUST FUND

ISSUES RELATED TO DETERMINING HOW BEST TO FINANCE FAA

STATEMENT OF JOHN H. ANDERSON, JR.,

GAO/T-RCED-97-59

February 5, 1997

Mr. Chairman and Members of the Committee:

[1] We appreciate the opportunity to testify on issues related to the financing of the Federal Aviation Administration (FAA). On December 31, 1996, the government's authority to collect the taxes that finance the Airport and Airway Trust Fund, which has historically provided about three-quarters of FAA's funding, lapsed. In December 1996, we reported to you, Mr. Chairman, and other members of the Senate and House on the status of the Trust Fund and on a proposal by a coalition of the nation's largest airlines to replace the tax on domestic airline tickets, which has been the Trust Fund's primary source of revenue, with fees on domestic operations. /1/ The coalition airlines /2/ contend that they pay for more than their fair share of the costs incurred by FAA in running the airport and airway system and that competing low-fare airlines underpay.

[2] Our testimony today discusses the (1) status of the Trust Fund, (2) issues raised by the coalition's proposal, (3) potential effects of the coalition's proposal on domestic competition, and (4) potential competitive impacts of alternative options for financing FAA. Our main points are as follows:

-- On December 9, 1996, we reported that, based on estimates

-- To the extent possible, commercial users of the nation's

-- While many factors drive FAA's costs, such as the number of

-- If the Congress decides to replace the ticket tax with a

BACKGROUND

[3] The Airport and Airway Trust Fund was established by the Airport and Airway Revenue Act of 1970 (P.L. 91-258) to finance FAA's investments in the airport and airway system, such as construction and safety improvements at airports and technological upgrades to the air traffic control system. Historically, about 87 percent of the tax revenues for the Trust Fund has come from a tax on domestic airline tickets. Before it lapsed at the end of 1996, the tax was 10 percent of the fares paid. The remainder of the Trust Fund was financed by a $6 per passenger charge on flights departing the United States for international destinations, a 6.25 percent charge on the amount paid to transport domestic cargo by air, a 15-cents-per-gallon charge on purchases of noncommercial aviation gasoline, and a 17.5-cents-per- gallon charge on purchases of noncommercial jet fuel.

STATUS OF THE TRUST FUND

[4] In fiscal year 1997, under current law, the Trust Fund is to provide $5.3 billion (62 percent) of FAA's budget of $8.6 billion. /5/ FAA and the Treasury originally estimated that if the taxes that finance the Trust Fund lapsed on December 31, 1996, the Trust Fund would be about $1 billion short of the funding needed to finance its portion of FAA's budget. However, in late January 1997, the Treasury acknowledged that it had miscalculated the balance of the Trust Fund because the agency incorrectly assumed that airlines would pay most of the taxes that they owed for the last several months of 1996 by the end of the year. However, under a regulatory interpretation provided to the airlines by IRS, they do not have to make most of those payments until late February 1997, and most airlines have not as yet paid. When these taxes are paid, they cannot be transferred from the General Fund to the Trust Fund because the authority to do so lapsed at the end of 1996. As a result, the Trust Fund may be about $2 billion short of the funding needed to finance its portion of FAA's fiscal year 1997 budget.

[5] FAA and Treasury officials are still attempting to determine the precise amount of the shortfall. However, based on FAA and Treasury data, a shortfall of $2 billion would mean that in order to pay its workforces through the end of fiscal year 1997, FAA would have to stop making new capital commitments about March 1997. Reinstating the authority to move tax receipts from the General Fund to the Trust Fund by March would provide FAA with money to fund new capital commitments to late July 1997. If the Congress reinstates the taxes (or some other financing mechanism) by July, the Trust Fund should be able to fully finance its portion of FAA's fiscal year 1997 budget.

WHETHER TICKET TAX RESULTS IN USERS PAYING THEIR FAIR SHARE OF THE

[6] FAA is responsible for a wide range of functions, from certifying new aircraft to inspecting the existing fleet to providing air traffic services, such as controlling takeoffs and landings and managing the flow of aircraft between airports. Over the past decade, the growth of domestic and international air travel has greatly increased the demand for FAA's services. At the same time, FAA must operate in an environment of increasingly tight federal resources. In this context, we have generally supported FAA's consideration of charging commercial users for its services and believe that the various commercial users of the nation's airspace and airports should pay their fair share of the costs that they impose on the system. /6/ In particular, we have previously suggested that FAA examine the feasibility of charging fees to new airlines for the agency's certification activities and to foreign airlines for flights that pass through our nation's airspace. In addition, to ensure full cost recovery, we have suggested that FAA consider raising the fees that it charges for the certification and surveillance of foreign repair stations.

