Part Two: Reforming U.$. Airports

July 8, 2001

Reformaing U.S. Airports: Part II

If efficient, competitive airports is the goal, change is the answer

By Dr. Frederick R. Warren-Boulton, Economist

July 2001

In our June issue, economist Rick Warren-Boulton offered insights into why the U.S. transportation system and consumers would benefit from the economic deregulation of airports. Here, he looks at the forces in play that could affect airports’ reactions regarding rates and charges and why long-term anticompetitive behavior would likely not occur. He also looks at the impact slots and government operation of ATC have on competition and technology investment.

One can readily think of four criteria that should be sufficient to ensure that regulation of landing fees is unnecessary or undesirable.
First, at unconstrained "profit-maximizing" levels, would revenue to the airport still be insufficient to cover costs, so that some subsidy from the public is necessary for its existence? If so, regulation can only distort the efficient financing of that deficit.
Second, and critically, passengers can only benefit from allowing higher landing fees at slot-constrained or slot-controlled airports. Allowing airlines to control slot allocation instead of using landing fees to allocate excess demand benefits only airlines, and costs passengers dearly. As long as slot constraints are binding, higher landing fees only transfer revenue from the airlines to the airport, with no short-run effect on ticket prices. In the longer run, shifting to higher landing fees results in both the ability and incentive of all parties (now including the airlines) to increase capacity, leading to reduced ticket prices without congestion.
Third, does the airport face sufficient effective competition from other airports to make regulation inefficient? Note that an airport does not need to have a "perfect" substitute available with excess capacity (indeed, given fixed costs, the availability of a perfect substitute would probably indicate that one or the other should be shut down). Rather, the question is whether there is sufficient substitutability between airports to constrain an airport from pricing significantly above its short-run marginal cost, or from not expanding capacity when it would be desirable.
This can only be answered on an individual airport basis, but it is not a difficult or unprecedented task. One precedent is the Department of Justice’s proposal for deregulation of the vast majority of oil pipelines. (An industry viewed as the classic example of a "natural monopoly" needing regulation, and which thus, on its face, would appear to be among the least viable candidates for deregulation. However, a closer examination of individual pipeline markets revealed far stronger competitive constraints than anyone had anticipated.)
A similar approach would require looking at each airport on a case by case basis, using a common and accepted criteria and methodology (e.g., the DOJ/FTC Merger Guide-lines). The goal: Determine whether an airport has monopoly power in a relevant market that could be expected to be exercised to the harm of consumers if regulation were relaxed or eliminated.
It should be noted that airports face two competitive constraints on the demand side:
• the ability of passengers to switch; and
• the ability of airlines to switch flights.
While related, each has a separate constraint on the pricing of airport services. In many cases, the combined effect may be enough to allay concerns that even a profit-maximizing airport would raise prices above efficient levels.
Fourth, assume an analysis concludes that certain airports have monopoly power. Public ownership of airports should be sufficient to ensure non-monopoly behavior as long as the burden of any supra-competitive pricing falls on the citizens of the political unit that owns or controls the airport.
A high proportion of traffic by residents should suffice to allay concerns, especially if the airport cannot discriminate between residents and non-residents. And, even if "outsiders" account for a high share of traffic, other airport stakeholders may bear the burden of supra-competitive pricing.
For example, even though most of the passengers traveling in and out of Las Vegas may not be residents and voters in Las Vegas County, as long as Las Vegas competes with other destinations for tourist traffic, any increase in ticket prices that resulted from higher landing fees would ultimately be born by hotels or other providers of the Las Vegas package that competes with similar packages available at other destinations. In antitrust terms, that means that the city itself needs to have monopoly power.
The combined set of airports that are either revenue-inadequate, slot-constrained, in competitive markets, or are publicly owned by competitive cities, should be quite large. Indeed, it is hard to think of an airport that does not satisfy at least one of those criteria. At the very least, a sizable share of airport capacity could be deregulated without concern, and the experience from those changes used to induce further reform.

Partial deregulation, on the other hand, has generally been less successful. The best example was eliminating many of the direct controls over the Savings & Loan industry while preserving government insurance of deposits — a lethal recipe for what economists call "moral hazard", and most people think of as massive fraud and theft. Most recently, deregulation of electricity generation has resulted, at least temporarily, in surprisingly high prices in California during peak periods.
Both are examples of what economists call the "theory of the second best" — that a policy change which moves one aspect to its ideal level may not improve welfare if all related variables are not also at their ideal levels.
While airlines have been deregulated since 1978, airport controls by the FAA/DOT remained in place. These include:
• rates and charges;
• policies on diversion of revenue and privatization; and
• limited slot controls.
In addition, air traffic control is provided by FAA and does not provide efficient pricing to encourage air traffic control investments.

