Contract Design, Financing Arrangements and Public Ownership—An Assessment of the US Airport Governance Model

Abstract

US airports negotiate legally binding contracts with airlines and finance large investment projects with revenue bonds. Applying insights from transaction cost economics, we argue that the observed variation in contractual and financing arrangements at US airports corresponds to the parties' needs for safeguarding and coordination. The case evidence presented reveals that public owners set the framework for private investments and contracting. We suggest that airline contracts and capital market control result in comparative efficient investments and act as a check on the cost inefficiency typically linked to public ownership.

Acknowledgements

Johannes Fuhr gratefully acknowledges financial support from the Erich Becker Trust, Fraport AG. The paper has benefited from comments by David Starkie and David Gillen. Any remaining errors are our own.

Notes

1. The theory's predictions have been corroborated in a large number of industry studies. See Macher and Richman (2006) for a recent and comprehensive survey on the empirical TCE literature.

2. TCE assumes economic actors to be bounded in their rationality, i.e., “intendely rational, but only limited to do so” and to be opportunistic, i.e., “self‐interest seeking with guile” (Williamson, 1985, pp. 45–47).

3. The early TCE literature identified three generic governance forms: market, hybrid and hierarchical governance. Market coordination of a transaction results in high‐powered incentives, leads to autonomous adaptation via the price mechanism and relies on classical contract law (enforcement through courts). Transactions within the firm (hierarchy) rely on administrative controls (low‐powered incentives), coordinated adaptation and law of forbearance (enforcement via management fiat, as courts ‘forebear’ to hear internal conflict within organizations). Hybrid forms, such as long‐term contracts or joint ventures, are located on the continuum between market and hierarchy and share characteristics of both polar forms. Private ordering, e.g., arbitration, supplements court enforcement in these arrangements (Williamson, 1985, 1991). More recent research has explored subclasses of hybrid modes, for example, supply contracts or strategic alliances (Ménard, 2004; Eckhard and Mellewigt, 2006; for recent surveys).

4. Asset specificity takes the form of site specificity, physical asset specificity, dedicated asset specificity, human capital asset specificity, brand capital specificity (Williamson, 1985), temporal specificity (Masten et al., 1991) and contractual specificity (Pirrong, 1993).

5. Airports must in particular seek local political and public support to overcome the NIMBY (not‐in‐my‐backyard problem). The NIMBY problem occurs when a development is locally undesirable (increase in pollution), but socially beneficial. In a world of positive transaction costs, mechanisms for efficient bargaining might be precluded, requiring specialized institutional arrangements (Richman and Boerner, 2006).

6. A slot is a time window in which the airline is entitled to use the runway of a congested airport.

7. Unlike contestability theory—the primary theoretical foundation of US airline deregulation—TCE magnifies conditions of asset specificity. Williamson (1985, p. 31) argues that existence of firm‐specific assets is widespread, turning hit‐and‐run entry in most cases infeasible.

8. The hub‐and‐spoke network structure is considered an outcome of airline deregulation. Hub‐and‐spoke network structures allow carriers to exploit economies of density (Caves et al., 1984; Brueckner and Spiller, 1994; Berry et al., 2006) and to offer a differentiated product in terms of high connectivity to business travellers (Berry et al., 2006).

9. In their formal model on procurement contracts, Bajari and Tadelis (2001) show that cost‐plus contracts with their superior adaptation capabilities economize on the procurement of complex construction projects.

10. According to the FAA/OST Task Force Study (1999, p. 3), 54.2% of commercial airports in the USA are directly owned and operated by cities or counties, followed by regional ownership (22.7%), state ownership (9.3%), multi‐jurisdictional authorities (6.2%), specialized (air)port authorities (4.1%) and other ownership forms (private, etc.) with 3.1%. So far, only Stewart Airport has been leased under a 99‐year lease contract to a private operator under the FAA pilot privatization programme.

11. The structure of AIP funds distribution reflects the national priorities and objectives of assuring airport safety and security, stimulating capacity, reducing congestion, helping fund noise and environmental mitigation costs, and financing small state and community airports. Small airports obtain approximately 60% of their funding from federal grants, while medium and large airports obtain less than 10% (GAO, 2003).

12. The total expenditure from the Aviation Trust Fund in the fiscal year 2005 amounted to $11 156 million, with $3531 (32%) dedicated to the AIP (GAO, 2005, pp. 2–3).

13. The FAA/OST Task Forces study (FAA, 1999, pp. 7–9) shows that in particular residual and hybrid agreements include MII clauses: 84% of all residual agreements, 74% of all hybrid agreements and 20% of all compensatory agreements include these clauses in the presented survey.

14. Preferential lease arrangements are heterogeneous as some include ‘use‐it‐or‐lose‐it’ rules or ‘use‐it‐or‐share‐it’ rules. While most preferential agreements come close to exclusive agreements, some preferential agreements are short term in nature and thus allow for a periodic reallocation (FAA, 1999, pp. 35–42).

15. Some smaller US airports have issued general obligation bonds, which are backed up by ‘the full faith and credit’ of the issuing local municipality.