[7] Because the airline ticket tax is based on the fares paid by travellers and not on factors that relate to system costs, it may not fairly allocate costs among the users of the airport and airway system. For example, two airlines flying the same number of passengers on the same type of aircraft from Minneapolis to Des Moines at the same time of day will impose the same costs on the airport and air traffic control system. However, because the ticket tax is based on the fares paid, the airline that charges the lower fares will pay less for the system's use. Citing such examples, the coalition airlines contend that they pay for more than their fair share of the system's costs and that competing low-fare airlines underpay.

[8] However, comparing the relative share of airlines' payments under the ticket tax to some common measures of domestic system usage does not provide conclusive evidence that the ticket tax is unfair. As figure 1 shows, the coalition airlines accounted for almost 80 percent of the total payments made under the ticket tax in 1995. Their percentage of system use was lower than this for some common indicators of system use such as domestic departures, passenger enplanements, and miles flown. However, the coalition airlines accounted for 81 percent of the fuel consumed by commercial airlines in domestic operations in 1995, another indicator of system usage. Airlines that compete with the coalition airlines, such as Southwest Airlines and America West, accounted for about 17 percent of the payments made under the ticket tax in 1995 but accounted for 21 percent of all domestic departures and 22 percent of enplanements. On the other hand, their share of miles flown and fuel used was the same as their share of ticket tax payments. Reaching definitive conclusions based on these comparisons is further complicated by the fact that most major commuter carriers are owned by or affiliated with one of the coalition airlines.

FIGURE 1: COMPARISON OF THE RELATIVE SHARE PAID UNDER THE TICKET TAX

[figure omitted]

[9] Currently, FAA has insufficient cost information to show whether the ticket tax or any of the system usage indicators shown in figure 1 would be good proxies for fairly allocating FAA's costs among commercial users. The Congress in October 1996 directed that, among other things, an independent assessment of FAA's costs be completed by February 1997 and that a national commission recommend to the Secretary of Transportation by August 1997 how best to finance FAA in light of the independent assessment. /7/ Additionally, the Congress required that we assess how costs are allocated between FAA and DOD and that we report to the Congress by April 1997. Because about 18 percent of DOD services are provided to civilian users, according to DOD, information regarding DOD's costs may also be relevant in assessing financing alternatives for FAA. As a result, better information should be available later this year on FAA's costs that will allow for an evaluation of the ticket tax and potential alternative options for financing FAA.

PROPOSAL BY LARGER AIRLINES WOULD INCREASE THE SHARE PAID BY OTHER

[10] Motivated by their belief that the ticket tax unfairly subsidizes their low-fare competitors, the coalition airlines in May 1996 proposed that the ticket tax be replaced by user fees on domestic operations. Under the proposal, airlines would pay fees for domestic operations according to the following three-part formula: (1) $4.50 per originating passenger; (2) $2.00 per jet seat on aircraft with 71 or more seats and $1.00 per seat on jets and turboprop aircraft with 70 or fewer seats; and (3) $0.005 per nonstop passenger mile. /8/

[11] By using two factors in particular -- originating passengers and nonstop passenger miles -- the formula tends to favor the larger airlines, which operate hub-and-spoke systems, at the expense of the low-fare and small airlines, which tend to operate point-to-point systems. This relationship can best be shown by example. Consider the two possible routings between St. Louis and Orlando shown in figure 2. The hubbing airline first takes the passenger from St. Louis to a hub, such as Chicago's O'Hare Airport, to connect to another flight to Orlando. The point-to-point carrier takes the St. Louis passenger nonstop to Orlando.

FIGURE 2: COMPARISON OF POTENTIAL HUBBING AND POINT-TO-POINT SERVICE

[figure omitted]

[12] The airline that picks up a passenger in St. Louis and then lands at O'Hare to transfer the passenger to another flight to Orlando has twice as many takeoffs and landings as the airline that flies nonstop between St. Louis and Orlando. As a result, the costs imposed by the hubbing airline on the air traffic control system are greater. However, by charging $4.50 per "originating" passenger the airline that flies the passenger from St. Louis to Orlando via Chicago O'Hare would pay the same amount as an airline that flies the passenger nonstop between St. Louis and Orlando, even though the hubbing carrier puts a greater burden on the system.