Economists would argue that there are potentially two considerations relevant in determining socially optimal prices: efficiency and equity. The set of efficient prices would be the set that maximizes the size of the pie (in economists’ terms, the set of prices that maximizes social welfare). Once anyone knows what their economic role is, of course, they are likely to prefer a set of prices that makes them better off than the efficient set. A better test of equity or justice would be what criteria we would choose before we knew what our role was (i.e. ex ante rather than ex post). Asked under those conditions, many of us might prefer a set of prices and incomes other than the set that maximizes total income.
For example, many of us might have voted for a tax system that redistributed some income from rich to poor before we knew whether we would be rich or poor.
But any income redistribution that results from "inefficient" pricing by airports (e.g., for landing fees at slot-constrained airports, from city taxpayers to airline shareholders) is not, I would assert, one that anyone would vote for ex ante, and thus cannot be defended as "just" or "equitable". Rather, those who favor deviations from efficient pricing do so because those prices are in their realized interest — where they stand depends on where they sit. One can conclude that the efficient set of prices is also the most just and equitable.
Thus, while someone with a private plane might greatly favor subsidies to general aviation, that same person would probably have voted against such subsidies if asked before he knew he was to become the owner of a private plane.
Persuading a large set of individuals to accept change when each of them has veto power, however, requires what economists call a "Pareto" standard: the change must make at least some better off without making anyone worse off.
In principle, of course, every move to a more efficient set of prices could be a Pareto improvement, since the winners could compensate the losers and still be better off. But how to compensate the losers?

For congestion tolls, often the key is to credibly assure those paying the fees that the revenue will benefit them directly. It can easily be shown that a congestion toll will reduce the welfare of users if they do not benefit from the toll revenue. (As can be shown, the cost of the toll is always greater than the benefit from the reduction in congestion). Users can thus be expected to oppose a congestion toll unless they believe that the resulting revenue will be used to increase capacity, which in turn will reduce congestion and subsequently the toll.
By "slot controlled" I mean that someone has the right to determine usage — i.e., to allocate scarce landing rights when there is excess demand. This includes the formally slot-constrained airports (O’Hare, LaGuardia, Reagan National, and JFK), and any airports where one or more airlines can effectively allocate landing rights directly, rather than through a first-come-first-served process where the value of those rights is dissipated through congestion.

A critical component of any such plan is the elimination of airline property rights in slots. Slot constraints, with slots granted to airlines, are very much a second-best solution to congestion problems. While better than allowing the value of landing rights to be fully dissipated in congestion, granting slots to airlines deprives airports of the funds to expand capacity. Slots also give the receiving airlines a strong incentive to resist capacity expansions, which would inevitably reduce the value of their slots. Given the influence of airlines over airport decisions, it’s hard to imagine a more effective scheme to prevent critical capacity expansions at airports.
As noted, the air traffic control system is provided by FAA and does not provide efficient pricing to encourage air traffic control investments. Efficient pricing of scarce air traffic control resources and effective increases in air traffic control facilities — both of which might be facilitated through privatization — could greatly increase the effectiveness and productivity of airport investments in additional capacity. As with airport capacity, efficient provision of air traffic control services requires both demand and supply management.
* * *
Since capacity expansion tends to come in lumps, efficient pricing will result in an accounting deficit in the earlier years, and an accounting surplus in later years, as congestion rises before the next capacity increase. Thus, the relevant questions for financing are ...
• whether the present value of the marginal-cost-based pricing revenues is at least equal to the present value of the costs of capacity expansion; and
• whether the flow of revenues is sufficiently secure (in terms of both economic projections and legal rights of the airport) to allow outside financing of the expansion.
For any such rule, total costs should be evaluated at their economic costs (i.e., incremental, forward-looking, opportunity costs) rather than on the basis of historic or accounting numbers.

About the Author
Dr. Frederick R. Warren-Boulton is a principal of MiCRA (Micro-economic Consulting and Research Associates, Inc.), a Washington-based economics consulting and research firm specializing in antitrust litigation and regulatory matters. He holds a B.A. degree from Yale University, a Master of Public Affairs from the Woodrow Wilson School of Princeton University, and M.A. and Ph.D. degrees in Economics from Princeton University. He has served as an expert witness on consultant mergers and other anti-trust matters, starting in 1981 as an expert witness for the DOJ in U.S. v. AT&T, and recently, for the States and the DOJ in United States of America v. Microsoft .