16. The ‘conduit’ may be a municipality, a city, or a specific public agency so as to qualify for tax exemption. As interest on municipal bonds is tax‐exempt under federal tax law, these airport bonds have more favourable interest rates than similar securities.

17. In our selection of case study airports, we decided to focus on large airports as these offer a larger variety of contracting and financing arrangements. Our results might thus not be extended to smaller airports.

18. Concessions and parking revenues amounted to $132.0 of a total $341.0 million in airport revenues in 2004.

19. Jetblue, American and Northwest Airlines have signed long‐term or revolving contracts with the airport.

20. In its bond prospectus, the airport points towards its experiences in the liquidation of Eastern Airlines in the late 1980s. All routes served by Eastern Airlines were subsequently taken over by other legacy carriers.

21. Expanding low‐cost carriers in the BOS market (AirTran, Jetblue and Southwest) considered the premium facility (designed before September 11) and its high rental rates to be in violation of their low‐fare business model. Even alliance members of Delta decided against relocation in the face of the rental rates and the costs of relocating their operations.

22. Under the Logan preferential use policy, the airport may schedule arrivals and departures at the gate of the tenant for any period that the tenant is not using the gate.

23. American Airlines, for example, must keep its gate utilization above 75% of Logan's average number of domestic aircraft movements per gate. If American's gate utilization should drop below the threshold, the carrier can evade recapture of its gates by bringing back its gate utilization above the threshold in the following 12 months (cure period). Even if Massport is entitled to relet American Airlines gates to another carrier, it may only do so for a maximum of 12 months, during which American may repossess the gate if it achieves a certain gate utilization level. Other long‐term leases follow a similar structure, but deviate in the specified dimension above (cure periods, threshold values, etc.). According to the lease signed with Delta Airlines, for example, Massport may not recapture any gates at all in the first five years of the new Terminal's operations. The authority may, however, force Delta to sublet up to four gates to other airline tenants or new entrants.

24. The Port Authority and airlines have recently completed the negotiation of a new compensatory master use‐and‐lease agreement with 20‐year duration (2004–23).

25. The sole exception used to be the International Airline Building, which was operated by the Authority up to 1997, before being replaced with a private third‐party development.

26. The shareholders are Schiphol USA LLC (40% equity interest), LCOR JFK Airport LLC (40%) and Lehman JFK LLC (20%). Schiphol USA LLC is an indirect subsidiary of N.V. Luchthaven Schiphol, a government‐owned company running Amsterdam Airport. LCOR JFK Airport LLC is an indirect subsidiary of a large real estate developer in the USA. Lehman JFK LCC is an indirect subsidiary of Lehman Brothers, a large US investment bank.

27. Terminal 1 and the international airline terminal (Terminal 4) are run by lean operating companies with primary responsibility to procure services, monitor quality, maintain financial control and retain/acquire sublessees. American Airlines, on the other hand, has opted to outsource a minor portion of its terminal operations (e.g. a master concessionaire agreement for its non‐aviation business).

28. Several terminals were under temporary management of the authority because primary tenants had exited the market (e.g., Easter Airlines, Pan Am and TWA). In most of the current lease arrangements with primary tenants, the Port Authority has the option/right to relet the facility if the primary tenant defaults.

29. United Airlines' market share decreased from 23.3% in 1990 to 14.6% in 2005. Delta Airlines' market share dropped from 17.2% in 2001 to 8.9% in 2005 as the carrier ceased its PDX‐based hub operation.

30. The PFC‐approved project volume ($681.8 million) included the (1) terminal expansion south, (2) terminal enplaning road, and (3) the light rail extension to the airport and the light rail station at the airport.

31. The sole exception is the outdated Concourses A, whose rental rate is 20% below the equalized general rate. Non‐signatory carriers, accounting for less than 0.1% of all enplaned passengers at PDX, pay a 25% premium on terminal rental rates and landing fees.

32. The airport is able to lower the revenue share from non‐airline cost centres if it lowers its operating and maintenance expenses in the airline cost centres that are below its budget.

33. The current master use‐and‐lease agreement was preceded by a five‐year agreement with differentiated terminal rates (2001–05), a ten‐year agreement without revenue share formula (1991–2001) and a 20‐year residual agreement (1971–91).

34. At the time of writing, there has been no indication, that any airline or consortium of airlines has approached the Authority to arrange for a specialized arrangement comparable to the one with Northwest Airlines.

35. The estimate is taken from the airport consultants report in the bond prospectus of the 2005 GARB series.

36. The Director of Design and Construction at Northwest Airlines has commented on the Northwest‐headed collaborative arrangement as follows: “It was really a collaborative effort where SmithGroup [the architect], Northwest, and Wayne County worked hand‐in‐hand, developing the conceptual phases through schematic and design developments, and finally construction drawings. … As a result of our customer knowledge, and through simulations based on the projected flight schedules in the future, we were able to look at how everything would affect customer short‐ and long‐term. Customer waiting is minimal; everything flows smoothly; and connections, both domestic and internationally, work extremely well” (Monroe, 2002, pp. 40–41).