[13] In addition, by charging $0.005 per "nonstop passenger mile" -- or the straight-line distance between the points of origin and destination -- the formula does not charge the hubbing airlines for the circuitous routings that are common to their hub-and-spoke operations. As a result, the airline transporting a passenger 297 miles from St. Louis to O'Hare and then flying that passenger 1,157 miles to Orlando would be charged the same as an airline flying a passenger nonstop from St. Louis to Orlando, even though the hubbing carrier placed a greater burden on the air traffic control system.

[14] Because the seven largest airlines operate hub-and-spoke systems and most low-fare and small airlines operate point-to-point systems, the proposed user fee would shift the fees for using the system away from the larger airlines and onto their competitors. As shown in appendix I, for example, if this proposal had been in place in 1995 instead of the ticket tax, the cost to the nation's seven largest airlines would have been nearly $550 million less while the cost to Southwest Airlines, America West, and other low-fare and small airlines would have been about $500 million more. In addition, the coalition's proposal would charge commuter carriers $1.00 per seat while charging airlines $2.00 per seat. Because most major commuter carriers are owned by or affiliated with one of the coalition airlines, the proposal would thereby provide an additional benefit to the coalition airlines by charging commuter carriers less per seat.

[15] Implementing a proposal that would shift about $500 million in costs from one segment of the industry to another could have substantial competitive impacts. For Southwest Airlines, for example, the increased amount paid would represent about 7 percent of the airline's total passenger revenue. According to the Department of Transportation (DOT), competition from low-fare airlines such as Southwest influences airfares in markets that account for about 40 percent of domestic passengers. In addition, according to DOT, these passengers tend to be the most price sensitive. As a result, such a substantial increase in costs would likely force Southwest and the other low-fare and smaller airlines to raise their fares and could result in a reduction in passenger demand in those markets, which tend to be in the West and Southwest. To the extent that these airlines stopped serving markets that were no longer profitable, competition would be reduced. On the other hand, consumers in the East and upper Midwest, who have not experienced the entry of low- fare airlines to the same extent, could pay relatively less than they did under the ticket tax and may benefit from an increase in airline competition that may result from any increase in passenger demand, if the larger airlines passed their reduced tax payments onto consumers by reducing ticket prices.

[16] While the ticket tax might provide a competitive advantage for low-fare airlines, other public policies favor some large carriers. For example, a few large airlines control nearly all the takeoff and landing slots at the four slot-controlled airports, /9/ which give them an advantage over their competitors. Simply eliminating the potential "subsidy" to low-fare airlines created by the ticket tax, while leaving the other policies in place that provide a competitive advantage to some large airlines, might result in higher fares and a reduction in service options for consumers.

IMPACTS AND TRADEOFFS ASSOCIATED WITH THE NUMEROUS ALTERNATIVE

[17] Determining how best to finance FAA is a complex problem that requires careful study and good cost data. FAA's costs vary depending on the amount, type, and timing of various airline operations. /10/ For example, hubbing operations at congested airports increase the peak service demands on the system and increase FAA's costs. However, this cost has not yet been quantified and neither the 10-percent ticket tax nor the large airlines' proposal accounts for these costs. A financing system that doesn't take such factors into consideration could result in costs not being fairly allocated among system users. As a result, any potential financing mechanism for FAA should be assessed from the standpoint of the data currently being developed on FAA's actual costs.

[18] If the Congress ultimately decides to replace the ticket tax with a different fee system, numerous financing options are available for it to consider. Possible options include taxing one or more of the common indicators of system use, such as departures, passenger enplanements, seats flown, fuel consumed, or a combination of these indicators. However, the potential competitive impact of using these indicators as a basis for allocating FAA's costs varies greatly depending on which indicator is used. For example, if a tax on passenger enplanements were adopted and designed to generate about the same amount of revenue as the ticket tax, the amount paid by the coalition airlines would decline by about $251 million while the amount paid by the competing airlines would increase by $269 million and commuter carriers by $61 million. /11/ (See app. II.) In contrast, a fuel tax would keep the amount paid by each airline group about the same as each paid under the ticket tax. (See app. III.)

[19] The impact of the financing options also varies among airlines within the coalition and competing airline groupings. For example, under a system that taxed both fuel use and passenger enplanements, the amount paid by four coalition airlines would decrease but would increase for the other three coalition members. Similarly, under a financing system that taxed departures and aircraft miles, the amount paid by Southwest Airlines would increase by about $135 million but would decrease by about $7 million for the other airlines in the competing airlines grouping. In general, such variances result from differences between airlines in operating factors, such as type of operation, average age of their aircraft fleet, and average distance of their flights.

[20] The various financing options for FAA also present tradeoffs between their ease of administration, impact on how efficiently the airport and airway system is used, and ability to produce an equitable system in which users pay their fair share. For example, a formula that combines several of the common system usage indicators might provide the most exact method to ensure that all users pay their fair share of system costs. However, such a formula may also be so complex that it would be difficult to administer. Similarly, taxing airlines for their use of the most congested airports may result in a more efficient use of the nation's airspace. However, because the coalition airlines are the primary users of these airports, this approach may not produce the most equitable result from their point of view.

[21] Such tradeoffs and the potential competitive impacts of a new fee system will need to be carefully studied over the next year by the national commission and the Secretary of Transportation. The financing alternative that is finally selected should be relatively easy to administer and help ensure that, in the long term, FAA has a secure funding source, the nation's airports and airways are used as efficiently as possible, commercial users of the system pay their fair share, and a strong, competitive airline industry continues to exist. Ultimately, it will be a policy call for the Congress to decide on how to achieve these goals.

* * *

[22] Mr. Chairman, this concludes our prepared statement. We would be glad to respond to any questions that you or any member of the Committee may have.

APPENDIX I

[graph omitted]

APPENDIX II

[graph omitted]

APPENDIX III

[graph omitted]

FOOTNOTES

/1/ Airport and Airway Trust Fund: Issues Raised by Proposal to Replace the Airline Ticket Tax (GAO/RCED-97-23, Dec. 9, 1996).

/2/ The coalition comprises the seven largest airlines -- American Airlines, Continental Airlines, Delta Air Lines, Northwest Airlines, Trans World Airlines, United Airlines, and USAir.

/3/ The Federal Aviation Reauthorization Act of 1996 (P.L. 104- 264). On November 18, 1996, FAA contracted with Coopers & Lybrand to conduct the independent cost assessment. As of late January 1997, the national commission had not yet been formed.

/4/ The extent to which airlines were able to shift some or all of the costs associated with the ticket tax to consumers depended on consumers' sensitivity to changes in airfares. Prior studies have shown that consumers' sensitivity to fare changes varies and that in some cases small fluctuations in fares can have a large impact on an airline's ridership. Thus, redistributing taxes among airlines could have substantial competitive impacts depending on the subsequent effects on fares and ridership.

/5/ Department of Transportation's Appropriations Act for Fiscal Year 1997 (P.L. 104-205).

/6/ Certification of New Airlines: Department of Transportation Has Taken Action to Improve Its Certification Process (GAO/RCED-96-8, Jan. 11, 1996), and Management Reform: Implementation of the National Performance Review's Recommendations (GAO/OCG-95-1, Dec. 5, 1994).

/7/ Under the Federal Aviation Reauthorization Act of 1996, after receiving the national commission's report, the Secretary of Transportation is required to consult with the Secretary of Treasury and report to the Congress by October 1997 on the administration's recommendations on how best to finance FAA.

/8/ Air Traffic Control User Fees: A Proposal by the Coalition for Fair FAA Funding, revised June 7, 1996. The proposal defines originating passenger based on the beginning point of the trip, irrespective of the number of take offs and landings made during the journey.

/9/ To minimize flight delays, FAA limits the number of operations (takeoffs and landings) that can occur during certain periods of the day at four key congested airports -- Chicago O'Hare, Washington National, New York Kennedy, and LaGuardia. The authority to conduct a single operation during these periods is commonly referred to as a "slot."

/10/ The issue of how various users of air traffic and other FAA services impose costs on the system is complex. Past studies of FAA's costs have found that the nature of how air traffic and associated services are produced entails many costs that are "common" -- that is they cannot be allocated to any one type of user. As a result, a full allocation of system costs may require a mechanism for assigning these common costs.

/11/ A tax of $10 per enplanement would generate about $79 million more than was generated under the ticket tax in 1995.

END OF FOOTNOTES